^tnvg„ THE UNIVERSITY of EDINBURGH Title Holocene environmental change: a palaeolimnological study in Belize Author Breen, A.M. Qualification PhD Year 2002 Thesis scanned from best copy available: may contain faint or blurred text, and/or cropped or missing pages. Scanned as part of the PhD Thesis Digitisation project http://librarvblogs.is.ed.ac.uk/phddigitisation Holocene Environmental Change: A Palaeolimnological Study in Belize 7186 l4C years BP) and over the next 1000 years the climate became increasingly wet. • This trend was reversed between 5050 and 4050 BC (6121-5281 l4C years BP). Although the lake waters were saline during this time period the climatic conditions were still considerably wetter than in the preceding millennium. • The wettest period of the Holocene was at 3850 BC (5085 l4C years BP) in this area. • These conditions were rapidly reversed and drier conditions are detected around 3450 BC (4691 l4C years BP). • From 3450 BC to 1800 BC (4691-3519 14C years BP) the trend was to wetter conditions. • From this point onwards drier conditions have prevailed to the present day. • There is no evidence of human disturbance on this lake system, with the diatom record indicating oligotrophic to mesotrophic conditions throughout. • This record lies on a climatic gradient with Lake Sayaucil and Lake Coba. San Jose Chulchaca receives the lowest annual rainfall and does not appear to have any archaeological evidence of sustained human settlement. This is the direct opposite to Lake Coba. The comparison of records along such gradients enables a great deal of useful information to be gained concerning the dynamics of an area. The dates were converted to 14C years BP from Stuiver and Pearson (1993) by the author of this thesis. 20 4. Location: Lake Punta Laguna, Quintana Roo Author: Curtis et al. (1996) Proxies: Oxygen isotope analysis Altitude: 14 metres a.s.l • The patterns of change that are found in this 3500 year record, are decadal scale cycles of alternating wet and dry conditions which are superimposed on millennial scale shifts in mean climate conditions. • The period from 3310 to 1785 l4C years BP was wet. • Dry conditions persisted between 1785-930 l4C years BP. During this time, exceptionally arid events are recorded at 1510, 1171, 1019, 943, 559 i4C BP. • These coincide with two periods of key interest in Mayan history: the hiatus and the collapse. These are both periods of cultural discontinuity which had profound effects on the way in which Mayan society operated. • By 1100 AD (942 l4C years BP) more humid conditions returned. The next key dry period occurred between 1368 to 1429 +/- 50 cal years AD (651-506 l4C years BP) which coincides with Mayan reports of famine, cold and drought and the simultaneous abandonment of cities in the Yucatan Peninsula such as Uxmal, Chichen-Itza and Coba. The dates in the last time period discussed were converted to radiocarbon years from Stuiver and Pearson (1993) by the author of this thesis. 5. Location: Lake Sayaucil, central Yucatan Peninsula Author: Whitmore et al. (1996) Proxies: Diatoms and oxygen isotopes Altitude: 25 metres a.s.l • Following the initial filling of the lake, water levels were low between 3050 to 2000 l4C BP. • The lake then rapidly became fresher as water levels rose. It was during this time frame that human disturbance is apparent in the watershed. 21 • Lake levels and human disturbance have remained fairly constant since this time. There are a number of different factors which have influenced the amount of environmental change that is apparent in the sequences from the Yucatan Peninsula. In general terms the records follow the trends expected with a dry Lateglacial, moist mid Holocene and dry late Holocene. Complications arise in the interpretation of the sequences due to the influence of sea level and human impact on the systems. 2.1.2 Guatemala 1. Location: Lake Quexil, Peten Lake District. Authors: Deevey (1978), Deevey et al. (1983), Leyden (1984), Leyden et al. (1993), Leyden et a/.(1994). Proxies: Pollen and oxygen isotope Altitude: 110 metres a.s.l • This record provides valuable evidence about the climate changes that took place in this region before the transition to the Holocene, which is extremely useful in placing the records from later time periods in context. • From 36-24,000 l4C years BP the lake level was lower than today. The vegetation was much more temperate which suggests that the climate was cool and moist, with temperatures 4.7 to 6.5°C less than today. • After 24,000 14C years BP an extremely dry period persisted with temperatures being 6.5 to 8°C cooler than today. • With the onset of deglaciation around 12,500 l4C years BP the dry climate conditions began to ameliorate. • The lake filled rapidly after 10,500 l4C years BP but the cool temperatures lingered averaging 3-4.7°C less than present. • Lowland tropical forest did not properly form in the watershed until after the onset of warmer, moister conditions around 9000 l4C years BP. • Drier conditions are thought to have developed after 5000 14C years BP when the forest became less dense. This is however more likely to be due to human influence on the system and the record after this point is clearly a cultural sequence, which limits the climatic inferences that can be made. Location: Lake Peten-Itza, Peten Lake District Authors: Islebe et al. (1996a), Curtis etal. (1998). Proxies: Pollen and oxygen isotopes Altitude: 7 metres a.s.l This multi-proxy record suggests that in the earliest Holocene, prior to 9000 l4C years BP, the climate was relatively dry. Water levels rose by 9000 i4C years BP which is consistent with Lake Quexil which is just south of Lake Peten-Itza. Evidence from pollen indicates that widespread tropical forest had been established by 8560 l4C years BP which is in line with the expected moist early Holocene conditions. The oxygen isotopes suggest, however, that the climate was relatively dry from 9000-6800 14C years BP. Records from other lowland tropical lakes reveal similar trends to both interpretations from Peten-Itza suggesting that neither reconstruction is anomalous (Leyden et al., 1993; Vaughan et al., 1985). It is likely that this juxtaposition was a result of the hydrologic budget of Peten-Itza exerting a greater control on lake water 5I80 than the regional evaporation/precipitation signal, hence the positive oxygen isotope signal during this time period. The oxygen isotope signal shows a steady and irreversible decrease beginning around 6800 l4C years BP, suggesting a change from a wetland environment to a deeper lake. This occurred during the early to middle Holocene moist period that has been observed elsewhere. The concurrence of the 8lsO signal with the regional record at this point suggests that there has been a significant change in the controls over the lakes isotopic signature. The climate signal is confused from this point in the record due to the increasing influence of humans. 23 • Increased abundance of high forest taxa after 1000 l4C years BP marks the recovery of the vegetation. This coincides with the time of Mayan collapse (see Chapter 3). • This record is comparable to the records that have been produced for Lakes Quexil, Salpeten and Sacnab in Guatemala. 4. Location: Lake Salpeten, Peten Lake District Authors: Brenner (1994), Rosenmeier et al. (in press) Proxies Pollen and stable isotopes Altitude: 104 metres a.s.l • Maya occupation of the watershed is marked by a decline in high forest and increased erosion beginning around 1400 BC (3118 l4C years BP). • From 1300-700 BC (3045-2469 i4C years BP) there is a signal which may relate to a wet climate in this time frame. • In AD 130 (1879 l4C years BP) and AD 520 (1582 14C years BP), the times of the Preclassic abandonment and hiatus, there is a signal for decreased precipitation and forest recovery. These periods in Mayan history will be explained in more detail in Chapter 3. • There is reduced soil erosion and a drying signal after AD 850 (1216 l4C years BP) which coincides with the collapse of the Mayan civilisation. • Forest recovery after AD 1400 (581 i4C years BP) coincident with the time of final watershed abandonment. • This record provides a clear indication of human activity during this time frame which may provide a general indication for the patterns in the surrounding regions. The dates were converted to l4C years BP from Stuiver and Pearson (1993) by the author of this thesis. 24 5. Location: Laguna Tamarindito, southwest Peten Authors: Dunning et al.{ 1998) Proxies: Pollen and gastropods Altitude: 120 metres a.s.l • At the onset of the Holocene period the lake appears to be a deep water system due to the presence of deep-water gastropods. • By 7500 l4C years BP the region was colonised by tropical deciduous forest species indicating that the system was responding to the moist mid Holocene conditions. • The lake experienced two periods of lower water levels between 6500 and 4900 l4C years BP suggesting significant dry episodes. • There is only equivocal evidence for significant drying of the local climate from 300 BC (2222 14C years BP). The date in the last time period discussed was converted to radiocarbon years from Stuiver and Pearson (1993) by the author of this thesis. The key difference between the Peten and the Mexican Yucatan Peninsula is the apparent amount of human impact that is recorded in the sequences. This may, however, be due to the reliance on pollen evidence in Guatemala. An understanding of how the records from the different proxies relate to one another is therefore essential to the correct interpretation of the sequences. The shift to lacustrine sedimentation in Guatemala occurred at the same time as the Yucatan Peninsula suggesting that the mechanism behind this change was regional. There is also evidence for change around 1000 years BP in these sequences. 25 2.1.3 Costa Rica 1. Location: La Chonta bog (Cordillera de Talamanca, south Costa Rica) Authors: Hooghiemstra et al. (1992); Horn and Sanford (1992); Horn (1993); Islebe et al. (1995); Islebe et al. (1996b) and Islebe and Hooghiemstra (1997). Proxies: Pollen and charcoal Altitude: 2430 metres a.s.l • This record covers the Pleistocene, which was essentially a dry climatic interval in this area. • The first pollen zone in this sequence represents the period from 80,000-50,000 years BP and provides evidence for four warm periods separated by three cold intervals. These episodes are thought to have been of regional importance. • The second zone from 50,000-13,000 years BP is a stable period with a 7-8°C cooling at the time of the Last Glacial Maximum. • Between 14,000 and 11,000 14C years BP moist montane oak forest was well developed in this region. • During the Younger Dryas chronozone (11,080-10,500 l4C years BP) a cool period is recognised in Costa Rica and has been given the name La Chonta stadial. A temperature drop of 1.5-2°C has been inferred in this time period and it coincides with a southerly shift of the northernmost position of the ITCZ. The change to warmer Holocene conditions was not a rapid one occurring between 10,400 and 9800 14C years BP. • From 7000 to 4500 14C years BP, during the mid Holocene, the climate was increasingly humid. The period between 6000 and 5000 l4C years BP was extremely stable with dense vegetation covering the area. • Pollen of species which represent forest disturbance by humans, appear in the record from 4900 l4C years BP. • Dry conditions prevail from 4500-1500 l4C years BP with two distinct dry periods during the Late Holocene: 2430 l4C years BP and 1110-1180 l4C years BP. 26 This record provides very clear evidence of a Younger Dryas interval in this region. This is important because it suggests that there are periods of climatic history where changes occur at the same time over a very wide area, but are manifested in different ways. In a similar manner to Guatemala and the Yucatan peninsula, the mid Holocene period was wet; dry conditions appear to occur earlier from 4500 years BP. This may however be an artefact of human impact on the system through vegetation clearance. 2.1.4 Honduras 1. Location: Lake Yojoa and Petapilla Swamp, western Honduras Authors: Rue (1987) Proxies: Pollen Altitude: 635 metres a.s.l • This lake is located in the southeastern periphery of the Southern Lowlands (Figure 2.2) and therefore provides an interesting comparison with the human activity records from the Peten, Guatemala which is at the heart of this Mayan region. • This sequence does not find any evidence for climate change during the late Holocene. The trends that are apparent appear to be due to human induced modification of the landscape. • Clearance of natural vegetation appears to have occurred from 4500 14C years BP and from this point to 3000 l4C years BP agricultural intensification is evident. After this phase until modern times there are few identifiable trends and the region appears to be ecologically stable. 27 2.1.5 Panama 1. Location: Laguna Volcan, Cordillera de Talamanaca, Panama Authors: Behling (2000) Proxies: Pollen and charcoal Altitude: 1500 metres a.sd • Evidence at 2860 l4C years BP suggests that the landscape at this time was open within lower montane rainforest. • The evidence of man is apparent through forest clearance throughout this record. • Zea mays is found relatively late compared to other sites in Panama at 1790 14C years BP. • Climatic changes such as the late Holocene dry period are difficult to identify due to widespread human impact. 2. Location: El Valle, lowland Panama (central Panama near the canal region) Authors: Bush and Colinvaux (1990) Proxies: Pollen. Altitude: 600 metres a.s.l • The climate before 127,000 years BP was considerably colder than at present. • 127,000-95,000 years BP was a period of near to present day temperatures in a warm and wet environment. • The period from 95,000-50,000 years BP indicates cool, windy and dry conditions. • From 30,000-10,000 years BP the climate was deteriorating and temperatures were 4°C lower than present. This may have been interrupted by a warm period. From 14,000 to 10,000 l4C years BP a cooling of 5-6°C has been inferred. • The warm and wet conditions of the present day were established in this record from 10,000-9000 14C years BP. 28 3. Location: Lake La Yeguada, (central Panama) Authors: Piperno et al. (1990), Piperno et al. (1991) and Bush et al.( 1992). Proxies: Pollen, phytoliths and charcoal. Altitude: 650 metres a.s.l • From 14,300-10,800 l4C years BP the lake appears to be under a strongly seasonal regime which was manifested in fluctuating lake levels. • The period between 13,000 and 10,000 i4C years BP is recorded as being dry but increased precipitation and temperature conditions occurred from 10,800-10,500 14C years BP. The onset of these changes was abrupt and a new climate regime was fully established during this time frame. • The early Holocene until 8600 14C years BP represents cooler and wetter conditions than at present. The rapid vegetation succession associated with Northern Hemisphere deglaciation ends by this point. • There is evidence of a dry phase between 8200 and 5500 l4C years BP with the period from 7000 to 5000 l4C years BP representing a dry Holocene maximum. • Human disturbance is also prevalent throughout the Holocene period. 4. Location: Gatun Basin, Panama Canal zone (central Panama) Authors: Bartlett and Barghoorn (1973). Proxies: Pollen Altitude: 30 metres a.s.l • This is a record of sea level change, climate variability and human induced modification of the landscape. • Temperatures from 11,300-9600 14C years BP appear have been at least 2.5°C cooler than today. • From 9600-7300 14C years BP a well-developed mangrove swamp was deposited, which suggests that this was a time of marine influence on the system. • From 7300-4200 14C years BP a transition from mangrove to freshwater swamp occurred due to a decrease in the rate of sea level rise. Seasonality along the Atlantic coast is thought to have been high at this time. 29 • All marine influence on the system ceased by 4200 l4C years BP. This coincided with the development of agriculture resulting in the record from this point becoming an unreliable indicator of purely natural change. 5. Location: Lake Wodehouse, Darien (south Panama near the border with Colombia) Authors: Bush and Colinvaux (1994) Proxies: Pollen and diatoms Altitude: 500 metres a.s.l • This record produces evidence for distinct climatic events in the last 4000 years. • The key dry periods were from 3800-3700 l4C years BP, 3400-2500 l4C years BP and 1900 l4C years BP to the present. These were not severe enough to cause the lake to fully dry out and are thought to be regional in their significance. The records from this area highlight the influence of Pacific rather than Atlantic driven climate systems. The key manifestation of this is the dry conditions that prevailed in this area during the mid Holocene and the Younger Dryas signal which is characterised in this region by an increase in moisture and temperature. This is very different to the signal in Guatemala and the Caribbean. The impact of sea level is felt in this region which hinders the climatic interpretations that can be made. There is evidence for climatic drying in the late Holocene but these periods do not appear to be as severe as those found further north. 30 2.1.6 Haiti 1. Location: Lake Miragoane, (south Haiti) Authors: Hodell et al. (1991), Brenner et al. (1994), Higuera-Gundy et al. (1999) Proxies: Oxygen isotope and pollen data. Altitude: 20 metres a.s.l • The oxygen isotope and pollen records indicate a dry climate from 10,500 to 10,000 l4C years BP. This interval forms the latter part of the Younger Dryas chronozone where cold conditions returned to Europe and North America. • From 10,000 to 7000 l4C years BP there is a general trend to higher lake levels and more humid conditions, with decreasing salinity and increasing temperatures. This trend is not smooth and two key wet periods are noted at 9100 and 8100 l4C years BP. The rising water level is attributed to both increased precipitation and rising sea level. Although water level was generally rising during the early Holocene, the ratio of evaporation to precipitation was still high between 10,000 and 8400 l4C years BP. The pollen data also indicates widespread dry conditions until 8200 i4C years BP. • Lake levels were high between 7000 and 5300 l4C years BP. • Evaporation increased slightly to precipitation at 5200 l4C years BP but lake levels remained generally high between 5200 and 3200 l4C years BP. • Greater fire frequency occurred in the moist mid Holocene. This was because although seasonality was at a maximum the early Holocene, it remained high into the mid Holocene. This resulted in there being large amounts of forests and dry winters. • Between 3200 and 2400 i4C years BP there is a two step increase in 5I80 indicating a trend to lower lake levels and increased aridity. The key dry period lasted from 2400 to 1500 l4C years BP, but, this may include a temporary increase in moisture around 1700 l4C years BP. • Wetter conditions are found between 1500 and 900 l4C years BP. These have been followed by a general increase in aridity until the present day. 31 2.1.7 Jamaica 1. Location: Wallywash Great Pond Authors: Street-Perrott et al. (1993) and Holmes et al. (1995) Proxies: Oxygen isotope Altitude: 7 metres a.s.l • From 120,000-106,000 years BP the climate was humid and warm in Jamaica. This was a complex episode that encompassed three successive lacustrine episodes. • The second warm and humid phase was prolonged and stable and lasted from 106,000-93,000 years BP. • Cold and arid conditions combined with lower sea levels during the Middle and Upper Pleistocene caused the lake to desiccate. • The Holocene consisted of alternating wet and dry conditions. The three highstands during the Holocene were at 10,000, 4400 and 1200 l4C years BP and were a response to a wetter climate and rising sea levels. The S180 values at these times indicate that the precipitation/evaporation levels were moderately high. 2.1.8 Bahamas 1. Location: Church's Blue Hole, Andros Island, northwest Bahama Archipelgo. Authors: Kjellmark (1996) Proxies: Pollen and Charcoal Altitude: not given • This record provides preliminary evidence for a period of dry climate prior to 4630 l4C years BP when the rest of the Caribbean was wetter than at present. 32 • A second dry period is noted and is tentatively dated from 2980 l4C years BP until 1530 l4C years BP. Human impact is clear in the record from 740 ,4C years BP with a peak in charcoal concentration at this point. • The patterns of pollen and charcoal also trace more recent environmental change related to the documented movement of peoples on the island. In order to ensure that this review is truly regional for the whole of the circum- Caribbean it is important that both the Florida Peninsula and north Venezuela are considered. 2.1.9 The Florida Peninsula 1. Location: Lakes Tulane and Sheelar, South-central Florida Authors: Watts and Hansen, 1994 Proxies: Pollen Altitude: 36 metres a.s.l • Full glacial conditions in this area are characterised by open vegetation and are different at all times from conditions exhibited in the Holocene even though there is variation throughout the Peninsula. • Many lakes in Florida began to accumulate sediment for the first time around 8000 l4C years BP as a result of rising sea level. • Between 7000-5000 l4C years BP pine became the dominant vegetation on the Peninsula as a result of the modern climate of predominantly summer precipitation being established in this time frame. 33 2.1.10 Venezuela 1. Location: Lake Valencia, north coast Venezuela Authors: Bradbury, 1979; Bradbury et al., 1981; Binford, 1982; Leyden, 1985; Rull, 1996, Curtis et al., 1999 Proxies: Stable isotopes, pollen, sediment geochemistry and diatoms Altitude: 402 metres a.s.l • From 12,600-10,000 14C years BP conditions were dry apart from a short lived eutrophic, freshwater event at 10,900 l4C years BP. • Effective moisture increased from 10,500-9800 l4C years BP which matches the Younger Dryas chronozone. • During the early to mid Holocene (8200-3000 l4C years BP) conditions were moist. Within this time frame there were two periods of lower lake level from 7600-6700 14C years BP and 3300 14C years BP. • By 2140 l4C years BP the lake was experiencing lowered water levels and increased salinity. The interpretation of conditions over this period have changed through time. 2.2 How do the records compare? For the Last Glacial Maximum, Leyden et al. (1993) inferred a 6.5-8°C drop in annual mean temperature for lowland Guatemala and dry conditions. These values agree well with those postulated for mountainous Costa Rica. The records from Costa Rica and Lake Miragoane, Haiti from glacial to Holocene transition suggests a gradual change to Holocene environmental conditions rather than the abrupt changes recorded in Greenland ice cores (Islebe and Hooghiemstra, 1997). Early Holocene records all indicate either warm and wet or warm and dry conditions. The Caribbean and Mexican Yucatan Peninsula records show dry conditions from 10,000-8200 l4C years BP and wet conditions from 8200-7000 l4C years BP. This too is shown in the records from Guatemala, with moist conditions being established 34 around 9000-8000 l4C years BP. Panama, however, was warm and wet from 10,000- 9000I4C years BP and dry from 8200- 5500 l4C years BP. This reflects the strong influence of the Pacific Ocean. The mid Holocene appears to be moist in Costa Rica (7000-4500 l4C years BP), the Caribbean (7000-5300 l4C years BP) and the Yucatan Peninsula. The latter region has records showing wet conditions from 7100-3000 l4C years BP, however, the record from San Jose Chuluchaca, Yucatan Peninsula has an intermittent dry period within this wet period from 6800-5270 l4C years BP. Records from Guatemala and Panama during this time cannot be purely related to natural variation. Open forests, which many of the records show existed at this time, may not be a result of a drier climate, but due to the impact of man. During the late Holocene conditions were dry. Costa Rica was dry from 4500-1500 l4C years BP with two key arid intervals at 2700-2000 l4C years BP and 1110-1180 14C years BP. Panama was dry from 3800-3700 l4C years BP, 3400-2500 l4C years BP and 1900 l4C years BP to the present. The Caribbean region is dry from 3200- 2400 l4C years BP, 2400-1500 14C years BP and wet from 1500-900 l4C years BP, and dry to the present. Curtis et al. (1996) provide evidence for decadal scale arid intervals in the Yucatan Peninsula from 1041-586 +/- 50 cal year AD. The key dry period noted by Curtis et al. (1996) has been found in other records from Central America. Lake Patzcuaro, Mexico shows a strong drying signal from 1100- 1200 l4C years BP (Metcalfe et al., 1994). The Zacapu Basin shows the same signal at 1000 l4C years BP (Metcalfe, 1995). La Piscina de Yurira also shows signs of desiccation from 1500-900 l4C years BP (Metcalfe et al., 1994). Preliminary evidence also exists from Lake Zirahuen for a drying signal around 900 years BP (Davies, 2000). Horn and Sanford (1992) noted that 1180-1110 l4C years BP was a time of increased fire frequency due to an arid climate. These connections are apparent over a wide geographical area. It has also been postulated that peaks in microparticle concentration in the Peruvian Quelccaya ice cap, that occurred in AD 535-665, 855-985, 1384-1410, can be regarded as a proxy for increased aridity in the 35 Andean Antiplano (Thompson et al., 1988). This tentative correlation between records from the Northern and Southern Hemisphere's suggests that these events may be related to large-scale departures in atmospheric and oceanic systems. 2.3 Mechanisms of Climate Change: It is clear that the modern-day weather patterns of Belize are characterised by distinctive seasonal cycles (section 1.3). In order to understand the changes to the climate of Belize through time, the controlling mechanisms governing the seasonal cycles need to be examined. These mechanisms operate on a variety of time scales and influence climate systems in a number of different ways. It cannot be assumed that the climatic changes of the past were caused by single forcing factors, it is however more likely that a number of different factors were involved. In order to unravel the mechanisms of climatic change in this region a number of different factors need to be considered. Over long time scales, changes in orbital forcing are influential. Milankovitch developed the idea that variations in the earth's orbit and axis occur over three distinct cycles. These variations result in seasonal insolation changes that modify the intensity of the annual cycle and therefore influence the global climate including the circum-Caribbean (Hodell et al., 1991; Leyden et al., 1994). The manifestations of these changes are geographically specific. In the northern tropics dry conditions in the late Pleistocene and moist conditions during the early Holocene were, in part, a response to changes in both precession and tilt (Curtis and Hodell, 1993; Curtis et al., 1998). In the tropics of North America dry conditions in the Lateglacial have been noted in: 1. Florida (Watts, 1975) 2. Peten District, Guatemala (Leyden et al., 1993; 1994; Brenner, 1994). 3. Panama (Bush and Colinvaux, 1990; Piperno et al., 1990; Bush et al., 1992). 4. Haiti (Hodell et al., 1991; 1995; Higuera-Gundy, 1999; Curtis and Hodell, 1993) 5. Jamaica (Street-Perott et al., 1993; Holmes et al, 1995) 36 The Younger Dryas is a key climatic event noted in records from higher latitudes (Lowe and Walker, 1997). Records from high latitude North Atlantic, Greenland and Europe all show a period of rapid warming from 13,000 to 12,600 l4C years BP followed by an abrupt reversal to colder conditions at 11,000-10,000 14C years BP (Leyden, 1995). A change at this time has been noted in records from the Caribbean and the Gulf ofMexico, but of a different magnitude and extent to that shown in high latitudes (Leyden, 1995). There are four records which cover this time period in Central America: La Chonta Bog, Costa Rica (Hooghiemstra et al., 1992; Islebe et al., 1995); Lake Quexil, Guatemala (Deevey et al., 1983; Leyden, 1984; Leyden et al., 1993; 1994; Brenner, 1994); Lake Miragoane, Haiti (Hodell et al., 1991) and Lake La Yeguada, Panama (Piperno et al., 1990; Bush et al., 1992). These sites show a shift in climatic conditions during the Younger Dryas period. Both La Chonta Bog and Lake Quexil show a return to cooler conditions and an increase in moisture. The record from Lake Miragoane indicates a drier climate. Lake La Yeguada has a more ambiguous environmental signal showing an increase in moisture and warmer temperatures. The key parameters enabling climatic links between high and low latitudes include: seasonality of insolation, tropical SSTs and sea level rise. During the Lateglacial the seasonality of insolation increased (Berger, 1978) causing an increase in precipitation associated with the annual movement of the ITCZ. The western tropical Atlantic and the southwestern tropical Pacific were 5°C cooler in the Lateglacial (Beck et al., 1992; Guilderson et al., 1994). It has also been found that SST fluctuations during the Lateglacial-Holocene transition correspond to sea level rise as a result of glacial meltwater inputs (Guilderson et al., 1994). This cooling in North America, as a response to meltwater forcing in the North Atlantic, may have generated more persistent or more frequent incursions of winter cold fronts to Central America (Leyden et al., 1994). This is compatible with cooler temperatures without a decrease in precipitation. Panama would have remained unaffected if the fronts only extended to Costa Rica as they do presently. In short, the Younger Dryas was a cool signal with an increase in precipitation occurring between 11,070-10,400 l4C years 37 BP in Costa Rica and immediately prior to 10,300 l4C years BP in Guatemala. The drying signal seen in Haiti occurred from 10,500-10,000 years BP. The intensity of the annual cycle in the northern tropics reached a maximum during the early Holocene (Curtis et al., 1999). The early to mid Holocene was moist. This has been noted in: 1. Yucatan Peninsula, Mexico (Covich and Stuiver, 1974; Hodell et al., 1991; 1995; Curtis and Hodell, 1993). 2. Peten District, Guatemala (Deevey et al., 1983; Leyden 1984; Islebe et al., 1996a; Curtis et al., 1998). 3. Panama (Piperno et al., 1990). 4. Costa Rica (Islebe et al., 1996b). During this period, due to changes in the earth's orbit, Northern Hemisphere summers were much warmer and winters were much cooler than at present. This led to the ITCZ travelling further north and south of its present-day maxima due to the large differences in insolation between the seasons. The consequences of this were wet conditions in low elevation sites in the northern tropics (Curtis et al., 1999). 2.3.1 The Role of the Ocean: It is apparent that the oceans play a significant role in modulating the climate of surrounding land masses. The strong marine influence over Central America suggests that changes in oceanic conditions should be reflected in the terrestrial record (Hastenrath, 1976). The North Atlantic operates as a conveyor system with water moving northwards in the upper levels of the ocean, sinking around latitude 60°N to form a deep water mass known as North Atlantic Deep Water (NADW). The return limb of the conveyor transfers deep water to the southern oceans. At a global scale, it has been postulated that differences in salt concentrations between the Atlantic and Pacific 38 Oceans ultimately drive this global conveyor (Broecker et al., 1985). The key issue is whether the ocean-atmosphere system has more than one mode of operation. The implication of a change to the system would be that the earth, under the same solar regimes, would have quite different climates (Broecker et al., 1985). The role of glacial meltwater from the North American Laurentide ice sheet is thought to be very significant as a mechanism for millennial scale climate changes such as the Younger Dryas. Periods of increased freshwater flow to the North Atlantic occurred at the same time as reductions in the formation of NADW (Clark et al., 2001). Recent evidence from the Bahama Banks suggest that both deep (NADW) and glacial North Atlantic Intermediate Water (NAIW) form. These two states wax and wane alternatively over orbital and millennial time scales (Marchitto et al., 1998). The replacement ofNADW by a less efficient system such as NAIW would have far- reaching implications for global climate. For example, if NAIW were to form during glacial conditions cooler SSTs in the tropics would result, as suggested by data from Barbados and Brazil (Webb et al., 1997). It has also been postulated that on longer time scales the Pacific and Atlantic Oceans had an equally significant role in producing millennial scale climatic cycles. This idea focuses on the fact that unstable ocean-atmosphere interactions in the tropical Pacific change tropical Pacific SST distribution. The locus of atmospheric convection lies over the warmest waters and therefore would have to move simultaneously. Moving this convection centre alters telecommunication patterns and therefore has far-reaching implications (Cane, 1998). The El Nino-Southern Oscillation (ENSO) cycle is the most familiar instance of instability in the tropical Pacific with global implications. The ENSO cycle is sensitive to the seasonal cycle in the tropics and is therefore likely to be influenced by orbital variations (Cane, 1998). It has been found that increases and decreases in ENSO warm/cold events occur as a result of modelled changes in orbital insolation. Changes in the mean position of the warmest SSTs accompany these variations (Cane, 1998). ENSO, and variability in the subtropical North Atlantic high sea level 39 pressure (SLP), are known to affect rainfall in the Caribbean region (Giannini et al., 2001). It has also been found that atmospheric circulation is shaped by the competition between the sea level pressure system associated with the North Atlantic subtropical high and the eastern Pacific ITCZ. These influence air mass convergence on seasonal and interannual timescales. Anomalously high SLP in the region of the North Atlantic high translates into stronger trade winds, cooler SSTs and less Caribbean rainfall (Giannini et al., 2000). Studies have also shown that rainfall in the Caribbean is dependent on the interaction between SST anomalies in the tropical Atlantic and the tropical eastern Pacific. An example of this is the rainy season in lower Central America, which starts early and ends late in those years beginning with warm SSTs in the tropical North Atlantic. The end dates are also delayed when the eastern equatorial Pacific is cool (Enfield and Elfaro, 1999). The tropical Atlantic region unlike the tropical Pacific, is not dominated by any single mode of climatic variability (such as ENSO) rather, it is subject to multiple competing influences of comparable importance (Sutton et al., 2000). The interactions between oceans and the atmosphere need to be examined further to enable the climate of the circum-Caribbean to be better understood. 2.3.2 The Late Holocene Dry Period One of the central aims of this study is to reconstruct the climatic changes that have occurred in Belize through time. One of the key events which ties climatic and human history together in the circum-Caribbean is the late Holocene dry period. Evidence for this event has been found in a wide range of sites (e.g. Horn and Sanford, 1992; Metcalfe 1995; Curtis et al., 1996) and has been implicated in the collapse of the Mayan civilisation, which occurred about AD 850 (e.g. Hodell et al., 1995; Curtis et al., 1996). Research efforts have concentrated on defining the extent and intensity of this drought rather than the causal mechanisms behind it. 40 The drought is part of a general drying trend which occurred in this region from 3000 l4C years BP onwards (Curtis et al., 1999). This general trend is thought to have been caused by a reduction in the intensity of the annual cycle whereby the perihelion shifted to the winter resulting in warmer conditions, and the aphelion shifted to the summer resulting in cooler conditions during the respective seasons. The movement of the ITCZ was, therefore, restricted and precipitation amounts declined (Curtis et al., 1999). Hodell et al. (1991, 1995) determined that the main driving mechanism behind the climate changes seen in Haiti and the Mexican Yucatan Peninsula were due to orbital forcing and the subsequent affects on the intensity of the annual cycle. The authors did not believe that the abrupt shift to dry conditions c.1000 years BP could be explained by this mechanism. Curtis et al. (1996) however suggest that the short- term variability in precipitation and evaporation balance shown in their record is a result of large-scale departures in atmospheric and oceanic circulation which must ultimately be due to orbital forcing. Brenner et al. (2001) suggest that there must be other, as yet unexplained, forcing factors which are at work in this region. The drought occurred in both Central America, the Sahel (Street-Perrott et al., 2000) and Ethiopia (Bonnefille and Mohammed, 1994). This linkage is not suiprising because both regions are influenced by the seasonal migration of the ITCZ (Hastenrath, 1976). Over the last century, variations in Sahel rainfall have been correlated with the degree of temperature contrast between the North and South Atlantic. Cold northern oceans and warm southern oceans are generally associated with dry years in the Sahel; Caribbean and Central America and vice versa (Hastenrath, 1991). Further explanations for the dry period include the reduced northward heat transport by the oceans as a result of either the freshening of the water column or feedback processes stimulated by changes in tropical land surface. Such events have been found to occur during the summer in the tropics and during the winter at temperate latitudes in association with cooler SSTs in the North Atlantic (Lamb et al., 1995). 41 Variations in the output from the sun are well established to have a major influence on climate change (Van Geel et al., 1998). GCM simulations suggest that a 2% decrease in solar output would result in a 4°C drop in earth surface temperatures (Hansen et al., 1984). The principal mechanism through which the output from the sun varies is through sunspots. Observations over the last two centuries have confirmed that these have a periodicity of ~11 years (Harvey, 1980). Furthermore it is thought that higher levels of solar activity increase the strength of the solar wind (the stream of protons and electrons emitted by the sun), which deflects cosmic rays and results in a decrease in the production of cosmogenic isotopes. (Van Geel et al., 2000). This has direct implications for the amount of 14C which is produced. In the most recent record which has been produced from Lake Chichancanab, Yucatan Peninsula, Hodell et al. (2001) discovered a 208-year cyclicity in the §180 record. This has also been identified in the record from Lake Punta Laguna, Yucatan Peninsula and fits well with the 200-year cycle isolated in the Cariaco Basin. This latter cycle has been attributed to solar forcing. A 206 year cycle is apparent in records of cosmogenic nuclide production. The 5I80 record from Lake Punta Laguna has been compared to l4C production records and for the last 2000 years they appear to be negatively correlated with times of higher 5lsO values coinciding with lower l4C production. This implies that the late Holocene dry period occurred during a period of increased solar activity. Changes in solar output are believed to affect global mean temperature, humidity, convection and the intensity of Hadley Circulation in the Tropics (Hodell et al., 2001). As demonstrated earlier, mean annual rainfall varies significantly throughout the Yucatan Peninsula and thus any change to the position of the Hadley Circulation or tropical convective activity is likely to affect rainfall patterns (Hodell et al., 2001). It is also possible that the late Holocene dry period could be the result of internal changes to the system such as volcanic eruptions. Evidence has been found for an eruption of El Chichon between AD 676-788 which could have been partly responsible for the climatic drying seen in AD 800-1000. El Chichon has erupted frequently over the last 8000 years and therefore it could have played an important 42 role in the development of environmental conditions in this area (Espindola et al., 2000). 2.3.3 Summary The mechanisms behind climatic change appear to be interrelated, with a variety of different forcing factors operating over a number of time scales. This has resulted in certain mechanisms having greater, lesser or even different implications for relatively nearby areas. This is especially important when considering the climates of North, Central and South America. On longer time scales it is apparent that orbital forcing has a profound influence on how wet or dry the climate is. Climates of Central America were directly influenced by the end of the Pleistocene glacial period through the input of meltwater into the oceans. This influenced both ocean and atmospheric processes. The climate of the Holocene was relatively stable. However, changes in the late Holocene have been more abrupt in nature and the forcing mechanisms behind these are not as clear. One of the most recently proposed ideas is solar forcing. In summary, the climate of Belize is affected by two key factors: the prevailing subtropical wind system and the associated ocean circulation patterns (particularly in terms of fluctuating SSTs). 2.4 The Impact of humans on the Environment It is apparent that the signal of climate-induced variability is blurred by the influence of humans on their environment. This makes it difficult to differentiate which factors are forcing change. There are three main ways in which this issue can be addressed. A detailed history of human occupation and activities can be gathered from archaeological and documentary sources. This will allow the results gained to be placed against a background of information which may enable a climate signal to be teased out. Alternatively, only sites that are free from human occupation could be studied which would ensure that any environmental fluctuations are climatically 43 induced. Alternatively, proxies such as oxygen isotopes could be employed which respond mainly to climatic forcing. The evidence for human impact is plentiful in the northern tropics (Deevey 1978; Deevey et al., 1979; Vaughan et al., 1985; Leyden, 1987; Rue, 1987; Bush et al., 1989; 1992; Hansen, 1990; Piperno et al., 1990; Bush and Colinvaux, 1994; Northrup and Horn, 1996). These impacts are not just restricted to modern human activity. For example O'Hara et al. (1993) convincingly showed that the impact of traditional agricultural methods was just as destructive to the environment as the plough agricultural systems introduced by the Spanish. One of the main problems encountered in the quest to unravel the impact of the Maya is that the chronological framework of lacustrine sequences is imprecise. Hard-water effects (Deevey and Stuiver, 1964) and the redeposition of carbonate-rich basin soils (Vaughan et al., 1985) result in inaccurate radiocarbon dates. Relative chronologies have therefore been relied upon through the correlation of pollen and archaeological evidence (Vaughan et al., 1985; Brenner, 1994). This process is hindered by the fact that pollen records are influenced by the activities of humans. This is especially influential when one considers that human impact is clear in Panamanian vegetation records from 11,000 years BP (Piperno et al., 1991; Cook and Ranere, 1992). The oxygen isotope record was thought to have been a purely climatic signal, but dramatic changes to watershed hydrology may actually produce isotopic shifts (Brenner et al., 1994; Rosenmeier et al., in press). These points highlight that as much information as possible must be gathered from the area being studied, before conclusions can be drawn from the data sets. The actions of humans will be influenced by the climate conditions within which they are found. Although it cannot be assumed that there is a causal relationship between climate change and the development of maize cropping, the latter occurred at the onset of drier conditions (7000 years BP). Increased seasonality (i.e. an augmented difference between the wet and dry season) and lower precipitation levels 44 would have expanded the range of exploitable habitats for cropping (Piperno et al., 1991). "....It appears that a major role for archaeologically inspired palaeoecology will be the provision of information on aspects of cultural process and change in tropical forests that is often difficult to extract from archaeological sites; dynamic environmental changes and their possible relationship to the evolution of subsistence strategies, organisation of labour and demographic trends". Piperno et al. (1991: 244). The key concept is therefore that the imprint of humans on their environment can be very significant and vice versa. This dynamic interplay needs to be unravelled before a true understanding can be gained of the environments in question. 2.5 Environmental Change Records from Belize There is a paucity of knowledge concerning the environmental history of Belize compared to neighbouring countries in Central America. A variety of environments in Belize have been studied, but the work presented in this thesis is the first palaeoecological study of its kind to be undertaken. The information which has already been published on Belize, provides a useful framework into which this study can be placed. Both the New River Lagoon and Honey Camp Lagoon are surrounded by marsh systems. Rejmankova et al.( 1995) studied the freshwater wetland plant communities of northern Belize. The climate of Belize is dominated by its seasonality and thus, although the marshes are waterlogged all year round, it is thought that they must experience periods of desiccation over longer time scales (Rejmankova, et al., 1995). Rejmankova et al. (1995) investigated the northern coastal plain within Orange Walk and Corozal districts of northern Belize (Figure 1.1). The hydrology of this area is not completely understood, but it is thought that the marshes are fed primarily by springs. Some surface runoff occurs during storms but this quickly enters the aquifer. The water table fluctuates in the order of lm a year (Siemens, 1978) and the 45 water level in the marshes is about 20 cm higher in an average wet season (Rejmankova et al., 1995). In modern times, there has been no attempt to modify the marsh environment in this area. Marsh fires are, however, common due either to accidents as a result of burning cane fields or are deliberately started to gather fish and turtles (Rejmankova et al., 1995). These marshes are not only important as the dominant ecosystem of the region, but they also have great archaeological significance. Through remote sensing techniques polygonal patterns have been distinguished in the marshes suggesting that the Maya utilised this environment for agriculture (Adams et al., 1981). Most of the marshes cannot be cultivated today because dry-season water levels and soil salinities are too high (Rejmankova, et al., 1995). Evidence suggests that ancient agricultural levels lie well below the modern marsh surface and the main period of activity was in the Preclassic Mayan period (2000 BC to AD 250). The subsequent rise in groundwater appears to have been the result of a rise in sea level (High, 1975) which ultimately resulted in the abandonment of the wetland fields (Pohl, 1990). Two important wetland sites in Belize have been studied in great depth: Cobweb Swamp and Pulltrouser Swamp (Figure 1.1). An 8000 year record has been obtained from Cobweb Swamp which is the location of the Mayan site of Colha (Alcala- Herrera et al., 1994; Jacob and Hallmark, 1996). Species of ostracods, foraminifera and molluscs were used to reconstruct salinity changes. This evidence suggests that there have been marked changes in salinity throughout the history of the swamp. The history of the site is as follows: 1. The period from 5630-4790 l4C years BP was one of rising sea level that resulted in a marine influenced estuarine environment. 2. The rate of sea level rise began to decrease at 4790 l4C years BP allowing the colonisation of marine tolerant species such as mangroves. 3. Deforestation in Cobweb Swamp began around 4500 years BP. This would have had a substantial affect on the swamp's hydrology. After 3370 l4C years BP water levels began to rise again and a new freshwater lagoon formed. There are three reasons why this may have occurred: 46 • In humid areas, such as Belize, trees are the main evapotranspirative pumps in the system. If deforestation occurred in the Classic period when population levels were high then the formation of a freshwater system could have been the result. From 1700-1000 years BP population densities in this area reached 200- 300 people/km2 (Rice and Culbert, 1990). • Substantial subsidence. • The construction of a dam by the Mayan population. 4. There is no evidence of renewed sea level rise that would have had an impact on Cobweb Swamp after 3370 14C years BP. 5. The system stabilised and swamp vegetation formed after 520 l4C years BP. The Mayan civilisation had collapsed by this point which meant that forest would have been able to develop (see Chapter 3). (Alcala-Herrera et al., 1994; Jacob and Hallmark, 1996). In order to develop a regional perspective on the changes that have occurred in north Belize, Alcala-Herrera et al. (1995) compared the records from Laguna de Cocos, Albion Island with Cobweb Swamp. Laguna de Cocos has been a freshwater lake for almost 4000 years whereas, during this time Cobweb Swamp has changed from a mangrove community to a brackish and then freshwater swamp. At the present time, the limnetic Laguna de Cocos contrasts with the oligohaline Cobweb Swamp (Alcala-Herrera et al., 1995). When the records are compared at 4820 l4C years BP (which is the date for Laguna de Cocos), both systems show similar mesohaline environments. Neale (1988) defines these salinity classifications as follows: • Mesohaline: 5-18%o • Oligohaline: 0.5-5%o • Limnetic: 0-0.5%o. Hansen (1990) showed (using pollen evidence from Laguna de Cocos) that from 5000-6000 years BP a dense forest surrounded the lagoon with high concentrations of pollen suggesting that there was little erosion of soils into the lake. Both sites also record the presence of Zea mays in the period 3000-3500 years BP suggesting that they were both impacted upon by man at the same time. Pollen data show a transition 47 from a closed forest to an open, disturbed agricultural area with savanna. The system reverted back to forest after the collapse of the Mayan civilisation in approximately AD 850. The evidence suggests that at the climax population in Albion Island, maize was planted in monoculture (Pohl et al., 1990). There is a difference in the timing of the removal of marine influence between the two areas. Laguna de Cocos was free by 4700 years BP and Cobweb swamp by 4000 years BP (Alcala-Herrera et al., 1995). This is not surprising as Cobweb Swamp is much closer to the sea than Laguna de Cocos and therefore its influence is likely to be much more pervading. Bradbury et al. (1990) attribute the freshening of Laguna de Cocos to the development of a moister climate at 5000 years BP. Alcala- Herrera et al. (1995) found evidence that present-day precipitation patterns were established 4000 years BP. The discrepancies between the two records are due to the poor chronological control and the different environments. Evidence for increased levels of moisture from 1700-1600 years BP and a drying trend around AD 1000 have been found which coincide with a drop in agricultural indicators (Bradbury et al., 1990). Pulltrouser Swamp has been extensively surveyed in terms of its vegetation, soils and Mayan history (Turner and Harrison, 1983). From this work a number of conclusions can be made. The region was undoubtedly affected high by sea level which stabilised at 2000 years BP (High, 1975). There is evidence of drier conditions before this time. The edges of the swamp show evidence for fluctuating water levels. The cause of this has not been determined as it could be due to human modification of the system or rainfall changes. It is thought, however, that major changes in rainfall would be needed to change appreciably the surface water characteristics (Turner and Harrison, 1983). 48 2.6 Conclusion It would be problematic at the moment to relate our knowledge of climatic change in Belize to the rest of Central America with confidence, due to lack of dating control. General trends can be delimited though. Sea level rise appears to have been the main control on lagoonal stratigraphy in lowland northern Belize during the Holocene (Alcala-Herrera et al., 1994). Its effect was also felt in the Mexican Yucatan Peninsula as it enabled the beginning of lacustrine sedimentation. There is also preliminary evidence for a drying trend in Belize c AD 1000. The effects of man are clear in the Belizian records through deforestation and the development of agriculture. The information presented in this chapter provides an important context within which the results of this thesis can be placed. In order to understand the patterns of human development in Belize, the next chapter will discuss the Maya. 49 31 crq c 0> ro H D- o> O o p o' D O |-h O 3 P C~t- CD >-5 O) o o »-1 O- t/3 5' r~+ ZT o ►-$ 0 c 3 1 n B. & cr CD p: 3 Figure 2.2 The Southern Lowlands (modified from Jones, 1991) Chapter Three: The Maya 3.1 Introduction Both human and physical forces shape the environment. In order to develop an understanding of the modern landscape the role of both these forces needs to be appreciated through time. This thesis aims to reconstruct the environmental changes that have affected Belize through time. An understanding of the human history of the area is therefore essential to the successful completion of this aim. The Maya are the native peoples of southern Mexico, Guatemala, Belize and Honduras having occupied this area for the last 5-6000 years. The region of interest to this investigation is the Southern Lowlands which comprises the Peten region of Guatemala stretching northwards to Tabasco and southern Campeche, Mexico; Belize; the Rio Motagua of Guatemala and a narrow portion of western Honduras (Coe, 1997) (Figure 2.2). With the exception of the Olmec and their associated groups, the Maya were the only new world high civilisation to arise and flourish within a tropical lowland forest ecosystem (Brenner, 1983). The development of the civilisation can be charted through a series of distinct periods (Coe, 1997): Archaic tol800BC Early Preclassic 1800-1000 BC Middle Preclassic 1000-300 BC Late Preclassic 300 BC- AD 250 Early Classic AD 250-600 (this includes the Hiatus: AD 534-593) Late Classic AD 600-800 Terminal Classic AD 800-925 Early Post Classic AD 925-1200 Late Post Classic AD 1200-C.1530 People have been fascinated by the Maya ever since Stephens and Catherwood rediscovered the sites of Central America in 1839. From this time on, there have been numerous excavations of Mayan sites. By the Late Preclassic, regional societies 52 of thousands of people had developed, in the Early Classic this had risen to tens of thousands and in the Late Classic to hundreds of thousands. The total population in the Maya Lowlands is believed to have peaked at between 9 and 14 million (Adams, 1991). Tainter (1988) estimates a population density in the Southern Lowlands of c. 200 people per km". This would have made the Lowlands one of the most densely populated areas of the pre-industrial world. Maya settlements have been regarded traditionally as religious centres but some sites were more inclined to this purpose than others (e.g. Copan, Honduras). The Maya economy has been described as elite controlled and ceremonial on the one hand and redistributive and dependent on markets/trade on the other. The society was patrilineal, patrilocal and patriarchal. Maya subsistence was both extensive and intensive (Pyburn et al., 1998). All these seemingly contradictory statements are correct as they can be applied to different areas at distinct times. The analysis of regions cannot, therefore, be undertaken using simple rules as these will not capture the essence of the society and how it developed through time with different power struggles and different centres of influence. One of the key research issues concerning the Maya is how their Classic civilisation came to an end in approximately AD 850. Popular literature describes this period as a collapse, which implies a rapid and catastrophic end. With the shift in archaeological focus in the 1950s away from ceremonial architecture to settlement surveys, the magnitude of the population decline at the time of the collapse was more fully appreciated (Willey, 1982). During the six centuries of the Mayan florescence the Maya created a society which was characterised by elaborate art and architectural styles, complex calendrical and hieroglyphic systems, long distance trade networks and spectacular ceremonial centres (Healy et al., 1983). Today hundreds of these sites lie as ruins across Guatemala, Belize and the Yucatan Peninsula highlighting both the wide distribution of the culture and their dramatic collapse. There are three sets of literature which can be investigated: anthropological, archaeological and more recently the palaeoclimate literature, to gain ideas about the collapse. One of the 53 most interesting aspects of Maya research is the development of the theories concerning the collapse and how ideas have changed through time. The collapse of the Mayan civilisation has been one of the most persistent research issues to have occupied archaeologists and anthropologists alike. The collapse was preceded by a period known as the hiatus (AD 534-593) which was a time of severe cultural decline. The collapse and its implications will be explored in more detail in section 3.5. The collapse is unlikely to have uniformly affected the whole of the Southern Lowlands at one time. Gill (1994) estimated that it actually extended from AD 760-910. The abandonment of the practice of erecting monuments bearing long count dates (the calendrical system devised by the Maya) was traditionally taken as the evidence for the dramatic collapse of the Maya. However, the withdrawal of this process was only one symptom of a widespread process of change that operated over several centuries. No single event or process was responsible for all the changes seen at the close of the Classic period and the sequence of contributing events varied from region to region and even site to site (Adams, 1973; Sharer, 1994). The traditional view concerning the 9th Century was that this was a period where the Mayan people experienced a time of dramatic upheaval that resulted in the collapse of the society. Through the increasing amount of work that is being carried out at Mayan sites, it has become apparent that different areas coped with the changes that occurred at this time in very different ways. One certainty is that Classic period culture did undergo a major transformation in the 9th Century AD. Tikal and similar sites appear to have experienced a true collapse whereby people left these high- density urban areas to join or create low density settlements that were near water- bodies. The fact that the Mayan civilisation collapsed in its entirety is a myth that is commonly retold in the general literature and by the media. Key sites that did not collapse were Lamanai and the coastal sites of Belize (Graham, 2001). 54 Adams (1973) defines the collapse in terms of the failure of the elite combined with the abandonment of palaces, cessation of temple and stelae construction and the diminishing use of calendrical and writing systems. This all occurred during a time of rapid depopulation of the countryside and ceremonial centres. The collapse of the Classic Mayan civilisation began in c. AD 760 and continued for the next 80/100 years with hundreds of major sites being abandoned over at least half the lowlands area and in the remaining half, the site occupation and construction declined markedly. The pattern of events for the Southern Lowlands is clear but in the Northern Lowlands the collapse is not thought to have been as severe (although much less research has been undertaken in this area). The Mayan civilisation was a substantial achievement especially considering that the Lowlands are not optimal lands for agricultural production. One of the main premises behind this thesis is the investigation of the climatic changes that occurred in the Holocene and more specifically the enhanced understanding of the late Holocene dry period. This arid period coincided with the collapse and it has been postulated that these two events are connected (e.g. Curtis et al., 1996). It is important to go through the changing schools of thought regarding magnitude, extent and causal mechanisms behind the collapse as it provides a context for the presently accepted ideas. The aim of this chapter is to develop an understanding of the human forces that have shaped the landscapes of Belize. This will be achieved by developing an understanding of the Mayan society and how it both utilised and impacted upon its environment. This, combined with a knowledge of the general trends in society (achieved through the study of the main Mayan sites in north Belize) will enable an appreciation of both the role and importance of humans in the history of Belize. 3.2 Rural activities and environmental impact The Maya exploited a wide range of agricultural techniques to deal with the vast array of physical environments in which they were based. The original view of the Mayan civilisation was that they were supported by shifting agriculture, had only a limited population and lacked true urban centres (Adams, 1991). Research since the 55 late 1960s has completely disproved this viewpoint and it is now known that although the Maya began by practising shifting agriculture, because of population pressure in the Late Preclassic, they began to employ more intensive techniques. It was this transition that was associated with the development of cities. The Mayan society at its inception was based on a hunter-gatherer system along the Caribbean coast. Settled village life is thought to have begun around 4200-3300BC (Sharer, 1994). The first type of agriculture that was employed was extensive i.e. large areas of land were utilised but the yields produced were low. The oldest method of intensive agriculture that has been found is in the form of household gardens. This system was based on a plot of land adjacent to the house and the soil fertility was maintained by the addition of household rubbish. The key intensive systems that were employed at the height of the Mayan success were terraces and raised fields which enabled much more land to be used productively. These are rarely found today. Investigations have shown that maize was the staple crop from Preclassic to Historic times for the Maya, although consumption appears to be highly influenced by cultural change (White et al., 1993). The Maya were constantly altering their environment resulting in the creation of a highly productive setting. During the initial colonisation period of the Maya (up to c 1000 BC) the coastal zone was the only main area that could support a settled society. With rising populations, people gradually moved inland following rivers and settling in swamp margins. These depressions ('bajos') often held water throughout the dry season and therefore could be used for agriculture. Shifting agriculture was not found to be an efficient way of utilising land because after three years of cultivation, land had to be lain to fallow for four to eight years to ensure continuing fertility (Adams, 1991). Maize was always a staple agricultural crop but frequently polyculture was undertaken which used a number of tropical domesticates (Scarborough, 1998). There is also evidence for a wide range of root crops, tended tree crops and cotton (Hammond and Miksicek, 1981). The strategy employed is vital to the success of the system because for example crop losses occur due to weeds, insect infestations and plant diseases. These become much more significant 56 the longer a field is under successive cultivation. This is especially severe if only one crop is planted. Farmers found that they could counteract drops in productivity by cultivating more land, but because of the increased labour input which was needed per hectare with such a scheme, there was a finite amount of land that could be cultivated at one time. Successful agriculture therefore mimicked nature with a variety of crops being planted at the same time. Evidence of permanent settlement is not found until the Late Preclassic period with the appearance of food storage pits. By the Late Preclassic sizeable populations were established throughout the Lowlands, all of which demonstrated localised techniques in dealing with the seasonal climate and the wide variety of ecosystems (Scarborough, 1998). The role and types of agriculture that were undertaken within the Maya communities is very important to determine in order to develop a clear idea of how the various societies operated through time in Belize. It was the discovery of rectangular canal patterns on the margins of riverine wetlands that led to the suggestion that the Maya may have developed intensive agricultural techniques in the Classic period. The patterns observed are thought to represent intensive hydraulic systems to improve crop-soil-water relations and to extend cropping into the dry season (Turner and Harrision, 1983). Within wetland cultivation, drainage channels, raised fields and water control devices were employed. These required a great deal of labour, planning and materials. It is thought that every major swamp, watercourse and lake edge in the Southern Lowlands was exploited for agriculture by the end of the Classic period (Adams, 1991). Controversy still exists however about the nature, chronology and extent of the modifications (Adams et al., 1981; Turner and Harrison 1983; Pope and Dahlin, 1989; Pohl, 1990 and Culbert, 1991). A raised field is an agricultural feature created by transferring earth to raise an area above the natural terrain. An example of where they were found is Pulltrouser Swamp and Barber Creek near Lamanai (Turner and Harrison, 1981). Raised fields and associated canals provided dry, cultivable land in areas otherwise subject to inundation. Terracing allowed directed drainage on both steep and shallow hillsides, trapped silt and created fertile areas 57 (Tainter, 1988). Water channelling and storage involved a variety of techniques (not all of which related to agriculture) but were all labour intensive. The Lowlands as a whole are deficient in surface water, especially during the dry season therefore canals, dams, reservoirs and small wells were built. Caracol (south Belize) had man made reservoirs, one of which even holds water in the dry season today. Due to the thin soils in this area terraces were also employed. The terracing systems at this site showed a high degree of planning. The upper elevations of the slope were untouched and lacked terracing. This treed zone may have acted as a biological reservoir and a source of wood, food and medicinal plants. The vegetation cover would have intercepted some of the rainfall which would have fallen on the area, reducing the amount flowing down the slopes. The water that did flow through the system would have picked up nutrients and brought them to the lower slopes (Healy et al., 1983). Evidence from the 'Programme for Belize' region in north Belize suggests that terraces were constructed through the modification of natural contours with the new features averaging 10 metres in width and 1.5 metres high (Walling, 1995). The purpose of these structures appears to have been to amend slightly degraded land rather than improve the worst areas. Dry season water supplies appear to have been crucial in the development of complex societies in the Lowlands (Adams and Valdez, 1995). It has been suggested that water control was the means by which the initial Mayan leaders were able to shift religious power to political power. Northern Belize has more reliable sources of groundwater than the central Peten region and therefore there would have been less opportunity to monopolise resources. This may be part of the reason why large political centres did not develop in lowland Belize. Communities that were larger than villages had to have reliable storage facilities that could provide the community with enough water to last for about 100 days without rain in the dry season. In conjunction with this further water management strategies would have had to control erosion and flooding during the heavy downpours in the wet season. Such features have been noted in Guijarral which is located in Programme for Belize land. This settlement is located on high land surrounded by a high frequency of reservoirs. The 58 reservoirs were designed to allow water to escape gradually in times of very wet conditions by having one side higher than the other. The lower lip of the reservoir in Guijarral consistently faces downslope and away from the settlement. The terrace systems below would have slowed the movement of water, preventing excess soil loss (Adams and Valdez, 1995). This intrinsic understanding of water management highlights the central place that this resource had in Mayan society. In the Late Preclassic water management was a 'concave' micro-watershed adaptation i.e. sites were low lying and relied upon natural slope runoff (i.e. Cerros, north Belize). In the Classic period sites like La Milpa (north-west Belize) practised 'convex' micro-watershed adaptations. This allowed communities to move away from the traditional areas in which they had been located. Sites were positioned on natural hills so that the quarried surface for construction fill and the resultant reservoir and rainwater catchment surfaces could best be utilised. This system allowed for much greater control but also shows the great awareness of the Maya about the best management of their surroundings (Scarborough et al., 1995). The development of these extensive water management systems implies that economic and political power was being centralised into these large cities. 3.3 The sites investigated in this study: Introduction In order to gain a real appreciation of the human history of Belize it is important that knowledge of the main archaeological sites is unravelled. This will provide two key sets of information: 1. The primary development areas of Belize; the shifts in political power; agricultural zones and the general affects of the postulated collapse around AD 850. 2. A context for the two key areas under consideration in this investigation (New River Lagoon and Honey Camp Lagoon). The aim of the following section is to provide a history of the New River Lagoon and Honey Camp Lagoon through the work that has been carried out by archaeologists. 59 This is essential in order to unravel the records of environmental change, that have been produced in this investigation (see Chapters 6,7 and 8). The history of human occupation and activity will provide a context and a relative chronology for the lake sediment sequences. Knowledge of the sites can also be related to the other centres in the country enabling an assessment of the importance and role of the New River Lagoon and Honey Camp Lagoon. These sites are shown in detail on Figures 3.1 and 3.4. 3.3.1 Lamanai The first published investigation of this site was by Pendergast and others in 1974 (Pendergast, 1975). The first documented visit to the site was, however, by Thomas Gann in 1917 (Gann, 1928). Lamanai is unique in Belize because of its unbroken span of occupation which lasted for 2000 years. The site name in Mayan means drowned insect or crocodile. The site had an extensive and well-developed population by 300 BC and it was an extremely important Preclassic centre (Pendergast, 1981). The founding of the site is thought to have been as early as 1500 BC. The key part of its historical development was the fact that it survived the Classic collapse (Pendergast, 1981). The post-collapse period shows a transformation of social organisation and technology without a change in the local population (Masson, 1997). An area of 6Km2 has been studied around the main site. The main structures form a compact zone in the centre, outside of which the scale and density of buildings rapidly diminish (Loten, 1985). Lamanai does not have a standard settlement pattern which consists of more than one ceremonial precinct plaza. This is due to the proximity of the lagoon, which has lead to a strip-like development along the shore (Pendergast, 1981). A shift in the affiliations of the site has been noted in the architecture and ceramics. In the Preclassic and Early Classic the links appear to have been with the Peten region of Guatemala. This weakened during the Late Classic as the Peten reached its cultural peak. By the Postclassic, Lamanai was linked to the northern Yucatan region (Loten, 1985). It has also been noted that there 60 are architectural links between Lamanai and Cerros (which is located at the mouth of the New River) (Pendergast, 1981). Lamanai like many other Late Classic centres was under very strict elite control (Pendergast, 1981). Information on population levels through the site's history is sparse. It is believed that the peak of the population was in the Preclassic and into the Early Classic and this may have been as high as 10,000 (Lambert et al., 1984). Ridged field systems have been found 32 km northeast of the site and these are thought to have been utilised by the people of Lamanai, suggesting that they undertook intensive agriculture. Evidence has also been found for the use of maize as an agricultural staple (Lambert and Arnason, 1978). By AD 1637 the population of the site was 72 (Jones, 1989). An indication of changing population levels can be gained from the time periods in which the major temples were built (Figures 3.1 and 3.2): 1. N10-43: This is the largest Preclassic structure in the Maya area and to build it would have required a considerable workforce that was under the control of a well-maintained hierarchy of elite. 2. P9-2: Late Preclassic structure. 3. P8-12: This structure was both built and modified in the late Preclassic 4. N10-9: This building was completed by the early Classic, modified both in the late Classic and the early/mid Postclasssic. 5. N9-56: This was constructed at the end of the early Classic. 6. N10-7: This is a Classic structure that was heavily modified in the mid- Postclassic. 7. The Ottawa Complex: Classic period construction but during the late 9th or early 10th Century it was massively modified (Pendergast, 1981; Loten, 1985; Graham, 2001). It appears as if there were two key phases to the construction activities in Lamanai. The major monument building phases occurred in the Preclassic and the Classic. The late Classic and Postclassic appear to have been periods of building modification 61 rather than construction. Both these phases would have had different impacts on the environment. The building of N10-43 in the Preclassic not only required a large workforce and an organised elite it would have also required trained architects, designers and astronomers (for temple location) as well as reliable food supply mechanisms either through an agricultural workforce or through trade networks. The construction of this building is thought to have been completed by 100 BC. The key period in Mayan history, between AD 850-925, was vibrant in Lamanai. No evidence has been found in the ceramics or architecture for a break at the beginning or the end of the Terminal Classic. This community survived the collapse and continued throughout the Post Classic while neighbouring sites were gradually being deserted. According to Pendergast (1987) there are three reasons, which may have contributed to this. Firstly, Lamanai is situated at the edge of a large lake meaning the inhabitants would have had a richer and more varied diet than was available at land locked sites. Secondly, the New River Lagoon provided the inhabitants with an open means of communication with the northern Maya area and other parts of Mesoamerica which would have enabled trade routes to continue. Thirdly, undetectable factors such as the strength and personality of the community's leaders. These would have provided a force of stability at a time of crisis. There is archaeological evidence in the Postclassic for mass production of pots using moulds. This is clear evidence that Lamanai was part of a trade network during this time frame. There is also evidence that copper was being imported from AD 1200 and by AD 1400 it was being made on site (E.Graham, pers.com. 2001). The site, at the time of the Spanish arrival in 1540, was concentrated in the southern half of the area with a northern satellite town (Pendergast, 1987). The transition to the south is thought to have occurred in the 11th Century (Pendergast, 1985). The settlement was directly adjacent to the lagoon edge, in an area that was formerly a ceremonial zone in Classic times. There is much evidence throughout the site of a move from a ceremonial to residential use of buildings and zones during the 11th 62 Century (Pendergast, 1985). The important point to note is, however, a continual use of the lagoon edge. The first church was built at the site in AD 1567 and was a Mayan building adapted for Christian use. The second, built near the end of the 16th Century, was European in style. This latter church was much more impressive and may signify a strengthening of the Spanish hold over the area (Pendergast, 1984). The church at Lamanai first appeared on the church list of AD 1582 (Roys, 1957) and the site was first visited by Fathers Bartolome de Fuensalida and Juan de Orbita in AD 1618 (Pendergast, 1981). In AD 1640 the Christianised Maya desecrated the church and allied themselves with the citizens of Tipu in central Belize. Incidentally, Tipu had exactly the same style church built in the late 16th Century (Pendergast et al., 1993). Famanai at this time was not, however, entirely abandoned (Loten, 1985) with the few inhabitants left still identifying the church area as a sacred space and undertaking some Christian practices (Pendergast et al., 1993). Famanai appears on European maps from 1622 suggesting that its Spanish rulers regarded the site as important. This is not surprising due to its strategic position at the headwaters of the New River Lagoon, which is the principal water route into the southern interior of the country. In maps dating from the late 19th to early 20th Century, Lamanai is not recorded. Indian Church, the modern day settlement in this area, is marked on most of these suggesting that it was this site and not the original territory that was being utilised during this time period. In the 19th Century a sugar mill was in place near Lamanai. The fields appear to have been concentrated in the area south of the main ancient centre and west of the church. This was closed down in the 1870s (Pendergast, 1975). There is also evidence of 19th Century British buildings and activity along the shoreline (Pendergast, 1986). 63 3.3.2 Hillbank There are thought to have been 15 Spanish colonial towns in Belize. One of these is Colmotz, which is believed to have been located at the modern day site of Hillbank (Pendergast, et al., 1993) (Figure 1.1). The terrain of the west bank of New River Lagoon is attractive for settlement, especially in comparison with the low-lying swamps on the east side (Figure 3.3). Its elevation protects against flooding and provides a good view of the Lagoon which may have been useful in terms of defence. These same benefits would also have been felt in Lamanai. Due to the extensive modification of this site in modern times any archaeological information which was present has been destroyed. The small amount that has been found suggests that the site was existence during the Terminal Prehistoric or early Spanish period (Pendergast et al., 1993). Hillbank was the site of extensive logging operations in the 19th and 20th Centuries. The site is noted as early as 1867 and on the maps dating from 1938-1952 the New River Lagoon is named Hillbank Lagoon. It is still referred to as such in Wright et al. (1959) even though logging is thought to have stopped in this area some time before 1949. This suggests that this was an area of significance. The name 'bank' in Belize predominates in northern river sites and it relates to those areas exploited for logwood transport (Duncan, 1966). The Lagoon is the only source of permanent water in this area. 3.3.3 Honey Camp Lagoon The original name for this Lagoon was Laguna de On or Lake of the Alligator Pear. Honey Camp was an experimental station for the intensive study of the Sapote tree and the logging camp associated with this was situated to the southwest of the lagoon on a high bank (Gann, 1928). Although the Lagoon is self-contained, the adjacent Doubloon Bank and Button Lagoons did connect to the Caribbean Sea through Freshwater Creek which is now silted up (Masson, 1993) (Figure 3.4). This is thought to have been a route to the Caribbean Sea during Classic and Postclassic 64 times (Masson, unpub). A location near this route would have been of great strategic importance. There are two small islands at the south end of the Lagoon which have two metre high platforms made out of lagoon sediment upon which Postclassic artefacts have been found. The western of the two islands appears to have been connected to the shore by a bridge of sand (Masson, 1993). Two key Mayan communities inhabited this site. The first population was located on the shore during the Terminal Classic/ Early Postclassic and the island population was founded here during the Postclassic (from AD 1100) (Masson, 1995). It is not known whether this latter population was a replacement or a displacement of the original shore community (Masson, 1996). The origin of the Honey Camp community is a critical issue in trying to resolve the extent of the collapse in this area. There are two scenarios which could have occurred: 1. The island population could have been migrants from a different area with the natives dying in situ. 2. The native population could have migrated to Lamanai (the nearest site that was populated during this period) or indeed taken refuge on the island suggesting that the collapse was a time of downscaling and consolidation (Masson, 1993). This site, like its much more powerful neighbour Lamanai, occupied a strategic position in the Postclassic between the Chichen Itza, Mayapan and Peten regions (Masson, 1995). It appears to be a general trend in north Belize that settlement patterns in the Postclassic moved towards defensible island or shoreline locations. Both Honey Camp and Lamanai fit into this category (Masson, 1997). Although current evidence points to two distinct island and shore communities in this area, it has been discovered that during the Early Postclassic ceramics were made using less intensive labour techniques, and thus it is possible that these may have been overlooked in archaeological investigations. This would mean that any Postclassic population on the shore may be severely underrepresented in the archaeological record (Masson, 1995). 65 Archaeological evidence from the main island suggests intensive and elaborate landscape modification was employed to improve the island's contours to make it more suitable for habitation. This includes the development of a dock (Masson, 1996). This has given a fortified appearance to the island. It is unsure where the fill for all these structures came from, whether from the island itself or the lagoon shore. If the latter is the case then this would have created significant disturbance (Masson, 1996). During the Classic period large animals are scarce or absent from the forests in this area (Scott, 1980; 1982) and pollen records show extensive deforestation (Rice, 1986). Agricultural production appears to have been the mainstay of both communities in Honey Camp Lagoon (Masson, 1993). Evidence has been found for the cultivation of corn, beans, squash, cotton and cacao (Masson, 1999). The period from AD 1000 to 1400 appears to have been a time of growth and development without disruption. From AD 1250 onwards there appears to have been a change in the community structure with a greater emphasis being placed on ritual and community leadership (Masson, 2000a). From evidence of skeletal remains it is apparent that forest dwelling animals were present in this area in the 12th-14th century suggesting improved ecological conditions during this timeframe as compared to the Classic period (Masson, 1996). 3.3.4 Summary Although the level of knowledge concerning these sites is not as great as some of the major sites of the Southern Lowlands, the information presented provides a valuable context within which results concerning environmental change can be placed. The two sites appear to have responded in different ways to the AD 850 collapse period. Lamanai appears to have been stronger in terms of resources, leadership and connections, which may have contributed to its ability to survive. Both sites record different phases to their histories demonstrating that the impact/influence of the Maya was not consistent through time or space. Both indicate the importance of location i.e. use of lagoon and island and show clear human impact both in Mayan 66 and colonial times. With the knowledge gained from the archaeology it will enable a much greater appreciation of the palaeoenvironmental records produced from these lagoons. The differences in these sites will provide this study with a greater applicability throughout the country. 3.4 Mayan sites in northern Belize The sites which will be discussed in this section were chosen because each site succinctly demonstrates a key feature of either Mayan society or their interaction with the environments in which they inhabited. The focus is on sites from north Belize as this is the specific area of interest to this thesis. These sites are all marked on Figure 3.5. Cuello is a Preclassic site (Tourtellot et al., 1993) and occupation is thought to have taken place at this site before major architectural construction began (Hammond and Miksicek, 1981). It is apparent that by the end of the Archaic period the environment surrounding this site had already changed as a result of human manipulation. This trend continued into the Middle Preclassic with increasing levels of open woodland and cultivated land. Hammond (1991) calculated Cuello's population dynamics. It is thought that in the early Middle Preclassic 296-370 people occupied this site. This grew to 571-711 people in the late Middle Preclassic, rising to 2200-2600 in the Late Preclassic. Cuello at this point could be regarded as a small town with a clear social structure, rather than a village. The site reached its peak in the Early Classic with a population of 3400 people, falling to 1100 people by the Late Classic. It is therefore in the Late Preclassic that this site experienced the beginning of a rapid and massive rise in population, site size and density. It is believed that intensive agricultural techniques were also developed because evidence for channelled and raised fields has been found (Hammond and Miksicek, 1981). The animal bone remains at Cuello contain a high proportion of mammals suggesting that they were a food source for the population. This site demonstrates the typical pattern of both population growth and provides evidence as to how the Maya manipulated their environment. 67 Nohmul was a major Maya centre which was founded in the Early Preclassic. It has one of the longest occupation sequences in the region. During a time span of two millennia it was a successful community in terms of population size and architectural development. This site had a bimodal peak of population and monumental construction in the Late Preclassic and the Late Classic (Hammond, 1985). There was also a final burst of activity during the Terminal Classic/ Early Postclassic in the 10th Century. During the bulk of the Classic period, when most of the Maya centres were flourishing, Nohmul was abandoned. Channelled and raised fields are found dating to AD 700-1000 which coincides with the second florescence of Nohmul. This event was due to the reoccupation of the site by people from a non-local architectural tradition. The same pattern is demonstrated in Seibal, Altun Ha and Colha and has been attributed to the southward movement of the Itza people. This population was located at the site of Chichen Itza in the Yucatan Peninsula. The group which reached Nohmul were numerous and organised enough to build a series of public buildings inside the plaza of the Terminal Preclassic and the Early Classic ceremonial centre (Hammond, 1985). Hammond et al. (1987) regard Nohmul as a major population centre at a time when many other sites were failing. It perhaps served as a regional centre of government that flourished briefly before the collapse. Pulltrouser Swamp forms the eastern boundary of the Nohmul settlement and the earliest cultivation of fields in this area has been dated to the middle Late Preclassic. It has been estimated that only 20% of the population could have been supported by the fulltime cultivation of all the known field complexes around Nohmul (assuming that the staple crop was maize). It is, however, possible that some of the field systems could have been used for the growth of cacao for trade (Turner and Harrison, 1983). Evidence from Nohmul and Pulltrouser Swamp provide further evidence on the type and scale of agricultural activities conducted in Belize. The site demonstrates an alternative history of settlement to that found in Cuello. Colha is a Middle Preclassic to Post Classic site located on the edge of Cobweb Swamp (Jones, 1991). It was important as a production rather than ceremonial centre. There is evidence that occupation at this site goes back to Archaic or even 68 Pre Ceramic times, but there are no radiocarbon dates to substantiate this claim (Jones, 1991). The earliest permanent house structure dates back to 700 BC and lithics/ceramics of Middle Preclassic age are abundant suggesting that the site was permanently occupied by this time. Jacob (1995) concludes that the reticulated field features discovered along the margin of Cobweb Swamp represent human modification of natural channels and islands for agriculture. This is thought to have begun as early as the Middle Preclassic (800 BC). Disturbance of this site is however noted from 2500 BC onwards, involving large-scale forest clearance with maize and manioc cultivation. This is based on pollen evidence (Hester et al., 1981). The Late Preclassic shows a dramatic increase in population, which was matched by an intensification of forest clearance and agriculture. At this time at least 31 of the 89 known workshops were active. This signifies a huge production of tools, as each workshop would have been capable of producing 75,000 a year. Simplified tools were also produced which suggests that a mass production market was being catered for (Hammond, 1982). Tools produced by Colha in the Late Preclassic have been found as far away as Moho Cay (Hammond, 1982). The Late Classic (AD 600-850) saw a second florescence of Colha as an industrial centre (Hammond, 1982). Colha, however, no longer had the monopoly over the market, due to sites such as Altun Ha entering into tool production (Hammond, 1982). The cultural links between Colha and northern Belize were very strong in the Late Preclassic. These had weakened by the Late Classic as Colha became increasingly under the influence of Altun Ha to the south (Hammond, 1982). The final occupation of Colha appears to have come to a dramatic end with the execution of the ruling elite by the Itza invaders from the north. The archaeological evidence to support the northern origin of the new lords of Colha is consistent with other sites from the region (i.e. Nohmul and Altun Ha). The site of Colha was therefore a major industrial site of the Late Preclassic proving that the Preclassic Mayan civilisation was a firmly layered and highly organised hierarchy. By AD 1350 the site was abandoned (Jones, 1991). The site of Colha demonstrates the wide range of activities which the Maya undertook and consequently the wide range of impacts they would have had on their environment. 69 At Altun Ha there is evidence for contact with Central Mexico as early as the 2nd Century AD. At this time the population of Altun Ha increased by 50% and the area occupied by the site increased by 32% (Pendergast, 1971). Altun Ha was not a likely choice to be influenced by Teotihuacan because it is located near the Caribbean coast of Belize, and it is much further away from Teotihuacan, Central Mexico than other sites in Peten. This contact marks the beginning of a two-way flow of ideas between the Southern Lowlands and Mexico (Pendergast, 1971). The most significant part of Altun Ha's history was after its decline at the end of the Classic period. It is not thought that the site was completely devoid of population at this time and there is evidence that there might have been trade between a remnant population and Lamanai between the 12-14th century. Altun Ha was revived in the 15-16th Century and this population represents a break with the past and was not related to the previous group that inhabited this site. It is thought that the occupiers were pilgrims. Although Altun Ha was not a major centre of population it lay on important communication routes and its history emphasises the importance of the coastal zone for the Maya (Hammond, 1981). The influence of this site has been noted in a number of surrounding centres of population. This site provides an interesting point of comparison with Lamanai as it was believed to be of similar importance but it had quite a different history. A large tract of land in the northwest of Belize is owned by the conservation charity 'Programme for Belize'. From 1992-1994 the archaeology of this area was intensively surveyed (Adams and Valdez, 1993; Adams, 1994; Adams and Valdez, 1995). The major site in this area is La Milpa, but there are 18 other smaller sites. The larger sites are located on high ground surrounded by swamp with smaller sites being found on the periphery about 500 to 1000 metres from the swamp edges. These smaller sites played an integral role in the administration of intensive agricultural activities (Robichaux, 1995). This pattern is also seen at Tikal, Guatemala. The occupation of the area began around 900 BC with a few scattered pioneer families. The hiatus was a very clear event in this area and lasted for 120 years. The peak in 70 population levels occurred during the Late Classic. In more recent times this area was heavily utilised for logging purposes (Barnhart and Hargrove, 1995). La Milpa was a major Late Classic Lowland Maya site located mid way between Rio Azul (Guatemala) and Lamanai (Tourtellot et al., 1993). The ceramics which have been found, link La Milpa to sites in the Peten Lake District, Guatemala rather than to Nohmul or Lamanai. It appears that La Milpa was the northeastern limit of Peten regional culture. The site was a major regional power as early as the Late Preclassic and throughout the Classic. At its peak it is thought to have been the same size as Lamanai (Guderjan, 1991). It appears to have operated as a pilgrimage site in the Postclassic (Hammond and Bobo, 1993). The coast was a very important location for the Maya in terms of communication , trade and resources. Cerros is one of the most important coastal sites as was the island site of Ambergris Caye. Within the archaeology literature Cerros is regarded as a good example of a site whose major occupation was confined to the Late Preclassic. There are traces of earlier initial settlement and, after a hiatus in the Classic, a substantial Postclassic occupation. This pattern is repeated in Santa Rita, located across the bay (Hammond, 1981). The key feature of Cerros was its location at or near the mouth of the New River on Chetumal Bay, which would have made it a coastal lookout station. Cerros has major public architecture dating to the Late Preclassic period. It appears that during this time the community underwent a dramatic transformation from an egalitarian fishing and trading community to a cosmopolitan, political and hierarchical society (Freidel, 1986). There appears to have been a shift towards the coast in the settlements between the Middle and Late Preclassic accounting for Cerros' development as a major site at this time. The centre of this site is much larger than the one at Colha, suggesting that it was a significant trade centre, but, this is very hard to establish from archaeological data alone (Freidel, 1986). The relationship of Cerros to the coast in Late Preclassic times differed from the present situation. The shore of Corozal Bay has eroded through time cutting into the 71 nucleated village remains. The implication is that the New River mouth was closer to the site at the time of occupation, than its present position some 3.5Km to the south (Pohl, 1985). Friedel (1986) notes great similarities between Cerros and sites in Central Peten which suggests that there must have been communications between these areas. The occupants of Cerros would have had access to a variety of aquatic habitats. Modern residents of the area exploit streams, seasonally flooded savannas, fresh to brackish mangrove-lined lagoons and the sea. The terrestrial environment includes well-drained areas in which the climax vegetation is a deciduous tropical forest. The amount of vegetation that was present at the time of occupation would have depended on the extent of agricultural land clearing. Guderjan and Garber (1995) conclude that large-scale maritime trade was a Classic and Postclassic phenomenon with Late Preclassic Cerros being an interregional port (Garber, 1989) implying that trade may have been an important mechanism in the development of the Mayan civilisation. Classic period trade provided goods which reinforced the authority of the elite, but, gradually by the end of the Classic period these goods found their way to household level, which would have contributed to the disruption of elite control of society (Guderjan and Garber, 1995). Ambergris Caye was also important in terms of trade networks in Belize (Guderjan and Garber, 1995) (Figure 1.1). The occupation of Ambergris Caye began during the Late Preclassic while Cerros flourished as a trade centre. The Caye was permanently occupied by the Early Classic and formal architecture appeared by AD 600. Although the site was intensively occupied around the coast this did not extend to the interior of the island (Guderjan and Garber, 1995). Communities were centred either on natural or artificial harbours. There is evidence for Terminal Classic trade with the Yucatan Peninsula, Campeche, south and north Belize. By the end of the Terminal Classic (AD 1000), however, all the towns with the exception of Marco Gonzalez were abandoned. (Pendergast, 1990). 72 The site of Marco Gonzalez on Ambergris Caye has been found to have had great links with Lamanai from AD 1100-1300 (Graham and Pendergast, 1989). This may account for its ability to survive when nearby sites were in decline. It is thought that the New River would have served as the principal means of communication between the two sites. The amount of 'Lamanai inspired' pottery in Marco Gonzalez in the Postclassic suggests that this was the site of a pottery works. The tie between Lamanai and Marco Gonzalez does not appear to have extended to the intervening sites along the New River suggesting that these sites did not flourish during the Postclassic (Graham and Pendergast, 1989). This is surprising because the island would have been a much more fragile environment than the inland sites (Graham and Pendergast, 1989). Albion Island had its time of maximum population in the Early Classic (Pyburn et al., 1998). The occupation of this site began in the Late Preclassic and the population density has been calculated as 775 persons/km2 (Dahlin, 1977). This is very high and the houses, which have been investigated, are small and densely clustered. This suggests that it was not extended families that resided here, but perhaps that it was the site of a large-scale commercial farming enterprise. This arrangement highlights the diversity of tactics and survival mechanisms that were put into place in the Maya Lowlands (Pyburn et al., 1998). The site is thought to have been within the sphere of influence of the Peten region, with trade taking place between agricultural goods from Albion Island and urban goods from Guatemala. The farming methods employed in Albion Island show a response to environmental change through two main phases of ditching. Firstly in the Late Preclassic / Early Classic in response to rising sea level (which is also seen in the Colha records) and secondly in the Terminal Classic/Early Postclassic after water levels had stabilised/receded and the floodplain of the Rio Hondo had silted up (Pyburn et al., 1998). Albion Island is the key example of organised agriculture on a wide scale. This highlights not only the enormous environmental impact that this would have had but also it portrays a clear picture of the manner in which Mayan society was conducted through elite control of the masses. 73 Ford and Fedick (1992) have studied the Upper Belize River area. This area is thought to have been one of the first to be occupied within the Lowlands as a geographical link between the Caribbean and the interior of Belize. It is also thought to have served as a corridor that allowed the population expansion at Tikal, Guatemala (Rice, 1976). The sites located in this area are examples of the rural population in Belize and illustrate not all populations were centred around large monumental sites. Excavations at Barton Ramie have demonstrated that this site was occupied throughout the course of prehistory including during the decline of the Classic Maya (Gifford et al., 1976). Pactibun was occupied from 900 BC to AD 900 (White et al., 1993). This site was first settled as a small farming community in the Middle Preclassic and by the Late Preclassic the construction of monuments had begun suggesting that the site had attained a degree of regional prominence. Increased prosperity is noted in the Early Classic with the discovery of exotic trade goods and the production of carved monuments. Cultural development reached a peak in the Late Classic (AD 550-700) and it was during the transition from Late to Terminal Classic that the site's population reached its climax being four times higher than in the Early Classic (White et al., 1993). It was during the Terminal Classic that the majority of agricultural terraces were constructed. The site was abandoned by AD 900 (Culbert, 1973). The site of Pactibun, would have been extremely reliant on maize as it is located inland and did not have the advantage of a location near a prominent watercourse. Maize consumption reached a peak during the Late Classic and then declined after this phase of maximum population and agricultural intensification. The dietary shift to low amounts of maize just prior to site abandonment suggests that one of the reasons behind this event was over exploitation of the environment due to agricultural intensification (White et al., 1993). 74 3.4.1 Summary Through the investigation of the main sites in lowland Belize it has become apparent that a number of different trends have operated through the country's history. After the initial sites were set up (Cuello and Nohmul) in the Early Preclassic the main phase of site development was in the Late Preclassic. Most sites reached their peak in the Early Classic confirming the general trend of the Classic period being the most prosperous for the Maya. Nohmul and Colha are two sites which did not follow this pattern, having a bimodal pattern in the Late Preclassic and Late Classic. The timing of site abandonment varies through the country with records being confused by the movement of Itza people into the region. It has become apparent is that sites were either genuinely abandoned or in decline at the time of the collapse highlighting that this was an episode of some severity. Through investigating the sites it has become obvious that there were an array of strategies which were employed through time, a number of different ways in which the environment of Belize was influenced by the Mayan civilisation and a number of different histories to the occupation patterns for the sites. These examples highlight both the common and divergent trends that are apparent in Belize. 3.5 The collapse Evidence for collapse has been shown in a number, but not all of the Mayan sites in north Belize. Explanations for the collapse can be broadly divided in terms of whether the driving forces were internal; external; social or natural processes. Up until 1967 the collapse was explained in terms of internal processes only and was centred on agriculture (Morely, 1920; Sabloff and Wiley, 1967). It was believed that maize agriculture was destructive to soil resulting in the creation of man made savannas. Sanders (1962, 1963) came to the same conclusion that the overuse of the shifting cultivation system lead to agricultural and, therefore, general cultural collapse. Ricketson and Ricketson (1937) developed early ideas on soil exhaustion. They found evidence that present day swamps were once shallow lakes which have 75 been filled up with eroded soil from agriculture. This suggested to them that the agricultural system, which was being operated, was not a sustainable one. Further studies have since challenged these ideas e.g. Cowgill (1961) showed that the shifting cultivation system is actually highly adaptive and efficient in specific tropical forest areas. These theories all assume that maize was the main crop within a swidden agriculture system in a similar manner to modern Mayan agriculture systems. Many of the ideas surrounding the collapse seem to question why indeed the Mayan civilisation evolved in what appears to have been a very unsustainable environment. Meggers (1954) developed the idea that the Maya lowland jungles by their very nature place a limit on cultural development, especially through agricultural constraints. She believed that the civilisation must therefore have had its origins elsewhere and hence why it collapsed. Coe (1957) refuted this theory due to evidence of in situ development of the Mayan people. The theory of Mayan development instigated by Meggers (1954) follows the concept of environmental determinism i.e. that the agricultural potential of an area limits the level of cultural development that can take place. The Mayans were not the only society to evolve in a harsh environment for example the Chiripa (1500-200 BC) and Tiwanaku cultures (400 BC to AD 1100) developed in very hostile environments in the Bolivian-Peruvian Altiplano. These examples defy environmental determinism (Brenner et al., 2001). Thompson (1954,1966) took the primary cause of the collapse away from agricultural failure. He believed that it three main events triggered violent internal dissension which destroyed the social system: 1. The presence of economic motivations for peasant revolt due to an increase in onerous tribute burdens. 2. The introduction of central Mexican mercenaries for the purposes of social control. 3. Moral decay resulting from the associated and introduced ideological shifts. 76 Altschuler (1958); Rathje (1971) and Adams (1973) all developed this idea that social structure theories should be included as part of multi factor explanations. The need for households to obtain their basic needs creates a bridge between ecology and socio-political organisation. The key problem with these social theories is that they do not account for the depopulation, which occurred at the time of the collapse. Sabloff and Wiley (1967) believed that non-Classic Mayan peoples invaded the Southern Lowlands in the 9th Century building on the invasion theories which were first postulated by Cowgill (1964). The human-nature balance in the Southern Lowlands is fragile and thus the invasion could have pushed the system above a key threshold. All evidence points to Classic level populations being very high and thus it is likely that the strain of over population would have disrupted the system allowing for an invasion to completely upset the balance. A month or two of widespread neglect of the maize fields in a crucial season could have been enough to cause famine among the large Late Classic population, prompting people to move north (Culbert, 1973). Coe (1966) believes, however, that invasion could only have had such a profound affect on the Mayan people if their society had already deteriorated markedly. As discussed in the site histories of northern Belize evidence has been found for architectural links between the sites of Chichen-Itza, Mexico and Nohmul, Belize (Chase and Chase, 1982) and Colha, Belize (Jones, 1991). This is very important evidence for the involvement of an outside force as an integral part of the collapse. These northern sites could be where resettlement of the remnants of the Classic period people occurred which goes some way to explaining the depopulation of the Southern Lowlands at this time. The site of Seibal, Guatemala also experienced invasion in the terminal part of its history but this was of a different nature implying that it was caused by different invaders to those that arrived in Belize (Sabloff and Wiley, 1967). It is thought that the invaders in Seibal were from the from the Puuc and Gulf Coast, western Yucatan Peninsula areas rather than from Chichen-Itza. This implies that there were two separate and contemporary events in the Southern Lowlands, which ultimately had the same impact. 77 The collapse was not the only time that the civilisation was put under enormous strain. During the latter half of the sixth century (AD 534-593) the Classic Maya of the Lowlands underwent a marked slowing down. This period is known as the hiatus and is marked by the decrease in stelae (carved sculpture) production. The hiatus was a time of cultural re-orientation dividing the Early Classic from the Late Classic (Willey, 1974, Culbert, 1991). How do the hiatus and the collapse compare? During the hiatus stelae production stopped, but, in the collapse all major architectural activities ceased. Both these point to a weakening in the central socio-political structure of the culture. Although population levels declined during the hiatus, these cannot be compared to the levels of decline seen in the collapse (Willey, 1974). Morley (1938) believed that the hiatus was a time when the major centres were devoting their energies to building up the provinces; this is, however, unlikely. The hiatus shows that although there were inherent weaknesses in the society it was also an adaptable system because the civilisation was able to thrive after the event (Willey, 1974). Rathje (1971) developed a theory by drawing on the experiences of the hiatus. He believed that the core area of northeast Peten was lacking in the natural products that are needed to support large complex societies i.e. hard stone for grinding corn, obsidian and salt. This therefore led to the development of long distance trade. The gathering of these resources requires organisation, which must inevitably lead to the development of a hierarchical society. Rathje named the area around the core as the buffer zone and he believed that this area served an intermediate relaying role, supplying raw materials to the core in return for luxury goods. Rathje hypothesised that with increasing sophistication and the profit derived from their more favourable geographical location, the buffer zones would have usurped the trade control from the core area precipitating the downfall of its cities. This theory is backed up by the fact that some sites located in the buffer zone have stelae which date to the hiatus time period e.g. Tonina (Chiapas) and Tulum (Yucatan Peninsula). This is not true for all buffer sites though. A further complication is that core areas did thrive after the hiatus and therefore they were not totally out-competed by sites in the buffer 78 zone. Sharer (1973) also refutes Rathje's theory because many of the same resources are actually found in the core and buffer areas. At the time of the collapse both buffer and core areas were affected equally. Willey (in Sharer, 1973) modified Rathje's theory by postulating that a new outer buffer zone developed from Non Classic or Mexican Maya states along the western edge of the Southern Lowlands. This would have severely diminished trade for both the buffer and core areas. (Sharer, 1973). Webb (1964, 1973) proposed that by the Late Classic other Mesoamerican societies such as those in the Yucatan were developing in much more favourable environments i.e. rich in resources, near to new land/sea trade routes and thus they continued to develop their populations and organisational complexity. The Lowland Maya were relatively isolated from these events and being unable to compete their society collapsed. The Lowland Maya society at conception was therefore very different from the society at the end. Adams (1973) cites Spinden (1928) "there is good reason for believing that the sudden appearance of yellow fever may have had a part in the catastrophe." It is well known that yellow fever, malaria, syphilis and smallpox were principal disease factors in the biological catastrophes that overtook the native new world populations between the 16th-18th Centuries. Disease may have had an influence in the collapse, because it would have weakened the community. Such a chain of events could be set into action through malnutrition. Haviland (1967) showed that skeletal material from Tikal had the poorest levels of nutrition at the end of the Classic period. This evidence of decline spans over a 600 year time period. Cowgill and Hutchinson (1963) also found evidence for skewed sex ratios in favour of men. These results have been supported by research by Dunn (1968) who suggests that complex ecosystems such as tropical rainforests support more disease vectors than simpler environments. Sheets (2000) looked at the relationship between environmental stress (i.e. volcanic eruptions) and the ability for a society to survive. It became apparent through this study that complex societies are less able than simple societies to evolve and adapt to 79 change. This can be related to the Mayan collapse because at that time the society had a complex political structure, a large investment in architecture and sedentary agriculture. 3.5.1 Summary Many of the theories presented are a result of work which was conducted many years ago within research paradigms that are now out-moded, but, these ideas are still being drawn upon in discussions of the Maya. The most recent research acknowledges multi-factor explanations which can be manifested in different ways in different sites. Between AD 300-800 the ancient Maya developed a complex hierarchical society in an environment which is highly susceptible to degradation. Santley et al. (1986) believe that it is the very success of the Maya in developing within this environment which lead to their collapse. The system was locked in a downward spiral with the need to sustain progressively larger populations placing severe constraints on the food production base, which precluded further development and ultimately lead to the collapse. These have been summarised into three key reasons behind the collapse: 1. Demographic instability. The population levels in the Classic period rose to very large numbers. The question is whether it was population instability i.e. the exponential rate of growth rather than the absolute numbers that lead to the collapse. 2. Agricultural failure (i.e. farmers making rational short-term decisions that were dysfunctional on a long-term basis). 3. The absence of regional resource links (e.g. trade networks). It has become apparent that several factors played a role in the collapse which all resulted in a weakening of the system. Overpopulation, excessive urbanisation, soil degradation and food security problems on one side with social conflicts and a disintegration of the political structure on the other (Sabloff, 1991). Evidence 80 suggests that the society had reached a threshold which acted as a trigger to the collapse of the whole cultural-demographic system. (Messerli et al. 2000). The collapse needs to be viewed in terms of the general organisation of the Mayan civilisation and its pattern of change and decline. Subsistence patterns developed from shifting to settled agriculture; increased diversification; heavier yielding crops; dependence on imported food crops and finally to a tighter control on land use and trade. Responses to agricultural stress included a movement away from milpa/maize agriculture to intensive agriculture based on ramon, root crops and kitchen gardens. Cultivated areas began to be developed in swampy areas around sites. The creation of cities allowed more land to be used for cultivation. Ridged fields in Belize are associated with low population densities. It is thought that these were plantations that were exploited to produce food for cities in the Southern Lowlands (e.g. Tikal). Many of the new and expanded forms of agriculture were more vulnerable to climatic shifts, natural disasters, disease, pest problems and soil exhaustion. A further source of vulnerability was the dependence of these new systems on long distance trade networks. The Maya held deeply seated beliefs in the supernatural believing that the problems which the society suffered were related to the anger of the Gods. It is possible that a great deal of agricultural labour could have been redirected towards ritual that would have been counter productive. Such a fragile system would have been easy to invade. Population densities were rising; there was a decline in per capita income; increasing local specialisation in crops; heavier reliance of the core area on the periphery for basic materials and more highly organised trade. The hierarchy of the system changed from tribal all the way up to four level states (Sanders, 1973). These developments link ecological and social theory which is an important step to making a coherent theory concerning the collapse. 3.6 The relationship between the Maya and climate One of the key aims of this thesis is to determine whether climatic changes can be isolated at the time of the collapse of the Mayan civilisation. The traditional view in the literature is that humans had more impact on the environments of the past than 81 climate (Cowgill, 1964; Lambert and Arnason, 1978; Rue, 1987; Vaughan et al., 1985). The concept that climate could be involved in the collapse of the Mayan civilisation first appeared in the literature in Gunn and Adams (1981) and was built upon by Dahlin, (1983); Folan et al. (1983); Messenger (1990); Gill (1994) and Curtis et al. (1996). Such ideas had been directly dismissed in the past, for example: "...the theory that marked changes in climate have had pronounced effects upon the Mayan civilisation in Yucatan...(has been)...completely discarded by many climatologists because of the absence of direct information and it is doubtful whether there is any satisfactory evidence indicating that important climatic changes have taken place in the Yucatan Peninsula subsequent to the fifth century AD..." (Page (1933) cited from Dahlin, 1983). In order to reach a conclusion on this topic, the magnitude and extent of the climate and cultural change at the time of the collapse therefore needs to be ascertained. This can then be used to determine whether or not the history of the Mayan people is linked to wider changes in the natural world. The uncertainty lies with the establishment of direct causal factors that can be proven to have influenced the Maya through time. A key line of research that is being pursued is the idea that climatic changes occurred at that time and were involved in the collapse. What is known is that the end of the Classic period represents a key time of change in the Mayan civilisation. It has also been established that a number of different factors may have been involved in this change. 3.6.1 Documented climate change The records highlighted in Chapter 2 demonstrate that Belize is situated in a climatically diverse area and therefore the manifestations of large-scale climate changes will be different from region to region. The record produced by Curtis et al. (1996) for Punta Laguns, Yucatan Peninsula is the most high resolution sequence that has been published in this area. Curtis et al. found that the beginning of the Classic period was marked by a shift to relatively drier conditions about 250 cal year 82 AD. This raises the question as to whether the Classic Maya culture evolved in response to the climatic drying or whether climate changed because of the environmental impact of the expanding Maya population i.e. forest clearance may have reduced rainfall (Lean and Warrilow, 1989). The record from Punta Laguna, Yucatan Peninsula suggests that the Maya hiatus coincided with the first major dry phase at 585 cal year AD. The next two hundred years saw the civilisation flourish during a moist interval in between two prominent droughts at 585 and 862 cal year AD. This latter event coincides with the collapse (Curtis et al., 1996). 3.6.2 The relationship How much are humans influenced by the environment in which they are situated? Is it just coincidence that times of climatic change have occurred at the same time as cultural demise or florescence? The importance of climate and weather are apparent in Mayan culture with there being much evidence that they predicted weather as part of their activities (Folan and Hyde, 1985). Amongst sites in the northern Mayan Lowlands there are many monuments to the rain god Chac demonstrating the importance of rain in this society. It also suggests that the society may have been surviving in a marginal environment. This is also demonstrated in the modern day precipitation values (Figure 1.6). If a dry phase affected Belize it is important to determine how the environments of Belize would have changed during this phase. Lakes and river networks, which were relied upon to be part of the transport network and a source of fish, may have dried up or become impassable. Water would no longer have been available for the agricultural systems (Gunn and Adams, 1981). Classic sites were built on hills and had man-made water management systems. These would not have worked with low levels of water throughout the year and therefore the sites which had the best chances of survival were those either near water or at the base of a watershed. The effects of such changes would not have been uniform and for societies adapted to the drier conditions of the northern Yucatan it was an environment which they would have readily been able to adapt to. This therefore provides an indication as to how the 83 population from Chichen-Itza would have been able to invade the Southern Lowlands (Folan and Hyde, 1985). Correlations between human society and climate have been discovered all over the world. The present day climatic regime was established in low latitudes during the mid-Holocene c. 4070-3070 BC just before the construction of the pyramids in Egypt and the emergence of several old world civilisations (Roberts, 1996). The time period between AD 900-1250 is known as the Medieval warm period and evidence for this is widespread outside Europe (Grove, 1996). This time of climate change enabled the colonisation of Greenland by the Vikings (Roberts, 1996). A general drying of Caribbean climates occurred between (3200-1500 BP) being noted in Haiti (Hodell et al., 1991) and subjectively in Puerto Rico (Burney et al., 1994) the main implication of this is thought to have been the delay of human colonisation in the Bahamas. Although evidence is strong for the existence of the extremely dry period at the time of the collapse, it is very important that its geographical extent is established. This will provide an index of its severity. As discussed in Chapter 2 evidence for this period has been found throughout Central America (e.g. Horn and Sanford, 1992; Metcalfe et al., 1994; Metcalfe, 1995; Curtis et al., 1996; Davies, 2000). The Inca of the central Peruvian Andes were affected by the AD 900 arid event (Chepstow-Lusty et al., 1996) and the breakdown of the Tiwanaku culture in the Titicaca basin coincided with the onset of a prolonged dry period which began AD 1100 (Binford et al., 1997). Climate is, however, only one of the variables which affect human societies with its influence being strongest in those areas which are marginal for agriculture. The interplay is very much dependent on the adaptability of the society involved (Grove, 1996). To assess the true implication of climate change it therefore needs to be placed within the context of the human societies which would have been directly influenced by it. 84 3.6.3 Summary Through the years a great deal of information has been amassed concerning the Mayan Civilisation, however, "...further study is needed to determine the full magnitude and geographical extent of the dry period that occurred between AD 800- 1000, and to explain the intraregional pattern of the collapse of the Mayan civilisation" (Hodell et al. 1995). This study in Belize will go towards these ideas providing a valuable extension to our knowledge of the dynamics of Central America. Evidence suggests that some of the key cities that were part of the Mayan civilisation had reached a critical point whereby the impact of a severe drought could have acted as a trigger, which lead to the collapse of the whole cultural system. It is therefore the combination of a whole set of attributes in the natural and human systems, which resulted in such severe changes. The concept of time scales is important because, for example, human degradation of the landscape acts over a long time period whereas climate change can be very short lived. Even if the drying was regional in its extent the ecological and anthropological effects would have varied spatially depending on the magnitude of sub regional climate change and the sensitivity of the natural and cultural systems to environmental change (Hodell et al. 1995). 3.7 Conclusion Even in the Preclassic, Mayan society was complex and sites were linked over great distances (Culbert, 1991). Populations developed rapidly which required the instigation of more intensive forms of agriculture and the production of huge amounts of stone tools in sites such as Colha. By the Early Classic this had extended to contacts in Mexico with the Teotihuacan population. Evidence for this is seen in Altun Ha. Near the end of the Early Classic there was a period of decline which Willey (1974) termed the hiatus. This period of cultural disruption did not occur evenly throughout the Lowlands with the east being least affected. It did, however, have devastating affects with civil wars and revolts and it was not until AD 650 that 85 the Mayan Civilisation began to function again (Adams, 1991). The Late Classic was the peak period of Mayan population and this was also the time of most elite interaction. By this time every major swamp, watercourse and lake edge in the area was being exploited (Adams, 1991). From the 8th Century the population of the Maya Lowlands was estimated to be 9-14 million (Adams, 1991). The Late Classic was a time of changing political conditions and after AD 830 there was a burst of sites that began to construct monuments close to major sites. This suggests that there were groups of elite who were taking advantage of disintegrating political situations (Culbert, 1991). The expansion of all parts of the Mayan society made it vulnerable to many internal and external stresses and ultimately to its collapse (Adams, 1991). The forces of the collapse were so great that they hindered the ability of ordinary people to survive. This implies that households were dependent on one another for their everyday needs (Blanton et al., 1993). The analysis of the palaeoenvironmental characteristics of a region and the human ecology of a prehistoric culture is complex. While the natural components can be determined, human perception and reactions to environmental change are harder to assess. The history of the Mayan civilisation in Belize highlights the importance of location (in terms of agricultural potential, defence and trade) and the power dynamics which can be developed. There has been a great deal of variability between sites due to their differing abilities to deal with changing fortunes through time. This was most apparent during the Classic period. If the climate was changing at the same time as this, the effect would be to compound the hardship that was felt by communities and collapse therefore becomes inevitable. "Studies in palaeoenvironment are much needed to understand the Classic to Postclassic transition in northern Belize. Sediment cores and pollen samples would certainly be able to provide evidence concerning the abandonment of this region, the rate of reforestation and the degree to which the area was under cultivation from AD 850-1300" Masson (1993) 86 Figure 3.1 Lamanai (modified from Pendergast, 1981) Figure 3.2 Lamanai temples Figure 3.3 This photograph illustrates the difference between the west and east banks of the New River Lagoon. ■«iamy Doubloon Bank Lagoon r v / v-y c V V v V v -Y v V ■ V. V v. v V V V Chiwa Lagoon 9 km to El Cacao Honey Camp Lagoon (Laguna de On) © © & 7 4 km to Kichpanha V V BAH *91 V V vy V V 0 1 ess Kilometers N Marsh and Swamp Bodv of water ^ Postciassic Archaeological Site Classic Archaeological Site Figure 3.4 Honey Camp Lagoon (from Masson and Rosenwig, 1996) Lagoon Button Lagoon Chiwa Q Lagoon ^Honey Camp Lagoon Doubloon Bank Lagoon Cobweb ^Si^amp^Kate's Lagoon Crooked Tree Wildlife Sanctuary Altun Ha"Small CrocLagoon"/Rio Bravo Conservation Ar$a LAMAmi New River (Lagoon Laguna Sec Botes Lagoon Lemonal Harry Jones I Creek Laguna Nrardi Fabers Lagoon Almond Hill [NorthedtLagoonl Wagner Lagoon' Belize River Nohmul-V 20 km CUELLO COLHA LA MILPA Belize City Caribbean Sea Figure 3.5 Location map of northern Belize showing the places referred to in this thesis. Chapter Four: Techniques and Methodology 4.1 Introduction The aim of this chapter is to describe and evaluate the methodology employed in this investigation. The two principal methods (diatom and stable isotope analyses) are discussed in detail focussing on the background behind the techniques, the development of knowledge and ideas and the specific applications to this project. Additional data were collected in a parallel project to this thesis. The results have been used to supplement this study where appropriate. The methodologies of the additional techniques are explained in this chapter. The justification of site selection and methodology are also explained as these are important steps to understanding the essence of an investigation. 4.2 Rationale This thesis has a key set of aims which were explained in Chapter 1. The following section provides the justification for the overall choice of research approach and the two main methods that were used. Palaeolimnology is the study of past lake environments. Through the responses of a lake, information can be gathered about the external forcing factors such as climate. Lakes can be an ideal tool for this purpose, because some respond in a very sensitive manner to change and this is registered in the sediments. The responses of lakes are particular to the lake type, the climatic setting and the extent of human modification to the system. In particular it is the hydrological characteristics of a lake which determine its sensitivity to changes in climate. This has been investigated by Street-Perrott and Harrison (1985) who concluded that closed amplifier lakes were the most responsive to change. These are systems where runoff provides nearly all of the inputs to the system and there are no outlet streams. Past environments can be reconstructed by analysing the biological and geochemical elements which have been preserved in the lake sediments. These proxies are controlled by a number of different factors, which need to be understood before meaningful conclusions can be drawn (Curtis et al. 1998). Investigations rely 92 on the assumption that present day conditions and associations can be used to interpret the past. One of the key variables to affect environments is climate change. This influences lakes in many different ways. The key changes are those which affect radiation and water balance. Radiation controls light and temperature which in turn affects the depth, duration and intensity of water column stratification. This influences many chemical and biological processes such as pH and nutrient cycling. Changes in water balance will affect lake-levels and residence times which in turn influences the structure of marginal habitats, the distribution of sediment and the ionic composition of the water body. Residence time changes will affect the availability of nutrients. All these variations are compounded by the affects that climate change have on the lake catchment as these will feed into the lake system through changes in vegetation and weathering regimes (Battarbee, 2000). The choice of methodology is crucial to the success of the reconstruction. Diatoms and stable isotopes were chosen in this investigation for very specific reasons. Diatoms respond to changes in their environment. These changes can be due to both humans and climate change. The differences between these two forcing factors are often very difficult to unravel (Deevey, 1978; Deevey et al., 1979; Vaughan et al., 1985; Wiseman, 1985; Rice, 1986; Wiseman, 1990; Hansen, 1990; Bradbury et al., 1990; Leyden et al., 1996; Whitmore et al., 1996; Islebe et al., 1996a; Dunning et al., 1998). Through previous diatom work in Belize (Breen, 1998) and the results of the first field season of this investigation, it was found that diatom preservation could not be relied upon to be consistent through space or time. A second technique therefore had to be employed to ensure that complete records from each site could be obtained. Stable isotopes were deemed to be ideal due to the high percentage of calcium carbonate in the sediment and secondly because they respond to different forcing factors from diatoms. This proxy has the potential to considerably widen the knowledge gained from the sediment cores analysed. Oxygen isotopes respond solely to climate while carbon isotopes provide additional information on the productivity of the lake (e.g. Curtis et al., 1998). 93 4.3 Field methodology The basis for this study was the collection of core material, modern diatom samples and water chemistry data during two field seasons (1999 and 2000). A total of five different water-bodies were cored and 31 sites were sampled over a variety of habitats for modern diatoms and water chemistry. 4.3.1 Coring The sampling strategy for each field season was different. The aim of the first visit was to collect cores from a wide range of environments to enable a much more targeted programme during the second trip after the analysis of this material. The criteria for selecting the sites to return to in the second field season was primarily based upon whether diatoms were preserved or not in the sediment sequences. This applied to Hillbank and Lamanai in the New River Lagoon. Preliminary stable isotope measurements were taken at this stage and the record from Honey Camp Lagoon demonstrated a high degree of variability. This suggested the potential for a highly sensitive record of environmental change. Consequently Hillbank, Lamanai and Honey Camp Lagoon were selected for further investigation in the second field season. A range of corers were employed to enable the successful collection of material from each site. A percussion corer was used on the shore of Honey Camp Lagoon (1999), Booth River and Aguacaliente Swamp (Figure 1.1). These were the 'solid' ground sites. In order to operate the coring device, a small motor was placed on top of the core barrel, which pushed the barrel into the sediment. A one metre core barrel was used with a plastic tube of the same length placed inside. This enabled the samples to be removed intact ready for transportation back to Edinburgh (Figure 4.1). A Livingstone corer was used to retrieve long records from the lake systems. This corer comprises a one metre core barrel which had a one metre square rod inside it with a piston at the base attached to a length of rope. This was used in Hillbank, 94 Lamanai, Honey Camp Lagoon (Figures 1.1 and 4.1). A Kullenberg corer was used to retrieve shorter cores with the sediment/water interface intact. This was extremely important because the top sediments in the New River Lagoon were extremely unconsolidated. The coring system could be adjusted to capture more or less of the sediment/water interface. Cores were also retrieved from Honey Camp Lagoon and Laguna Verde (Figures 1.1 and 4.2). On arrival in Edinburgh all the core tubes were X-rayed intact at the British Geological Survey, Edinburgh. They were then cut in half lengthways using a device designed by Dr Antony Newton and Dr Malcolm Murray, University of Edinburgh. The sediment stratigraphies were then carefully noted, colour descriptions were taken from a Munsell chart and photographs were taken of the cores to form a permanent record. Half the core was then wrapped intact while the other half was cut into 1cm sections and stored in petri dishes. Both were stored in the cold store. The Livingstone core collected from Hillbank (Hillbank 1998) was collected by a team from the Natural History Museum, London. This core had only a very preliminary stratigraphy taken and unfortunately some of the one metre sections of the core have gone missing. It is the general consensus in the literature that if a single core is to be taken then it should be from the middle of the lake where sediment disturbance is minimal and where sediment depth is likely to be at a maximum. This latter quality will increase the resolution of the record ensuring that a detailed record of change can be obtained. Subtle changes in lake volume may be amplified in marginal lake cores, although hiatuses in sedimentation are more likely (Lamb, 2000). For the two Lagoons studied in this investigation over both field seasons, cores were taken from both these locations in order to enable inferences to be made concerning whole lake dynamics. This is especially important in the New River Lagoon as it is such a large system. 95 4.4 Diatoms: The diatom cell - general structure and development Diatoms are unicellular algae, which belong to the class Bacillariophyta. They are pigmented and photosynthesise. The key characteristic of diatoms is their cell wall which is highly ornamented and heavily silicified. The wall consists of two valves which are linked by girdle elements. All diatoms secrete polysaccharides which either diffuse into the medium surrounding the cell; form a capsule around the cell or create threads, pads or stalks for attachment. The two valves and the siliceous parts of the cell wall are collectively known as the frustule (Round et al., 1990). Each frustule has one valve, which was formed just after the last cell division, and an older valve which may have existed for several cell cycles. The older valve, together with its girdle elements, are known as the epitheca and the newer valve and its girdle elements are known as the hypotheca (Round et al., 1990). New parts of the wall are formed within protoplasts and then added to the wall by a form of exocytosis (whereby matter is released by the living cell). The epitheca overlaps the hypotheca and cell growth occurs in one direction, forcing the epitheca and hypotheca apart (Round et al., 1990). The production of new frustule components within the parental cell results in the decline of the mean cell size. Size is restored via an auxospore, which is a special cell that expands in a highly controlled way before producing a new frustule. Auxospore formation is associated with sexual reproduction (Round et al., 1990). These processes result in the size of diatoms within one species being highly variable. 4.4.1 Diagnostic features of the diatom cell There are two general forms that diatoms take: centric and pennate. Pennate diatoms are characterised by one or two longitudinal slits which run through the valve. These slits are known as the raphe, a feature which is intimately associated with movement. Some pennate diatoms are araphid and they have a silicified rib (which does not contain punctae) called a pseudoraphe. Pennate diatoms are bilaterally symmetrical and centric diatoms are radially symmetrical. Apart from these three major 96 divisions, it is the shape, size and valve ornamentation of the cell that are key features in the identification of diatoms. In terms of valve ornamentation it is the pores or punctae that are important. These can either be arranged linearly to form striae or they can form concentric or sector arrangements. 4.4.2 Ecology Diatom species have specific ecological requirements. This means that particular species can be regarded as diagnostic for different types of environmental conditions. This is not a universal concept as some species are cosmopolitan being found in a wide range of environments. Due to their short cell cycle and ability to immigrate diatoms are able to respond quickly to environmental change. Factors which they are sensitive to include: habitat, water depth, nutrient availability and water chemistry. In general terms, centric diatoms are planktonic and pennate diatoms are benthic. Planktonic diatoms are free floating and due to the density of their frustule will often sink in the water column. Diatoms have a variety of mechanisms to counteract this tendency and as water bodies are rarely completely still, species are kept in suspension and within the photic zone. Benthic diatoms are associated with other substrates such as vegetation (epiphyton), rocks (epilithon), sand (episammon) and sediments (epipelon). Some species are even found in sub-aerial habitats such as in soils or damp rock faces and in moving waters (Round et al., 1990). Metcalfe (1988) highlighted the importance of habitat in distinguishing between diatom populations that were from different sites with similar water chemistries. Shifts in species that can be related to habitat changes may indicate changes in the lake's morphology e.g. a shift from planktonic species to epiphytic species may indicate a drop in lake levels. Water depth is important to diatoms as they are photosynthetic and therefore need to reside within the photic zone. Species that prefer abundant light are found in the plankton or shallow littoral zones (Patrick, 1977). The turbidity of the water body will affect this variable. Such sensitivities in the diatom record have the potential to provide an excellent addition to the record produced from the 5lsO record. As water 97 levels fluctuate (due to wider climatic changes) corresponding changes occur in habitat, light, stratification, mixing and chemical conditions. This will therefore affect the diatom species that are present (Wolin and Duthie, 1999). The chemical composition of lake waters is of key importance in influencing the diatom species that are present. Gasse et al. (1995) found that it is not only ionic strength (conductivity) that diatoms respond to but also ionic composition. This allows the distinction to be made between for example carbonate/bicarbonate and chloride/sulphate systems. This therefore creates the potential to enable the reconstruction of brine evolution as waters follow particular geochemical pathways in response to evaporative concentration (Eugster and Hardie, 1970). It can be difficult to specify one factor that has the overriding influence over a diatom population. For example Haberyan et al. (1997) found that magnesium concentration was very influential in determining the distribution of diatom species found in Costa Rican lakes. This element could not, however, be singled out as the most important variable because cation concentration, hardness, pH and temperature were also isolated as influential factors. Podzorski (1984) studied the diatom communities in Broad River, Western Jamaica. Many factors appeared to have an influence on community structure but again no single factor appeared to have overriding influence. Oxygen, Silica and P-PO4, organic nitrogen and mainstream flow-rate were deemed important. This highlights the significance of investigating a wide variety of variables in the modern day environment so that systems can be fully characterised. Different diatoms have particular nutrient requirements. This is especially pertinent in nutrient enriched environments where specific changes to species assemblages can be meaningfully related to nutrient changes (e.g. Bennion, 1994; Cooper et al., 1999). It has been determined that phosphorus, nitrogen and silicon have a key influence on phytoplankton community structure and biomass. Silica utilisation by diatoms is high and diatoms can deplete natural levels to very low concentrations. Kilham (1971) postulates that diatoms need at least 0.5mg/l of silica and if levels 98 drop below this, other forms of algae will replace diatoms in the ecosystem. The ratio between phosphorus and silica is also a variable which species can be differentiated by (e.g. Cyclotella meneghiana is dominant at low levels and Fragilaria and Synedra species are dominant at higher ratios). Diatoms are rare if Si:P ratios are too low (Tilman et al., 1982). Such changes may be part of a seasonal succession which lakes follow. Due to the ecological preferences of diatoms, workers began to classify species by variables under very general headings e.g. temperature (Hustedt, 1956); pH (Hustedt 1937-39) and salinity (Kolbe 1927). In modern studies this has been done in a more quantitative manner through the development of transfer functions (e.g. Birks et al., 1990; Fritz, 1990; Gasse et al., 1995; Reed 1998a; Davies et al., in press). Information from such studies can then be used to infer climatic and other changes (e.g. Fritz et al., 1991; Laird et al., 1996; Gasse et al., 1997; Laird et al., 1998). The changes which are found in diatom assemblages correspond closely to shifts in other biotic communities such as other algae, zooplankton and aquatic macrophytes (Dixit et al., 1992). This means that diatoms can be used as a reliable indicator of changes throughout the different ecosystems in a lake. The diatoms that are present in sediment sequences therefore provide a snapshot of environmental conditions that reflect the sum of a whole host of environmental conditions that existed at that time (Bradbury, 1999). Care, however, needs to be taken because if sample thickness is too large or sedimentation rates are extremely slow, the assemblage which is analysed may have not been an actual assemblage that existed at a particular time in the past. It will be a time-averaged sample which limits the environmental interpretations that can be made. 4.4,3 Modern sampling In order to enable the successful interpretation of a fossil diatom record the factors which influence the distribution of the species need to be known. Ecological information is published in the literature but this is often very limited. The best way 99 that ecological information can be gained is through the sampling of modern diatom and water chemistry variables. This enables an understanding of the modern limnological characteristics of Belize which consequently ensures that the reconstructions of the fossil environments are as insightful as possible. To enable full characterisation of the water bodies visited samples were collected from a variety of habitats including surface sediments, vegetation and the water surface. Samples were collected and kept in zip lock bags with alcohol. The plankton samples were collected by skimming a clean water bottle across the surface of the water and stored with alcohol. Site information was also gathered including the depth of the water and the surrounding vegetation types. Water chemistry was undertaken in the field using a Palintest, which measured alkalinity, nitrate, chloride, phosphate, magnesium, sulphate and silica. In order to analyse magnesium and calcium (1999 field season) on an atomic absorption spectrophotometer, the samples were stabilised with 2ml of nitric acid and the analyses were carried out back in Edinburgh. Separate probes were used to test for pH and conductivity. The location of the sites visited for modern samples are shown on Figure 4.2. 4.4.4 Laboratory methodology for preparing diatom samples In order to prepare samples for diatom analysis, 0.5 g of dried sediment or modern material was placed in a conical flask and 10% Hydrochloric acid was added (to remove carbonates) until the sample stopped reacting. The samples were then placed on a hot plate and 25ml of 30% Hydrogen Peroxide was then added to the samples to remove the organic matter. Once the reaction had stopped and the levels of liquid in the flasks were at a minimum more Hydrogen Peroxide was added. This continued until the reaction ceased to occur on the addition of the acid (Battarbee, 1986). The samples were then left to cool overnight. The samples were made up to 50ml with distilled water and centrifuged three times for 10 minutes at 1500 RPM. The final sample was decanted down to 30ml. 400pl was pipetted onto a 19mm diameter coverslip and left to dry overnight at room temperature. The coverslips were then permanently mounted onto slides using naphrax. The remaining solution was archived in sealed glass vials. 100 400 diatom valves were counted on each slide using xlOOO oil immersion lens on either an Olympus BX50 or BX40. Pictures of the diatoms were taken using a Matrox Intellicam digital imaging system. Diatoms were identified using Patrick and Reimer (1966); Gasse (1986); Krammer and Lange-Bertalot (1986, 1988, 1991 a, b); Hustedt (1930 a, b, c); Germain (1981) and Foged (1984). Scanning Electron Microscope (SEM) analysis was carried out in the Department of Geology and Geophysics, University of Edinburgh on a Philips XL30-CP. Samples were prepared by pipetting solution from the archive material onto 13mm coverslips. These were then secured onto SEM stubs with carbon coated adhesive and coated in gold. The typical beam current was 60nA and the gun current was 20 KV. 4.4.5 Taxonomy When diatoms are used in palaeoenvironmental studies there are a number of different issues that need to be considered. Two, which were of particular importance to this study were: taxonomy and dissolution of diatom valves. Diatom taxonomy and nomenclature is extremely complex and is compounded by lack of consensus on species names. Diatoms have been recorded and classified for over two centuries but it was not until the late 19th Century that systematic and taxonomic investigations began to collect details about the distributional ecology of species (Battarbee, 1986). In the 1920s researchers began to realise the value of diatoms in palaeolimnological studies, due to their wide distribution and numbers. The correct identification of diatom species is vital to the success of a palaeoecological reconstruction of a lake system. Identification is complicated by the existence of two main schools of thought, those that take a narrow view of species and those who take a rather broader view. The range of morphotypes (key characteristics) that are included within a species is of great importance. Confusions arise when changes in the shape and pattern of the valve that may occur in a diatom's life cycle, are not taken into consideration. 101 Furthermore, it has been noted that some species undergo changes in response to environmental gradients. For example, a change in salinity or silica availability influences the degree and pattern of silification in the valves (Round et al., 1990). These changes are difficult to define or account for. Descriptions of species in floras usually note the extremes of the variation in length, breadth and striae density. This may not, however, be adequate if species are being identified outwith the geographical type-site area. It is also impossible to define the point at which one closely related species grades into another and thus taxonomy will always have inherent within it, bias and distortion (Round et al., 1990). The working groups involved in Anonymous (1975) produced a valuable transcript proposing standardised diatom terminology. Such studies are vital in the harmonisation of taxonomic work. As stated earlier diatom species are often described as cosmopolitan (Mann and Droop, 1996). This implies that diatom species can be found anywhere in the world where their ecological tolerances can be met. This raises a number of issues: 1. In the natural world the broad ecology of systems may be similar, but variations will exist at the micro scale. What is the response of diatoms to the specific habitats and environmental conditions in which they are found? If species respond in slightly different ways (which may be manifested in morphological differences) can these species really be regarded as the same? 2. Do truly endemic species (i.e. those which are restricted to very specific geographical areas) actually exist? 3. Do the ecological preferences of diatoms persist over space and indeed time? The only way in which such issues can be resolved is through the extensive study of diatom species in a variety of geographical settings coupled with the collection of water chemistry, habitat and detailed taxonomic information. This should also extend to the revision of previous studies to ensure that the information available is coherent (Stoermer, 2001). This work would be a great benefit to palaeoenvironmental studies as it would ensure that accurate ecological reconstructions are being made. 102 A species is one or more groups of individuals which can interbreed within the groups but not with other populations. A species can be made up of groups in which members do not exchange genes such as at the extremes of a continuous geographic range. As long as some gene flow does occur along the continuum then the formation of a new species is unlikely to occur. Where physical barriers exist, this reproductive isolation may lead to genetic drift resulting in morphologically distinct subspecies developing. If the species and subspecies were introduced they would be able to breed with one another. Once this ability ceases then the two may be regarded as distinct species (Allaby, 1994). It is this definition which needs to be borne in mind when judging when a variety becomes a species in its own right. There have been no accurate estimates of the number of diatom species. The most common figure quoted is 10,000 (Guillard and Kilham, 1977). Mann and Droop (1996) believe that with the application of modern species concepts this would be raised to 100,000. The implication of this is that there are potentially an enormous number of undiscovered or misclassified diatoms which may have a bearing on ecological reconstructions. In the last three decades new taxa have been described at a rate of 400 per year (Stoermer and Smol, 1999). A key example of the study of varieties is Sellaphora pupula (Mann, 1989). Mann determined there to be at least three varieties (e.g. lanceolate with broadly rostrate ends, linear-lanceolate and linear). In the past they have been identified as belonging to the same species (Hustedt, 1930c; Krammer and Lange-Bertalot, 1986) which implies that the differences exhibited between them have little significance. Observations by Mann (1984, 1989 in 1994) have established that these varieties are in fact reproductively isolated gamodemes i.e. isolated biological species. In terms of palaeoecological studies the significance of this is upheld if it can be determined that the varieties respond to different environmental variables and therefore represent different ecological niches. 103 Within a palaeoecological framework such detailed analysis of species is not possible unless collaborative projects are undertaken. This would be especially valuable where studies are being undertaken in new areas where very little is known about the modern diatom populations. Close taxonomic studies would ensure that the palaeoecological interpretations which are made, are as accurate as possible. Mann (1999) believes that different forms of species which are separated geographically are probably separate species even though some stages of the respective life cycles may be indistinguishable. The affinities of these intermediate forms can only be determined with any certainty if the whole life cycle is known or if the taxonomist has a good understanding of the biogeography and ecology of the area. Such associations would move diatomists away from the tendency to fit their taxonomy into European and American floras enabling a true understanding of species to be developed. Closely related taxa do often differ ecologically (Gasse, 1986) and thus it should not be deemed unlikely that varieties of the same species could also exhibit differences in the field. The ecology of taxa (especially Aulacoseira) studied by Haberyan et al. (1997) in Costa Rica differ from those found in the United States and East Africa. Davies et al. (in press) found that derived conductivity optima for species in Mexico generally compared well with the African data set (Gasse et al., 1995) but there are some differences, especially at the higher end of the conductivity range. For example, Cyclotella meneghiana 2,980pS cm-1 (Mexico) and 6,010 pS cm"1 (Africa); Navicula halophila 19,050 pS cm"' (Mexico) and 2,980 pS cm"' (Africa) (Davies et al., in press). These could simply be a function of the differences in the data set sizes or actual variations in the ecology of the species which have evolved due to the geographical distinctions between Mexico and Africa. Differences are also noted between Spain and Africa (Reed, 1995). For example, Nitzschia amphibia has an optimum tolerance of 8300 pS cm"1 in Spain whereas in Africa it is much lower at 500 pS cm"1. The three studies from Africa, Mexico and Spain are transfer functions i.e. they can be used to reconstruct environmental conditions. If species are found to have different tolerances in distinct areas it means that the same data set will have different reconstructed conditions depending on the transfer function used. This 104 highlights the point that species thrive in potentially divergent conditions in different areas. In order so that the reconstructions that are made are accurate, the transfer function applied should therefore be from either close to the study site or from a similar environment. The more modern diatom based studies that are conducted, the greater our understanding will become regarding the environmental requirements of species. All these issues are particularly relevant to this study because diatom work has not been undertaken in Belize before. Although the species encountered were found in the general literature, some differences were noted. It was therefore decided to conduct a more in depth study of one of the most common species Mastogloia smithii var. lacustris. The details of this are described in Chapter 5. 4.4.6 Dissolution A key assumption of palaeoecological studies is that the fossil record preserved in the sediments is a faithful representation of the community which existed at the time of deposition. There are, however, a host of factors which can interrupt this process, both in the water column and in the sediments. One of these processes is dissolution. The dissolution rate of a particular species can be explained by variables such as the degree of silicification, the extent of sculpturing and the specific surface available for dissolution (Barker, 1992). It has been found in many diatom studies that dissolution is a problem. It is very important to ascertain why this is the case especially if the level of preservation changes through time. The most common explanation given is post-depositional dissolution of frustules at times of high pH (greater than pH 9.0) (e.g. Bradbury et al., 1981). The diatom free sediments found in Bradbury's study of Lake Valencia, represented a time of low lake level. This may have allowed meromixis (lack of overturn in the water column) where the carbonate-rich, hypolimnetic water may have had a pH in excess of pH 9, which would have resulted in the complete dissolution of the diatoms. This is not a fixed rule because if there is high silica 105 availability, diatoms can survive high pH conditions. Poor diatom preservation is a problem in the record from Lake La Yeguada, Panama (Bush et al., 1992). The water chemistry throughout the history of this lake has been dilute and carbonate levels low. The reasons behind poor diatom preservation in this system are problematic. Lake levels at the time of poor diatom preservation were high meaning that solute concentration through evaporation is unlikely. The deep lake could have been stratified resulting in a permanently anoxic hypolimnion and biogenic meromixis but, such effects should be picked up in the chemistry of the sediments and biogenic meromixis in the absence of carbonates is more likely to lower pH than raise it. A drop in silica is unlikely as sponge spicules are present. It could be that strongly oligotrophic conditions resulted in diatoms being outcompeted by other phytoplankton groups. Such groups have not, however, been preserved to prove this hypothesis (Bush et al., 1992). This example highlights the importance of knowledge about the whole lake system so that reasonable hypotheses can be generated with regard to changes in environmental conditions. If there is a high quality of diatom preservation in the surface sediments this indicates that diatom silica dissolution within the lake waters, prior to deposition, is not an important forcing factor (Reed, 1995). If modern samples are fragmented, however, this will increase the chances of post depositional dissolution occurring because of the greater surface area available for dissolution. Fragmentation is caused by different processes than dissolution. The absence or poor preservation of diatoms in surface sediments implies that post depositional taphonomic processes operate very rapidly (Reed, 1995). The chance of this situation arising is greatly enhanced if the lake system is not permanent. As a general rule diatom preservation is better in acid rather than alkaline systems and it is the form of carbonate in these latter systems that is of key importance (Flower, 1993). Preservation should decline as the activity of the metal species increases. Dissolution is caused by hydrolysis of the metal carbonates, which produce hydroxyl ions that attack Si-O bonds in the diatom cell wall (Stumm and 106 Morgan, 1970). Preservation should therefore decrease from calcium, magnesium and sodium carbonate dominated systems (Flower, 1993). In general terms the factors which influence the dissolution of a diatom cell fall into two categories. The first are the characteristics of the water body itself. These include temperature, pH, ionic concentration and composition, the concentration gradient of the dissolved Si02 between the sediment and water, water depth, turbidity, diagenetic processes which are related to groundwater movement, dissolved silica diffusion rates, the interstitial dissolution of silica and the sediment accumulation rate. The coarseness of the sediment, grazing and bioturbation will also have an affect. In shallow lakes, mechanical breakage from turbulence and desiccation are key processes (Ryves, 1994; Reed, 1998b). Flower (1993) notes that diatom valves have a tendency to become highly fragmented in shallow, turbid waters and are better preserved in sheltered lakes with macrophyte beds and a higher sedimentation accumulation rate. The second category is the characteristics of the diatom cell. These include the silica content of the cell wall and the availability of polyvalent cations for adsorption to the cell wall to provide a protective coating (Battarbee, 1986). Diatom silica dissolution is a potential problem in all lakes, but it only becomes an issue where the parameters of dissolution rate and the reaction time are high. As described above there are a number of specific factors which predispose a system to dissolution, but it also occurs in situations where low specific kinetic rates of dissolution act over extended periods of time. In systems which have a high sedimentation rate, the preservation of the diatoms is promoted (Flower, 1993). Where diatoms sediment over large distance in undersaturated water or when the pore-water silica levels cannot be maintained at saturation levels then dissolution will be the main process which acts on the diatoms (Ryves, 1994). It is in lake systems such as these that extrapolating the relationship between the living flora and the modern sediment assemblage is not simple. This especially affects fragile taxa (e.g. Nitzschia frustulum) whose distribution is intimately linked with dissolution processes rather than environmental gradients (Ryves, 1994). With regard to 107 sediment records, problems begin when there is partial preservation of the sequence as this will lead to bias in the record towards the more hardy species. A key factor, which must be taken into consideration, is that techniques in the laboratory can contribute to the poor state of diatoms. Vigorous acid oxidation will destroy fragile diatom valves and make the valves more prone to dissolution (Flower, 1993). Rapid centrifuging and drying is known to affect long species especially and indeed the drying of sediments before these processes begin will result in damage, due to sediment shrinkage (Flower, 1993). Barker (1992) observed a number of stages which diatoms move through when being dissolved. Firstly valves will become relatively transparent as a result of the enlargement of the structural pores. Poorly silicified features are then lost (such as the valve margins) leaving behind the stronger central areas. The surface area:volume ratio and basic shape of a diatom cell influences its ability to dissolve, for example, Nitzschia species begin dissolution at their apices whereas in centric species, the delicate margins disappear first leaving behind a featureless disc (Barker, 1992). Structures with high surface area:volume ratios such as pore fields and areas with a high striae density tend to dissolve preferentially to those with low ratios such as central areas. Raphes often act as a line of susceptibility in the resistant apical zones (Ryves, 1994). It is difficult to follow the dissolution process because diatoms become much more prone to breakage once dissolution has begun. Ryves (1994) formulated a series of dissolution stages which particular diatom genera appear to follow. This is valuable information to gather when assessing the preservation of a sample and crucial to the creation of a meaningful palaenvironmental reconstruction. These stages can then be fed into dissolution indices that can be used to investigate the preservation of samples. These are Flowers DDI, Weighted Index and Square Weighted Index (Flower and Likhoshway, 1993; Ryves, 1994). Flowers DDI is a uniform index varying from 0-1 (the latter being perfect preservation). This index allows the comparison between any samples, but, it is not as sensitive to highly dissolved samples. Weighted and Square weighted 108 are only comparable for taxa which have the same number of dissolution stages. The square weighted index as compared to the weighted index emphasises the non- linearity between the degree of sample dissolution and the proportion of diatoms in the highest dissolution stages (Ryves, 1994; Ryves and Battarbee, unpub). Square Weighted = (a*l+ b*4+c*9+d*16)/ (a+b+c+d) Weighted = (a*l+b*2+c*3+d*4)/(a+b+c+d) DDI = a /(a+b+c+d) Where a = dissolution stage one b = dissolution stage two c = dissolution stage three d = dissolution stage four Dissolution indices can only be applied to taxa which have an identifiable end stage (d). This encompasses genera such as Cymbella, Mastogloia and Cyclotella. Other species either completely disappear or have featureless end stages that are impossible to identify and therefore cannot be used (Ryves, 1994). To ensure that dissolution indices were applied in the most appropriate manner it was decided that a small investigation should be carried out. Only the top 11cm of the Honey Camp 1999 sequence (Figure 4.3) preserved diatoms. Such a shift suggests that there has been a significant change in the system. The study of diatom dissolution provides a great deal of information in terms of the magnitude and timing of the change in the system. In more general terms it allows inferences to be made concerning the effects of the degree to which species are silicified and the implications which this has for using one species as a guide. The use of one species is employed on the longer records from the New River Lagoon. The investigation of had two main aims: 1. To investigate the difference in preservation status between different species. 109 2. To use this information to provide preliminary information about Honey Camp Lagoon. Although diatoms were preserved in the top 11cm there is a great difference between the top 5 and bottom 6cm of the record in terms of diatom concentration. Figure 4.3 is the graph of diatom concentration which clearly shows the higher levels at the top as compared to the bottom of the sequence. In order to gain a meaningful idea of the differences between species, dissolution indices were applied to the counts from the top 5cm. The dissolution indices are based on the amount of diatoms at particular stages of dissolution at each level. The stages for the species used in this investigation are shown in appendix 1. In general terms, there are four stages of dissolution that a particular species will go through. These stages are based on the morphology of the diatom and it has been demonstrated that species do follow distinct stages between perfect preservation and complete dissolution (Barker et al., 1990; Barker, 1992; Flower, 1993; Ryves, 1994; Ryves et al., 2001). Figure 4.4 shows the dissolution indices for the species investigated. The following points can be made immediately: 1. Mastogloia smithii var. lacustris, Denticula elegans and Navicula radiosa all show very similar results. 2. The record from Brachysira neoexilis is different with this species showing best preservation when the rest of the species were at their worst. 3. Denticula elegans is obviously slightly more robust than Mastogloia smithii var. lacustris and Navicula radiosa because it takes slightly longer to drop to lower preservation levels. 4. The DDI index appears to be the most sensitive because it highlights changes between the zones rather than the smooth curve of the weighted and square weighted indices. 5. The key problem with the DDI index is that it records a zero value when there are no perfectly preserved valves meaning that information can be lost. 110 In order to determine how the robust species are, the DDI score for each species was plotted against the percentage of that species in each level (Figure 4.4). This clearly shows that Mastogloia smithii var. lacustris is the most robust because it is able to reach high percentages at low DDI scores. From this information and the fact that it showed a similar pattern of dissolution to Denticula elegans and Navicula radiosa it was decided that Mastogloia smithii var. lacustris is a suitable species to provide a great deal of information about the general diatom preservation status of a record. Although dissolution rates are specific to species it has been demonstrated that a common species can be representative of the system (Barker, 1992; Barker, 2000 pers.com). 4.4.7 Data presentation Two diatom records were produced and the percentage diatom diagrams for the sediment cores have been created using TILIA and TILIAGRAPH (Grimm, 1992). CONISS (Grimm, 1987) was used to identify the diatom zones. This is a stratigraphically constrained clustering programme within TILIA. 4.4.8 Statistical methodology In order to quantify the trends which have been found in both the modern and fossil diatom data sets analysed, particular statistical techniques have been applied. Ordination describes multivariate techniques that arrange samples along an axis on the basis of species composition. This results in a graphical representation of the data where sites are represented by points in space. The aim of ordination is to arrange these points so those samples that are similar in species composition are located close together. This is a good technique for organising large data sets as it is very effective in showing relationships, reducing noise and identifying outliers (Gauch, 1982). For each axis an eiganvalue is produced which summarises the amount of variance which the axis accounts for. The axes are then ranked according to their eiganvalues, 111 with the first axis having the greatest value. The ordination diagram produced can only be interpreted in terms of what is known about the environment of the site. If explicit environmental data are not available, the interpretation must be done in an informal way. If, however, it has been collected, direct gradient analysis can take place. When using direct gradient analysis one is interested from the beginning in particular environmental variables and their specific influence on the system. This study employs two main types of ordination methods. Firstly, Detrended Correspondence Analysis (DCA) which is where the structure of a single data set is described. This is an indirect form of ordination. Secondly, Canonical Correspondence Analysis (CCA) which uses environmental data to explain the data set and is therefore direct gradient analysis. These can be can be accessed as part of the Canoco programme (ter Braak, 1987-1992). DCA was devised by Hill and Gauch in 1980 (Kent and Coker, 1992) in order to attempt to solve the 'arch effect' of Canonical analysis, by detrending (Jongman et al., 1995). DCA extracts the dominant pattern of variation in the community composition from the species data. CCA is different from other ordination techniques, in that it selects the linear combination of environmental variables which maximise the dispersion of the species scores. Thus it incorporates the relationships between species and environment in the actual ordination itself rather than by superimposing environmental data onto ordination plots. The joint plot of species points and environmental arrows is a biplot, which approximates the weighted averages of each species with respect to each of the environmental variables. 4.4.9 Transfer function In order to quantify the changes in the diatom population a transfer function was applied to the core data. For a number of regions around the world modern calibration data sets have been produced. These are created through the collection of modern sediment samples and water chemistry variables with species optima being estimated using regression techniques. These are now very sophisticated and include estimates of error and validation (Birks, 1998). This investigation is the first 112 comprehensive study of both modern and fossil diatoms in Belize. The water chemistry variables collected in association with this data were not adequate to develop a transfer function. Reed (1995) developed the model (which was employed in this investigation) for lakes in Spain. The reconstructed variable is conductivity. This transfer function was chosen because it contained most of the species which were found in the Belize data sets. Issues related to transfer functions were raised in section 4.4.5. These will therefore have to be considered in the interpretation of the results. The application of training sets to core material can also be problematic if for example there is poor diatom preservation (Barker et al., 1990; Ryves, 1994; Reed 1998b) or if the fossil diatoms are not present in the modern day environment (Davies, 2000). From the evidence presented in this chapter it is apparent that diatoms are a powerful tool in terms of reconstructing environmental change. They provide information concerning whole lake ecology e.g. water balance, vegetation change and chemical composition. With all this information a number of issues have to be taken into consideration: 1. The relationship between the modern environment and the palaeo environment. 2. Issues of species identification and the applicability of this over space. 3. The processes which influence the preservation of diatoms. The key point is that if all the issues are considered then diatoms are an extremely insightful tool. 4.5 Stable isotope analysis: Introduction Stable isotope analyses are being employed in this investigation to enable the reconstruction of changing levels of evaporation/precipitation (through oxygen isotopes) and lake dissolved inorganic carbon levels (through carbon isotopes). The aim of this section is to provide a background on the theory behind stable isotope analysis; the controls which act on the bulk carbonate/gastropod records and the behaviour of isotopes in the modern environment. This information will enable an informed interpretation of the results which have been gained in this study. 113 Isotopes are atoms whose nuclei contain the same number of protons but a different number of neutrons i.e. 16 O, l70, lsO. Thus, isotopes of the same element will have slight differences in mass and energy which results in differences in physical and chemical properties (Tucker and Wright, 1990). Results from isotopic studies are reported in ratios, as the absolute abundance of heavy isotopes is low. The values from the sediments are compared against baseline standards, which have been assigned zero per mil on the 8 scale. Low temperature carbonate and organic materials are compared against Peedee Formation Belemnite (PDB) (Craig, 1957). Water is compared against Standard Mean Ocean Water (SMOW) (Craig, 1961). The trends in isotope records are referred to as enrichment or depletion relative to the standards. The measured difference between the standard and sample is reported in terms of a 5-values where: 5 = (Rx-Rstd) / R.std-1000 R = isotope ratio i.e. I80/I60. (1000 converts 8-values to per mil (%o)). Urey (1947) first proposed that oxygen isotopes could be used to give an idea of palaeotemperatures. Freshwater carbonates were not thought to be suitable for such considerations. It has since been found that the change in the lsO content of water, due to climatic change, is much larger in freshwater reservoirs than it is in the ocean. This suggests that it is possible to use oxygen isotope ratios to estimate temperature changes and other climatic forcing in the freshwater environment (Stuiver, 1970). Carbonate minerals, which have been precipitated under equilibrium, will have an isotopic composition that reflects the isotopic content of the system. This relationship is known as isotopic equilibrium. Therefore, the stable isotope compositions of fossil limestones and shells contain within them information about past environments. 114 4.5.1 Controls on oxygen isotopes: The dominant control on the l80/l60 ratio in tropical lakes is evaporative fractionation which is controlled by temperature and humidity. Annual temperature ranges in the tropics are small (section 1.3) and therefore the effect of temperature on the isotopic composition of carbonate is negligible (Gat, 1980). Providing there are no changes to mineralogy then changes in 5I80 can be linked to the ratio of evaporation to precipitation (Fontes and Gonfiantini, 1967; Covich and Stuiver, 1974; Gasse et al., 1990; Talbot, 1990; Lister et al., 1991). The degree of change is very much dependent on the lake hydrology, with closed basins being the most responsive. Within open systems, groundwater flux will have a substantial affect (Bridgwater et al., 1999). In lakes with a rapid through-flow of water, the isotopic composition of the lake water may reflect the mean inflow, such as precipitation and groundwater. Such information is useful as the composition of meteoric water varies according to its origin and climatic evolution (Rozanski et al., 1993). The term meteoric water includes glaciers, groundwaters, surface and atmospheric waters (Leng, unpublished). During times of high evaporation to precipitation (i.e. a dry climate), the ratio of l80/160 in lake-water (and in the carbonate precipitated in equilibrium with the lake- water), will increase as the lake volume decreases. Conversely, periods of low evaporation to precipitation (i.e. a wet climate) will be marked by low i80/l60 ratios and increasing lake volumes (Curtis et al., 1996). 18 1 fi The key control on the O/ O in lake water is generally the oxygen isotope composition of the rainfall supplying the lake. The isotopic composition of the precipitation is controlled by the source area of the water vapour; the fraction of the water vapour remaining at the time of precipitation; the amount of evaporation that occurs as the rain falls through the atmosphere and the temperature of condensation (Curtis et al., 1996). Curtis et al. (1996) in their study of Lake Punta Laguna assumed that these affects had been small during the late Holocene relative to the 115 main control of lake water 5180. The main control on this is the fraction of the lakes water budget which has been lost to evaporation which, in turn, is affected by lake residence time (Stuiver, 1970). Changes to the level of precipitation and surface flow will also influence the transfer of dissolved and particulate material to the lake. Changes in the flux of materials to the sediments are a result of variations in the material output of the catchment. The key factor that would influence this is land-use change. Forest clearance and high rainfall accelerate alluviation and colluviation as well as altering the transport of soil nutrients, organic and inorganic matter to the lake. Forest removal will also decrease evapotranspiration and soil moisture storage in the watershed. This will increase the catchment water yield and the transport of isotopically light surface and groundwaters to the lake. To gain a complete understanding of the 5lsO record the role of vegetation in controlling the lacustrine hydrologic budget should be assessed (Roseinmeier et al., in press). 4.5.2 Controls on carbon isotopes: In the same manner as oxygen, carbon has three isotopes 12C, l3C and i4C. It is the ratio between ]2C and l3C that provides the 8I3C signal. Records produced from carbon isotopes are much more problematic to unravel than those from oxygen isotopes. Consequently, very detailed information on the modern day carbon sources 13 12 are required. Variations in the 'C/ C ratio of authigenic carbonates reflect changes in the l3C/l2C ratio of the Total Dissolved Inorganic Carbon (TDIC) pool. This in turn is dependent on the temperature of carbonate precipitation; the degree of equilibration with atmospheric CCF (which is determined in part by residence time); the ratio of aquatic productivity to organic matter decay within the lake; microbial activity and groundwater inputs (Durazzi, 1977; McKenzie, 1985; Chivas et al., 1993; Heaton et al., 1995). 12 Aquatic photosynthesis results in the preferential assimilation of C into organic 116 Imatter leaving the surface water TDIC pool enriched in C. CO2 exchange between the lake TDIC and the atmosphere will also enrich 8l3C. This process becomes more important as water residence time increases. Inputs of freshwater and the oxidation f 1 3of organic matter will result in the depletion of C in the TDIC reservoir (Lamb, 2000). There is also a tendency to equilibrate with atmospheric CO2 through gas exchange (8I3C -7%6) (Holmes et al., 1997). Lake volume changes may play a role on the effect which CCL exchange has on carbon isotope fractionation. A rapid increase in lake volume reduces CO2 exchange between lake water and the atmosphere which causes the lake 8I3C to approach steady state more slowly, evolving towards a value which is lighter than under average conditions (Li and Ku, 1997). Carbon isotope studies can also be undertaken on the organic carbon in the sediment. This provides an indication of the carbon source into the system. There are two photosynthetic pathways which plants can follow: C3 Calvin pathway (lowland forest trees) and the C4 Hatch-Slack pathway (grasses) (Meyers, 1994). These have discernible isotopic signatures with C3 plants occurring between 5l3C -22 to-33%o and C4 plants between -9 to -16%o. The signals, which come from submerged aquatic macrophytes and algae, are less distinct with both having large and overlapping ranges (Holmes et al., 1997). As already described if there is a shift in the catchment vegetation then this will influence the 5I3C signature of the organic matter. Oxidation of terrestrial organic matter generates CO2 that has an isotopic signature that is akin to the source material. Some of this CO2 will enter the groundwater and once this reaches the lake it will influence the isotopic ratio of the lake water DIC (Curtis et al., 1998). 4.5.3 The relationship between carbon and oxygen isotopes: Every lake is different and will respond to the same external forcing in diverse ways. The records from each lake will, in turn, be different. Inferring regional change from individual systems is therefore difficult. Through work which has been undertaken it 117 has become apparent that there are a number of trends which are common to particular types of lake (Talbot, 1990). One such trend is for the carbon and oxygen isotopic ratios to covary. Isotopic covarience is most typical of carbonates from lakes which are hydrologically closed (Talbot, 1990). The R2 value for the correlations in such systems is greater than 0.7. Carbonates precipitated during periods of high lake levels will plot towards the negative end of the trend and those precipitated during low lake levels will be towards the positive end (Talbot, 1990). The persistence of a covarient trend through time implies that the isotopic composition of the inflow to the system is stable, as is the response of the basin to changes in the precipitation to evaporation ratio (Talbot, 1990). What drives the covarient trend? As lake volume declines due to net evaporation, 5I80 will increase as l60 is preferentially lost from the system. It also results in the increase of 5I3C for the lake dissolved inorganic carbon. This increase is the result of three main effects: 1 31. Freshwater is usually more depleted in ~ C than lake water, therefore, in times of high evaporation, photosynthetic removal of organic carbon leads to the increase of lake 8l3C, even if productivity remains unchanged. 2. Strong evaporation raises pCCL of the lake, resulting in a net loss of CO2 to the atmosphere with lighter Sl8C than the lake water S13C. 3. The mixing across the thermocline or chemocline of a lake which supplies nutrients from deep water to the euphotic zone where phytoplankton grow. This increased vertical mixing enhances surface productivity leading to elevated 8I3C values for the lake DIC. (Li and Ku, 1997). These relationships do not hold true in open systems. Basins with rapid throughflow will reflect the composition of the inflowing water and hence often have narrow calcite 8lsO ranges. The variations which do occur reflect small changes in temperature and inflow-precipitation balance between the periods of carbonate precipitation (Talbot, 1990). 8l3C records show wider ranges due to variations in photosynthesis rates (Stuiver, 1970; McKenzie, 1985). 118 4.5.4 The contemporary environment: Knowledge of the present day dynamics is key to the understanding of the palaeoenvironmental record. Ocean waters have a §l80 value of 0 +/- 1%c. Values of surface waters around the globe differ from this due to evaporation, formation of sea ice and the addition of meteoric water. The physical processes which are responsible for the production/transport and condensation of atmospheric water vapour, cause large variations in the isotopic ratios of meteoric water. The total range of 8lsO in natural precipitation is +4 (tropics) to -62%o (Antarctic) (Leng, unpublished). 2 1 Hydrogen has two isotopes: ~H or D (Deuterium) and H. A water molecule containing D or i80 is heavier than a normal 'h'HI60 molecule. Water vapour forming precipitation will be depleted in heavy isotopes relative to ocean water. Condensation forming raindrops from a cloud reverses this process. The heavier molecule condenses first i.e. rain is isotopically enriched and the cloud moisture is subsequently depleted as the rainout continues. This series of isotopic fractionations is temperature dependent and therefore a water sample has a particular isotopic signature depending on the environmental conditions it has experienced. The climate of the Tropics is distinguished by the distinct wet and dry seasons which divide the year. In mid to high latitudes precipitation is isotopically depleted in the winter and enriched in the summer. These differences are due to a number of factors which include the changes in temperature between winter and summer and the seasonally changing source areas of storm trajectories. In general terms the precipitation in polar regions is isotopically lighter than in low latitudes. In the Tropics, however, the seasonal fluctuations in 8lsO and 5D have a different origin. The main correlation is with the amount of precipitation, with isotopically depleted precipitation occurring in the rainy season (Rozanski et al., 1993). This association was first recognised by Dansgaard (1964) and is referred to as the 'amount effect'. There is no relationship with temperature as this varies very little in the Tropics throughout the year. 119 The key point is that surface waters preferentially lose the lighter water molecules due to evaporation and thus are often enriched compared to the rainwater from which the surface waters were formed (Global Network for Isotopes in Precipitation, 2000). Craig, 1961 discovered that there is a linear correlation between 52H and 5lsO for meteoric water whereby: 82H = 8 8lsO + 10. This is known as the Global Meteoric Water Line (GMWL). Deviations from this slope are caused by evaporation. This occurs mainly in low latitudes and semi-arid regions. The position of rainfall along the GMWL is dependent on the timing and duration of rainfall as well as temperature, humidity and altitude. There will therefore be seasonal differences in the values (Lamb, 2000). Since 1961, the International Atomic Energy Agency (IAEA) and the World Meteorological Organisation (WMO) have conducted a worldwide survey of hydrogen and oxygen isotopes in precipitation (Dansgaard, 1964). From the collection of these data it has been possible to deduce circulation patterns and the mechanisms of global and local water movements. Data have not been collected for Belize, but have been in the neighbouring locations of Veracruz, Mexico; Havana, Cuba; Puerto Rico; Panama and the Dominican Republic. In order to enhance the interpretations which can be made from the !80 record it needs to be determined whether, in the modern day environment, there is a relationship between the lsO of precipitation and the amount of precipitation for the same period. Data from monthly precipitation totals and average lsO values have been plotted as scatterplots (Figure 4.5). The only area which has an R~ value of greater than 0.5 is the Dominican Republic. This implies that this is the only region where the 'amount effect' holds true. If this is the case then this has severe implications for the interpretation of isotope records because the controls over the isotope signal are not clear. There are however a number of reasons why the relationship does not hold in the other sites. It could be a function of the years which the data are from i.e. those years may not have had a particularly distinct wet and dry 120 season. The sites where the data have been taken from may have local environmental factors such as a mountain range which would affect the signal. The accuracy and reliability of the data are also not known. The values are averages (as these are the only data available) and therefore the signal is also likely to be dampened. This study highlights that a degree of caution needs to attached to interpreting results and shows the importance of collecting modern data so that these relationships can be determined for the area that is being studied. 4.5.5 Methodology In order to gain an idea of the modern water isotope values in the New River and Honey Camp Lagoons, modern water samples were collected. These were kept in a sealed water bottle which was kept as cool as possible. The samples were analysed for 5lsO and 5D by Mr Andrew Tait at the Scottish Universities Environmental Research Centre, East Kilbride. 4.5.6 Bulk carbonate record: The stable isotope results in this study have been obtained from both bulk carbonates and individual gastropods. In order to fully understand the results that have been derived, it is important to appreciate the governing mechanisms which control the isotopic signature from both the gastropods and the bulk carbonates. Authigenic calcites are precipitated mainly during the summer due to increased algae and macrophyte photosynthesis in the epilimnion. Algae have seasonal blooms during which time the dissolved C02 in the lake waters is depleted inducing carbonate precipitation. Carbonate ions in lake water form as a result of the dissolution of C02 which when partially hydrated exist as carbonic acid (H2C03). The disassociation of H2C03 produces C02, HC03 and C032" the relative proportions of which are dependent on pH. Precipitation of carbonate occurs either due to an increase in temperature or due to a decrease in the partial pressure of C02 (due to its removal from the lake). As 121 temperature has only a small affect on carbonate supersaturation, C02 removal is generally more important (Tucker and Wright, 1990). A common way to remove CO? is by photosynthesis and therefore most inorganic carbonate precipitation is biologically induced. The reduction of C02 from the lake will slowly be replaced via exchange with atmospheric C02 until equilibrium is restored. The isotopic signal, which is recorded, is that of surface conditions which will be more sensitive to changes during the annual cycle than material from deeper locations (Leng, unpublished). Biogenic and inorganic carbonates are authigenic if they have been precipitated within the lake itself and thus are the most useful material to use as they reflect conditions in the lake at the time of precipitation. Authigenic carbonate therefore records average trends but will be weighted towards summer conditions when, as explained earlier, most precipitation occurs. Bulk carbonates were used in this study and these are made up of: 1. Detrital carbonate derived from the catchment via rivers and shoreline erosion including reworked lacustrine carbonates which have become exposed during falls in lake level. 2. Biogenic carbonate derived from various organisms including molluscs. 3. Diagenetic carbonate produced by post-depositional alteration of other carbonate minerals. (Tucker and Wright, 1990). A complication to the bulk carbonate record is fractionation. This occurs when there is a change in the ratio of the two isotopes during a reaction or process (Tucker and Wright, 1990). The lsO and l3C ratios in water, bicarbonate and carbonate are related by equilibrium constants (isotope fractionation factors) which are functions of temperature. Thus, the l80/l3C variations measured in carbonates are not exactly the same as the original variations in the water, which must be taken into consideration when interpreting results. It is, however, generally assumed that primary lacustrine 122 carbonates precipitate in equilibrium with lake water and thus are related to the isotopic composition and temperature of the lake water at the time of precipitation (Epstein et al., 1953). The resulting fractionation is mineralogically dependent (Stuiver, 1970; Pearson and Coplen, 1978; Gat, 1995). The record that is produced from the bulk carbonates will therefore be an average signal representing the lake and its catchment. This may mean that changes in the isotopic signature of the system may be dampened with small-scale shifts being lost. As a result of this the trends that are seen in the record must therefore be very significant. 4.5.7 Biogenic carbonates: The record from the gastropods The record gained from gastropods is a function of both the species microhabitat and the length of its life cycle. Molluscan carbonate grows incrementally and therefore the results from one shell will be a time-averaged result for the life cycle of the species. This tends to be one year. Knowledge of the gastropod species ecology is therefore essential to the interpretation of the results. Fritz and Poplawski (1974) 18found that there is a good relationship between the O in mollusc shells and lake water during the growing season. Differences between species appear to be related to different growth periods and the temperature of their habitat (e.g. shallow versus deep water). From a palaeoecological perspective one needs to make the assumption that no evolutionary change has taken place that would have altered the relationship between the gastropod and its environment through time i.e. that its life cycle, habitat and feeding requirements have not modified. If this is the case then knowledge of the modern species can be applied through a core sequence (Goodfriend, 1992). One also must be aware that the point of shell deposition may not be where the species lived and thus the conditions which the shells represent may not be of the core environment, but, elsewhere in the lake system. 123 Equilibrium fractionation between oxygen isotopes of water and shell carbonate is assumed in studies of land snail shells. In aquatic molluscs, shell carbonate appears to be deposited at or near equilibrium with the water in which the gastropods live (Mook and Vogel, 1968; Fritz and Poplawski, 1974). Gastropod shells are made from aragonite and therefore there is a vital offset between the water and carbonate and also between calcite and aragonite. Tarutani, et al. (1969) established this to be 0.6 at 25°C for oxygen isotopes. Thus, aragonite concentrates lsO relative to calcite. Rubinson and Clayton (1969) worked out a fractionation of 1.8+/- 0.2 for carbon values resulting in l3C being enriched in aragonite as compared to calcite. For the Belize investigation vital effects were minimised by analysing only adult specimens of particular species because Curtis et al. (1996) determined that mature individuals of Pygophorus and Cochliopina species secrete their shells near oxygen isotope equilibrium. There are two main ways in which one can gain isotopic information from gastropods. One is to crush a shell and take a reading from a representative portion of the powder. This will provide an average for the individuals life cycle. Alternatively, measurements can be taken along the growth whorls which provide a much more detailed record of change (e.g. Feng et al., 1998). This requires shells to be quite large so that the growth stages can be distinguished. The shells used in this study were 2 mm - 3 mm and therefore unsuitable for the latter application. The controls on the stable isotopes in the gastropod shell are more complicated than 13 on bulk carbonates. The C of shell carbonate is affected both by the isotopic composition of the dietary carbon (plant matter and inorganic carbonates) and the degree to which this is modified by exchange with atmospheric C02 (Feng, 1998). Where variations are low in the carbon record it implies that all the proportions from the different carbon sources have remained constant. Factors affecting shell 13 18carbonate C values will also influence the O values most notably the metabolic rate relative to exchange with the environment (Leng et al., 1998). If metabolic and temperature induced variations can be discounted, the changes must be due to variations in the isotopic composition of the environmental water taken in by the 124 gastropod (Leng et al., 1998). It is important to have an understanding of the microhabitat changes, which will influence the isotope values recorded in the shells. An example of this is the photosynthetic activity of aquatic plants. In a similar manner to terrestrial plants, aquatic plants preferentially utilise isotopically light 13carbon which results in "C enrichment in the immediate vicinity of these plants. Two species were used in this study: Cochliopina sp. and Pygophorus sp. These were identified by Alan Covich (February, 1999). These have been successfully used in isotopic studies elsewhere in Central America (Covich and Stuiver; 1974; Curtis et al., 1996; Curtis et al., 1998). Both these species are present in Covich's (1976) study of molluscan species diversity in the Peten, Guatemala. Both Cochliopina and Pygophorus sp were also used in Curtis et al.( 1998) study of Lake Peten-Itza, Guatemala and the two species produced very similar isotopic records. Covich undertook a study of freshwater gastropod assemblages in Albion Island, Belize (Pohl, 1990). In this investigation environmental inferences were made on the basis of the species found at each level. Cochliopina sp. is typical of deep, permanent freshwater in Belize and it is also found in the Mexican Yucatan Peninsula. They feed on periphyton and bacteria and live on rocks and dead wood. Not much is known about the water quality tolerances but they need sufficient levels of dissolved oxygen as they are gill bearing prosobranchs (Covich pers. com. February, 1999). Pyrgophorus sp. is distributed throughout the Caribbean and coastal areas of the Gulf of Mexico. It is typical of permanent, relatively deep lakes (Covich, 1983). Live specimens were found in the littoral zone of Pulltrouser Swamp. There are two forms of this species smooth and spiniose. There is no consensus as to whether this difference represents an ecological variation or is a response to predation (Covich, 1983). In summary, both gastropod and bulk carbonate measurements were carried out in this investigation. This was because these two sources provide complementary 125 information to one another. The bulk carbonate provides a general record whereas the signal from the gastropods is specific to their habitat and will therefore be influenced by slightly different factors. The combination of these two will enable a much clearer idea of the lake dynamics to be developed. It will also enable an improved understanding of the inputs into and controls over the bulk carbonate record. 4.5.8 Methodology A carbonate bomb was used in order to measure the percentage of carbonate in the sediment. This provides information on not only changing inputs to the catchment but also determines whether the material is suitable for stable isotope analysis. 0.7g of dried sediment and a fixed volume of 6N hydrochloric acid were added into the vessel. As the vessel is shaken, the sediment and acid react to produce carbon dioxide. The amount of gas evolved is proportional to the percentage of carbonate in the sediment. Stable isotope analysis of the bulk carbonate material was carried out at Scottish Universities Environmental Research Centre by Mr Andrew Tait (SUERC) and the author. 1 mg of each sample and standard were weighed and loaded into glass tubes which were sealed with a cap containing a piercable septum. Once all the samples were loaded they were placed into a temperature controlled hot block at 70°C for 30 minutes. This ensured that the samples were at reaction temperature. The carbonate acid injector was then placed into each tube in turn. In the first phase ultra-pure helium was used to purge all the atmospheric gases from the tube, for two minutes. 7 or 8 drops of '103%' phosphoric acid were then added to the tube. This reaction was then left to proceed for at least 8 hours. Once the reaction had finished the tubes were transferred from the carbonate acid injector to the AP2003 Gas Prep Interface. This analysis consisted of a single reference peak followed by four sample peaks from one tube and then a final reference peak. Gas was extracted from the sample tube and was moved through a small room temperature gas chromatograph which separates the CO? from any other gas which may be in the tube. The resulting gas 126 then flowed into an AP2003 triple collector mass spectrometer which measured the 45/44 and 46/44 ratios. The 5 value for the C02 was then calculated from the ratios of the sample gas peaks and the reference gas peaks. The calculations are as follows: 45/44 (l3C + ,602) / (12C + 1602) 46/44 (12C + 160 + lsO) / (12C + 1602) 513C = 1.0676 5(45/44) - 0.0338 5lsO 5lsO = 1.0010 5(46/44)-0.0021 5I3C For the analysis of gastropods, shells were picked out from the sediment and cleaned in distilled water in an ultrasound bath until all the residual sediment had been removed from the shell. These were analysed by Mr Colin Chilcot (Department of Geology and Geophysics, University of Edinburgh) and the author. 0.5-lmg of shell powder was reacted with 100% orthophosphoric acid at 90°C in an automatic carbonate preparation system. The resulting C02 was then analysed on a VG Isogas PRISM III mass spectrometer. The standard MAB2B was analysed with each run with the standard deviation being +/- 0.0073 for 5I3C and +/- 0.071 for 5lsO. All the measurements were corrected for isobaric interference. 4.6 Chronology In order that the results of this investigation can be put into a meaningful context, the sediments need to be dated. There are several ways in which this can be done. In 210 this study radiometric methods were used namely radiocarbon and ~ Pb dating. 4.6.1 Radiocarbon dating Radiometric dating techniques are based on the time dependent radioactive decay which unstable isotopes undergo. Radiocarbon dating was one of the earliest 127 radiometric techniques to be developed (Libby, 1955). I4C is the radioactive isotope of carbon and eventually decays to form the stable element !4N. I4C atoms are 19 rapidly oxidised to carbon dioxide and along with CO2 become mixed throughout the atmosphere and gets stored in the atmosphere, biosphere and hydrosphere (Lowe and Walker, 1997). The activity of l4C is halved every 5730+/- 40 years (Godwin, 1962), this was a revision of Libby's 1955 estimation of 5568+/- 30 years. However, because a large number of dates had been published using the original estimation an internationally agreed constant has been agreed on. This is 5570 +/- 30 years (Mook, 1986). The l4C activity of a sample is measured by two methods: conventional radiocarbon dating (which involves the detection and counting of (3 emissions from l4C atoms over a period of time) and accelerator mass spectrometry (which uses particle accelerators to count the actual number of l4C atoms in the sample). The dates produced in this study have all been produced using the latter methodology, which allows for the more accurate dating of much smaller samples than the conventional method. 4.6.2 Issues of dating in limestone geology A large proportion of northern Belize is composed of a limestone terrain and thus the key issue in relation to radiocarbon dating is the introduction of 14C deficient carbon into the reservoir of dissolved inorganic carbon in surface and groundwaters (Leyden et al., 1998). When this is incorporated into aquatic primary producers; precipitating marl or the carbonate shells of aquatic organisms, the result is anonymously old radiocarbon ages. This phenomenon is known as the hardwater effect (Deevey and Stuiver, 1964). Terrestrial organic matter is free from this influence because it is in isotopic equilibrium with the atmosphere and is therefore the ideal dating medium. This, however, was sparse in the cores that were collected in this investigation. A strategy which has been used to overcome the lack of terrestrial organic matter in other studies in Central America, is the dating of gastropod shells. Curtis et al. 128 (1996) dated both terrestrial wood and aquatic gastropod species from Punta Laguna. It was found that the dates obtained from the shells were consistently older by 1200- 1300 years. This correction factor was then applied to the other gastropod dates because it was believed to represent the hardwater effect in the system. Hodell et al. (1995) found dates obtained from aquatic gastropods from Lake Chichancanab, were 1200 years older than those obtained from terrestrial organic material. Hodell et al. (1991) investigated Lake Miragoane, Haiti. The chronology for this sequence was based on dates from ostracod shells and two from organic carbon. One level had both an ostracod and terrestrial wood date. The difference between these two was 1025 years. It was assumed from this that the hardwater error was constant throughout the core and a correction of 1000 years was applied to the ostracod dates obtained. There are two strategies that have been employed in the literature with regard to the correction factor applied to gastropod dates: 1. Paired dates (i.e. terrestrial organic matter and a gastropod from the same level) can be obtained and the difference between the two dates can be regarded as the hardwater error 2. Where it is not possible for dates to be obtained from both media at the same depth, results from each can be plotted separately and the difference between the two gradients can then be used as the correction factor to be applied to the gastropod dates. From the diatom records, which have been produced from Belize, the species do not show any significant shifts with regard to pH or other water chemistry variables suggesting that the hardwater error is likely to have been constant through time. 4.6.3 Methodology When each core was opened any material which was thought to be suitable for radiocarbon dating, was removed and placed in a darkened petri dish. All material was stored in the cold store in the Department of Geography and the British 129 Geological Survey, Edinburgh. Once samples had been approved for dating they were brought to the NERC Radiocarbon Laboratory and prepared by Dr Charlotte Bryant and colleagues. The reference number for the dates used in this investigation is: 761.1298. The pre-treatment of the samples before dating depended on the type of material that was submitted. For plant macrofossils, any carbonaceous sediment was removed by soaking the samples in 2M Hydrochloric acid until the pH remained less than 7. The samples were then rinsed in distilled water and digested in a further aliquot of 2M Hydrochloric acid at 80°C for 8 hours. Samples were then rinsed with distilled water and dried to a constant weight. A pestle and mortar was used to homogenise the material. The total carbon in a known weight of the pre-treated sample was recovered as CCL by heating with CuO in a sealed quartz tube. The gas was converted to graphite by Fe/Zn reduction (C.Bryant pers com, 2001). Gastropod samples were soaked in hot distilled water to remove any sediment from the shells. The outer 20% by weight of the shells was removed by controlled hydrolysis with dilute Hydrochloric acid. The samples were then rinsed in distilled water, dried and homogenised. A known weight of pre-treated sample was hydrolysed to CO? using 85% orthophosphoric acid at 25°C. The gas was converted to graphite by Fe/Zn reduction (C.Bryant pers.com, 2001). 4.6.4 Lead 210 dating Radiocarbon dating can be used over the last 50,000 years (Lowe and Walker, 1997). The fine temporal resolution examination of recent sediments can be achieved 210 through ~ Pb dating which can be used over the last 150 years to date events (with a 222half life of 22.26 years). The radioactive decay of radon gas ( ~Rn) which is part of 210 the U-series decay chain produces a series of daughter nuclides, one of which is Pb (Lowe and Walker, 1997). This unstable isotope is removed from the atmosphere and accumulates in lacustrine, terrestrial and marine environments where it decays to the stable 206Pb. The ratio of 210Pb to 206Pb can then be measured and assuming that 130 210the atmospheric flux of Pb has remained constant, the time elapsed since the lead was deposited in the sequence can be established (Olsson, 1986). This methodology will therefore also allow the rate of sediment accumulation to be estimated. The main 9 10 problem is that most sediment contains a small amount of Pb derived from the 9 10 decay of uranium or its daughters. This is known as the 'supported' " Pb and must 9 I 0 therefore be calculated and subtracted from the 'unsupported' Pb which is produced in the atmosphere (Lowe and Walker, 1997). The 'supported' component is calculated by measuring the 226Ra activity. In a system that is accumulating sediment at a uniform rate the activity of 'unsupported' 210Pb will decrease exponentially with depth. There are two key methods for the calculation of sediment accumulation rates: the Constant Rate of Supply (CRS) model and the Constant Initial Concentration (CIC) model (Goldberg, 1963). The difference between the models is the assumptions that 910 they make. The CRS model assumes that unsupported Pb flux is constant but that 210the initial " Pb concentration in the sediment is variable, as is the influx of sediment. The CIC model assumes that all these variables are constant. As an independent means for testing the 2l0Pb age chronology, man-made radionuclides can be analysed (Eades et al., 1998). Due to the testing of nuclear weapons in the 1950s and 1960s, artificial radionuclides were released into the atmosphere. l37Cs reached its peak concentration in 1963 (Bonnett and Cambrey, 1991). A second peak is also found in 1986 due to the fallout associated with the Chernobyl disaster. Biological and physical activities which occur near the sediment-water interface result in the redistribution of sediments. Three major processes govern the concentration of radioactive nuclides within a sediment sequence: radioactive decay, sedimentation and mixing. The measurement of artificial radionuclides provides information on sediment accumulation rates. The key problem with this is that postdepositional mixing, erosion and redeposition processes destroy the original delivery pattern to the sediments (Krishnaswami and Lai, 1979). 131 Davies (2000) successfully demonstrated the use of 2l0Pb as a chronological tool in Lago de Zirahuen, Mexico. The calculated depositional flux for this basin was 48 Bq 2 1 2 1 m~ yr" which is much lower than the global average of 185 Bq m""yr" (Appleby and Oldfield, 1983). This implies that there maybe regional variations in the global flux 210 210of Pb (Davies, 2000). Pb has also been successfully used in Lake Miragoane, Haiti (Brenner and Binford, 1988; Hodell et al., 1991). This thesis found that the 210 2 1Pb fallout rate was very low (0.09 pCi cm~"yr~ as compared to a global average of 2 10.5 pCi cm " yr" ). The reasons for this discrepancy are not clear but it could be due to loss of sediments from erosional zones, the result of upward migration and 210solubilisation of sedimentary Pb under conditions of deepwater anoxia or low rates 222of "Rn flow from the sea surface and local soils (Brenner and Binford, 1988). 4.6.5 Methodology 4g of dry sediment from a 1cm slice of material was ground with a pestle and mortar and weighed into petri dishes. The petri dishes were then sealed with an epoxy resin to stop the radon diffusing. Samples were then left for 3 weeks to attain radioactive equilibrium. The activity of the radionuclides was determined by gamma spectroscopy. Mrs. A. Stewart and Dr. A. MacKenzie at the Scottish Universities Environmental Research Centre at East Kilbride and the author carried out the analyses. 4.7 Other methodologies employed Dr Malcolm Murray carried out the following analyses on a parallel investigation to this one funded by the Leverhulme Trust (F/158/BQ). This information has been used to supplement the methodologies already described. 132 4.7.1 Available phosphorus Studies have shown that one of the environmental responses to population growth in the Peten was an increase in the delivery rate of phosphorus to lakes (Brenner, 1978; Deevey et al., 1979; Vaughan, 1979). Brenner (1983) found that total phosphorus levels in the catchment soils around Lake Quexil were extremely similar to those found in the lake sediments. This suggests that soil movement is the principal means of nutrient transfer. Activities by humans in the catchment would have released phosphorus to soils where it would have been locked into insoluble compounds and removed by erosion (Deevey and Rice, 1980). Work in Peten has also shown that phosphorus is much more concentrated in surface soils than lower horizons and at the bottom of slopes (Brenner, 1983). Changes in phosphorus levels can therefore be regarded as a cultural signal. Evidence for the mobilisation and deposition of phosphorus is very important as it is essential to support ecosystem dynamics (Brenner, 1983). Deevey et al. (1979) believe that phosphorus is an element which is likely to have become deficient in the Mayan environment because soluble phosphorus is immobilised by calcium in a limestone terrain and is not replaced by the atmosphere. Available phosphorus is critical if most of the phosphorus is transferred to the sediments by runoff (Deevey et al., 1979). Available phosphorus levels in this investigation were determined using a HACH DR2000 portable spectrophotometer. The colour of the solution produced by this methodology is proportional to the amount of phosphorus in the sample. 4.7.2 Magnetic susceptibility The measurement of magnetic susceptibility is generally considered to be a proxy for catchment disturbance (Thompson, 1984). In limestone terrain, the presence of a strong magnetic susceptibility signal cannot be guaranteed, but changes may indicate inputs of clay or iron rich soils from the slopes. 133 One of the most common applications of this technique has been to use the magnetic signal to identify times of increased catchment erosion. This is picked up because topsoil has a greater magnetic signal than subsoil. Where clear peaks and troughs are exhibited in a record this method can be used to cross-reference cores from the same region. O'Hara et al. (1993) successfully used this technique as an indicator of human induced catchment disturbance in Lake Patzcuaro, Mexico. These measurements were taken with Bartington Instruments MSI and MS2 meters which were connected to 80mm or 125mm MS2C loop sensors. Mass specific susceptibility was measured using a Bartington MS2B dual frequency sensor. Samples were dried at 40°C and placed into 10cm3 pots and then weighed. 4.7.3 Particle size analysis Binford (1983) proposed that particle size changes down core could be used as an index of human activity in the catchment. A drop in mean particle size is thought to represent times of increased disturbance and can be used to cross-correlate cores from a region. Changes in mean particle size can also occur by natural changes such as climatic or hydrologic changes to the basin. The usefulness of this technique does depend on the geology of the area. Pohl et al. (1990) undertook particle size analysis in Albion Island, Belize. They did not come up with any meaningful conclusions due to the flocculation of clay particles and the high content of gypsum in the sediment. Analyses were undertaken using a Beckman Coulter LS230 laser diffraction particle analyser. Samples were pre-treated with hydrochloric acid (to remove the carbonates) and hydrogen peroxide (to remove the organic matter). The results therefore reflect the non-carbonate mineral fraction. Details of this methodology and the problems associated with particle-size analysis in carbonate sediments are discussed in Murray (in press). 134 4.7.4 X-ray diffraction (XRD) In order to gain an idea of the bulk mineralogy XRD was undertaken using a Phillips PW1800 instrument. Knowledge gained from this technique is very helpful to the interpretation of the stable isotope record. Fractionation of aragonite, dolomite and calcite is different and therefore if the system is a mixture of these or changes from one mineral to another this will affect the interpretation of the isotope results. Information from XRD allows one to make inferences about the sediment sources to the systems and the weathering regimes to which the system has been subject to. XRD analyses were undertaken in the Department of Geology and Geophysics, University of Edinburgh by Dr Malcolm Murray, Mr Geoff Angell and the author. A mixture of dried sediment and acetone was dropped onto a glass slide to create a thin covering which could then be analysed. For samples with a high clay content, the same analysis was repeated on just the clay fraction. This was separated out by producing a suspension in water and sampling from the liquid after sediment settling had occurred. 4.7.5 Loss on ignition This technique is used to determine the amount of organic matter present in a sample. l-2g of sediment was weighed out and placed in a furnace for two hours at 500- 550°C to ensure that all the organic carbon in the sample was completely oxidised. This method will provide another proxy for environmental change in the system which will be valuable for corroborating with the other records produced. 4.7.6 Carbon and nitrogen analysis The ratio of carbon to nitrogen is used to determine changes in the sources of organic matter to lakes. This is possible because the signatures from algae and terrestrial organic matter are distinctive (Meyers, 1994). This method will provide a valuable additional information to the 5I3C record. 135 Carbon and Nitrogen analysis was undertaken using a Carlo-Erba NA 1500 Elemental Analyser in the Department of Geology and Geophysics, University of Edinburgh. For this method 40mg of air-dried and ground samples of sediment were used. This amount is larger than is normally required but this ensured that reproducible nitrogen values were obtained. The methodology followed that employed by Nieuwenhuize et al. (1994). 4.8 Conclusions This chapter has provided key background information on methodologies employed in this investigation. This not only highlights the suitability of these methodologies to answering the main questions behind this project but also aids in the understanding of the results gained. The statistical techniques which have been employed in this study enable the rigorous classification of the data sets providing the basis for environmental interpretation. The trends which have been delimited in the data sets will be put into context by the chronology created by l4C and 2l0Pb dating. This is vital to ensure that the activities of humans in the areas studied can be assessed in terms of the environmental records collected. This project comprises two main components: the climatic and the human signal. The techniques which have been employed will tease out both of these, allowing a greater insight into the lake systems under investigation. Oxygen isotopes will provide the main climate signal. Phosphorus, particle size analysis and magnetic susceptibility will provide an indication of human activity. The diatom record is one of general environmental change which will be influenced by a number of different factors. XRD, carbon:nitrogen ratios, LOI and carbon isotopes will provide a suite of information which will also improve understanding of the lake dynamics. The key proxy which has not been mentioned is pollen. A preliminary record for the long core at Hillbank, New River Lagoon has been produced by Ms Viveca Persson and Professor Steve Blackmore while at the Natural History Museum, London 136 (present address for SB is the Royal Botanic Gardens, Edinburgh). This record is very coarse with many of the species being unidentified. This limits the interpretations that can be made but it does help to provide some idea of vegetation changes through time. 137 Figure 4.1 A. Percussion coring at Aguacaliente Swamp B. Livingstone coring at Hillbank Figure 4.2 Location map of northern Belize showing the places referred to in this thesis. Honey Camp 1999 Diatoms per g c 200000 400000 600000 800000 n AVJ 1 - 2 - 3 - 4 - E" 3, 5 - §" 6- Q 7 - 8 - 9 - 10 - 11 - Figure 4.3 Diatom concentration in Honey Camp Lagoon 1999 Pink: Weighted index Green: Square weighted index Blue: DDI index Mastogloiasmith ivarlacustr s DissolutionIndices 010 Brachysiraneoex lis DissolutionIndices 0510 Mastogloiamithiivarl cus ris Brachysiraneoex lis Figure 4.4 Denticulaelegans Navicularadios Disolutionindices 0510 Denticulaelegans DDI 0.2468 DissolutionIndices 102 \ \ °T • .i Navicularadios DDI 0.2468 Veracruz 10 PuertoRic DominicanRepublic Havana Panama Figure4.5180versusamountofpr cipitation(mm) Chapter Five - Results: The Modern Environment 5.1 Rationale One of the principal methodologies employed in this project to investigate the changes that have occurred in the environments of Belize through time was diatom analysis. This is the first time that this tool has been applied in Belize. In the previous chapter it was explained that in order to develop a meaningful interpretation of fossil records a knowledge of the modern diatom flora is essential. This chapter provides the interpretation of those results. In the final part of this chapter the results from the modern isotope survey will be discussed. Figure 4.1 shows the location of the water bodies sampled. The two sites in the south of Belize (Monkey Tail River and Aguacaliente Swamp) are shown on Figure 1.1. A number of different samples were taken from Hillbank and Lamanai. These are shown in more detail in Figures 5.2 and 5.3. 5.2 Water chemistry Water chemistry data were collected over three years as part of the author's MRes and PhD studies from a variety of sites in Belize. The methodology employed has been described in Chapter 4. The information in this chapter provides an indication of quantitative differences between sites. There are issues with the data set which need to be considered before interpretations are made. The data were collected from 1998-2000 at approximately the same time of year. It is not known however how conditions varied seasonally through this time frame, but values from the same lakes over the different years are comparable suggesting that the data are reliable. The number of water chemistry variables measured at the sites is not consistent because the water samples collected in the year 2000 were stolen at the end of the field season. This means that variables cannot be compared throughout the whole data set. The accuracy of the field equipment used has not been quantified. The same equipment was used throughout and therefore the results can be considered to be comparable within themselves. Both Honey Camp and Kates Lagoon have been measured for a variety of water chemistry variables in the past (Brenner pers. com. 143 (data from 1985); Jacob, 1992). These results fall within the same range as those collected in this study, which is encouraging in terms of the validity of this data set. The results are listed in Figure 5.4. 5.2.1 Discussion The triangular diagrams (Figure 5.5) contain cation and anion data for the sites visited on the 1999 field season. Water chemistry measurements from the other sites visited did not contain the full complement of variables. Although limited, the 1999 data set provides an indication of the general characteristics of water bodies in Belize. The sites fall into two clear groups and a group of outliers. The groups are: 1. Calcium-Sulphate dominated: Aguacaliente Swamp, Lamanai, Booth River and Small Croc Lagoon. The pH range for these sites is 6.82-8.1. The conductivity range is 0.6-1.63 mScm1. 2. Sodium-Chloride dominated: Honey Camp Lagoon, Progresso Lagoon, Fabers Lagoon, Jones Lagoon, Northern Lagoon, Southern Lagoon and Wagner Lagoon. The pH range for these sites is 7.92-8.35. The conductivity range is 1.6-23.34 mScm"1. 3. Calcium-Total Carbonate dominated: Kates Lagoon. This site has a pH of 7.76 and a conductivity of 0.27 mScirf1. 4. Magnesium-Total Carbonate dominated: Chiwa Lagoon. This site has a pH of 7.21 and a conductivity of 0.43 mScm '. 5. Calcium-Sulphate/Total Carbonate dominated: Crooked Tree Lagoon. This site has a pH of 8.06 and a conductivity of 0.6 mScm"1. From this information a number of key points can be made. The characteristics of the sites that fall into group 1 concur with Eugster and Hardie's (1970) second evolutionary pathway whereby the systems are enriched in alkaline earths and have low levels of carbonates. The second group in this study are further along the pathway falling in category IIB i.e. their water chemistry is more evolved. This 144 suggests that there is continuation of water chemistry throughout the lake bodies that have been studied. The three outlying sites (3-5) are all rich in total carbonates which suggests that these are true freshwater systems. Both the pH and conductivity ranges for group 2 are higher than group 1, which is a further indication that these are evolved systems. The two sites of principal interest to this investigation are Honey Camp Lagoon and the New River Lagoon. The water chemistry for the latter site can be represented by Lamanai. These two sites fall into separate groups and represent the two main types of lakes that have been studied in Belize. This difference is not surprising as the New River Lagoon is a large, open system and Honey Camp Lagoon is much smaller, closed system. This helps to explain why its water chemistry has evolved along the brine pathway. If one excludes the coastal sites (Almond Hill, Northern Lagoon, Southern Lagoon, Progresso Lagoon, Wagner Lagoon, Fabers Lagoon and Jones Lagoon) the range of conductivity between all the sites studied (Figure 5.4) is fairly low between 0.23-2.19 mScm"1. The pH range for all the sites is between 6.61-8.76 which again is not high. The sites that were measured for phosphates and nitrates do not show any sign of nutrient enrichment (Figure 5.4). 5.3 Modern diatoms Diatom samples were collected in conjunction with the water chemistry variables. The methodology involved has been described in Chapter 4. Through studying the modern diatom flora knowledge of species associations and preferences can be established. This will enable a better understanding of both the lake systems of Belize and in particular, improved interpretation of the New River Lagoon fossil record. This work is of particular value due to the paucity of knowledge concerning both the diatom flora of Belize and the lake systems. The data has been analysed through Canoco, which is an ordination programme (see Chapter 4). The main aim of this section is to present the results and preliminary analysis of this work. All the 145 diatom species mentioned are pictured in appendix 2. The total list of the diatom species and their abundance found in the modern samples can be found in appendix 3. The list of diatom samples taken from each water chemistry site can be found in appendix 4. Of particular note are the samples which did not preserve diatoms. These were mainly restricted to plankton and sodium chloride dominated samples. 5.3.1 The modern data set Modern samples were collected from a number of water bodies over the period 1998- 2000 (in the Spring) from a variety of habitats including surface sediments, plankton, reeds, cutting grass and aquatic algae. These were preserved in alcohol and brought back to the UK where they were digested in Hydrochloric acid and Hydrogen Peroxide (Battarbee, 1986). Permanent slides were made using the mountant naphrax and the samples were counted using an Olympus BX50 microscope at xlOOO resolution. This data set is made up of the 31 sites visited and the species counts (up to 400 valves) from the 73 samples in total. A further 25 samples were collected but did not contain diatoms. Two data sets were created and analysed. The first comprises all the species that were encountered at each site. In order to ensure that the associations between the sites, created in the canonical analyses, were firm and not just a function of rare species, a second data set was created which contained only the species that reached 2% or more of a population. These will be referred to as the full and dominant data sets respectively. Figures 5.6 and 5.7 show the first two axes created in DCA. The eigenvalues for the two axes in the full data set are high at 0.7973 and 0.6008. The closeness of these two values suggests that both axes are important in the explanation of the species/site scores. The eigenvalues for the dominant data set are 0.7916 and 0.606. The similarity between the eigenvalues for the two data sets suggests that the removal of the rare species does not strongly influence the amount of variance that the axes explain. This suggests that the rare species are not central to the associations seen in these data sets. The groups formed in the two data sets are also extremely similar. It is clear from the graphs that the sites and species form a gradient along the first axis. 146 This must represent a significant environmental gradient because the axis is very long. Using the dominant data set, the sites form five groups: (Figure 5.7) 1. Southern Lagoon (sed ) (circle) 2. Rio Bravo (sed), Monkey Tail River (stones), Aguacaliente Swamp (pla) and Irish Creek (sed) (filled circle). 3. Progresso Lagoon (sed) and Almond Hill (triangle). 4. Rio Bravo (pla), Monkey Tail River (sed), Irish Creek (pla), Lemonal Creek (pla), Chiwa Lagoon (sed) and Harry Jones Creek (sed) (filled triangle). 5. The fifth group (square) and this contains most of the sites sampled including Crooked Tree Lagoon, Booth River, Kates Lagoon, Lamanai and Honey Camp Lagoon. In the full data set, groups 1 and 2 were not distinct from one another. The removal of the rarer species has therefore emphasised the difference of Southern Lagoon. This is not unexpected as this site has a conductivity of 23.34 mScm"1 which is significantly higher than any of the other sites in group 2. The species that are associated with Southern Lagoon are: Naviculci florinae, Cocconeis placentula var eugylpta and Diploneis parma (circle). These can all be found in brackish conditions, especially Navicula florinae which is a coastal species. There is only one species that is common to the four sites in group 2 (filled circle): Navicula radiosa var tenella. Other common species for these sites include: Cymbella microcephala, Nitzschia palea, Gyrosigma acuminatum, Navicula cuspidata and Schistauron crucicula. These species can all be found in swamp or riverine environments and have wide ranging tolerances to chemical conditions. The sites in this group are in rivers and in southern Belize. These present quite different habitats for diatoms than found in the lake environments of north Belize. The lack of species consensus for these sites is not surprising because the river environment provides a great deal more opportunity for species specialisation than a lake due to 147 an increased number of habitats and constant changes to the ecosystems e.g. to the water flow. The species that separate group 3 (triangle) are low in their occurrence and include an unidentified species and Achnanthes minutissima. The latter species is found in epiphytic habitats, in well-aerated waters and can tolerate concentrated waters (Gasse, 1996). The first species could not be identified because it contains no distinguishing marking which suggests that it is highly dissolved. This suggests that this environment does not present optimum conditions for diatoms. The site Almond Hill (surface mud) which is the outlier to this group. This is because this site has significantly higher sulphate values than any of the other sites sampled (Figure 5.4). The species that are common to group 4 include (filled triangle): Brachysira neoexilis, Brachysira neoexilis var small, Brachysira neoexilis var capitate, Encyonema carina, Nitzschia amphibia and Nitzschia gracilis. Monkey Tail (sed) is dominated by Cymbella mesiana. These species can all be found in lakes and rivers often in the littoral zone. They are fairly widespread in their distribution. Samples that are found within this group are in sites that have already been mentioned in other groups. This is because different habitats were sampled and these must contain sufficiently different species assemblages to separate samples from the same system. Rio Bravo, Monkey Tail, Chiwa and Irish Creek are the sites where this is apparent. This is likely to be because these are (apart from Chiwa) river sites and therefore have a great diversity of ecological niches. The fact that most of the sites sampled occur in group 5 (square) suggests that they must share a number of key characteristics in terms of both the chemical and physical environment. The common species to these sites are: Denticula elegans, Mastogloia smithii var. lacustris, Gomphonema gracile, Brachysira neoexilis var. large and Achnanthes minutissima. The key gradient in this data set is along axis one. Through looking at the water chemistry of the sites which fall into groups 1-5 it is apparent that the axis is 148 representing the evolution from calcium-sulphate dominated systems to sodium- chloride dominated systems or pathway II to IIB according to Eugester and Hardie (1970). One would expect Southern Lagoon to be at the other end of the axis as it is in the same chemical group as Progresso Lagoon. This lagoon has however got extremely high conductivity values and is the most chemically evolved lake that was studied. This suggests that the diatom species present in this lake are quite different to those in the rest of the group and therefore it is quite separate on the graph (Figure 5.7). Most of the sites sampled were present in group 5. The most detailed water chemistry data is only available for the sites sampled in 1999. The 1999 sites, as explained earlier, could also be categorised through pH and conductivity. In order to determine whether the sites that fall into group 5 form a chemical group their pH and conductivity values were analysed: Table 5.1 Site PH Conductivity mScm"1 Kates Lagoon (x) 7.76-8.29 0.23 Chiwa Lagoon (x) 7.21 0.43 Crooked Tree 8.06 0.6 Booth River 6.82 0.92 Lamanai 7.29 1.07 Hillbank 7.88-8.44 1.07-1.29 Outpost 8.06 0.91 Monkey Tail (x) 7.55 2 Botes Lagoon 6.61 0.64 Doubloon Lagoon 8.52 1.48 Cobweb Swamp 7.26 0.77 Laguna Seca 7.77 0.32 Laguna Verde 8.65 0.45 Honey Camp Lagoon (x) 8.08 1.6 (x) denotes the sites where not all of the diatom samples collected, from that area, fell into group 5. According to the 1999 water chemistry data set Booth River, Lamanai, Hillbank and Outpost belonged to the calcium-sulphate group. Cobweb Swamp and Laguna Seca both fit into the pH and conductivity range for the calcium-sulphate group. Botes, Doubloon, Laguna Verde and Monkey Tail appear to be intermediate forms between the calcium-sulphate and the sodium chloride systems. Honey Camp Lagoon is a 149 sodium-chloride dominated system. It is therefore apparent that the sites in group 5 represent a range of water chemistries but the diatom populations are obviously sufficiently similar to bring the sites together. This suggests that the ecological tolerances of the diatom species found are fairly wide. The data set which has been analysed is quite large and in order that all the differences and associations can be fully explored between sites and species it is important to create subsets of this main data set. In the first instance the dominant main data set was separated by habitat creating Figure 5.8 in order to determine whether this variable is influential in creating groups within the data set. The sediment sites are represented by diamonds, filled diamonds are epiphytic and crosses are plankton samples. It is apparent that the sediment and plankton samples are most clearly associated with the environmental gradient represented by axis one as they are stretched out to a greater degree along this axis than the epiphyte sites. The sediment and plankton samples make up groups 2 and 4 and some of group 5 in Figure 5.7. These include sites that are generally in the calcium-sulphate dominated chemistry rather than the more evolved sodium-chloride systems. Groups 2 and 4 are also all either creeks, rivers or swamps. This suggests that in order for a plankton flora to flourish a combination of both chemical and physical factors are needed. 5.3.2 The role of water chemistry In order to develop a clearer idea of the role of water chemistry the diatom samples from the sites sampled for water chemistry in 1999 were analysed. Not all of these sites preserved diatoms but this data set enables a more specific understanding to be gained concerning the relationship of diatom floras and water chemistry. Figure 5.9 shows the sites. This highlights a clear separation of Southern Lagoon (filled circle), Aguacaliente (circle), Progresso (triangle) and Chiwa Lagoons (filled triangle). The species that differentiate those sites are highlighted on Figure 5.10 using the same symbols. 150 The species that differentiate these sites are: 1. Southern Lagoon: Achnanthes exigua, Cocconeis placentula var. euglypta, Navicula florinae and Fragilaria brevistriata. 2. Aguacaliente Swamp: Schistauron crucicula, Nitzschia palea and Achnanthes minutissima. 3. Progresso Lagoon: Fragilaria fasciculata, Denticula elegans and Nitzschia amphibia. 4. Chiwa Lagoon: Brachysira neoexilis var. small, Aulacoseira granulata and Diploneis ovalis. The clustering of the sites does not exactly match the water chemistry groupings highlighted in section 5.3. Aguacaliente, Booth River and Lamanai were all in the same water chemistry group (calcium-sulphate dominated) but the sites did not form a coherent group in the CCA analysis. This highlights that the diatom flora is affected by factors other than water chemistry such as habitat availability. This therefore emphasises the complexity of interpreting the factors that result in particular associations of diatom species. A CCA was also carried out to determine the influence that the specific water chemistry variables had on the site associations seen in the 1999 sites. Figure 5.11 is the biplot of the water chemistry and site data for this reduced data set. Although the eigenvalues for the analysis were high at 0.868 for axis one and 0.814 for axis two the water chemistry arrows are not long. This suggests that there may be other factors (that have not been measured) that contribute to the distribution of the sites. Sodium, Potassium, Chloride and conductivity are all very closely related. Other important variables are calcium and sulphate. These two groups of water chemistry variables match the two groups of lake-type highlighted in the 1999 water chemistry study namely calcium-sulphate and sodium-chloride dominated. 151 5.3.3 Reconstructed conductivity The following table shows the conductivity optima for the key diatom species found in the modern data set. These values are the weighted average based on the three sites where the species predominate. For the purposes of comparison the values reconstructed by Reed (1998a) are shown. Table 5.2 Species Conductivity mScm"1 Belize Conductivity mScm"1 Spain (Reed, 1998a) Mastogloici smithii var. lacustris 0.61 6.17 (3.37-11.33) Denticula elegcins 0.98 2.65 (1.63-4.3) Encyonema carina 0.67 / Achnanthes minutissima 2.80 1.88 (0.87-4.09) Brachysira neoexilis 1.23 1.38 (0.42-4.6) The values seen in Belize are much more similar to the lower end of the tolerance levels found in Spain. This is likely to be a function of the sampling strategy employed in the respective studies. The implications of this are explored in Chapters 6 and 7. 5.4 The New River Lagoon data set In order to gain a greater understanding of the New River Lagoon the samples from this area were analysed separately from the main data set. The axes of this reduced data set explain less variance than in the full data set with the first axis having an eigenvalue of 0.5048 and the second 0.2908. The axes are much shorter than in the main data set. This is not surprising because one would expect there to be less variability within one lagoon than between many lagoons. The sites fall into three groups, which relate to the habitat from which they were sampled (Figures 5.12). The most coherent group of sites are those from epiphytic habitats (filled diamond). There are two outlying sites which are sediment samples from Lamanai and 152 Hillbank. Both these sites were surrounded by the same mix of vegetation types, which were not present at the same level of abundance in the other sites. This highlights the direct influence of specific habitats on diatom populations. The sediment samples (diamond) do not form such a coherent group with there being a slight division between the samples from the main lagoon body and those from the side creeks of the New River Lagoon. This highlights the sensitivity of the New River Lagoon flora to slight changes in habitat. The plankton samples are the crosses. These also do not form a coherent group. This is not surprising because all the samples were taken from fairly near the lagoon edge and not in open water. This decreases the possibility of the flora being composed of true planktonic species and increases the risk of 'contamination' from other habitats. Care needs to be taken with this interpretation because the sampling sites were from shallow water and therefore may not even support a true plankton. Although habitats could not be so readily distinguished in the main data set it is apparent that on a more local scale this is an important factor in differentiating between diatom populations. Specific diatom species have therefore been isolated as indicative of a particular environments in Belize (Figure 5.13). 153 Table 5.3 Habitat Species Epiphyte Achnanthes minutissima Brachysira neoexilis var. large Gomphonema gracile Mastogloia smithii var. lacustris Sediment (outliers) Mastogloia elliptica var. dansei Fragilaria construens Nitzschia amphibia var. rostrata Sediment (Main Lagoon body) Achnanthes exigua Brachysira neoexilis var. capitate Brachysira neoexilis var. large capitate Denticula elegans Encyonema carina Navicula radiosa Nitzschia amphibia Sediment (Side Streams) Brachysira neoexilis var. small Gomphonema intricatum var. vibrio Plankton Brachysira neoexilis Fragilaria ulna Navicula radiosa var. tenella Nitzschia palea Nitzschia gracilis The two species highlighted in bold are very common in a whole range of sites. In general terms these habitat categorisations agree with the published literature (e.g. Gasse 1986; Patrick and Reimer, 1966). Some of the species, which were found in the plankton of the New River Lagoon and in the literature (Brachysira neoexilis, Synedra ulna and Nitzschia gracilis) are noted to be shallow plankton forms. The other species that were found in the plankton of the New River Lagoon are regarded as epiphytic or littoral species. This implies that these species have detached from these environments and have become part of the marginal lake water environment. The species found in the side streams are also not completely in agreement with the literature. This is not surprising because this environment is transitional being at the edge of the lake system and the beginning of a creek. The species found here are also mainly epiphytic species. Caution needs to be exercised when interpreting the sediment samples as these tend to be a reflection of all the environments that have an input into the system. One key point which the habitat divisions clearly show are the 154 differences between the nominate forms and their varieties. Brachysira neoexilis, Nitzschia amphibia and Navicula radiosa were all found in different habitats from their varieties. This is evidence that it is important to employ careful taxonomy to ensure that ecological information is not lost. The discovery of these variations will enable a much more informed interpretation of the fossil sequences from this area. 5.5 The role of habitat In order to investigate the role of habitat further the main data set was separated into the three habitat categories and these were analysed separately. The distribution of the epiphytic samples in the dominant data set is predominantly explained by the first axis which has an eigenvalue of 0.7327 (this is significantly higher than axis two) (Figure 5.14). The values for the full data set are very similar. The first group of outliers is composed of Almond Hill and Progresso Lagoons (circle). The species that differentiate these sites are Fragilaria fasciculata, Brachysira neoexilis and Denticula elegans (Figure 5.15). These species are not regarded as epiphytic species according to the New River Lagoon data set. Their inclusion suggests that these species can be found in different types of environment and the conditions/ habitats available in Almond Hill and Progresso enable these species to be present in epiphytic habitats. The conductivity values in Almond Hill and Progresso Lagoons are high compared to the other sites that preserved diatoms. These species could have a competitive advantage over others due to an ability to withstand higher conductivity levels and therefore they flourish. The samples were studied under the light microscope after they had been treated in acid (see Chapter 4 for details). This meant that it could not be determined whether the diatoms being studied were actually attached to the sampled vegetation i.e. it is possible that the samples could be contaminated with diatoms from other habitats. The rest of the epiphyte samples form a group (filled circle). This is likely to be a function of the similarity of vegetation types sampled. 155 Plankton samples were analysed and it is the first axis of the dominant plankton data set that explains most of the variance having an eigenvalue of 0.6919 (Figure 5.16). This is the same as in the full data set. The first axis is also very long which suggests that this is a distinct environmental gradient that the samples are responding to. The New River Lagoon is an open system and therefore it is likely that the system does not support what would be regarded as true planktonic species. The plankton may be composed of species that are faculatively planktonic or indeed the plankton may be dominated by other forms of algae. There are no distinct clusters in this data set. The site which is most different is Aguacaliente (circle). This sample has a very limited flora and is entirely dominated by Nitzschici palea which is not as common in any other site (Figure 5.17). Other species present include: Nitzschia calida and Schistauron crucicula. These are found in medium conductivity waters in epiphytic environments. This sample was from very shallow water in a vegetated zone so this might be expected. Gasse (1986) found Nitzschia palea, when in large numbers, an indicator of eutrophic conditions. Aguacaliente Swamp was the home for a number of cattle. These would have contributed to the creation of a eutrophic environment. This was not picked up in the water chemistry results because nutrients were not measured at this site. The sediment samples are generally stretched out along the first axis which has an eigenvalue of 0.82. The first axis is long which indicates that this represents a strong environmental gradient. The second axis also has a high eigenvalue at 0.68. This is also a long axis and is therefore a significant environmental gradient that the samples are responding to. The sites (Figure 5.18) are stretched out along the first axis with three obvious exceptions: Southern Lagoon, Monkey tail and Almond Hill (triangle). It is not possible to differentiate the species that are causing these sites to separate because there is such a large degree of scatter within the species data (Figure 5.19). Within the rest of the sites it is apparent that the sites that are towards the top end of axis one are more coherent than those at the lower end of axis one. These sites are shown by closed circles and include Hillbank, Kates Lagoon, Lamanai, Cobweb Swamp, Botes Lagoon and Outpost. The species that are associated with these sites include: Brachysira neoexilis, Navicula radiosa, Mastogloia elliptica var. dansei, 156 Denticula elegans and Nitzschia amphibia var. rostrata (Figure 5.19). These species are all found in the sediments groups in the New River Lagoon. The rest of the sites are shown by open circles. There is quite a degree of scatter within the species data for these sites. It is apparent from this investigation that habitat does play a key role in differentiating between species assemblages. The sampling system needs to be carefully managed so that each habitat is visited to enable the differences/similarities to be clearly defined. One way in which epiphyte samples could be categorised further would be through the identification of the plant samples to which they are attached. This data set would be improved with a much more detailed survey of the lagoons plankton to determine whether indeed it is comprised of facultative species or is dominated by other forms of algae. 5.6 Summary The analysis of the modern diatoms of Belize allows the following conclusions to be made: 1. The sites which were sampled contained a range of conductivities from marine to freshwater. Many of the former sites did not preserve diatoms within them. 2. The samples from lakes in northern Belize show a uniformity to the diatom species that are found. 3. The sites that were consistently different were: Southern (sed), Rio Bravo (sed), Aguacaliente (pla), Irish Creek (sed), Almond Hill (epi), Progresso (epi) and Monkey Tail (sed). 4. It is apparent from the 1999 water chemistry data set that the lagoons sampled can be approximately divided into three types: freshwater, calcium-sulphate and sodium-chloride dominated. 5. In order that a much more detailed understanding of the relationship between diatom species and water chemistry can be established for Belize a much more extensive collection of water chemistry data needs to be undertaken. 157 6. A checklist of species and the habitat in which they are found has been made. The coherence between the Belizian samples and the literature implies that, at least for habitat, there is a degree of uniformity over space. 7. Samples from the New River Lagoon highlight the importance of habitat in differentiating between diatom samples. 5.7 The relationship between the fossil and modern data sets One of the central reasons behind collecting the modern diatom samples was to enable a greater understanding of the fossil data. In order to achieve this aim a combined data set of the two long cores analysed for diatoms (Lamanai 1999 and Hillbank 1998 (Chapters 6 and 7)) and the modern diatom samples collected from all over Belize was created. In this section the complete data set refers to the data set comprising all the modern and core samples. The New River Lagoon data set is made up of the core data and only the modern samples collected from the New River Lagoon. 5.7.1 The complete data set The eigenvalues for this data set are 0.4992 and 0.3915 (Figure 5.20). These are fairly low which suggests that these axes do not explain all the variance in the data set and therefore the associations between the sites may not be stable. The three data sets all plot in different zones which suggests that they are not related. The modern (triangles) and Hillbank (filled circle) data form a continuum and the Lamanai (circle) data is much less coherent. This suggests that there has been much more variability in the Lamanai species assemblages through time. Samples from the top of the Hillbank sequence are slightly separated from the rest. It is these depths which are most similar to the Lamanai data. This is due to the dominance of Denticula elegans in these levels. 158 5.7.2 The New River Lagoon data set The two fossil data sets are both from the New River Lagoon. The modern diatom populations from this lagoon are therefore the most relevant to compare to the core material. Figure 5.21 highlights the distinct differences between the three data sets and the eigenvalues for the axes are low at 0.3169 and 0.2328. One of the Hillbank modern samples is found amongst the Hillbank core data. This sample was from deeper water than the core. The modern sample, which is most different from all the samples analysed, is a plankton sample from Lemonal stream that flows into the New River Lagoon and is far away from any other site sampled. As there is only one overlapping site this limits the inferences which can be made from the modern data set. This highlights the necessity of multiple coring and the collection of a wide range of modern samples so that a large system, such as the New River Lagoon, can be truly understood. Despite extensive sampling this may still not be possible because the modern environment may be very different from that experienced in the past. 5.7.3 The dominant diatom species data set In order to gain a clearer understanding of the main species which have driven the differences between the sites, a data set was created which contained species that were present at the 5% level or higher in either Lamanai, Hillbank or the modern data set (Figure 5.22). Although this resulted in the three data sets being much less coherent, the modern data set is more closely related to the two cores when just the dominant species are considered. The eigenvalues of the axes are 0.506 and 0.3661 which is about the same as the main data set. Sites which occur in the same zone as the Hillbank samples are from Kates, Progresso, Doubloon, Honey Camp and the New River Lagoon. Samples from Aguacaliente, Progresso, Monkey Tail and Irish Creek separated from the main cluster of modern samples. These have been stretched along the main axis in a similar manner to the Lamanai samples. This is unexpected as these are sites which were identified as outliers in section 5.6 and may be a 159 function of the limited number of species used in this data set which may have forced associations between sites. Figure 5.23 allows an understanding to be developed as to which species are driving the differences between the samples. Modern: Achnanthes minutissima, Brachysira neoexilis and Encyonema carina. The first two species are littoral/epiphytic species that have wide tolerances. The latter is present in high numbers throughout the data set. Hillbank: Navicula radiosa and Denticula elegans. These are littoral/bottom mud/marsh species. Lamanai: Cyclotella distinguenda, Nitzschia amphibia, Achnanthes exigua and Denticula tenuis. The first two species are faculatively planktonic/littoral species and Achnanthes exigua is a freshwater littoral species. Denticula tenuis is only found at depth in Lamanai and is not a common species. It is thought to be found in waters of medium conductivities in littoral environments. Mastogloia smithii var. lacustris is in the centre of the diagram suggesting that it is common to all the samples. The species in Hillbank and Lamanai are different, but they represent the same type of environments. Lamanai has a much more diverse flora in the fossil data and this is reflected in its more scattered distribution in the Canoco analysis. These results demonstrate that there are tangible differences between Hillbank and Lamanai. What are not clear are the forcing factors which have resulted in these variations. The most likely mechanisms are the amount and type of habitat availability because water chemistry values are very similar in both sites. It also possible that the forcing factor could be a variable that was not measured 160 5.7.4 The New River Lagoon dominant diatom species The same analysis was then undertaken on the New River Lagoon data set. This shows the same pattern of sample distribution as the full data set. The eigenvalues for the axes are 0.5131 and 0.332 (Figure 5.24). The slightly higher eigenvalues for this reduced New River Lagoon data set as compared to the full data set suggests that it is the rarer species outside the New River Lagoon in the modern environment that are decreasing the associations between sites. The modern samples themselves show a degree of spread, which is probably related to the different habitats from which the material was collected. The close relationship between the Hillbank and modern samples could be related to the bias in modern samples towards those from the Hillbank area. The outlying samples from the modern cluster are the ones from Lamanai. The same pattern of species occurs in the New River Lagoon data set to the full data set (Figure 5.25). This means that these species are the key species for explaining distribution in both modern day and historical Belize. 5.7.5 Implications The modern assemblages in the New River Lagoon do not appear to be related to the fossil samples studied. Analysis in Chapters 6 and 7 reveals that the top sediments in the two Livingstone cores, do not represent modern day conditions due to the unconsolidated sediment found in the lagoon. Kullenberg cores were also taken from Lamanai and Hillbank. These capture the sediment/water interface and therefore are more likely to represent modern day conditions onwards. These, however, do not have diatoms preserved within them which may suggest that the Kullenberg cores have not successfully collected the most recent sediments because the modern samples were found to contain diatoms. This suggests that the following changes have occurred through time: 1. Diatoms are preserved in the modern day environment although not in the plankton samples taken. 2. Conditions are currently not suitable for diatoms to be preserved in any more than the most recent sediments. This could be due to the mixing of the top 161 sediments which would encourage processes such as dissolution and fragmentation. 3. The system shifted back to 'preserving conditions' at some point in the past. The system has changed though because if conditions were the same, then these core samples would have plotted amongst the modern samples. 4. Through time the systems have moved between preservation and dissolution being the dominant processes. Conclusions can also be drawn concerning the relationship between Hillbank and Lamanai: 1. The degree of sample cluster in Hillbank is much greater than in Lamanai. This implies that the species assemblages have changed much less through time in Hillbank than in Lamanai. One reason for this might be that the cores are located on opposite sides of the lagoon with Lamanai being located by forest and Hillbank by marsh. Through time it is likely that the forest system has undergone changes that are both natural and human induced. Marsh environments are much harder to manipulate and therefore likely to have a more stable diatom population. 2. Although many of the main species are common to both cores, the mix of assemblages and the dominance of these main species differ enough to separate the sequences. This could also be a function of Hillbank's more stable past than Lamanai. 3. This study highlights the need for multiple cores in a basin as it cannot be assumed, especially in a lagoon as large as the New River, that one core will provide an accurate reflection of the changes which it has undergone through time. 5.8 Summary Through the analysis of the main data set it is apparent that there are sites which are outliers. Almond Hill, Progresso, Monkey Tail and Aguacaliente are four of those sites. There is no apparent relationship in the main data set between habitat and site 162 groupings. Through the analysis of the New River Lagoon it was possible to assign habitats to the species due to the clear divisions of the sites sampled. This was possible because samples were collected from different habitats in the same area allowing differences in diatom species to be more keenly recognised. The collection of water chemistry data has enabled some initial categorisations of the site which is helpful in terms of understanding the general limnology of north Belize. 5.9 Mastogloia smithii var. lacustris: A study in Belize The issues raised in Chapter 4 and in the previous analysis of the modern data set prompted a study of Mastogloia smithii var. lacustris which is one of the diatoms species common to the samples studied both in the modern and fossil environment. This investigation had three aims: 1. To determine the consistency of diatom characteristics within a variety of habitats and geographical locations. 2. To determine whether these characteristics match those cited in the literature (Hustedt, 1959; Patrick and Reimer, 1966; Krammer and Lange-Bertalot, 1986) and in Van Heurck's type material (N 47, BM 26358). 3. To develop a greater understanding of diatom taxonomy in northern Belize. Mastogloia smithii var. lacustris is a dominant form in the New River Lagoon reaching up to 36% of the individuals from a sampled population. This lagoon has been investigated in terms of its modern diatom flora (this chapter) and its palaeoenvironmental history (Chapters 6 and 7). These two studies are linked because through the unravelling of the species modern characteristics a greater understanding of the fossil record will be developed. Mastogloia smithii var. lacustris according to the literature is found in brackish waters in coastal areas and frequently in the littoral zone of freshwater systems (Krammer and Lange-Bertalot, 1986). Patrick and Reimer (1966) found it most commonly in freshwater lakes as a shore form; sometimes in springs and occasionally in slightly brackish waters. The key difference between the ecology of Mastogloia smithii var. lacustris and 163 Mastogloia smithii appears to be that the variety is found only in the littoral and in waters of lower conductivity than the nominate form (Hustedt 1959). 5.9.1 Methods Width/Striae measurements were taken on 40 diatoms from each site and on type slides from Mastogloia smithii var. lacustris (Van Heurck N 47 26358) and Mastogloia smithii (Coll WM Smith BM 24346). The width/striae ratio does not vary through a diatom's life cycle (D. Mann pers.com 2001) and therefore it is a good measure of difference between populations. Mann-Whitney tests were performed in order to determine whether there were significant differences between the groups measured. 5.9.2 Results Thirteen samples of Mastogloia smithii var. lacustris were studied. Each sample was tested against the remaining twelve. Groups of sites were created that were deemed by the Mann-Whitney test to be not significantly different from one another at the 95% level. The three tables created highlight: 1. The characteristics of the original type material of the species Mastogloia smithii and Mastogloia smithii var. lacustris (table 5.1). 2. The characteristics published in the literature (table 5.2). 3. The differences in width/striae characteristics exhibited in the modern habitats of New River Lagoon, Belize (table 5.3). Table 5.4- Type Material (illustrated in Figure 5.26) Authority Mean Width 00 Mean Striae (in lOp) Mean Ratio M smithii var. lacustris (VH) 10 20 0.5 M smithii (Smith) 13 21 0.61 164 Table 5.5 - Published Descriptions Authority Width (p) Striae (in lOp) Ratio Range Hustedt (1959) 8-14 15-18 0.53-0.78 Patrick and Reimer (1966) 8-11 15-16 0.53-0.69 Krammer and Lange-Bertalot (1986) 8-14 15-18 0.53-0.78 Table 5.6 - Belizian Modern Samples (illustrated in Figure 5.27) Site Mean Ratio Mann-Whitney Outpost Reeds 0.43 Group 1 Outpost Flat Grass 0.43 Group 1 Outpost Sediment 0.52 Group 2 Outpost Aquatic Algae 0.52 Group 2 HB 3 Reeds 0.45 Group 3 HB 3 Aquatic Algae 0.48 Group 3 HB 3 Sediment 0.5 Group 4 HB 1 Algae 0.43 Group 5 HB1 Reeds 0.45 Group 5 HB1 Sediment 0.45 Group 5 HB 2 Sediment 0.44 Group 5 HB 2 Reeds Sediment 0.45 Group 5 HB 2 Cutting Grass 0.45 Group 5 These sites are shown on Figures 5.2 and 5.3. 5.9.3 Discussion From studying these tables the following points can be immediately made: 1. The sediment and reed populations of Mastogloia smithii var. lacustris in the sites Outpost and HB 3 can be differentiated from one another. These are illustrated in appendix 5. 2. This relationship does not hold for sites HB 1 and HB 2. 165 3. The width/striae ratios of the New River Lagoon populations do not fall within the ranges published in the literature for this species. 4. The HB 3 sediment site is the only one which matches the Van Heurck type material. The New River Lagoon is a large freshwater system which eventually connects to the sea via the New River. The west shore is dominated by forest vegetation (Outpost and HB 3) and the east shore is dominated by marshland (HB 1 and 2). This is shown clearly in Figure 3.3. The key differences between the sites HB 1/2 and Outpost/HB 3 are the water depth and the role of the marshland. HB 1/ 2 are shallow water sites and therefore the chance of mixing between different habitats is increased. The marshland adjacent to the southwest shore (where HB 1/2 are located) is an integral part of the lake system due to the seasonal flooding of this area during the wet season. With no barriers to inhibit movement between the lake and the marsh the dominant input of diatoms are likely to be epiphytic species from the marsh environment. Thus, the Mastogloia smithii var. lacustris that enter from the marsh would be from a 'reed' population. Both the sediment and reed habitats sampled in HB 1 and 2 are in the range of the ratios exhibited by Outpost and HB 3 reed populations. The distinction in Outpost and HB 3 is clearly between a reed type and a sediment type. If one takes the mean measurements of the reed populations and the mean measurements of the sediment populations from these two sites, the following mean ratios are produced: 1. Reed: 0.44 (8.5/20) 2. Sediment: 0.53 (10/19) These ratios are clearly different suggesting that the two populations which have been delimited in this investigation, are distinct from one another. 166 A sediment core was collected in 1999 from Lamanai (Figure 1.1). The Mastogloia smithii var. lacustris, that were found in this sequence, fall into two width/striae ratio categories which equate to the reed and sediment populations found in the modern environment. This suggests that two populations of Mastogloia smithii var. lacustris have been present in the New River Lagoon for at least the last 3440 l4C years BP. A sediment core has also been taken from the east shore (Hillbank 1998) and the distinction between the two types of Mastogloia smithii var. lacustris, as one would expect, is not present. The ability to differentiate between two types of Mastogloia smithii var. lacustris provides additional insight in to the environmental changes that have occurred at the Lamanai site because it enables inferences to be made concerning the predominance of reeds in the vicinity of the site. The second aim of the investigation was to relate the characteristics of the populations found in the New River Lagoon to those cited in the literature (Flustedt (1959); Patrick and Reimer (1966); Krammer and Lange-Bertalot (1986)). The characteristics of the modern diatoms from Belize and the literature do not overlap. The reed population is distinctly different, but the sediment population is closer to the smaller end of the range cited. This raises two points. Firstly that the sediment population in the New River Lagoon is likely to be the 'true' Mastogloia smithii var. lacustris and the reed population is the evolved population. This line of argument could be followed up with the rationale that the reed population is specific to either the New River Lagoon or Belize. This cannot be substantiated because the material from which the literature characteristics were derived is unknown and it could be the case that it was just sediment populations that were studied. If the sites were visited again and sampled for reeds, the same distinction as found in Belize could well be present. Thus, further investigation of the original sites and other geographical locations is needed to determine how widespread the reed type is. The third comparison was with the type material. The modern sediment populations from Belize are of the same order of magnitude as the type material. No other information is available on the environment from which these type samples were taken and therefore no further inferences can be made. The ratio of the Mastogloia 167 smithii type material falls within the range for Mastogloia smithii var. lacustris characteristics cited in the literature. This suggests that a careful study of the reasons behind the division of Mastogloia smithii var. lacustris and Mastogloia smithii should be carried out. The question, which has not been answered by this investigation, is why such a difference would evolve? The type of variation between the 'reed' and 'sediment' population is of the same order of magnitude as for example two demes of Sellaphora pupula (David Mann pers com. 2001). This suggests that the explanation behind the two different populations of Mastogloia smithii var. lacustris may be due to a reproductive barrier; a genetic difference or indeed they may even be different taxa. If there are no barriers to sexual reproduction then the discontinuity between two forms may represent different episodes of sexual reproduction. Laboratory controlled experiments of modern populations would be needed to establish the reasons behind the different morphologies. There is evidence that many species vary in the same way as Sellaphora pupula (Mann, 1989) with species being divided into two or more morphologically distinct forms which may or may not be found in the same geographical area. An example of this is the work by Knudson (1953) who found three morphotypes of Tabellaria flocculosa coexisting, two in the plankton and one in the epiphyte. Mann and Droop (1996) state that the significance of minor variations in shape and form in unicellular algae is unknown. This study highlights that one reason for such difference may indeed be habitat preference. To conclude, this study has explored variation on the following scales: 1. Variation between samples. 2. Variation between habitats. 3. Geographical variation within one lake basin. 4. Long term variation. Inferences have been made concerning one species and its division into two distinct populations. Width/striae measurements were taken to represent the gross 168 morphology of the specimens. This ratio does not vary through the life cycle and thus the differences between the two populations can be deemed to be stable. The two populations are related to habitat with one being found in reed environments and the other in the sediment. This relationship does not hold for the entire lagoon highlighting the need for surveys of lake systems to be extensive. The characteristics of both populations do not match the literature or type material. It is the reed population which is most distinctive and it is not known whether this type of environment was sampled when creating the characteristics cited in the literature. It cannot therefore be concluded that this reed population is unique to Belize. What it does highlight is that the morphology of species may not be stable within different habitats or geographical areas. The concept of restricted geographical distribution of diatom species can occur on a number of different scales from being only present in one lake system up to presence in one continent. This can only be resolved if large-scale studies using a narrow species concept are undertaken. This study alludes to the fact that diatom species are not stable in their characteristics and thus to gain a true understanding of a species studies need to be both intensive and extensive. Diatom sample collection needs to be coupled with the collection of water chemistry and habitat information. This work would have a great deal of benefit to palaeoenvironmental studies, as it would ensure that accurate ecological reconstructions are being made. This study also highlights the need for a more specific species concept than exhibited in the Krammer and Lange-Bertalot series (1986, 1988, 1991a, 1991b). The existence of more than one morphological type is important to stress if one is concerned about the diversity of flora at a site. Such diversity will be underrepresented if the types are not considered. 5.10 Conclusions This chapter provides the first analysis of modern diatom populations in Belize. Through the work which has been undertaken knowledge has been gained concerning species assemblages, the controlling factors behind these, the 169 relationships between different lagoon systems, the connection between the modern and fossil environment and the specific factors which influence one diatom species in particular. The conclusions which have been reached are subject to modification due to the need to sample a wider variety of sites, habitats and a more through collection of water chemistry variables. 5.11 The Modern Isotope Results The following table compares the results gained from Belize with values that have been published in the literature for sites in the surrounding area. Table 5.7 Lake Authority Water Type 5I80 Belize Honey Camp Lakewater 2.91 Lagoon 5D 15+/- 0.5 New River Lagoon Lakewater -0.97 (Hillbank) 5D-5.2+/- 1.5 Yucatan Peninsula Lake Chichcanab Covich and Stuiver Lakewater 4 1974 Spring -4.7 Punta Laguna Curtis et al., 1996 Groundwater -3.92 Lakewater 0.93 Coba Whitmore et al., 1996 Lakewater 1.18 San Jose Chulchaca Whitmort et al., 1996 Lakewater 0.5 Sayaucil Whitmore et al., 1996 Lakewater 5.3 Lake Chichcanab Hodell et al., 1995 Lakewater 3.24 Guatemala Lake Peten-Itza Curtis et al., 1998 Lakewater 2.62 Lake Yaxha Curtis et al., 1998 Lakewater 2.33 Petenxil Curtis et al., 1998 Lakewater 2.71 Quexil Curtis et al., 1998 Lakewater 3.45 Paxcaman Curtis et al., 1998 Lakewater 3.21 Salpeten Curtis et al., 1998 Lakewater 3.52 Sacnab Curtis et al., 1998 Lakewater 3.85 Macanche Curtis et al., 1998 Lakewater 3.08 Regional Rozanski et al., 1993 Rainwater -4.0 Yucatan Peninsula Punta Laguna Curtis et al., 1996 Rainwater -3.91 Lake Chichcanab Covich and Stuiver, Rainwater -6.1 1974 Guatemala Peten-Itza Curtis et al., 1998 Rainwater -0.98 170 When liquid water and water vapour are in equilibrium, the vapour is isotopically lighter with respect to both D/H and l80/160 than the liquid. The isotopic composition of seawater, by definition of the SMOW scale is 0%c for both 8I80 and 8D. The 8D and 8i80 values in precipitation and hence freshwaters generally plot close to a straight line 8D = 8I80+10. The position along this line (the meteoric water line) of a particular rainfall depends primarily on the amount of precipitation that has occurred between the time the air mass left the ocean and the time of the particular rainfall. With respect to the published data it is apparent that Honey Camp Lagoon is very similar to Lake Peten-Itza and Lake Petenxil, whereas the New River Lagoon equates to rainwater measured at Lake Peten-Itza. This is extremely interesting because it highlights two key points. Firstly that the systems in Belize are more similar to those in Guatemala than those in the Mexican Yucatan Peninsula. This is most likely to be due to the similarity in the climate of Belize and Guatemala. Secondly the results from the New River Lagoon suggest that rainwater enters this system and the levels of evaporation are not strong enough to enhance its isotopic signature. This is likely to be as a result of its open nature. Figure 5.28 contains data from Mexico and Cuba (Metcalfe, unpublished data). The two results from Belize fall exactly on the Cuban trendline. This suggests that these systems have very similar controlling mechanisms. This is not surprising due to both areas being under the influence of the Caribbean Sea and underlain by carbonate geology. The comparison of modern isotope data enables a much greater idea of regional dynamics to be developed. This chapter has dealt with the modern results from Belize. The next three chapters will discuss the results from the sediment cores collected. 171 Hillbank, New River Lagoon Basemap: DOS Sheet IS, Edition S-GSGS (1993), Scale 1:50,000 Projection: Universal Transverse Mercator (Zone 16) Figure 5.2 Modern 1 - Hillbank 1 2000 2 - Hillbank 2 2000 2B - Hillbank 2B 2000 3 - Hillbank 3 2000 C - Centre Hillbank 1998 F - Far side Hillbank 1998 N - Near side Hillbank 1998 Modified from Murray (unpub.) Lamanai, New River Lagoon Basemap: DOS Sheets 15, Edition 5-GSGS (1993) & 10, Edition 5-GSGS (1994), Scale 1:50,000 Projection: Universal Transverse Mercator (Zone 16) Figure 5.3 Modern Sampling sites 2000: Modified from Murray (unpub.) R and L Figure5.4 Sites Year Calcium Magnesium Sodium Potassium Chloride Sulphate Bicarbonate pH ConductivitymS m•1 Phosphate Nitrate Silica BluewaterCreek 2000 / 6.42 / 4.16 6.45 7.56 2.19 0.2 0.04 25.8 BoothRiver 1999 2.43 1.81 0.51 0.01 0.12 6.68 2.75 6.82 0.92 / / / BotesLagoon 2000 / 0.86 / 5 1.60 6.61 0.64 0.16 0.024 8 ChiwaLagoon 1999 0.88 1.39 1.26 0.06 0.39 0.35 1.77 7.21 0.43 / / / CobwebSwamp 2000 / 1.28 / 3 2.50 7.26 0.77 0.24 0.12 8 CrookedTreLagoon 1999 0.68 0.44 0.52 0.02 0.13 1.21 1.19 8.06 0.6 / / / DoubloonLagoo 2000 / 3.95 / 4.16 1.80 8.52 1.48 0.03 0 3.2 Caledonia,N wRiverL goon 2000 / 6.42 / 4.16 3.46 7.65 3.54 0.14 0.02 14.8 IrishCreek 1998 / / / 0 / 6.39 8.76 0.86 0 0.0019356 4 LemonalCreek.N wRiv rag on 1998 / / / 0 / 2.16 / 1.07 0 0.00009678 4 RamgoatCreek,N wiv rL go n 1998 / / / 0 / 5.20 8.44 1.93 0 0.00004839 4 HarryJonesCre k 1998 / / / 0 / 3.36 8.22 1.37 0 0 4 HarryJonesCre k(side) 1998 / / / 0 / 1.24 8.7 1.31 0 0.00004839 4 Hillbank1,NewRiverLagoon 2000 / 2.80 / / 4.16 3.10 7.88 1.19 0.18 0.003 14.2 Hillbank2,NewRiverLagoon 2000 / 2.14 / / 3.37 3.94 8.12 1.29 0.1 0.009 14.4 Hillbankcentre.N wRiverLagoon 1998 / / / 0 / 2.96 / 1.12 0.006318 0 4 Hillbankfarside,NewRiverL goon 1998 / / / 1.81 / 2.96 8.44 1.15 0.0069498 0.00014517 4 Hillbanknearside,NewRiveLagooi 1998 / / / 0 / 2.96 / 1.12 0 0 4 Lamanair ft.NewRiveragoon 2000 / 2.30 / / 4.16 2.90 8.06 0.91 0.24 0.021 12.6 Lamanai,NewRivergoon 1999 1.79 1.17 0.4 0.07 0.23 4.83 1.31 7.29 1.07 / / / HoneyCampLagoon 1999 1.29 3.07 6.38 0.19 6.21 7.02 2.05 8.3 1.5 / / / HoneyCampLagoon 2000 / 5.92 / / 4.16 3.40 8.08 1.6 0.36 0.15 27.8 HoneyCampSpring 2000 / 5.26 / / 4.16 2.60 8.19 1.59 0.24 0.047 24 KatesLagoon 1999 0.82 0.3 0.58 0.03 0.14 0.15 0.98 8.29 0.23 / / / KatesLagoon 2000 / 1.03 / / 5 1.90 7.76 0.27 0.22 0.094 13.44 LagunaSec 2000 / 1.19 / / 3 1.90 7.77 0.32 0.1 0.003 16 LagunaVerde 2000 / 2.14 / / 3 2.90 8.65 0.45 0.16 0.17 16 SmallCrocLag on 1999 1.74 0.98 0.68 0.04 0.29 3.16 1.77 7.04 0.6 / / / AlmondHillLag on 2000 / 5.92 / / / 185 0 7.4 3.82 0.12 0.018 0 JonesLagoon 1999 3.85 23.91 180.69 4.08 212.42 29.23 1.48 7.92 23.34 / / / NorthernLagoo 1999 1.21 4.48 35.38 0.82 45.14 5.87 1.56 8.26 5.34 / / / ProgressoLa oon 1999 2.18 3.41 14.65 0.36 17.11 9.29 2.21 8.1 7.25 / / / SouthernLagoo 1999 3.54 22.37 193.18 4.44 162.49 23.69 2.67 8.35 23.3 / / / WagnerLagoon 1999 1.43 6.39 50.14 1.14 63 8.24 1.77 8.1 7.25 / / / FabersL goon 1999 3.16 17.56 146.49 3.32 46.08 17.97 1.84 7.95 16.99 / / / MonkeyTailRiver 1998 / / / / 0 / 0.84 7.55 2 0 0 4 AguacalienteSw mp 1999 4.32 3.49 0.27 0.03 0.06 19.57 0.92 8.1 1.63 / / / /signifiesnodata AllDataIsnMEQ Figure 5.5 Water Chemistry Data (1999) Ca 0 10 20 30 40 50 60 70 80 90 100 CI Figure 5.6 6 5 - 4 3 ^ O I- o 0 « 1 - >5 < n The Main Data Set: Dominant Species O Afl(A A^ A A °o° „• •• -1 -1 -2 - -3 - □ 6 ) 1 2 3 4 8 6 7 □ □ Axis One Figure 5.7 Key: Group 1: Southern Lagoon (sed ) Navicula florinae, Cocconeis placentula var. eugylpta and Diploneis parma (circle) Group 2. Rio Bravo (sed), Monkey Tail River (stones), Aguacaliente Swamp (pla) and Irish Creek (sed) Navicula radiosa var. tenella. Cymbetta microcephala, Nitzschia palea, Gyrosigma acuminatum, Navicula cuspidata and Schistauron crucicula. (filled circle). Group 3. Progresso Lagoon (sed) and Almond Hill. Achnanthes minutissima (triangle). Group 4. Rio Bravo (pla), Monkey Tail River (sed), Irish Creek (pla), Lemonal Creek (pla), Chiwa Lagoon (sed) and Harry Jones Creek (sed). Brachysira neoexilis, Brachysira neoexilis var. small, Brachysira neoexilis var. capitate, Encyonema carina, Nitzschia amphibia and Nitzschia gracilis, (filled triangle). Group 5. This contains most of the sites sampled including Crooked Tree Lagoon, Booth River, Kates Lagoon, Lamanai and Honey Camp Lagoon. Denticula elegans, Mastogloia smithii var. lacustris, Gomphonema gracile, Brachysira neoexilis var. large (square). Figure 5.8 The Dominant Data Set 3 Axis One Key: Sediment (diamonds), Epiphyte (filled diamonds), Plankton (crosses) This graph shows the distribution of samples collected in Belize highlighting the differences between diatom habitats. Figure 5.9 3 2.5 -I 2 1.5 - 1 - 0.5 - 0 0 ♦ ♦ ♦ Sites 1999 3 4 Axis One Figure 5.10 4.5 3.5 2.5 1$ 0.5 Diatom species (1999) -0.5 <() -1.5 mar t 4 A A ♦ * ♦ A %/ o <9o 4* Axis One Key: Water chemistry 1999 data set. 1. Southern Lagoon (filled circle): Achnanthes exigua, Cocconeis placentula var. euglypta, Navicula florinae and Fragilaria brevistriata. 2. Aguacaliente Swamp (circle): Schistauron crucicula, Nitzschia palea and Achnanthes minutissima. 3. Progresso Lagoon (triangle): Fragilaria fasciculata, Denticula elegans and Nitzschia amphibia. 4. Chiwa Lagoon (filled triangle): Brachysira neoexilis var small, Aulacoseira granulata and Diploneis ovalis. Figure 5.11 Water Chemistry 1999 Axis One -Ca Na -CI - HC03+C03 -Mg K -S04 -pH Water Chemistry 1999 1999 Water Chemistry data. The separation of the water chemistry variables match the two groups of lake- type highlighted in Figure 5.5. The second graph has had the outlying sites removed so that the distribution of the arrows can be seen more clearly. Figure 5.12 New River Lagoon Sites 3 2.5 2 1.5 - 1 - 0.5 - 0 0 0.5 1.5 2 Axis One 2.5 3.5 Figure 5.13 New River Lagoon Diatom Species I x * x aW © © ♦ ©♦©♦*© ^«©'©♦ ♦ ♦ -2 -1 % *©©v -1 - Axis One Key: Epiphyte (filled diamond), sediment (diamond), plankton (crosses). This figure highlights the clear differences between habitat in the New River Lagoon. The species are noted in section 5.4. Figure 5.14 Epiphyte Dominant Sites 2.5 -| 2 • o 1.5 • X 1 - • 0.5 ^ % 0 - a 0 0.5 1 1.5 2 2.5 3 Axis One Figure 5.15 4 V 3 • • •• 2 - 1 II- » Epiphyte Dominant Species . •. • . ° o • • -1 -0.5 -1 - -2 -3 - 0.5 1 • 1.5 2 2.5 3 3.5 J ' Axis One Key: Almond Hill and Progresso Lagoons (open circle). The species that differentiate these sites are Fragilaria fasciculata, Brachysira neoexilis and Denticula elegans. These two sites had much higher conductivity values than the other sites that preserved diatoms. Figure 5.16 Dominant Plankton Sites 2 1.8 1.6 1.4 1.2 1 0.8 0.6 - 0.4 - 0.2 0 0 3 Axis One Figure 5.17 Dominant Plankton Species I X < 5 4 3 -\ 2 o1 « -0- • • / * -1 6 2 • Axis One Key: Aguacaliente Swamp (open circle) This site is dominated by Nitzschia palea. The lack of clustering in this data set could be due to the fact that the samples were not composed of a true plankton flora. Figure 5.18 Sediment Dominant Sites o o /• o 3 Axis One Figure 5.19 6 1 5 - 4 - o 3 5 o2 .<2 1 - X < n Sediment Dominant Species O o 00 e ° ° ° °0 o°* 00 ° o s Vv 00 o° „ O ° -1 -1 ( -2 -3 0 ° ° _ to 1 2 ° 3° V 5 6 7 o Axis One Key: Southern Lagoon, Monkey Tail and Almond Hill (triangle) Hillbank, Kates Lagoon, Lamanai, Cobweb Swamp, Botes Lagoon and Outpost, these sites are dominated by Brachysira neoexitis, Navicula radiosa, Mastogloia elliptica var. dansei, Denticula elegans and Nitzschia amphibia var. rostrata (Closed circle). Figure 5.20 Co 5 - 4 3 - 4 A 2 mplele Data Set o o °o° °° o OQo o o ° °o -3 -2 -1 -3 - V 21 A n New River Lagoon O § * otfoP °o ° O ° oO 0 o ° -3 -2 • IT -3 5> rf?o op°° 2 3 4 5 o 1 Key: Modern Samples (triangle), Hillbank (filled circle), Lamanai (open circle) Even with the reduced data set all the samples from the New River Lagoon are still distinctly different from one another. Figure 5.22 Figure 5.23 5% Species Full data set 2.5 -, Achnanthes^iinutissima 2 1.5 1 -\ Brachysira neoexili^ q ^ Encyonema carina^ Denticula tenuisj^ti chnanthes exigua Nitzschia amph^ia Mastoqloia smithii var lacustris -1.5 -1 -0.5 ♦ Navicula radiosa -0.5 Denticula elegans ^ 0.5 +1 1.5 2 Cyciotella distinguenda 2.5 3.5 Key: Lamanai (circle), Hillbank (filled circle), Modern samples (triangle) When only the dominant species are considered the data sets are much less coherent, but the modern samples are much more closely related to the core data. Figure 5.24 5% Species New River Lagoon 4 n A 3 - A 2 - A V. -2 -1 f -2 ° ° °0°eo O o o 8 O b O Oj o® Figure 5.25 5% Species New River Lagoon Achnanthes minutissir&5 ♦ 2 1.5 1 Brachysira neoexilis 0.5 Encyonema car^a Navicula radiosa + -0.5 Denticula elegans♦ -1 J ^\chnanthes exigua Mastogloia smithii var lacustris Nitzschia amphibia ♦ 1 2 Cyclotella distinguenda Denticula tenuis Key: Modern samples (triangle), Hillbank (filled circle), Lamanai (circle) This shows the same sample and species distribution as the full data set. Figure5.26ExamplesofMastogloiaSmith i(c lWMithB24346)ndMastogloiait iva .lacus ris (VanHeurckN^726358). Figure 5.27 Mastogioia smithii var iacusiris family, New River Lagoon, Belize Modern Sediment type Modern Reeds type Fossil "Reeds" type (Outpost) Figure528CentralAmeric nwaters 50 30 10 -10 D to -30 -50 -70 -90 -150505105 s18o Diamonds:Belized ta GMWL/ Cuba y=6.0025x-1 1 63 R2=0.9902 Mexico y=5.5304x-1 .794 R2 =0.9984 Chapter Six: The Results from Hillbank, New River Lagoon 6.1 Introduction The three chapters which follow present the results from the three main coring sites: Hillbank and Lamanai, New River Lagoon and Honey Camp Lagoon. The Hillbank 1998 sequence covers the longest time period and therefore provides a context within which the other records can be placed. The impact of human activity through time is clearest in the Lamanai sequence. Honey Camp Lagoon provides insight into the wider manifestations of environmental change in north Belize. This chapter presents an analysis of the results gained from the diatom and stable isotope records from Hillbank, New River Lagoon. In order to appreciate both spatial variation and a more complete picture of the lagoon's history, two cores were analysed. These were a 14m Livingstone core from the southern shore and a 69cm Kullenberg core from 6-7 metres of water (Figures 6.1 and 6.2). This information provides a general appreciation of changes over a long time period, an in depth understanding of events in the more recent past, and an idea of spatial variability within this part of the lagoon. 6.2 Hillbank 1998 An earlier study (Breen, 1998) analysed this core for diatoms and a total of 71 depths were counted as part of this joint Natural History Museum and Edinburgh project (Blackmore, unpublished). Subsequently, 58 further depths have been analysed for diatoms and a stable isotope record has been created for the whole core from the analysis of both bulk carbonate and gastropods. The stratigraphy of this core can be found in appendix 6, which includes a key to the symbols shown in the diagrams. The stratigraphy for Hillbank 1998 can only be regarded as preliminary due to the problems highlighted in section 4.3.1. 191 There are five AMS radiocarbon dates for this sequence: Table 6.1 Code Depth Material I4C Years 5I3C pdb%O Calibrated 2 Sigma (cm) BP +/- 0.1 Ages range CAMS 1242 Organic 9840 +/- 60 -27.6 8778, 9138- -77197 Matter (OM) 8771, 8745 BC 8610 BC CAMS 990 OM 6020 +/- 50 4909, 5046- -45870 4872, 4855 BC 4744 BC AA- 400 OM 130+/-55 -28.0 AD 1689, AD 39722 1729, 1810, 1922, 1948 1656- 1954 AA- 153 OM 4752+/- 66 -30.3 3626, 3655- 39721 3586, 3527 BC 3367 BC CAMS 113.5 Gastropod 3990 +/-40 -7.9 / / -77198 (G) AA- 113.5 OM 2463+7-48 -27.6 756, 701, 789-402 42417 539, 526, 524 BC BC The hardwater lake error from the paired date at 113.5cm is 1527 years. The difference is a result of the input of old, !4C-deficient carbon into the dissolved inorganic carbon of the lake water (Deevey and Stuiver, 1964). Terrestrial organic material is free from this because it derives its carbon from atmospheric CO2. The estimated hardwater error for Hillbank fits with published values for Guatemala and the Yucatan Peninsula (Hodell et al., 1991; Leyden et al., 1993; Hodell et al., 1995; Curtis et al., 1996 and Leyden et al., 1998) (see section 4.6.2). The dates are all in chronological order apart from the one at 400cm. The radiocarbon samples from 990-153cm were taken by the team from the Natural History Museum, London. The author was not involved in these stages and thus the reasons behind the two dates which show a large reversal can only be postulated. There are three possible reasons behind the reversal. 192 1. Firstly these dates could be correct and the sediment cores (which were taken in one metre sections) could have been incorrectly labelled and the top half of the sequence has in fact been reversed. This is the worst case scenario, as it would completely change the interpretation of the sequence. 2. The radiocarbon samples could have been incorrectly labelled. 3. The sample could be contaminated in some way, for example through the addition of very young carbon. Due to the coherence of the rest of the dates it is apparent that the date at 400cm, which is anomalously young, should be rejected. Figure 6.3 shows the dates in stratigraphic context. This shows that there have been three quite different sedimentation rates through time. These vary from 0.066cm/year (9840-6020 l4C years BP) to 0.66 cm/year (6020-4752 l4C years BP) to 0.017cm/year (4752-2463 l4C Years BP). The immediate implication of this is that there is an order of magnitude difference in the rate exhibited in the middle period. This suggests that this was the greatest period of activity in the catchment. Such a finding is in good agreement with the record from Lake Peten-Itza where the mid-Holocene also shows a comparable increase in the sedimentation rate (Curtis et al., 1998). As a result of the changes in sedimentation rate, it would be an unsound strategy to extrapolate dates below 1242cm. The age at the base of the sequence is therefore greater than 9840+/-60 l4C years BP and it is not certain how much sediment is missing from the top of the sequence. In order to calculate the ages of important depths the record was split into three parts and the equation of each line was worked out. From this it was possible to estimate ages in radiocarbon years. These were converted to calendar years using Stuiver and Pearson (1993). The estimated dates are referred to as years BP. This core was collected as part of a project funded by NERC (GR3/10721) to the Natural History Museum to investigate the vegetational history of Belize through pollen analysis. From this a limited pollen diagram has been produced comprising of 17 depths (Figure 6.4). This record shows that the abundance of pine pollen varies 193 throughout the profile. This is interesting because today Hillbank is surrounded by pine and palm savanna. It is absent between 200-300cm and abundant from 200cm upwards. Chenopodiaceae pollen, which is an indicator of agricultural activity, reaches the peak in its abundance at 220cm. Broadleaf trees are more abundant than pine below 720cm and Rhizophora (the dominant genus in mangrove forest) is consistently present below 700cm. There is no evidence for maize pollen in this sequence (Blackmore, unpublished). 6.2.1 The Diatom Record: The diatom records from the New River Lagoon have been analysed in four stages: 1. The qualitative description/interpretation of the diatom diagram. 2. Dissolution indices were employed on one of the most common species: Mastgloia smithii var. Icicustris. The aim of this was to improve understanding of preservation in the system. 3. The diversity and concentration of diatom species at each level was calculated. 4. The transfer function created by Reed (1995, 1998a) was employed. This reconstructs conductivity drawing on a data set from Spain. This was deemed to be more suitable than the data sets from East Africa (Gasse et al., 1995) and Central Mexico (Davies et al., in press) because it contained the most species in common with the Belizian data sets. Table 5.2 highlighted the differences between the reconstructed conductivities for key species in Belize and Spain. This must therefore be taken into consideration when interpreting these results. The record from Hillbank has a very uniform diatom flora and there are no dramatic shifts in the species that are present (Figure 6.3). The key periods of change are those where diatoms are not preserved. All the species mentioned are pictured in appendix 2. The record can be split into 6 zones according to changes in the main diatom flora: 194 Zone 1 - 1381-1300cm: This zone is dominated by Denticula elegans, Mastgloia smithii var. lacustris, Encyonema carina, Brachysira neoexilis and Navicula radiosa. There are also small numbers of Mastogloia smithii, Cyclotella plitvicensis, Navicula pupula, Navicula radiosa var. tenella, Nitzschia amphibia var. rostrata and Achnanthes minutissima. There are no shifts between these species which suggests that this was a very stable period in the lagoon's history. These species are all littoral (Gasse, 1986; Krammer and Lange-Bertalot, 1986; 1988; 1991a and 1991b). The ecology of Cyclotella plitvicensis is unknown and it does not occur in any of the modern samples collected in Belize. This species is noted in the literature as often being confused with Cyclotella distinguenda. This is discussed in more detail in appendix 7. This assemblage without C. plitvicensis is found in Kates Lagoon, Lamanai and Hillbank sediment samples all of which are highly vegetated sites (Figure 4.1). The gap at the top of this zone is because sediment was not available to sample. Zone 2 - 1300-1200m: This is the most unusual zone in the whole sequence. It is barren of diatoms (signified by the dashed lines) apart from two layers. In these two layers only 200 diatoms were counted and therefore the description of these layers can only be regarded as preliminary. Aulacoseira granulata and its varieties, Denticula elegans, Mastogloia smithii var. lacustris, Encyonema carina and Brachysira neoexilis dominate the layers. Aulacoseira granulata is found in two modern day samples from Belize, in the Booth River wetlands and Chiwa Lagoon and these environments are both swamps (Figure 4.1). A. granulata is usually found in warm, nutrient rich, turbid conditions (Gasse, 1986). The rest of the species that are found in this zone have already been found in the sequence, which suggests that conditions have not changed as much as it might first appear. What is significant is that the species appear during a time of poor diatom preservation suggesting that the system is undergoing change. 195 Zone 3 - 1200-950cm: The gap at the beginning of this zone is because sediment was not available to sample. This zone is dominated by the same species that are found in zone 1, with the additions of very low levels of Cyclotella distinguenda, Cymbella muellerii and Mastogloia elliptica var. dansei. The slight increase in diversity suggests that the environment has now become more amenable to a broader sweep of species. There is very little trend to this zone apart from a general decline in the levels of Denticula elegans while levels of Navicula radiosa and Achnanthes minutissima increase. Mastogloia smithii is not common to modern samples but the main suite of species are found in Kates, Lamanai and Hillbank which are the same sites associated with zone 1 (Figure 4.1). Zone 4 - 950-720cm: The same species are present in this zone and Denticula elegans dominates. There are relatively more Brachysira neoexilis than Mastogloia smithii var. lacustris suggesting that conditions have changed slightly to favour the increase in Brachysira neoexilis. The sites that have a similar diatom flora in the modern environment are Hillbank 2 and 2B sediment, neither of these sites is dominated by vegetation and both have conductivities of 1.29 mS cm'1 which indicates that these sites are freshwater (Figure 5.4). Zone 5 - 720-280cm: A reciprocal relationship occurs within this zone between Denticula elegans and Mastogloia smithii var. lacustris whereby in the centre of the zone when Denticula elegans declines, Mastogloia smithii var. lacustris increases. This highlights what seems to be the pattern for this part of the lagoon - the shift between species rather than the replacement of one with another. This suggests that the diatom record is a reflection of changes to the lake shore rather than wider changes to the catchment. The levels of Brachysira neoexilis drop in this zone while levels of Encyonema carina remain high. Species such as Navicula radiosa and Mastogloia smithii are very consistent throughout the whole record which suggests that these species are less sensitive to change than others. The sites in the modern environment which 196 have more M. smithii var. lacustris than D. elegans are Doubloon Lagoon (epiphyte), Kates Lagoon (sediment) and Hillbank 1 (epiphyte) all these samples are from highly vegetated zones and low conductivities ranging from 0.23-1.48 mS cm"1 (Figures 4.1 and 5.4). Zone 6 - 280-0cm: This zone is punctuated by three zones that are barren of diatoms. The depths which have preserved diatoms each have the same species present which suggests that the system is fluctuating between two states. This implies that the lagoon is on the cusp of preservation and therefore it is not wider scale changes that are causing the lack of diatom preservation but small shifts to the chemical and physical environment of the New River Lagoon which enable the preservation of diatoms. The key shift in diatom species which is seen in this uppermost zone is the dominance of Denticula elegans, Encyonema carina and Mastogloia smithii var. lacustris in the central zone to Mastogloia smithii var. lacustris, Brachysira neoexilis, Denticula elegans and Nitzschia amphibia var. rostrata in the top section. These species are rather stable and therefore the shift is really once again between a Denticula elegans dominated environment to a Mastogloia smithii var. lacustris one. In terms of the environment this may involve a shift from higher to lower conductivities with an increase in the size and domination of vegetation. The dissolution indices highlight that the story for Hillbank is more complicated than would first appear (Figure 6.6). The main section of this record is a long period of diatom preservation (zones 3-5). The results of the DDI highlight that there is a great deal of variability in the conditions of the Mastogloia smithii var. lacustris through this period. The DDI index records a value of zero if there are no perfectly preserved diatoms in a sample. This means that information is lost because a value of zero implies that there are no Mastogloia smithii var. lacustris present, but there could be high numbers of specimens preserved at stages 2,3 and 4. The WI indices show that through zones 3-5 Mastogloia smithii var. lacustris is present. Preservation appears to switch between two states suggesting that preservation is not the natural state for the system. The key point that the indices highlight is that although diatoms are 197 found throughout zones 3-5 they are being influenced by forcing factors which affect their presence. Thus a seemingly stable part of the record is actually quite variable. Poor preservation is a phenomenon across Belize with all the other lagoons cored for this investigation failing to have diatoms preserved within them (see Chapter 4 for more details). There is also a good match between the number of diatoms in the sediment and the preservation levels, i.e. where there are a lot of diatoms they are well preserved. The average diversity of species for the Hillbank sequence is 13 (Figure 6.7). The average for the main section of the record is 16 species, which is slightly higher suggesting that during this time period conditions were more favourable to the preservation of greater range of diatoms. The reconstruction of conductivity (using Reed's 1995 Spanish transfer function) in the Hillbank sequence showed that the average conductivity for the record is 3.3 mS cm'1 (Figure 6.5). The value for each zone is: 1. Zone 1: 2.8 mS cm'1 2. Zone 2: no data 3. Zone 3: 2.7 mS cm"1 4. Zone 4: 3.3 mS cm"1 5. Zone 5: 3.4 mS cm"1 6. Zone 6: 3.5 mS cm"1 It would appear that there is a division in the record between zones 1-3 and 4-7 with the first zones having a lower reconstructed salinity than the top zones. This highlights that although there have not been any key changes in the flora, the relative changes in species abundance are significant. The second point is that the reconstructed conductivity values for the core data are much higher than the modern values for Hillbank (1.12-1.29 mS cm"') and are more similar to Caledonia, New River (3.54 mS cm"1) (Figure 4.2). This site is located further down the New River and is therefore much more likely to be influenced by 198 the sea than the lagoon. The difference between the reconstructed core values and the modern day conditions could be the result of several reasons: 1. The date for the top of the Hillbank sequence is not known and therefore conditions could have been different from those of the present day. This is compounded by the fact that the modern day assemblages in Hillbank are not exactly the same as those found in the core. 2. The conductivity range for the species found in the core is different in Spain and Belize. 3. The diatom species found in the core may not have been living in their optimum conductivity conditions through time in Hillbank. This may be because conductivity is not the most influential factor with regard to species distribution in the New River Lagoon. 4. The diatom species found in the core have wide tolerances. These points highlight the importance of creating transfer functions from the study area so that firm conclusions can be made with regard to species tolerances and optimum conditions. Through the work completed in this investigation on the modern diatom flora (Chapter 5), it is apparent that habitat is of key importance in differentiating the diatom flora of New River Lagoon and therefore conductivity may not be so influential. Chapter 4 emphasised the importance of geographical location as a factor which influences species distribution and also the issue that species may evolve to inhabit different physical and chemical environments in different areas. To summarise, the diatom record reflects subtle changes in conductivity and vegetation in a littoral record. Zone 1 (1381-1300cm) reflects medium conductivity (2.8 mS cm "') and well vegetated conditions, zone 2 (1300-1200cm) seems to indicate a drop in water levels with swamp-like conditions persisting. More detailed information would need to be collected on Aulacoseira granulata to verify this. Conditions revert back to those in zone 1 in zone 3 (1200-950cm). From 950-720cm (zone 4) conductivity levels are increase to 3.3 mS cmand vegetation levels drop. The opposite conditions occur in zone 5 (720-280cm) although conductivity values remain high. In zone 6 conditions shift from a Denticula elegans dominated system 199 (150-90 cm) to a Mastogloia smithii var. lacustris dominated system. The reconstructed salinities move from 3.2 to 4.2 mS cm'1. This increase is likely to be influenced by the increase in Nitzschia amphibia var. rostrata and Brachysira neoexilis at the top of the record. A DCA analysis was undertaken to enable the associations and differences between species and depths to be quantified. The eigenvalues are however extremely low at 0.135 and 0.1. This implies that the axes do not explain the associations seen in the core. This may be a function of the limited variation in the data set which is signified by the short axes. The graphs (Figure 6.8) both show scatter, suggesting that the differences between depths are shifts in the dominance of species rather than completely new assemblages. Such an interpretation is in agreement with the qualitative analysis of the diatom record. 6.2.2 The Stable Isotope Record: Oxygen and carbon were measured on both bulk carbonates and three species of gastropods (Cochliopina, Pygophorus and an unidentified species). 54 samples have been measured, approximately every 10 to 20cm. The sampling resolution was constrained by the availability of material due to the missing core sections. The oxygen isotope record has three main episodes of change at the beginning, middle and end of the sequence (Figure 6.9). The conditions represented by the isotopes are most different at the inception of the record. At the beginning of the oxygen isotope record conditions are very unstable with values shifting from -3.7 %o to -0.5 %o. This suggests that the system was responding to large, but short lived, forcing events. Values then have a more sustained excursion from 1287-1248cm where the least negative isotope values for the whole sequence are found. These are approximately 3%o higher than the mean value of the sequence. This is a highly significant difference. Values return to stable conditions by 1198cm and are maintained at around -3.5 %o until 630cm. This depth is the peak of a negative excursion where values reach -4.12 %o. A shift occurs at 378.5cm where values 200 remain stable, but are less negative than in the previous phase (584-378.5cm (average) = -3.6 %o, 338.5-203cm (average) = -3.3 %o). Conditions then remain stable until 193cm. This is the beginning of a very variable period where values oscillate to the top of the sequence. The level of fluctuation is not high but the values are more negative in this transitional phase than at any point previously encountered in the record. The level of variation is also the largest in the sequence with the exception of the base of the record. Gastropods have also been measured from this sequence (although at a much lower resolution) (Figure 6.9). The record from Cochliopina sp compares well with the bulk carbonate record showing the large swings at the base of the record, stability through the main section of the record with values centring around -3.5 %o and greater variability at the top of the sequence. Measurements were also taken from Pygophorus sp and an unidentified sp. The Cochliopina and Pygophorus sp match well. The unidentified sp shows a great deal more variation at the top of the record. This could be because the habitat which this species occupies is different from that of Cochliopina and Pygophorus sp which occupy similar areas (see Chapter 4 for details). The bulk carbonate results for the carbon isotopes show the same four zones of change as the oxygen isotope record (Base-1198cm, 1198-584cm, 584-378.5cm, 378.5-203cm and 203-0cm) but the shifts that are seen at the top of the sequence are of a greater magnitude (Figure 6.10). From the base of the record to 1248cm the 8I3C values are stable at around -2.5 %c. This suggests that even though the 8lsO record is very variable the inputs of carbon to the system are very stable. The system then undergoes a rapid change with the isotopes recording the most negative peak in the record at 1208cm at -8.1 %c. The 813C values increase after this point and become much more stable. Fluctuations are minor until 666 to 584cm. This is a period of change in the system which coincides with the negative excursion in the 8I80 record. At this time the SI3C records increase and then at 630cm levels drop. Levels then stabilise at around -2.9 %o. From 338.5cm values begin to decline but this is a time of very gradual change which again coincides with the 8I80 record. From 193cm 201 onwards the carbon isotopes record rapid changes suggesting that the system is now operating in a very different way to the past. The carbon isotope gastropod record does not relate as well to the bulk carbonate record as it does for the oxygen isotope sequence (Figure 6.10). The trends between the species are the same though, with Cochliopina sp and Pygophorus sp being similar in their magnitude and the unidentified species being both different and exhibiting larger variations in isotopic values. In interpreting the records one needs to be aware that the two isotopes could be telling different stories because the New River Lagoon is an open system. This is highlighted by a scatter plot of 5I80 versus 5I3C which shows that the system does not exhibit covariance (Figure 6.11). 6.3 Preliminary Interpretation: Of principal importance is the determination of the main driving forces behind the system, i.e. whether they are catchment- or within-lake processes. The New River Lagoon is a large open system which eventually connects to the sea via the New River. It is therefore under the influence of not only climatic change but also groundwater and catchment variations, which are likely to have more influence on the isotopic signature because of the open nature of the system. The isotope zones which have been identified date to: • >9840+/- 60 - 9200 years BP • 9200-5400 years BP • 5400-5100 years BP • 5100-4800 years BP • 4800 years BP -top Freshwater gastropods and diatoms are present to the base of the Hillbank sequence which implies that the system has held permanent water throughout. The Hillbank 202 record reaches its most positive 5I80 values between 1278 and 1248cm i.e. in the immediate period before 9840+/- 60I4C years BP. This is also a time of change in the diatom flora. Although the shift in the Hillbank record looks dramatic it actually occurred over a period of approximately 1000 years with the system reaching almost average conditions by 1228cm or c. 9600 years BP. The climate of Central America was dry in the Lateglacial and moist during the early to mid Holocene periods (Leyden et al., 1994). The record from Hillbank suggests that the Lateglacial was a very variable period and that conditions in the time period immediately before the Holocene were very different from the rest of the sequence. Total carbonate has also been measured on this sequence (Figure 6.12). This shows uniformly high levels throughout apart from at the base of the record where values drop to 55%. This matches the magnetic susceptibility record well (Figure 6.12). The period of highest levels and therefore a period of disturbance and changing inputs into the system is at the base of the record. This time period is therefore one of significant change. It is possible that this period equates to the Younger Dryas. If this is the case then the signal for Belize at this time is a dry climate. The transition to wetter conditions was very smooth and this implies that there was a clear climatic difference between the Lateglacial and early Holocene in Belize. The transition to full moist conditions was completed by c. 9600 years BP. Conditions remain stable and moist until 666cm or 5500 years BP. From this point until 584cm or c. 5381 years BP there is a negative excursion in the 5lsO record. This is a fairly short-lived event that coincides with a change in the 5nC record. This suggests that it is a period of catchment disturbance with low 5I80 values occurring at a time of enhanced 5l3Cvalues. The 5lsO record continues to be stable and moist with a shift occurring at 378.5cm or c. 5100 years BP to slightly less negative values. At c. 4800 years BP or 193cm the Sl80 record becomes more variable. The shift in the isotope record from a stable to a variable signal from this point suggests that this is the beginning of a new phase of environmental change in the region. 122cm equates to approximately 3000 years BP and is the end of the negative excursion in the 8lsO record. This is generally considered to be the end of the mid 203 Holocene moist period in the circum-Caribbean (Hodell et al., 1991). The late Holocene dry period occurred between 1300 to 1100I4C years BP. Due to the lack of knowledge concerning the sedimentation rates it would be unwise to postulate where this time period might fall on the Hillbank record. Although levels of 5'sO do become more positive, a distinct and severe arid period is not recorded in the Hillbank oxygen isotope record. It cannot however be guaranteed that the top of the record covers this event. It is possible that the large nature of the New River Lagoon renders it insensitive to short-lived events such as the late Holocene dry period. A further possibility is that north Belize was not affected by this event. The carbon record provides a good complementary record to the oxygen sequence. The base of the record is characterised by stable, but increasingly positive levels of 8I3C. This is likely to be a result of the relative proportion of DIC coming from the dissolution of limestone versus equilibration with atmospheric CO2. During this period the 8lsO record is moving towards the most positive values in the sequence. This shows good agreement between the two records with exchange with atmospheric CO2 becoming the dominant process in this increasingly shallow system. The large negative excursion which follows this, reaches a peak at 1208cm (c. 9300 years BP) which is when the transition to full moist conditions in the 5I80 record occurred. As the climate was becoming wetter, the 8I3C record became more negative. There are a number of reasons why the carbon record would be showing this signal: 1. Influx of freshwater in the system (influencing exchange with atmospheric C02) which would equate to the negative conditions shown in the 5lsO record. 2. The oxidation of organic matter. It is during this time period that lowland forest became established in Guatemala (Leyden et al., 1993) which suggests that the input of terrestrial organic matter into the catchment would have shifted to much more negative values (C4 plants = -12%0; C3 plants = -25%o) (Curtis et al., 1998). The oxidation of terrestrial organic matter generates CO2 that has an isotopic signature similar to the source material, some of which can dissolve into 204 groundwater and reach the lake, influencing the signature of the lake water DIC. In addition, oxidation of allochthonous organic matter within the lake produces CO2 which has an isotopic signature that also reflects the source material and this CO2 influences the isotope ratio of the DIC (Curtis et al., 1998). If this signal is the transition to a forested catchment then it occurred approximately over a thousand year period. The negative peak in the 8I3C record coincides with an interesting change to the diatom flora. It occurs during a phase of poor diatom preservation and the introduction of the diatom species Aulacoseira granulata. This species is found in the plankton of shallow lakes. Although the other species found in this part of the sequence are common throughout the record the introduction of this species suggests that this is a period of environmental change. This is compounded by the fact that such a change in the diatom flora or the 813C record is not found at any other part of the record. Levels of 8I3C rapidly increase reaching a steady state by 1048cm or c. 6887 years BP. It is unlikely that the forest taxa would have disappeared and therefore the increase in 8I 3C is likely to be due to a shift in the inputs into the system. This is the beginning of a more stable phase of diatom preservation suggesting that the environmental conditions have changed because diatom preservation is now continuous. With the moist conditions of the early Holocene, aquatic photosynthesis would have been high and therefore the 8I3C levels would have increased correspondingly. Values of 8I3C are fairly stable through the Holocene suggesting a productive lake which exists in stable climatic conditions. There is a negative excursion in the 8I80 record which peaks at c. 5450 years BP. The relationship between the two isotopes is not simple. Although levels of 8I3C do increase they drop at the peak of the Sl80 excursion. This suggests that there is an interplay in the 8I3C record between the influence of freshwater inputs (negative excursion) and the role/magnitude of aquatic photosynthesis (positive excursion). 205 The 8I3C record begins a negative trend at c. 5000 years BP which is where the shift to more positive values in the 5I80 record begin. The carbon record highlights a time of significant change to the system between 193-152cm or c. 4790-4750 years BP. The pollen data (Figure 6.4) for this period show a peak in Chenopodiaceae pollen which is an indicator of disturbance. This suggests that it may be during this time frame that the area around Hillbank began to be first disturbed by human activity. This is not enough evidence to make this claim but it is possible because during this phase of Mayan history sedentary agriculture is thought to have begun in Belize (Hammond, 1982). This time period is also one of increasingly negative 5I80 values, reaching values akin to those found at 630cm. Rosenmeier et al. (in press) looked at the 5i80 record for Lake Salpeten and related the changes in this isotope to variations in the forest cover of the lake catchment. Decreased values were associated with times of forest loss, as a result of greater surface runoff and groundwater inflow to the lake. This therefore adds weight to the argument that this is the time period when human activity is prevalent. Hansen (1990) documents dramatic deforestation in Albion Island, Belize around 2800 l4C years BP which equates to approximately 120cm. This is not a clear period of change in the Hillbank pollen diagram (Figure 6.4) but it is difficult to be certain because the top of the record is missing. This period occurred during a positive shift in the S13C record which suggests that this a highly productive phase and the catchment is being increasingly deforested. The 5180 values are also recording a positive trend. If deforestation was the main control over the system at this point the 5i80 signal would be negative (following the arguments of Roseinmeier et al., in press) which suggests that further climatic changes are occurring at this point in the sequence. The negative shift of the carbon isotopes from 60cm to the top of the record could represent forest recovery. How do the isotope and diatom data relate to one another? The diatom record is a reflection of changes to habitat availability and changes in conductivity. The changes that the record represents are not major but there are interesting links between the two sequences. The first is that the periods where diatoms are not 206 preserved equate to the most variable times in the isotope record (1201-1288, 214- 260, 152-193 and 22-80 cm). This implies that the lack of diatom preservation is symptomatic of a change to the environment of the lagoon. The periods where diatoms are not preserved coincide with periods of decreasing 513C. This could be a result of lower inputs into the system due to forest cover in the catchment, an increase in the oxidation of organic matter or fluxes of freshwater. All of these processes are likely to affect both the chemistry and physical make-up of the system. The percentage of calcium carbonate present in the core (Figure 6.12) only drops during the 1200-1300m phase which suggests that this period has different causal mechanisms or is more severe than the changes at the top of the sequence. This is not surprising because the changes at the base of the sequence are most likely due to changing inputs into the lake as a result of the climatic changes associated with the end of the Lateglacial and the beginning of the early Holocene. The isotopic changes to the top of the sequence are of a smaller magnitude and the inputs into the system have not changed. 6.4 Hillbank 2000 The location of this core (500 m from Hillbank 1998) is shown on Figure 6.1. This core was analysed for oxygen and carbon isotopes using both bulk carbonates and the gastropod Cochliopina sp. This is a 69cm record which was collected using a Kullenberg corer and was analysed every l-5cm. The stratigraphy for the core can be found in appendix 6. The core was dated using 2l0Pb. This was the only core suitable for this method due to the insufficient levels of 2l0Pb in the other Kullenberg records (Outpost 2000 and Honey Camp 2000). The analysis of these data (Figure 6.13) shows that the sediment in the New River Lagoon system is highly mixed 7 10 because the levels of Pb do not decay exponentially. Consequently, the only approximate point of reference is the date of AD 1960 at 12cm because this is where i37Cs can be first picked up in the record (see Chapter 4). 207 Although the oxygen isotope record is stable, there are definite trends to the data set (Figure 6.14). There are three negative excursions at 65cm, at 27cm and at 17cm with a slight positive excursion at 55cm. This, however, is very short-lived. The gastropod data from Cochliopina sp. are limited but provide a fairly good match to the bulk carbonate record apart from at 20cm and 64cm. The difference between the two results is not systematic and therefore it is difficult to draw any firm conclusions as to why this might be the case. The most likely explanation is the role of habitat in influencing the isotopic signature found in gastropod shells. The 5i3C values follow a negative trend from the base of the record to 33cm (Figure 6.14). The trend from the base to 33cm is gradual, but between 61-53cm there is a negative excursion where values reach a maximum of -9.57%o. The lowest 5I3C values are found at 33cm where they reach -10.03%o. This suggests that it is the oxidation of organic matter that is providing the largest input to the DIC pool. Conditions are rapidly reversed with the most 'positive' values of the record being found at 27cm at -5.05%c. After a reversal of this trend, the record from 21cm to 0cm is stable at around -6%o. The gastropod carbon isotope data matches the bulk carbonate data well in the first half of the record. The most recent gastropod data point is quite different from the bulk carbonate record. This could be a result of a number of factors including a change in carbon source. The percentage of calcium carbonate that is present in the Hillbank 2000 core can be separated into distinct zones (Figure 6.15). From the base of the record to 61cm the levels are steady at 80%, these then rapidly drop at 57cm to 47%. Conditions then revert back to the original state until 25cm where they drop again to 33%. Levels are then steady at 40-50% from 21cm to the top of the record. The two periods where levels of calcium carbonate drop are short lived, but they do represent a significant change to the inputs into the system. Magnetic susceptibility has also been measured on this core (Figure 6.15). The measurements were taken using a loop sensor and a mass specific technique. Both these records show the same pattern. Levels are stable and negative from the base of 208 the record to 32cm. From this point levels begin to rapidly increase reaching a maximum at 25cm. Levels begin to decline just as quickly as they rose and by 15cm the system had stabilised again. This suggests that the peak at 25cm represents a very significant change in the catchment. There are no mass specific results for the 210 top of the core because all the material was used for Pb dating. 6.4.1 Preliminary Analysis: A scatter plot of 8lsO versus 5I3C was undertaken for Hillbank 2000 (Figure 6.11). This shows a fairly coherent cluster of results but these are within a different zone from Hillbank 1998. This suggests that the controlling mechanisms over the New River Lagoon system have changed through time. The trend at the top of the Hillbank 1998 8I3C record is towards more negative values reaching -5.35%o at the top of the sequence. The base of Hillbank 2000 is -4.43%c which is further evidence that the two records are definitely representing different phases of the history of Hillbank. The low 513C values exhibited throughout the Hillbank 2000 core may be the result of its location. Hillbank 1998 was taken from the lagoon edge next to the extensive marsh system. Hillbank 2000 was from nearer the centre of the lagoon (in 6-7m of water) and therefore not so influenced by the marsh system. Aquatic photosynthesis enriches the TDIC pool and this is likely to be a much more prevalent process near the marsh system. This may therefore account for the more 'positive' values seen in Hillbank 1998. There are very low levels of diatoms in the Hillbank 2000 core and as a result a diagram has not been produced. The two species that 'dominate' the assemblages are Nitzschia amphibia var. rostrata and Mastogloia smithii var. lacustris. These are present episodically through the top 24cm of the record but not below this point. The appearance of these two species suggests that these must be very robust species that are able to persist in conditions that are obviously not suitable for the vast majority of species that are present in later New River Lagoon sequences. 209 The levels of magnetic susceptibility are low in this system (Figure 6.15). This is not unexpected because calcium carbonate gives weak or negative values of magnetic susceptibility. At 25cm there is a drop in carbonate levels which is coupled with an increase in magnetic susceptibilty implying that this is a time of significant catchment disturbance. The XRD results for Hillbank 2000 show that during the peak in magnetic susceptibility there is a change in the systems mineralogy, namely with the introduction of quartz, pyrite, dolomite and smectite (Figure 6.16). It is the latter three minerals that are more magnetic than calcite (the dominant mineral), especially pyrite. It is apparent therefore that the peak in magnetic susceptibility is due to the influx of clays into the lagoon. This influx of clays is most likely the result of catchment disturbance and represents a very significant disruption because it is an order of magnitude higher than anything else in the record. During this period there is a peak in the 8I3C values which can imply deforestation. It is therefore postulated that this time period is when colonial logging was at its peak. If 12cm is equal to 1960 that means that 40 years has accumulated in 12cm therefore the period of disruption between 31-25cm is approximately equal to 1897-1917 i.e. the peak of colonial logging. 210 • 210 The Pb record does not show a general decline in the levels of Pb through time (Figure 6.13). This suggests that this is a highly mixed system. The main implication of this is that the records produced from this lagoon will be an average of the actual conditions experienced and thus signals will not be as strong as conditions at the time warranted. Thus, the changes that are seen in the record must be judged to be very significant as they are likely to be an understatement of the actual magnitude of change. The mixing regime which has been highlighted at Hillbank is unlikely to be unique to that area and it therefore should be taken into account when studying the records from Lamanai. 6.5 Summary of Hillbank, New River Lagoon From this investigation of Hillbank a number of key conclusions can be made: 210 1. The oxygen isotope record shows that the Pleistocene was a variable time period in terms of climatic conditions. This is not reflected in the diatom record which shows this to be a stable freshwater littoral system. This suggests that the type of changes that were occurring did not affect the factors which most influenced the diatom populations at that point e.g. habitat. 2. The transition to the Holocene was arid according to the oxygen isotope data and this is matched by a time of very poor diatom preservation. With improved dating control this may be proved to be a Younger Dryas signal. 3. From c.9200-5100 years BP the oxygen isotopes record stable and moist conditions. This is the most stable phase of diatom preservation being characterised by changes in species dominance rather than shifts in species. 4. 5500-5400 years BP is a distinct negative shift in the bl80 record and a positive shift in the SI3C record. This is the first preliminary evidence for catchment disturbance. From 6020-4752 l4C years BP the highest rates of sediment accumulation occur. This is further evidence that this period was one of change. 5. From 5000 years BP to the top of the record diatom preservation is very variable as are the signals from the stable isotope data. From approximately the same time period there is an increase in the disturbance pollen and a negative trend in the 5lsO data which suggests that this is when the influence of humans began in this area. 6. From 4800 years BP the abundance of pine pollen increases which implies a drier climate. This is backed up by a shift to a positive trend in the 8lsO record at c. 3000 years BP. This suggests that the shift to late Holocene dry conditions was gradual. 7. Evidence from Hillbank 2000 highlights the impact of colonial activity in the catchment in terms of catchment disturbance. 211 Hillbank, New River Lagoon Basemap: DOS Sheet 15, Edition 5-GSGS (1993), Scale 1:50,000 Projection: Universal Transverse Mercator (Zone 16) Figure 6.1 Modified from Murray (unpub.) Figure 6.2 Hillbank, New River Lagoon Hillbank 1998 11000 10000 - n 9000 - m 8000 to 7000 1 co 6000 £ 5000 O 4000^ 3000 -j 2000 - 1000 - f 0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 Depth (cm) Figure 6.3: Radiocarbon dates for Hillbank 1998. The points joined up by the line are from terrestrial organic matter. The single point is the date from the gastropod. The error ranges can be found in Table 6.1. Hillbank1998 Figure6.4PollenDiagram SandGastropo sOrga icClay Hillbonk1998 Figure6.5Diatomi gra J" °150- 100- 150- 200- 250- 300- 350- 400- 450- 500- 550- 600- 650- 700- 750- 800- 850- 900- 950- 1000- 1050- 1100- 1150- 1200- 1250- 1300- 1350- 1400 ✓ 2# <3\ ,<355<£■ o->> '<0 2^2SQVQ&> # ^« , jr/ o? » -1Vo$ -t.4 ♦^».6 -0.8 Figure6.11B plotsofStabl IsotopeResults Figure6.12H llbank1998:Magneticsusceptibilityandpercentagetot lcarbor ults Figure6.13H llbank2000210PbResults Hillbank2000 0+- 10 20 E30 Q. HI O40 50-I 60 70J ■'///////////////. S/////S/////S///. '///////////////. ■'///////////////. '///////////////. ■'///////////////. Hillbank^OOO Figure6.14H llbank200018and3CReco s OpenDiamonds=Cochli pinas . Hillbank^OOO G Hillbank2000 Hillbank2000 0% Carbonate 10%6 80%1 0% '/////////////St '/////////////St '/////////////// '/////////////// 70J Hillbank2000 MagneticSusceptibility Blue:MassSpecificResults Black:LoopSensorR sults Figure6.15H llbank2000Carbonatea dM gneticSusceptibilityesul s Figure 6.16 Hillbank 2000 XRD Percentage 0 25 50 75 100 Blue: Calcite Pink: Quartz Green: Pyrite Black: Dolomite Chapter Seven: The Results from Lamanai, New River Lagoon 7.1 Lamanai 1999 This chapter presents and discusses the results from Lamanai, New River Lagoon. The location of the first core (Lamanai 1999) is shown in Figures 7.1- 7.2 and the stratigraphy of this core is detailed in appendix 6. Two cores were taken from this area in a similar manner to Hillbank which is located due south of this site. There are four radiocarbon dates for the Lamanai 1999 sequence (which are also shown in Figure 7.2): Table 7.1 Lab Code Depth Material 14C Years b'^CpDB^O Calibrated 2 Sigma (cm) BP +/- 0.1 Age range CAMS- 312 OM 3440+/- -28.6 1741 BC 1880-1636 77195 40 BC AA- 259 OM 3070 +/- -27.1 1374, 1338, 1433-1132 35786 50 1319 BC BC AA- 38.5 OM 810+/- 40 -32.4 AD 1224, AD 1161- 35787 1231, 1239 1283 CAMS- 38.5 G 2470+/- -6.9 / / 77196 40 The dates were calibrated according to Stuiver et al. (1998). The hard water lake error from the paired date at 38.5cm is 1660 14C years BP. This, in a similar manner to Hillbank, is in keeping with published estimates (see section 4.6.2). This evidence supports the idea that a general correction factor can be applied to lakes in this region (Hodell et al., 1991; Leyden et al., 1993; Hodell et al., 1995; Curtis et al., 1996 and Leyden et al., 1998). 7.2 The Diatom Results from Lamanai 1999: All the diatom species mentioned in this chapter can be found in appendix 2. The diatom diagram (Figure 7.3) has been split into 7 zones based on the major shifts in the flora: 228 Zone 1 - 318-22lcm: The base of the sequence is mainly barren of diatoms except for alternating zones which are dominated by two different assemblages of diatoms. The first assemblage comprises Denticulci tenuis, Nitzschia amphibia var. rostrata, Mastogloia smithii var. lacustris and Mastogloia smithii var. lacustris (fine). This latter species was discussed in greater detail in Chapter 5 and it represents sedimentary rather than epiphytic environments. In general terms, this assemblage of species suggests a littoral lake environment. Denticula tenuis is thought to be indicative of medium conductivities and is found in littoral environments (Krammer and Lange-Bertalot, 1988). Denticula tenuis does not occur in conjunction with the species in this zone in the modern environment and is only found in very low levels in three of the sites sampled in Belize: Booth River (plankton); Progresso Lagoon (epiphyte) and Crooked Tree Lagoon (plankton) (Figure 4.1). These sites all fall into different water chemistry groups. Booth River is a calcium-sulphate system; Progresso Lagoon is sodium-chloride dominated and Crooked Tree Lagoon is calcium-sulphate/total carbonate. This implies that the species must be quite cosmopolitan and therefore easily out-competed by species that are experiencing their optimum conditions. Denticula elegans, Mastogloia smithii var. lacustris, Brachysira neoexilis and Nitzschia amphibia var. rostrata dominate the second assemblage of this zone. This is a slightly different assemblage and because Denticula elegans is thought to be a marsh species this suggests that the times when this species is prevalent is during low water level conditions. Nitzschia amphibia var. rostrata is not common in the sites sampled in Belize. The site which has the closest population to this assemblage is Hillbank (3) sediment (Figure 5.2). This is a shallow water site that is surrounded by algae and vegetation which supports the interpretation of Denticula elegans. Zone 2 - 221-181.5cm: This zone is the first prolonged period of diatom preservation and it is dominated by Denticula elegans, Nitzschia amphibia var. rostrata, Mastogloia smithii var. lacustris, Mastogloia smithii, Mastogloia smithii var. lacustris (fine), Brachysira neoexilis and Cyclotella distinguenda. The fact that diatom preservation occurs over 229 an extended period suggests that the system is more stable during this phase. Cyclotella distinguenda was not found in any of the modern sites and Mastogloia smithii is very rare. This implies that the environmental conditions which this zone represents are different to those that are found in the modern areas sampled. Reed (1998a) found Cyclotella distinguenda in her samples from Spain, in oligosaline, littoral environments. She classified it as facultatively planktonic. The ecological differences between Mastogloia smithii and its variety are not clear, but the contrast in their distribution suggest that they do respond to different variables. The highest percentage of population of M. smithii is in Honey Camp Lagoon (algae) sample at 4% (Figure 4.2). This is also the site where the highest percentage ofM. smithii var. lacustris occurs at 48%. This site had (in relation to the other freshwater sites sampled) high conductivity and sodium-chloride levels. This suggests that, in terms of measured variables their optimum conditions are the same, but there must be a reason why M. smithii var. lacustris is able to so effectively out-compete the nominate form. Zone 3 - 181.5-172.5cm: This zone is barren of diatoms. Zone 4 - 172.5-74cm: This zone is an extended period of diatom preservation where new species are introduced into the record. It is dominated by Nitzschia amphibia var. rostrata, Achnanthes exigua var. exigua, Achnanthes minutissima, Brachysira neoexilis, Cyclotella distinguenda, Encyonema carina, Gomphonema gracile, Mastogloia smithii var. lacustris, Mastogloia smithii var. lacustris (fine) and Nitzschia amphibia. These species are again found in littoral environments (Krammer and Lange- Bertalot, 1986; 1988; 1991a; 1991b and Reed 1998a). The diversity of species is high in this zone which suggests that the environmental conditions at this time were amenable to more diatom species. Although Cyclotella distinguenda is not found in any of the modern samples, the rest of the species dominate the modern site Hillbank (1) sediment. This site is on the opposite shore from Hillbank (3) i.e. the marsh side (Figure 5.2). Sediment samples are an average of all the different inputs into the 230 system and therefore represent all the environments in the area. This suggests that the environment of this zone, like Hillbank 1, was a marsh edge which would have had a great diversity of habitats in which different diatom species could flourish. Zone 5 - 74-40cm: The number of diatom species drops in this zone and there is an abrupt transition to a new diatom assemblage. Denticula elegans dominates this sequence with Cyclotella plitvicensis. Other species that are present include Brachysira neoexilis, Mastogloia smithii, Mastogloia smithii var. lacustris (fine), Nitzschia amphibia var. rostrata and Sellaphora papula. This is a clear shift in the flora and therefore represents a time of environmental change. Cyclotella plitvicensis is not found in the modern sites sampled and Mastogloia smithii and Navicula papula are also not common species. This zone therefore represents an environment that has not been sampled in modern day Belize. The modern site which contains the highest numbers of Denticula elegans is Progresso Lagoon (epiphyte). This lagoon has a conductivity of 7.25mS cm _1 and is dominated by sodium-chloride. Nitzschia amphibia var. rostrata and Mastogloia smithii var. lacustris (fine) are littoral mud species and Sellaphora papula is widespread in its distribution. The ecology of Cyclotella plitivicensis is unknown. These species suggest that this was a time of shallow waters, high conductivity and very little large vegetation at this site (Krammer and Lange- Bertalot, 1991a). This cannot be substantiated without ecological knowledge of C. plitivicensis. Zone 6 - 40-9cm: This zone is barren of diatoms. Zone 7 - 9-0cm: Within this very small zone the diatom population moves from one which is dominated by Denticula elegans and Brachysira neoexilis to one which has a much more diverse flora including the varieties of Brachysira neoexilis, Achnanthes minutissima, Mastogloia smithii var. lacustris, Nitzschia amphibia var. rostrata, Navicula radiosa and Navicula radiosa var. tenella. This is not an assemblage that 231 has been present in the past. The section dominated by Denticula elegans and Brachysira neoexilis is similar to Hillbank (2b) sediment (Figure 5.2). This was a site which was away from the large reeds of the marshland and in slightly deeper water. This does not support the interpretation that Denticula elegans is a marsh species. The second assemblage is not one which is found in the modern environment suggesting that in a similar manner to Hillbank, the top of the core is not the present day. This is not surprising because the top sediments in Lamanai were extremely unconsolidated. The graph of reconstructed conductivity (Figure 7.4) contains only the depths that have 70% or more of the species in common with the modern data set. The division which is immediately obvious is between the diatom zones 1 and 2 and zones 4 to 7. The first two zones have an average reconstructed conductivity of 3.6 mS cm and the top zones have an average of 1.94 mS cm"1. This is a significant difference and its suggests that zone 3, which is barren of diatoms, is a distinct event which separates two different states in the lagoon's history. The conductivity reconstruction is based upon the species that are present at each depth. The value produced is therefore dependent on those species. The average value for the top zones (which represents the period from c. 2152 years BP onwards) has a reconstructed value that is closer to the modern measured conductivity values for Lamanai (0.91-1.07 mS cm"1) than the previous zones. From this information it is likely that the reconstructions are probably in the right range and that the lagoon was quite different to the past. This provides more confidence in the use of the Spanish training set. The species diversity in zones 4-7 is higher than zones 1-2 with an average of 22 species as compared to 13 species per depth (Figure 7.4). This suggests that the environmental conditions in the top part of the record were more amenable to a greater number of species. The higher conductivity in the lower half of the sequence may be part of the reason why diatoms are fluctuating between presence and absence. Unlike Hillbank 1998 species diversity does not appear to be so clearly linked to diatom concentration. This suggests that the environmental changes that are influencing diatom numbers are not severe enough to result in an environment in which only a limited number of species can survive. 232 Figure 7.5 shows the dissolution indices. The DDI clearly shows the division between zones 1 -2 and 4-7. The lower zones have either no preserved diatoms or very badly preserved specimens. In a similar manner to the results from Hillbank this means that information has been lost. The upper zones still exhibit very variable preservation but higher proportions of well-preserved valves are found. The transition to zone 4 coincides with the period of highest diatom concentration in the entire sequence (Figure 7.4). A clearer picture of preservation changes in this sequence is gained from the weighted and square-weighted indices. In a similar manner to the DDI they show the division between the top and bottom half of the record. Zones 1 and 2 are characterised by three clear cycles where preservation is poor but is punctuated by short periods of improved preservation. The upper half of the record is much more variable but on the whole the diatoms are better preserved. The use of preservation indices has enabled a greater appreciation of the variable preservation levels through the sequence even through zones of seemingly stable conditions. In order to gain a greater understanding of the relationships between depths a DCA analysis was undertaken on the core data. This methodology is explained in more detail in Chapter 4. These results are shown in Figure 7.6. The eigenvalues for the axes are not high with axis one being 0.5258 and axis two being 0.1606, but the axes are fairly long (in comparison with the Hillbank axes shown in Figure 6.8) which suggests that the axes represent some form of environmental gradient. This is directly related to the species that are found in the depths. The species form two groups. These are not tightly clustered and thus the only conclusion that can be made at this stage is that the species in 'group' one are associated with each other because they are not associated with the species found in 'group' two. 'Group' one: Achnanthes exigua, Brachysira neoexilis, Cocconeis placentula var. placentula, C. placentula var. incisa, Cyclotella distinguenda, Denticula elegans, Encyonema carina, Gomphonema gracile, Navicula radiosa (diamonds). 233 'Group' two: Denticula tenuis, Mastogloia smithii, M. smithii var. lacustris (open squares). The main point that these groups highlight is the division between Denticula elegans and M. smithii var. lacustris which adds weight to the argument that the shifts between these species, that are seen in the record, do represent times of environmental change. The depths do not appear to form any form of coherent grouping apart from those shown as open circles. These are all depths where Denticula elegans is extremely dominant as compared to Mastogloia smithii var. lacustris. To summarise, the diatom record from Lamanai can be split into three periods. From the base of the record to 181.5cm diatom zones 1 and 2 are found. These zones are characterised by low species diversity, poor preservation and the diatom species reflect 'high' salinities (3.49-3.7 mS cm"1). The environment is one of a littoral lake edge community that is fringed with vegetation. From 181.5-172.5cm there are no diatoms preserved. This represents a significant event because it forms the transition zone between two different lagoon states. From 172.5cm to the top of the record diatom zones 4 to 7 are found. These are characterised by high species diversity, variable preservation (which does reach quite high levels) and low reconstructed salinity (1.55-2.6 mS cm "'). The key change in the environment, which the diatoms show, is an increase in the amount of vegetation through time with a more marsh-like ecosystem predominating. 7.3 The Stable Isotope Results: Both oxygen and carbon isotopes were measured from this core from bulk carbonate samples every 5-10cm (Figure 7.7). The amount of variability throughout the whole record is low with a 5I80 range of 2.2%o. The 5i80 record is characterised by a constant series of small scale shifts which are forceful enough to stop the system from maintaining a steady state. At particular 234 points in the record the shifts are larger than 'normal'. The points that exhibit more than +/- 1 SD of the average change are 39, 42, 174 and 178cm. The most significant of these episodes is at the top of the sequence where S180 values reach -5.3%o (42cm). This is the most negative excursion in the whole sequence. The 813C record (Figure 7.7) shows greater shifts than the 8I80 record suggesting there have been significant modifications to the carbon sources in the system. The record can be split into three zones which match the 8I80 record (Figure 7.7): 1. 307-178cm 2. 178-42cm 3. 42-0cm Discussing each zone in turn: Zone 1: The scale of the shifts in this zone are similar to those at the top of the record. The general values are however more negative. Zone 2: This zone can be split into three phases of prolonged negative departures in the 8I3C values. After each negative excursion the 8I3C values return to approximately the same value between -3.9%c and -4.5%o. Zone 3: This zone is characterised by small and short-lived shifts in 8I3C. In general terms it is apparent that at the beginning and end of the record, the Sl80 and the 813C records are responding in a comparable manner with the level of variability between layers being of a similar magnitude. The beginning and end of the second phase in the carbon record is bracketed by the two largest shifts in the Sl80 record. This suggests that these two shifts in the 8I80 record are very significant and that zone 2 in the 813C record does indeed represent a time when the system operated in a different manner. 7.4 The Results From Further Proxies: Lamanai 1999 has also been analysed for percentage calcium carbonate, Loss On Ignition, available phosphorus and C:N ratios (see Chapter 4). 235 The percentage of calcium carbonate in the core is uniformly high but there are trends which can be recognised (Figure 7.8). The record can be effectively split into three zones: Zone 1: Base - 171cm: This zone shows a gradual trend increasing levels of calcium carbonate. Average value = 74% Zone 2: 171-42cm: Within this zone the shifts are much more long-lived and the lowest value in the sequence is found at 143cm (60%). Average value = 74% Zone 3: 42-0cm: Levels in this zone are very steady and average the highest in the sequence at 80%. Trends are difficult to identify in the Loss On Ignition record (Figure 7.8). The values in this record are low ranging from 3.5-11.3%. The key interval that stands out is from 174-113cm which contains peaks at 134 and 144cm. This is the only period of extended change in the system. LOI is a measure of the amount of organic matter in the system and this peak is matched by a drop in the percentage of calcium carbonate in the system. The available phosphorus record can be split into four zones (Figure 7.9): Zone 1: Base - 171cm: This zone is characterised by short-lived but large excursions in the amount of available phosphorus. An extended excursion occurs between 234- 210cm. After this point the excursions are smaller in magnitude. Zone 2: 171-90cm: This period begins with very stable values which rapidly reach a peak at 140cm. The changes in the values after this point are short-lived. Zone 3: 90-38cm: The values are very stable through this zone. Zone 4: 38-Ocm: This zone records a rapid drop in values reaching zero at 18cm. By the top of the sequence the values exhibited in zone 3 have been restored. The fact that this record can be categorised into the same zones as the isotope records suggests that these really are periods when the system was undergoing radical changes. This record does not however provide the definitive answer to phosphorus 236 patterns in the area. This is because it is the bulk movement of soils that is the key mechanism by which phosphorus reaches lakes and therefore total phosphorus would provide a clearer idea of changing phosphorus levels in the system (Brenner, 1983). C:N ratios were measured on this core ( Figure 7.9). The ratios suggest that the key input into the system is terrestrial organic matter throughout the entire sequence because the values are all greater than 20 apart from at the very top of the sequence (Kaushal and Binford, 1999). This record can be split into four phases: Zone l:Base-269 cm: This zone is characterised by the two large positive excursions at 294cm and 274cm. These suggest that the system was experiencing disturbance during this time frame. Zone 2: 269-180cm: This zone is characterised by very stable levels of C:N ratios apart from an excursion at 205cm. This coincides with the initial prolonged excursion in the 8I3C record. Zone 3: 180-45cm: This zone is also characterised by stable values but these are all less positive than in the previous phase. There are two large positive excursions at 175 and 90cm. Zone 4: 45-0cm: This follows a negative trend to the top of the record. From 8cm the least positive values for the whole sequence are found. The X-Ray Diffraction record is very stable showing one period of change in the system (Figure 7.10). At 177cm there is a rise in the amount of aragonite in the system which is during the zone 3 in the diatom record. This change in the sediment confirms the significance of this key transition zone. Aragonite is the stable form of calcite in hypersaline conditions which is preliminary evidence that this is an arid phase. An alternative explanation is that this is a period of time where there is a high number of gastropods in the sediment. 7.5 Preliminary Analysis: Clear zones of change have been identified in the Lamanai 1999 core. The record which shows the most prolonged periods of change is the 8I3C sequence and more 237 specifically zone 2 within this. In a similar manner to Hillbank, the Lamanai 1999 core was taken from the edge of the lagoon and is therefore most likely to be influenced by changes to the adjacent catchment. The changes in zone 2 are cyclical in their nature which suggests that these might represent periods of increased and decreased activity in the catchment. The 5I3C record has three distinct positive excursions in zone 2. These are estimated to run from: 1. 178-143cm: c.2200-1900 years BP 2. 143-102cm: c. 1900-1400 years BP 3. 102-42cm: c. 1400-830 years BP The three events in zone 2 are different in their nature. The first rose to a peak and then returned to previous values, the second has two peaks and the third is a prolonged, but less severe, excursion. As described in Chapter 4 the 5I3C signature of carbonates reflects the 8I3C ratio of lakewater DIC which is controlled by several factors including lake primary productivity, 5I3C of atmospheric C02, methanogenesis and the 813C of dissolved bicarbonate from the watershed. This latter factor can be affected by the relative abundance of C3 and the isotopically heavier C4 plants in the catchment since these influence the 5I3C of soil CCL. This equates to a shift from a forested to a tropical grassland catchment. The Lamanai 1999 core was taken from the edge of the lagoon and is therefore likely to be highly influenced by changes to the catchment. The shifts that are seen in the carbon record in zone 2 are positive. One causal mechanism behind these changes may be deforestation events because this would result in a change in the isotopic signature of the incoming material in to the lagoon (i.e. a shift from C3 to C4 plants). The peaks of these events occurred at 174cm (c.2180 years BP/ 196 BC), 139cm (c.1820 years BP/ AD227), 112cm (c.1540 years BP/ AD 541), 82-46cm (c. 1240-870 years BP/ AD 782-1192). In Chapter 3 the construction history of Lamanai was explained. In short, two main phases were defined: the Preclassic and the Classic (1500 BC-AD 600) as the time of monument building and the late Classic and Postclassic (AD 800-1200) as the time of monument modification. It is therefore apparent that the first three peaks in the 5i3C record coincide with the period of 238 highest activity in the catchment. The magnetic susceptibility record shown in Figure 7.11 shows three main periods of catchment disturbance. It is the second peak that coincides with the AD 227 event and implies that this had the most impact on the catchment. The peak from 82-46cm is much more subdued and is therefore matches the lowered activity in the catchment during the modification phase. The building of temples would have resulted in a removal of trees from the area which was to be built upon. This would have been compounded by the need to make the lime plaster which covered every temple. An enormous number of trees would have had to be cut down to heat the kilns. Experiments are being undertaken at the moment to determine how much wood had to be burnt to make the amount of plaster that was used at this site (E.Graham pers.com, 2001). In more specific terms monument N10-43 was completed by 100 BC; P9-2 and P8-12 were completed by AD 250; N10-9 and N9-56 were both completed by the end of the 6th Century (Figure 3.1). The changes in the 8I3C record therefore match the archaeology record exactly. A pollen record from this core would confirm changes to the vegetation and therefore should be regarded as a priority for future research. The record of percentage calcium carbonate provides further evidence that through time the inputs into the lagoon have changed. The times of variation in the 8I3C record match the times of change in the percentage calcium carbonate of the sediment. The drops of 5I3C are matched by drops in the percentage calcium carbonate and vice versa. The peak in the LOI record coincides with the key period of change in the 5I3C record, with the peaks in the 8i3C record being matched by troughs in the LOI. This suggests that the inputs into the lagoon during the building phases were affecting the sedimentary make up of the system. This shows that the different parts of the system are responding to perturbations which must be external to the system as they are all changing at the same time. Deforestation is characterised by an increase in the C:N ratio (Kaushal and Binford, 1999). From the 5I3C record it appears that there are three main phases of deforestation in the catchment. The C:N ratio during zone 2 is lower than the 239 previous phase but there are two positive peaks. These coincide with the first and last increases in 8I3C at 175 and 90cm i.e. c.2200 and 1300 years BP. The second peak of 5I3C in zone 2 does not coincide with an increase in the C:N ratio and therefore may have a different causal mechanism. This is compounded by the fact that this second peak at c.1820 years BP or AD 227 is the only one that is picked up by the magnetic susceptibility record. This could be related to the fact that this temple is located right by the harbour and therefore sediment would have had a direct route to the lagoon and hence the enhanced magnetic susceptibility levels at this time. The other two temples must have been built in much more forested areas and therefore the predominant signal was that of sedimentalogical change. The C:N ratios also show that conditions are different from at any point in the past towards the present day. The 'collapse' period equates to 82-46cm or c. 1240-870 years BP (see Chapter 3). Although this is a peak in the 813C record it is not of the same level of magnitude as seen in the past. This suggests that this period was one of prolonged disturbance rather than severe disruption. This is backed up by the archaeological evidence as this was the period when the Ottawa Complex was being modified (see Chapter 3 and Figure 3.1). The two previous periods of change in zone 2 occurred over approximately 300 and 400 years respectively. This latter phase takes about 570 years. This is a big difference and is additional weight to the idea that this was a time of significant change in Lamanai. Archaeological evidence suggests that Lamanai was not affected by the collapse (see Chapter 3), the carbon isotope record in this study suggests that this was indeed a period of change which was of a different type and magnitude to that experienced in the past. The collapse is thought to have occurred at the same time as the late Holocene dry period seen throughout Central America. This time period in the 5I80 record in Lamanai does not reflect arid conditions, but this is the period of greatest stability in the sequence. The fact that this is not registered as a dry phase is not surprising because of the large size of the lagoon. The evidence from the sediments therefore suggests that the human activities in the catchment have left an imprint on their environment. The verification of the link between the archaeological and palaeolimnological records is reliant on further dating control and more evidence concerning the human society 240 such as population levels. The relationships discussed cannot therefore be regarded as proven. Changes in the diatom record coincide with those in the isotope records. As described earlier the diatom record can be effectively split into two equating to pre- and post-human impact as determined by the isotope record. The period before human impact was one of high reconstructed conductivity and low species diversity. There is then a period where diatoms are not preserved. The XRD data shows this to be a period of increased aragonite in the sediment which may suggest dry conditions. The reconstructed conductivity decreases and species diversity increases after this point. The period from 82-46cm is a time of significant change to the diatom flora. The two species that dominate this zone are Denticula elegans and Cyclotella plitvicensis. The reconstructed salinity for this phase is not high but Denticula elegans is regarded as a high salinity species and is found in marsh environments. The ecology of Cyclotella plitvicensis is unknown and it is not found in the modern environments sampled in Belize. This alone implies that the environmental conditions represented in this zone are ones that are very different to the present day. A scatter plot of 5180 versus 8nC (Figure 6.11) shows that all the points were found in the third quadrant. The inorganic carbon is therefore derived from plant material (as the 5nC values are light) which adds weight to the argument that the 5I3C record is a reflection of the changes to the catchment's land cover over time. As there is not a correlation between Sl80 and 5L3440-2200 years BP show fairly stable conditions. The diatom and §l80 data suggests that the core environment was a littoral well vegetated zone and that the surrounding climate was moist. 2. The key period within which the lagoon appears to have undergone a significant transformation was between c.2200-2150 years BP. During this phase diatoms are not preserved, the 5I80 signal is a positive trend and there are sedimentological changes. It is during this period that the first peak in the 8nC record occurs. This is at c.2180 years BP or 196 BC and coincides with a 245 building phase in Lamanai. This produced the largest Preclassic temple in Belize (N10-43). 3. The second event in the 5I3C record occurred from c.1900 -1400 years BP with a peak at c. 1820 years BP or AD 277. This again equates to a building phase in Lamanai (temples P9-2 and P8-12). This is also a period of change in the diatom record with a drop in the reconstructed salinity and an increase in species diversity. Evidence of catchment disturbance is apparent during this phase. 4. The third peak in the 5I3C record is not as pronounced but it occurs over a much longer time interval than the previous events. This is from c. 1240-870 years BP or AD 782-1192. During this period there was the most pronounced shift in the diatom flora and the bl80 record indicates a drying climate. 5. This period of change is also noted in the Outpost 2000 record with a negative excursion in the 513C record centred on c.1100 years BP or AD 970. This is a higher resolution record and therefore it is not surprising that different signals would be picked up. The key point is that both records show this period to be a time of change. This is also the end of a positive phase in the 5lsO record for this sequence. 6. A drying trend is noted from c. 970-640 years BP which may represent the late Holocene dry period. 7. The Outpost 2000 record also shows preliminary evidence for catchment recovery from AD 1640 onwards (when the site was first abandoned) and possible evidence for disturbance between AD 1862-1917. This is when a sugar mill was operating at the site. 8. The climate shows evidence of drying from 360 l4C years BP to the present day. 246 Lamanai, New River Lagoon Basemap: DOS Sheets 15, Edition 5-GSGS (1993) & 10, Edition 5-GSGS (1994), Scale 1:50,000 Projection: Universal Transverse Mercator (Zone 16) Water Bank GILLETTS LAMANAI 1999 Indian Church £ OUTPOST •!# 2000 Burnhom 'Hill lamana i -p i' ■ ■ l968000 m— 000 m E Figure 7.1 Modified from Murray (unpub.) Figure 7.2 View from a Lamanai temple, looking south along New River Lagoon towards Hillbank. Lamanai 1999 Depth (cm) 3750 3500 - 3250 - 3000 - 2750 - Q. 2500 - " 2250 - ro 2000 - ® 1750 - O 1500 - - 1250 - 1000 - 750 - 500 - 250 - 0 - Radiocarbon dates for Lamanai 1999. The points joined up are from terrestrial organic matter. The single point is the date from the gastropod. The error ranges can be found in Table 7.1. tamanai,NewRiverLagoon Figure7.3 810+/-40- 3070+/-50- 3440+/-40- 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 -C 150 CL CL) 160 Q 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 M A® \p Litholoqy.V1 NV.-»^ ^'/»V- NK-.>s- ✓VK-n.. •V>y \AK v # c/ c-yo c°° O-VV _ #■ 20 0 20406 <o jf #//c/ -1 ♦ « ♦ 0.5 1 1.5 2.5 3.5 Axis One Lamanai 1999 (Samples) 2.5 n 2 ? o 1.5 - o o r>0 ♦ ♦ 1 1 J ♦ ♦ ♦ 0.5 h ♦ ♦ 0.5 1.5 Axis One 2.5 Figure 7.6 Lamanai 1999 Detrended Correspondence Analysis The species fall into two groups: 1. Achnanthes exigua, Brachysira neoexilis, Cocconeis placentula var. placentula, C.placentula var. incisa, Cyclotella distinguenda, Denticula elegans, Encyonema carina, Gomphonema gracile and Navicula radiosa (diamonds). 2. Denticula tenuis, Mastogloia smithii, M.smithii var. lacustris (open squares). The samples shown by open circles are those where Denticula elegans is extremely dominant as compared to M.smithii var. lacustris. Figure7.7Lamanai1999180and3CReco s Figure7.8Lamanai1999PercentageB lkCarbo ta dossIgnition Lamanai1999 Figure7,9Lamanai1999:Availablephosphorus C:NRatio 0,010.02 ,345 . Figure 7.10 Lamanai 1999 Percentage 0 25 50 75 100 Key: Pink: Calcite Orange: Aragonite Blue: Quartz Green: Smectite Lamanai 1999 Magnetic Susceptibility (LF) SI -10 12 3 Lamanai 1999 Magnetic Susceptibility (HF) SI 0 12 3 4 Figure 7.11 Lamanai 1999 Magnetic Susceptibility Figure7.12Outpost2000180and13CReco s Blue:Cochliopinas Green:Pygophorissp Red:MixofPygophorussp Outpost2000 Outpost2000 Carbonate Outpost2000 MagneticSuscep bility 0i 5 360+/-410 15 20 25 „30 1.35 &40 o 45 50 55i 60J 65j 70 75J •■/////////////. ■'/////////////. ■'/////////////. '/////////////. S////////S////. 72%4 -0.75 -0.5 -0.25 Figure7.13Outpost2000BulkCarbonatea dM gneticSusceptibility Chapter Eight: The Results from Honey Camp Lagoon This chapter presents the results from Honey Camp Lagoon. Cores were collected over two field seasons (1999-2000) using all three coring methods referred to in Chapter 4. The locations of the cores are shown in Figure 8.1 and the stratigraphies are detailed in appendix 6. 8.1 Honey Camp Lagoon 1999 This core was taken from the edge of the southwest shore of the lagoon on solid ground (Figures 8.1 and 8.2). This is the first of four cores that were analysed from this lagoon. Oxygen and carbon isotopes on both the bulk carbonates and two species of gastropod (Cochliopina and Pygophorus sp) were measured on this core. This record is 280cm long and has been analysed every 5- 10cm. Three radiocarbon analyses have been made for this sequence: Table 8.1 Code Depth Material l4C years BP SI3C pdb %c Calibrated 2 sigma (cm) +/- 0.1 age range AA-45637 39 G 1670+/-45 -1.5 / / (GU-9580) AA-39724 49 OM Modern -28.2 / / AA-39723 237 OM Modern -26.7 / / There is no clear reason why two of the samples would be modern and discussions with Dr. Charlotte Bryant at the NERC Radiocarbon Facility have not been able to resolve this. It is possible that the material sampled may not have been in situ to the core and therefore could have fallen into the core during sampling. Alternatively, the material sampled could have been root matter, although it did not have this appearance. The date at 39cm is from gastropod and therefore will be affected by a hardwater error. This error can only be established by having dates from terrestrial organic 260 matter to compare with. A paired date was gathered from core LI and is discussed in section 8.4. The hardwater error that was established from core LI is 1105 years. The hardwater estimates from the two cores in New River Lagoon were comparable and because this is a much larger system it is feasible to apply the hardwater error calculated from LI to Honey Camp Lagoon 1999. The oxygen isotope record is much more variable than the records from the New River Lagoon (Figure 8.3). From the base of the record to 75cm the 5i80 values are variable but a prolonged trend is not apparent. The most positive point, in this part of the record, is at 260cm where values reach -1.3%c. The bl80 record gradually returns to more negative values and stabilises around -3.5%o. From 140cm the record becomes more noisy which culminates in a negative peak at 70cm where values reach -7.5%o. This is a very short-lived event and by 65cm average trends have been restored. Levels fluctuate around zero to the top of the record. These are the most positive values in the whole sequence. Gastropods are only present in the top 40cm of this core (Figure 8.3). The values recorded from Cochliopina sp and Pygophorus sp are both similar to the bulk carbonate record and to one another. The gastropod records are more variable than the bulk carbonate values. This is probably a function of the habitats in which they live and the shorter time scales that are represented by the shell data. The carbon record can be split into three main zones (Figure 8.4). From the base of the record the carbon signal is extremely stable with the mean value being -10.3%c and the standard deviation being +/- 0.6. At 70 and 55cm there are two short lived events which show a great depletion in the carbon values to -19.6 and -17.4%c respectively. 8I3C then rapidly increases and is stable around 0%o to the top of the record where there is a small positive peak from 25 to 15cm. Values from Cochliopina and Pygophorus sp. are consistently more negative than the bulk carbonate record apart from one value at the base (Figure 8.4). The Cochliopina sp. 261 record is the most variable and apart from at the top of the sequence, the species seem to be recording a similar signal. The values found in the LOI record (apart from at the top of the record) are extremely low and do not reach higher than 5%. This is lower than the values seen in the Lamanai 1999 sequence. This is evidence that the two systems are different in their sedimentary make up. The Honey Camp record can be split into three zones which are based on clear shifts in the amount of organic matter that is present in the sediment (Figure 8.5): Zone 1: Base - 252cm: Although this zone has a very small range of values there are fluctuations between depths. Zone 2: 252 - 22cm: This is a very stable zone and levels are clearly lower than in the previous section. In a similar manner the values fluctuate between depths but within a very small range. Zone 3: 22-0cm: The uppermost sediments show a massive increase in organic matter. This is a function of these sediments not being lacustrine (see stratigraphy in appendix 6 for details). The particle size record can be divided into two zones (Figure 8.5). From the base to 131cm, the record fluctuates between high and low values. This variability suggests that inputs into the catchment are highly changeable. From 131cm onwards the record is very much less variable apart from a large peak at 60cm. This peak is of the same order of magnitude as changes seen lower in the core. The shifts in the mean particle size of the sediments suggests that the inputs into the lagoon have changed over time. This matches the XRD record which switches between quartz and calcite domination in the first half of the record before becoming much more stable. In this stable section, quartz dominates. Levels of calcite increase at the same time as the major negative excursion in the isotope records at 70cm (Figure 8.6). The magnetic susceptibility record is quite different from the rest of the proxies (Figure 8.7). There are two clear peaks at 280-260cm and 195-187cm. From 160cm 262 to the top of the record the values of magnetic susceptibility are very much lower. This record is not mass specific and therefore cannot be compared to the other magnetic susceptibility records. The percentage calcium carbonate record has two phases (Figure 8.7). From the base of the record to 160cm the values climb steadily from 0% to 80%. By 152cm the levels have dropped to 8% and remain low until 101cm where values reach 49%. This is a short-lived peak and levels have dropped to low values by 99cm. Values remain low until 50cm after which they increase to 40% by the top of the record. From the base to 256cm of the available phosphorus record (Figure 8.8), the values follow a declining trend. From 256-211cm values rise and fall in a concave pattern. There is a peak in available phosphorus at 206cm (6.3 ppm) but these have dropped by 198cm. From 193cm upwards, values although fluctuating, are very stable and low. Values rise to 5 ppm at 53cm falling gradually to 18cm. From then on they rise rapidly to the top of the sequence and at 8cm the highest values in the sequence are found at 8.1 ppm. This rise at the top of the sequence is very similar to the pattern exhibited in Lamanai. Throughout the whole sequence of Lamanai values are much higher than those in Honey Camp Lagoon. The C:N ratios of the sediment are extremely variable in this sequence fluctuating between very high (150) and low values (0) (Figure 8.8). This suggests that the inputs into the system have been extremely variable through time. This is most apparent in the lower half of the record from the base to 128cm. This matches the enhanced variability of the particle size record in this period and is further evidence that the inputs to the lagoon were different in this period to more recent times. The limiting variable in this sequence is nitrogen. 263 8.2 Preliminary Analysis: For the purposes of comparison and in order to create a preliminary chronology for this record, a constant sedimentation rate was assumed through time. The problems of doing this are highlighted most clearly in Figure 6.3. Due to the paucity of dates from this record it was decided that this was the best strategy to employ because it provides the basis for comparing between records. The extrapolated dates should therefore be treated with caution. All the dates referred to have been corrected for the hardwater error of 1105 years after the extrapolations took place. The base of the isotope record is 280cm which equates to 10,900 years BP. From this it appears as if this record covers the same time period as Hillbank 1998. This raises a number of important points: 1. Both Hillbank 1998 and Honey Camp Lagoon 1999 cover approximately the same period of time even though they are very different lengths (2.8m as compared to 14m). 2. The signal for the Lateglacial in Hillbank 1998 represents the driest conditions seen in this record. The base of Honey Camp 1999 (280-240cm) is also drier than the main period of the sequence (240-75cm) but this period of time does not represent the driest conditions seen in this record (40-0cm). The two periods of time where conditions are thought to have been dry in Central America are during the Lateglacial and the late Holocene. The Hillbank 1998 core does not cover this latter period. It is likely that the late Holocene dry period is covered by the top of the Honey Camp Lagoon sequence. This will therefore allow a preliminary assessment to be made concerning the relative severity of these events. 3. Both records demonstrate a stable climate for the main period of the Holocene (240-75cm or 9200-2100 years BP). The values that are seen in the 5180 sequence from the base of the record to 75cm in Honey Camp Lagoon 1999 are very similar to those in the New River Lagoon (Honey Camp average: -3.7%o; Lamanai average: -3.9%o; Hillbank average: -3.5 %6). This suggests that during this period the systems were more similar than they are today. 264 From the base of the record to 130cm (c. 4500 years BP) the particle size and C:N ratios are at their most variable. All this evidence points to this period being one where the system is most influenced by the catchment rather than within-lagoon processes. If the lagoon was an open system through the main period of the record (like the New River Lagoon) then the inputs into the lagoon would be very dependent on the inputs provided by rivers and the surrounding marshland. These are likely to be changeable which therefore accounts for the variable particle size and C:N ratios in the first half of the record. This is also shown in the XRD results which show a great deal of variability during this period between the amount of quartz and calcite in the sediment. The magnetic susceptibility record suggests there were two key disturbance events in the catchment at 192 and 280-260cm. The latter peak coincides with the most positive signal in the 6I80 record at the base of the sequence and dates to approximately 10,900-10,000 years BP. The second peak dates to approximately 7100 years BP. Changes in the calcium carbonate or XRD records do not match those in the magnetic susceptibility. This suggests that the peaks are not due to changes in the type of sediment entering the system as has been noted in the records from the New River Lagoon. The negative shift at 70cm is very significant and it represents the systems ability to rapidly change. Its importance is accentuated because the system does not return to its previous S180 condition and enters a new state, which is totally different from that experienced in the past. The shift at 70cm occurs at approximately 1900 years BP. One issue that is not clear concerning Honey Camp Lagoon is its past connection to the sea (see Chapter 4). 70cm could perhaps mark an event which led to the closure of the lagoon resulting in an isotopic signature much higher and more sensitive to change i.e. what one would expect from a closed system. The 70cm event could be a flooding of the system as a result of the closure of Freshwater Creek which may have 265 temporarily increased the water volume in the lagoon. This is supported by the two negative excursions in the 5I3C record. The 813C values exhibited in this record are very different from those in the New River Lagoon. In comparison levels are very depleted which could be for a number of reasons: 1. Groundwaters from well-vegetated catchments are more influenced by the decomposition of isotopically light plant material than those from drier, thinly vegetated catchments. The former tend to produce runoff with more negative 513C than the latter (Talbot, 1990). 2. Climatically induced floral differences (e.g. the shift from C3 dominated to C4 dominated systems) will also influence the 5I3C of the inflow producing isotopically negative inflow from humid catchments. 1 o 3. High level of oxidation of organic matter which results in high levels of C in the TDIC pool. The atmosphere has a §I3C of -7%o and at the surface water interface (at approximately 20°C) this carbon will be transformed to bicarbonate and then precipitated to carbonate. This will be in equilibrium with the 13 atmosphere and will therefore have a C value of 0%o. The highly negative carbon values suggest the carbon being measured has been oxidised from organic matter. 4. The two very negative shifts could represent times when the carbon has been completely derived from aquatic plants. This is not however reflected in the C:N ratios. 5. High rates of flushing also explain low carbon levels as the TDIC would not have time to equilibrate with atmospheric CO?. 6. There must have been a change in either the source or the mix of sources to result in the base of the carbon record being so stable and the top so variable. The change in the system at 70cm must therefore be a very significant time in the history of the lagoon. 266 On the Sl80 versus 5nC plot there is a correlation between the two variables with an R~ of 0.6622 (Figure 6.11). The data does not form a continuum along this line and appear to be composed of two separate populations. This relationship is driven by the significant difference between the top of the record and the rest of the sequence. This is therefore clear evidence in favour of the idea that the event at 70cm resulted in the lagoon taking on a very different form. The shift to the most positive 5I80 conditions begins at 55cm or c. 1200 years BP. This is an isotopic shift of 2.8%o. This is highly significant and coincides with the late Holocene dry period. 8.3 Honey Camp 2000 (L4) Core L4 is located near the western shore of the lagoon, north of the Honey Camp Lagoon 1999 core (Figure 8.1). It is a 38cm Livingstone core taken in approximately 8 metres of water. L4 has been analysed for oxygen and carbon isotopes in both the bulk carbonates and Pygopliorus sp. measured every l-2cm. Diatoms were not preserved in this sequence. Due to the unconsolidated nature of the sediments in the New River Lagoon the cores retrieved all covered different time periods. The sediments in Honey Camp Lagoon are much more coherent and therefore it is more likely the cores will cover much more similar time periods. The top sediments in Honey Camp Lagoon are a very fine material with scattered gastropods. This extends for 12cm in Honey Camp 2000 (K), 3cm in LI and 1cm in L4. The records may therefore only be slightly offset. Table 8.2 Code Depth Material l4C years 5I3C pdb %o Calibrated 2 (cm) BP +/- 0.1 age sigma range AA-45636 17 G 3705 +/- 55 -4.4 / / (GU-9579) The implications of this date will be discussed further with core LI. 267 8.3.1 General description: The oxygen isotope record is stable with all the values falling close to zero apart from a positive excursion to 1.5%o at 14cm (Figure 8.9). These values match the top of Honey Camp 1999. The excursion at 14cm is a very significant change in the system although it is short lived. Limited Pygophorus sp. shells were analysed from this core and these show a large offset from the bulk carbonate record. This core was taken from the main body of the lake and this species is generally found in littoral environments and thus the offset could be a function of the different environments the two records represent. The carbon isotope record is much less spiky than the oxygen isotope record, but in a similar manner to Honey Camp 1999 the values fluctuate around zero (Figure 8.9). Similar to the oxygen record, the Pygophorus sp. record is offset with the values being consistently more depleted than the bulk carbonate record and exhibiting a larger range of values. This is most likely to be a response to the different carbon sources which would be available to the gastropods. Both the carbon and the oxygen records from 10cm to the top of the record are very stable which suggests the very recent past in Honey Camp Lagoon has been so. Figure 6.11 is a scatter plot of 5lsO versus 5I3C. These two variables are not correlated but the values are scattered in all four quadrants suggesting clear changes have occurred throughout the core's history. The removal of the outlying point at 14cm does not alter this. L4 has also been analysed for percentage calcium carbonate and magnetic susceptibility (Figure 8.10). The carbonate record is fairly steady apart from a sharp temporary drop in levels at 22cm to 20% from approximately 75%. Levels remain fairly steady apart from a further small drop at 12cm. The amount of variability is less in the top half than in the bottom half of the core. The magnetic susceptibility 268 record can be separated into three phases. From the base of the record to 27cm the values gradually decline, this is reversed between 27-17cm. Values drop by 15cm and from this point to the top of the record values increase to the highest in the sequence. The highest value is found at 2cm reaching almost ISi. 8.3.2 Preliminary Analysis: As described earlier if the system is in equilibrium with the atmosphere it will therefore have a SI3C value of 0%o. If this holds true then the 5I8C of the water should be 0%c unless it has been enriched through evaporation. The key episode of change in this sequence is at 14cm. This is significant because it is the largest excursion in the sequence, representing an isotopic shift of 2.6 %o. The oxygen isotope signal at this time therefore represents a time of enriched evaporation and the carbon record has higher levels of 5l3C which can indicate atmospheric exchange or increased aquatic productivity. 8.4 Honey Camp Lagoon 2000 (LI) This core has been analysed for oxygen and carbon isotopes and is located approximately 700 m north of core L4 (Figure 8.1). It is 36cm long and has been analysed every l-2cm. Three radiocarbon dates have been obtained for this sequence: Table 8.3 Code Depth Material 14C Years 5I3C pdb%C Calibrated 2 Sigma (cm) BP +/- 0.1 Age range AA- 19 OM 3315+/- 53 -28.2 1604 BC 1738- 42419 1455 BC CAMS- 19 G 4420+/- 40 -3.5 / / 77199 AA- 7 OM 764+/-37 -27.7 AD 1276 AD 1214- 42418 1295 269 From the paired date at 19cm the hardwater error for Honey Camp Lagoon is 1105 years. This is 422-555 years less than in the New River Lagoon, but is in the same realm suggesting that there is a consistency throughout this region. The oxygen record shifts from positive values from the base of the core to 19cm and then to negative values to the top of the sequence (Figure 8.11) within a range of 5i80 values between 0.58 and -0.65%c. Both Cochliopina and Pygophorus species were analysed in this core. The differences in oxygen isotope values between these two species is marked at the top of the core but diminishes towards the base, suggesting that the relationship between these two species has changed over time. The Cochliopina species used in this record were very small and thus they may have been juvenile. The record from gastropod shells is time averaged over their life cycle and therefore a younger specimen may not be recording the same time period as an older one and thus the records may not be comparable. The carbon isotope record shows the opposite trend from the oxygen isotopes, moving from negative values at the base to positive values at the top of the core (Figure 8.11). The switch over occurs at the same point in the record at 19cm suggesting that this is a significant time in the sequence. The range of values is large from -3 to +2%c which is greater than in L4. The gastropod species are consistently more negative than the bulk carbonate results and the offset between the two species is much less than in the oxygen record. This suggests that the two species have similar carbon sources which is a factor that is unlikely to vary through a life cycle. Both percentage calcium carbonate and magnetic susceptibility (Figure 8.12) have been measured for this sequence. The magnetic susceptibility record is similar to that produced for L4 although they cannot be strictly compared because they are not mass specific records. The base of the record is negative and values rise to a peak by 17cm and drop by 15cm. From 15cm to the top of the sequence values rise steadily. The calcium carbonate record can be divided into two halves with the base of the record to 21cm being stable, apart from a peak in values at 33cm. From 21cm to the top of 270 the record the values follow a concave pattern before increasing from 5cm to the top of the sequence. C:N ratios have been determined for this core and values are stable at around 17 apart from an excursion between 20-5cm (Figure 8.12). Values increase to the top of the record. During the excursion, the inputs into the system are from terrestrial organic matter. For the rest of the sequence values are intermediate between algae and land plants suggesting that both are contributing to the system. This matches the S|3C record which switches from negative to positive values and back to negative at the top of the record. The positive phase coincides with the increase C:N ratio levels at 16cm which is a signal from terrestrial plants entering the system. An increase in 5I3C implies either an increase in aquatic photosynthesis (which does not tally) or a change in the inputs to the catchment. 8.4.1 Preliminary Analysis: In terms of the stratigraphy for LI and L4 two points are important to help in the understanding of how the two cores match together. Firstly, as has been explained earlier there is more of the top sediment in LI than L4 which suggests that L4 is approximately 2cm offset from LI. Both records demonstrate a clear banded structure, but the L4 sediment is much more detailed suggesting that the sedimentation rate in this part of the lagoon is higher than in the LI region or it is more variable. The base of both LI and L4's 5lsO record are characterised by short excursions, both following the same positive to negative trend. The records are approximately 2cm offset and thus the large positive excursion in L4 between 15-13cm is matched by a small positive excursion in LI between 17-14cm. The L4 excursion occurs during a pale sedimentary band in the sediment. The previous positive excursions at 22 and 28cm in the L4 record also occur in pale bands. These bands are not present in the LI record suggesting that the signal may be dampened. 271 The 513C record in both sequences is less variable than the 5180 sequences. The record from L4 is much more changeable and does not display the clear trend seen in LI. There is a clear change in LI between 17-14cm and this is matched in L4 with a similar positive trend from 15-13cm. A positive signal in 5I3C implies there has been an increase in CCL exchange with the atmosphere. This occurs at the same time as a positive excursion in the Sl80 record. The synchronicity of both isotope changes in both cores is clear evidence that this represents a drying event. This is supported by the similarity between the two magnetic susceptibility records from L4 and LI which suggests that this proxy is responding to changes in the catchment. This is further evidence that the two core locations have been influenced by the same environmental changes through time. LI has two dates which are surprisingly old for the length of sediment retrieved. There is both an organic matter and a gastropod date at 19cm. The date from L4 at 17cm is 2600 years BP when it has been corrected for the hardwater error. This is comparable with LI and confirms that the records are approximately 2cm out of phase. It is possible there could be a break in the sedimentation between 7 and 19cm. There is however no obvious point in the sedimentary record where this could have occurred apart from a section of fine organic bands around 17cm in LI which may indicate shallowing (S.Metcalfe, pers.com., 2000). Both records therefore have evidence for a dry phase and if this was severe enough it could have resulted in a dramatic decrease to the lagoon's sedimentation. In order to determine whether or not this is the case, very close interval sampling as well as dating would have to take place. This however would prove difficult with these cores due to the lack of terrestrial organic matter available in the sediment. Gastropods could be dated but the one paired date would not be sufficient to apply a correction factor when such high resolution information would be required. The drying event in cores LI and L4 can therefore only be ascribed with any certainty to be older than 764 l4C years BP which means that it could be a signal from the late 272 Holocene dry period. The period of time where the driest conditions are noted in the Punta Laguna record are 1785-920 14C years BP with maximum conditions occurring between 1225-930 l4C years BP (Curtis et al., 1996). The human occupation of Honey Camp Lagoon is thought to have occurred in two phases: 1. Postclassic: AD 1100 onwards (940 years BP onwards). 2. Terminal Classic/early Postclassic: AD 750-1100 (1300-940 years BP) The dates were converted to radiocarbon years according to Stuiver and Pearson (1993). It could be postulated that these two communities centred around the dry phase seen in the record but this cannot be substantiated with the evidence available. The scatter plot of §180 versus §13C for LI has an R value of 0.6128. This is highly significant (Figure 6.11). This suggests the shift from positive to negative values in this sequence is a real trend and represents a shift over time in the dynamics of the lagoon. The most interesting point is that this is not so clear in the record from L4. This is most likely to be due to core location. LI was taken from a narrower portion of the lagoon and therefore it could be argued this would be a more sensitive area to change. 8.5 Honey Camp 2000 (K) This Kullenberg core was taken from the same position as LI in Honey Camp Lagoon (Figure 8.1). The core was analysed for oxygen and carbon isotopes in both the bulk carbonates and gastropod species. The species analysed were Cochliopina and Pygophorus sp and the latter was investigated for the difference between a single gastropod and a mix of specimens in a similar manner to Outpost 2000. This record is 78cm long and was analysed every 4-13cm. Diatoms were not preserved. There are at present no dates from this sequence. 273 The analyses for this core are at low resolution and therefore the changes in the system may in reality be more complicated, but it does give an idea as to the general changes that have occurred in the most recent period of Honey Camp Lagoon's history. The 5I80 record comprises mainly slightly negative values with two positive excursions at 40 and 10cm (Figure 8.13). Cochliopina and Pygophorus species were both measured in this core (Figure 8.13). With the latter species both a single specimen and a sample from a mix of specimens was analysed. The relationship between the species is not consistent through time. At the base of the record the mixed Pygophorus assemblage and single Cochliopina are synchronous, with the single Pygophorus following an opposite trend. From 40cm upwards single Pygophorus and Cochliopina are synchronous and the mixed Pygophorus assemblage shows an opposite trend. The most interesting point is that the mixed Pygophorus assemblage and the single Pygophorus specimens do not show the same record suggesting the mixed assemblage maybe from individuals that are not exactly the same age and thus the record is an average over a wider time scale. The records from the gastropods are much more variable than the bulk carbonates. This is a trend noted in other sequences. The 5i3C record shows stability from the base to 55cm, values then continue to be stable but are much more negative. From 30cm to 15cm there is a prolonged positive trend which is reversed to the top of the record. This record is echoed by the Pygophorus sp record (Figure 8.14). Magnetic susceptibility has also been measured for this sequence (Figure 8.15). The values are all negative apart from at the base of the sequence. This record should be treated with caution because it is likely these measurements are at the limits of the sensor's capability due to the number of points that have the same value. C:N ratios were measured on this core and the record falls into two distinct phases from the base of the record until 36cm where values are stable around 33; there is then a 274 transitional phase where values drop and then from 26 to 5cm values are stable at 16 rising to 19.5 at the top of the record (Figure 8.15). 8.5.1 Preliminary Analysis: Both the 5I3C and the C:N record show a change in state from 35cm onwards suggesting there has been a change in the source of organic matter reaching the system. This same reciprocal relationship is seen in LI. The distinctive part of the LI and the L4 records is the banding of the sediments. This banding is only apparent between 12-30cm in the Honey Camp 2000 record. The rest of the sequence is uniform grey sediment. This has two implications: 1. If there has been a break in sedimentation in LI and L4 it affected the whole lagoon as the three cores all show approximately the same amount of banded sediment. 2. The oxygen and carbon isotopic signature for 12-30cm in Honey Camp 2000 is akin to LI, L4 and Honey Camp 1999 for the equivalent depths which is evidence that the cores may be recording the same events. The record from Honey Camp 1999 undergoes a large isotopic shift at 70cm. This is not apparent in the Honey Camp 2000 record at the same depth. Through looking at the relationship between LI and L4 it is apparent that although sequences may cover the same time period, the recording of the events may not be the same. The base of the Honey Camp 2000 5lsO sequence is the most negative for the whole sequence implying this period is a time of change even though the signal is not as large. This is not unexpected as the Honey Camp 2000 sequence was taken from 4 metres of water and therefore it may be less sensitive to change than the shore record. The dry period apparent in LI and L4 is not picked up in the Honey Camp 2000 record but this is more likely to be a result of sampling resolution. The C:N record also does not register change through this period. A positive excursion does occur at 275 40cm. This is where the shift to dry conditions is completed by in Honey Camp 1999 and dates to 565 l4C years BP (after hardwater correction). The two Livingstone cores taken in Honey Camp Lagoon (L4 and LI) are very short. In both instances the corer was unable to penetrate the sediment any deeper. Several other cores were also taken from the lagoon and these were of a similar length which suggests there is a lagoon wide sedimentary layer which is much denser than the sediments above it. The cores do not however record this, as the corers did not penetrate the layer. It is important to try and work out how Honey Camp Lagoon 1999, LI and L4 match. The links can only be regarded as preliminary due to the poor dating control on the cores. The transition to positive 8lsO values begins at 55cm (1251 years BP) and is complete by 40cm (565 l4C years BP). The peak in positive conditions in LI and L4 at around 15-17cm can only be said to occur at some point after 764 l4C years BP. The hypothesis is that this peak in L1/L4 coincides with this transitional zone in Honey Camp Lagoon 1999. The idea was postulated that the L1/L4 have been affected by a break in sedimentation because the date for 19cm of 3315 l4C years BP is extremely old for such a short depth. The equivalent time period is equal to 77cm in the Honey Camp Lagoon 1999 record. One idea is that the break in sedimentation occurred after the peak in the isotopic signature in LI and L4. The break in sedimentation therefore encompasses the large negative shift in the 5180 record at 1894 years BP. This suggests this was a very significant event and may be part of the reason behind the postulated break in sedimentation. 8.6 Summary of Honey Camp Lagoon The key difficulty in interpreting the records from Honey Camp Lagoon is the lack of a reliable chronology within which to place the changes. The dates that are quoted should therefore be regarded as preliminary. The following summary points can be made: 276 1. It is postulated that the Honey Camp Lagoon 1999 sequences covers the entire Holocene period. The transition to the Holocene in the Hillbank 1998 record is marked by the most positive 5I80 values for the entire sequence. The values for the same time period are more positive in Honey Camp, but they are not the most positive in the whole sequence (10900-8300 years BP). Further dates are needed to clarify that the record does indeed cover this time period. 2. The main part of the Honey Camp 1999 5I80 record is stable (8300-2100 years BP). The values shown are akin to those in the New River Lagoon. 3. The 5I3C results in the same period are however very different suggesting the carbon systems for the two lagoons are driven by different causal mechanisms. 4. The first half of the Honey Camp Lagoon 1999 record (up to c. 4500 years BP) shows highly variable mean particle size and C:N ratios which is evidence the system is being influenced by the catchment rather than just within-lagoon processes. This is more confirmation that Honey Camp Lagoon in this time period was more similar to the New River Lagoon than it is at present. 5. The key period of change in both the 5I80 and the bl3C record in Honey Camp Lagoon 1999 is at 70cm (c. 1900 years BP) which is marked in both as a large negative excursion. The causal mechanisms behind this are likely to include a change in the catchment vegetation, an increase in the oxidation of organic matter and high rates of flushing through the system all within a much moister climate. 6. It is postulated that LI and L4 have been affected by a break in sedimentation which is likely to have occurred at the time of the large negative shift in the Sl80 record from Honey Camp Lagoon 1999. This is further evidence this was a very significant time period in the lagoon's history. 7. There is a second negative peak in the 5I3C record at 55cm (c. 1200 years BP). This is also where the transition to the most positive 5I80 values in the sequence begins. 8. The levels of both 5lxO and 5I3C are very much enhanced in the top 40cm of the record. This is the only zone where gastropods are preserved. LI, L4 and Honey Camp 2000 all have similar isotopic values through this period. 277 9. LI shows clear evidence that this part of the lagoon has operated as a closed system throughout the time period covered by the core. The difference between LI and L4 highlights the importance of core location in the dissemination of trends. 10. The key point of change in the LI and L4 record is a positive excursion in both the 5I80 and 5nC records at some point before 764 l4C years BP. The coherence of both cores and proxies suggests this was a severe event and therefore could be related to the late Holocene dry period. 11. The human occupation history of Honey Camp Lagoon occurred in two phases. It is possible that the interruption was the late Holocene dry period and therefore the Mayan people in this area may have been influenced by the climate to a greater extent than the population at Lamanai. 278 Honey Camp Lagoon Basemap: DOS Sheet 5 Edition 3-GSGS (1994), Scale 1:50,000 Projection: Universal Transverse Mercator (Zone 16) 343000mE 44 45 46 47 48 49 3S0000 Figure 8.1 Modified from Murray (unpub.) Figure 8.2 Honey Camp Lagoon HoneyCamp1999 HoneyCamp1999 180 -8.06420.02 HoneyCamp1999 180 7^ ■V;:,.v■,s■ >:■;, ■: *:; ■■■ ;,.1.,V■IVs" -4.03-210.012 BulkCarbonate Blue:Cochliopinas Green:Pygophorussp Figure8.3Hon yCampLagoon1 0reco ds(N tchangefs alinstr podc r ) HoneyCamp1999 HoneyCamp1999 HoneyCamp1999 -20.0155.0.5 ■■■ ;; ,■■■ ■;j-;■ ■; ;■■■ 1■> ■; ;1-■■.v ■■■■■j" Blue:Cochliopinas Green:Pygophorussp Figure8.4Hon yCampLagoon13Creco ds(N tchangeis alastr podc rd) HoneyCamp1999 -VV ,■i■.11%%«i ■■ ■VWr'iW. ,■v%I_V-v«', HoneyCamp1999 LossOnIgnition 0.0%510.0 15.0% HoneyCamp1999 ParticleSizeAnalysis(m crons) 01002345 Figure8.5Hon yCampLa oonssIgnitindP rticlSizeAnalysi Figure 8.6 XRD Honey Camp 1999 Percentage 0 25 50 75 100 Key: Quartz: Blue Calcite: Pink Smectite: Orange Green: Kaolinite Figure8.7Hon yCamp1999Ma neticSusceptibilitydPerce tagea bo a es HoneyCamp1999 0 25- 1670+/-45 50- 751 100- —125- E £150- Q. d) Q175- 200-I 225 250- 275- 300- HoneyCamp1999 AvailablePhosphorus 246 10 YMWA ■;■■ AVjWtfrta; ■-■13S Figure8.8AvailablePhosphorusndC:NR tioiHon ympLa o n1999 HoneyCamp1999 C:NRatio 025507100255 Figure8.9Hon yCampL4St blIsotoperesults Opendiamonds:Pygophorussp re8.10HoneyCampL4a bo atedMagneticSusc ptibilityResul s Figure8.11Hon yCampLSt blIs topeResu ts Blue:Cochliopinss Green:Pygophorussp Figure8.12Hon yCampLc rbonate,magneticsuscep bilityd:Nr tioe ul s Bulkcarbonatelu ;Cochli pinas Green:Pygophorussp Red:MixofPygophorussp Figure8.13Hon yCamp2000stableiso opresults Figure8.14Hon yCamp2000St blisotopresults Blue;Cochliopinas Green:Pygophorussp Red:MixofPygophorussp HoneyCamp2000(K) C:NRatio 010234 JI Chapter Nine: Discussion and Conclusions This chapter aims to bring together the results of this thesis from both the modern and the past environments of Belize and sets the findings within the context of regional changes. 9.1 The modern environment From the modern limnological studies it is apparent that the lakes studied in Belize fall along a geochemical gradient from calcium-bicarbonate (freshwater) to calcium- sulphate to sodium-chloride dominated systems. Diatom preservation was not good in the latter group following the pattern set out by Flower (1993). In terms of the fossil diatom records, preservation was changeable in the New River Lagoon (calcium-sulphate dominated) and very poor in Honey Camp Lagoon (sodium- sulphate-chloride dominated). In order to ensure that diatoms are preserved through the sequence, cores from the freshwater sites could be taken (i.e. from Kates Lagoon, Crooked Tree Lagoon and Chiwa Lagoon (Figure 4.1). Access to the latter two sites is difficult. Silica availability is a key variable important in diatom preservation. This would also need to be monitored to determine whether silica availability is a limiting factor for diatoms in Belize. Results from this would also aid site selection for future diatom work. The characterisation of the water bodies visited was not sufficient to enable transfer functions for Belizian diatoms to be created. This must be regarded as a priority for future research in order to enable quantitative diatom species ecology in Belize. Such information is particularly pertinent with regard to the ecological relationship between different geographical areas. It has become apparent in the samples investigated, that habitat is one of the important factors controlling distribution of the diatom species. This may, however, be a function of the sampling strategy employed. Further water chemistry analysis would enable an improved understanding of the role of this variable in controlling species distribution. One way in which .294 water chemistry data could be more intimately connected to the diatom samples would be through the collection of water chemistry variables in the weeks running up to the collection of the diatom sample. These data would then provide an indication of the range of conditions to which the species are adapted. Studies in the modern environment are essential to the understanding of the forcing factors behind change. The role of habitat in the evolution of two varieties of Mastogloia smithii var. lacustris could not have been determined without such a study. This is especially important as both of these varieties are present in the fossil record. The environmental reconstruction from the core Lamanai 1999 is much more tightly constrained as a result. The study of Mastogloia smithii var. lacustris also highlighted the point that species found in the tropics may not fall neatly into the categories assigned to European species. There is the potential for a great deal more taxonomic work in Belize with samples currently only taken from a limited number of habitats at one time of the year. To enable the true characterisation of the water bodies and their diatom populations in Belize, samples would have to be taken throughout the year and with an increased sampling density in lakes so that more detailed ecological information can be gained. 9.2 The past environment From the work that was carried out at Hillbank, New River Lagoon it is apparent that the climate of the late Pleistocene was very variable. Despite the large variations through time in the stable isotope records, the diatom species present in this period are uniform throughout. This suggests that the climatic changes occurring did not influence the factors controlling diatom distribution during this time. It could therefore be presumed that because habitat is an important forcing factor in the present day, it was also important in the past. The transition to the Holocene is marked by very poor diatom preservation and the driest conditions in the entire Hillbank record. This represents a clear shift in climatic conditions and, with improved dating control, this may in fact prove to be a Younger Dryas signal. Drier 295 conditions are also postulated to have occurred at this time at Honey Camp Lagoon. These are not as severe as those in Hillbank, but again improved dating control is needed to verify the exact timing of these changes. For the main period of the Holocene, the records from Hillbank and Honey Camp show stable and moist conditions. The next key period of change begins at c. 4922 years BP in Hillbank. Around this time the isotopic signal and diatom preservation become much more variable. There is also an increase in disturbance pollen. The coincidence of all these factors suggests that this represents the beginning of a new period in Belizean history. It could be suggested that these factors are related, with climatic changes occurring at the same time as the first potential signs of human activity in the area. The shift to late Holocene dry conditions was gradual with the amount of pine pollen in the Hillbank catchment increasing from c. 4800 years BP. From c. 4500 years BP a new phase also begins in Honey Camp Lagoon. At this time the lagoon shifts from being dominated by catchment processes to within lagoon processes. All the evidence suggests that this is an important period of environmental change in this area. This may represent the point at which Honey Camp lagoon shifted from being an open to a closed system. The driving force behind this change could be climatic such as a shift to drier conditions. The oxygen isotopes in Hillbank show a drying trend from c.3000 years BP. At this time there is a significant increase in reconstructed conductivity values from the diatoms at Hillbank. It is also during this period that the Lamanai 1999 record begins. The isotopic values in Lamanai and Hillbank around 3000 years BP are very similar, suggesting that they are recording similar conditions. The Lamanai sequence records clear episodes of change which could be related to human activity in the catchment. These are represented by three main shifts in the carbon isotope record. These coincide with periods of building activity, as deduced from archaeological evidence. These events approximately date to: 296 1. 2180 years BP (196 BC). This is during the period when temple N10-43 was being built. This is the tallest building in Lamanai and would have required a considerable workforce plus associated societal hierarchies to organise the site at this time. 2. 1820 years BP (AD 277). This coincides with the building of P9-2 and P8-12. These are located north of N10-43. P8-12 is located next to a harbour. 3. 1240-870 years BP (AD 782-1192). The peak in the carbon isotope record through this period is not as high as in the past but it is much more prolonged (c.300-400 years cf. c. 570 years). This coincides with the considerable modification of a group of buildings known as the Ottawa Complex. This group is located at the south end of the site. The Lamanai 1999 core was taken parallel to the Ottawa Complex which may account for the prolonged signature in the isotope record. The apparent consistency of the archaeological and sediment records is very encouraging because it clearly demonstrates the impact of humans on the environment. The creation of a pollen record in this area would be beneficial, providing evidence for changes to the amount and type of vegetation in the catchment. Such information is essential to prove the connection between carbon isotope fluctuations and human activity. Evidence from archaeologists relating to population estimates would be invaluable in order to develop a better understanding of human dynamics in this area. Around 1900 years BP there is a large negative shift in the oxygen isotope record from Honey Camp. Significantly, the system does not return to its original condition after the change. The two short cores from this lagoon (LI and L4) are thought to have a break in sedimentation in their sequences during this time period. This sharp change in sedimentation rate may have been a response to the significant environmental changes that were affecting the system at this time. 297 Of particular note is the fact that the modification of the Ottawa Complex in Lamanai occurred during the period commonly referred to as the 'collapse' of the Mayan Civilisation. Clear evidence of disturbance in this period is proof that a society in Lamanai existed throughout this time. Although this was not a building phase, 20,000 tonnes of stone are estimated to have been moved as part of this process (Graham, 2001). This therefore represents significant manpower and organisation in the Mayan society. What is apparent is the changing priorities of the society during this time. This may have been a response to the difficulties that nearby sites were experiencing. The core Outpost 2000 was taken 1.5 km south of Lamanai 1999 and could be suggested that the human impact signature at this site may not be as pronounced. During the 'collapse', the carbon isotope record in Outpost 2000 shows a period of catchment recovery centred around c. 1100 years BP (AD 970). This highlights an important methodological point. This time period has been covered by two records which both represent different sequences of events. This demonstrates that it is important to obtain a good geographical spread of coring locations in order to ensure that the events can be correctly understood. The collapse of the Mayan Civilisation has been attributed to the late Holocene dry period (e.g. Curtis et al., 1996). This period is identified in Lamanai by a significant change in the diatom flora. Diatoms are influenced by a number of different factors and therefore it is likely that this change could be related to catchment disturbance resulting from building modification of the Ottawa Complex at Lamanai during this time. The 5lsO record in Lamanai is remarkably stable from 1240-870 years BP. This is unusual in the context of the record and therefore is significant. In cores LI and L4 from Honey Camp Lagoon there is a distinct positive excursion in the §l80 records that occurred at some point before 764 l4C years BP (AD 1265). It can therefore be postulated that this change is related to the late Holocene dry period. There is also clear evidence from Honey Camp for a drying trend from 1200 years BP onwards. The records from Hillbank do not cover this time period. 298 From the results gained in this study it is difficult to judge the severity of late Holocene dry period. The shift to dry conditions is very pronounced in the Honey Camp 1999 record, but it is not a discrete episode and dry conditions are maintained to the top of the sequence. The shift in LI and L4 is, however, a distinct event. This difference is likely to be a result of the higher resolution of the latter records. Both the Outpost 2000 and the Hillbank 2000 records show evidence for colonial activity in their catchments. The archaeological site of Lamanai was abandoned in AD 1640 and from this point the Outpost sequence shows signs of preliminary catchment recovery. From AD 1862-1917 a sugar mill was operating in this area and this period is clearly one of disturbance in the catchment. In a similar manner Hillbank 2000 records disturbance between AD 1897-1917. This is most likely to be due to the colonial logging that was prevalent in this area. The New River Lagoon was a principal site for the logwood industry in Belize. Logwood exports reached a peak in 1895 at 35,500 tons which equates to the beginning of disturbance in the sediment sequences. By 1920 exports had dropped to 1,500 tons primarily due to the development of synthetic dyes (Duncan, 1966). This demonstrates that the colonial settlers also had an impact on the environment. 9.3 The records from Belize Although limited (see Chapter 2), there is information available concerning the environmental history of Belize. In Cobweb Swamp there is evidence for sea level rise between 5600-4800 l4C years BP. Laguna de Cocos, Albion Island also records mesohaline conditions during this period. This is not however a period of enhanced reconstructed conductivity in the Hillbank 1998 diatom record. Throughout the whole record the diatoms in Hillbank record oligohaline conditions. Further research is therefore needed to determine the nature and extent to which sea level rise affected Belize away from the immediate coastal zone. 299 Zea mays is found in Cobweb Swamp and Albion Island between 3000-3500 l4C years BP and is an indicator of early agriculture. Zea mays was not picked up in the Hillbank pollen record. This does not necessarily suggest that it was absent in this area of Belize, owing to the low resolution of the Hillbank sequence. The evidence for the possible impact of humans in Hillbank occurs at c. 5000 years BP with a peak in Chenopodiaceae pollen - from herbs that grow in response to agricultural disturbance. This matches with the pollen evidence from Laguna de Cocos, Albion Island where human activity is indicated by Zea mays (Hansen, 1990). There are two studies which have drawn conclusions on the palaeoclimate of Belize. Bradbury et al. (1990) found evidence for a drier climate 5-6000 years BP. In order to substantiate this, more evidence on the patterns of change in the Rio Hondo (the river in which Albion Island lies) is required. The climate is then believed to have become increasingly moist in the period beginning around 5000 years BP. At this time Laguna de Cocos moved from being sodium chloride dominated to a freshwater system. There is no evidence from the records gathered in the present study for a drier climate from 6000 years BP. Between c. 5500-5400 years BP there is a shift to moister conditions in the Hillbank record. However, this is not a sustained excursion. The New River Lagoon is part of the same system as to the New River, which eventually links to the sea. Due to the proximity of the New River to the Rio Hondo (Figure 1.1) it would be expected that both would be affected by the same external events and respond in a similar manner. Hillbank is located a considerable distance away from the New River which means that the magnitude and extent of any change to this system will be less than in areas such as Albion Island. This perhaps accounts for the drying signal noted at Albion Island c. 6000 years BP and yet not at Hillbank. A further increase in moisture is seen in the Albion Island record from 1700-1600 years BP. It is hypothesised that the negative excursion in the Honey Camp Lagoon 1999 record at c. 1900 years BP could be the manifestation of this event. From AD 1000 Bradbury et al. (1990) note a drying trend on Albion Island. This is likely to be the late Holocene dry period signal detected at Honey Camp and Lamanai. 300 Despite northern Belize being a small area, it is apparent that there is variation in the environmental histories that have been produced, reflecting the influence which the local environment has on how a signal of change is recorded. This observation highlights the importance of not only a multiple coring strategy, but also a region- wide strategy so that a better grasp of the area's dynamics can be gathered. This needs to be coupled with an understanding of the present day interconnections between areas, so that differences between sites through time can be explained. The importance of high-resolution records is also key so that the magnitude and severity of events can be more faithfully assessed. 9.4 The wider picture In order to place the records from Belize within a meaningful context, the results gained in this study need to be compared with those already produced for the circum- Caribbean. These records are described in Chapter 2. A summary of the results found in Belize is shown in Figure 9.1. Figure 9.2 highlights the links between Belize, Guatemala and Haiti. The record from Hillbank shows evidence for climatic drying in the period immediately prior to the Holocene. It is postulated that this is a Younger Dryas signal. The climatic signal for this period in the Tropics is geographically specific. At this time Lake Miragoane, Haiti has a clear drying signal (Hodell et al., 1991). This is different from nearby Guatemala and Costa Rica which show a cool, moist signal (Deevey et al., 1983; Leyden, 1984; Hooghiemstra et al., 1992; Leyden et al., 1993; 1994; Brenner, 1994; Islebe et al., 1995). These results suggest that during this period the climate of Belize was most similar to the climate of the Caribbean. This is probably because of the particular ocean-atmosphere interactions driving climatic change at this time. These causal mechanisms account for the signal seen in Guatemala and Costa Rica - increased seasonality, decreased SSTs and rising sea level - resulting in cool, wet conditions. How can the dry signal in Belize be 301 accounted for? Due to their geographical proximity all the records are likely to have been equally affected by the shifts in the ITCZ movement. The records from Belize and Haiti are located much closer to the Caribbean Sea than Lake Quexil, Guatemala and La Chonta Bog, Costa Rica and are therefore more likely to be affected by changes to the Caribbean Sea. One hypothesis is that lower SSTs will decrease the amount of evaporation over the sea, therefore leading to a drier climate in coastal regions such as Belize and Haiti. The transition to full moist conditions was completed at Hillbank by c. 8900 years BP which coincides with the rise in water level by 9000 l4C years BP in Lake Peten- Itza (Curtis et al., 1998). In general terms, there is a period of lake filling in Guatemala and the Yucatan Peninsula ranging from before 7230 l4C years BP (San Jose Chulchaca, Yucatan Peninsula, Leyden et al., 1996) to 9000 14C years BP (Lake Peten-Itza, Curtis et al., 1998). This change has been attributed to increased moisture availability and sea level rise which enabled freshwater aquifers to rise (Fairbanks, 1989; Watts and Hansen, 1994). The oldest date in the Hillbank 1998 record is 9840 l4C years BP and the system is lacustrine for the whole of the record (inferred from the presence of diatoms). This implies that the New River Lagoon has been a permanent water body for much longer than many of the lakes in the area. It therefore has great potential to produce an even longer record than analysed in this thesis. The longevity of the water body is likely to be a function of the low-lying nature of the lagoon which is presently surrounded by marshland and the faulted nature of the substrate geology. The middle Holocene in Peten-Itza (6800-4800 l4C years BP) and Lake Miragoane (7000-5300 years BP) is characterised by wetter conditions matching those in the Hillbank record. The wettest period of the Holocene in San Jose Chulchaca centred on 5085 l4C years BP (Leyden et al., 1996; Whitmore et al., 1996). This is likely to be coincident with the increase in moisture seen at Hillbank between 5505-5381 years BP. After 5000 l4C years BP there is a clear cultural signal in Guatemala, Honduras, Panama, Costa Rica and indeed Belize. The changes as a result of human 302 impact in the Yucatan Peninsula, Mexico are only found in much more recent times. In Peten-Itza this becomes more pronounced by 2800 i4C years BP. However, this is not seen in the Hillbank pollen record. Records show a clear signal for the improvement of the forest ecosystem at around 1000 years BP in the Peten which is attributed to the collapse of the Mayan Civilisation. Although the late Holocene dry period is thought to be wide in its extent (Hodell et al., 1991; Horn and Sanford, 1992; Metcalfe et al., 1994; Metcalfe, 1995; Hodell et al., 1995; Curtis et al., 1996), there is only one record in Guatemala from Lake Salpeten (Rosenmeier et al., in press) that has preliminary evidence for climatic drying at this time. This period is not detected at all in Lake Peten-Itza (Curtis et al., 1998). Curtis et al. believe that this is due to the large size of the lagoon which would render the sequence much less sensitive to fairly short-lived shifts in climate. This would also account for the limited signal in the New River Lagoon and the stronger signal in Honey Camp Lagoon, which is a closed basin. The Hillbank record does have evidence for drier conditions persisting from 3000 14C years BP which is similar to Lake Miragoane, Haiti where dry conditions began 3200-2400 14C years BP (Hodell et al., 1991). Clearly, the results of this study highlight the importance of geographical location when considering the signal of climatic change that is recorded. The times of dry climatic conditions in Belize occur at the same time as those in the Caribbean. However, when conditions are wet in Belize they coincide with those in Guatemala. This must be a function of the combination between the dominant forcing factors and how areas respond. It is very difficult to judge the relative severity of the Lateglacial period and the late Holocene dry event. Due to lack of chronological control in the Honey Camp record it cannot be said with certainty that both events are found in the same sediment record. The Lateglacial period in Hillbank is represented by an extremely large isotopic shift which is much greater than any seen later in the sequence. The late 303 Holocene period in Honey Camp is also significantly drier than the rest of the record, but the episode is not discrete because these conditions are maintained until the present day. The two records which show this as a discrete event in Honey Camp have a much smaller isotopic shift than is seen in the Hillbank Lateglacial signal. The preliminary conclusion is therefore that for Belize the Lateglacial was drier than any episode recorded in the Holocene. 9.5 Methodology: In this investigation three types of corer were used. The aim of any coring exercise is to collect an undisturbed and continuous sediment sequence. It was hoped that the records collected in this investigation would provide material from the present day back. It is apparent from the lack of overlap between cores that it is vital that the type and behaviour of the sediment is investigated first to ensure that the most suitable methods are employed. Devices such as glew corers could be employed as these collect very short cores which would increase the possibility of the most recent sediments being captured. In order to ensure an overlap between the Kullenberg and the Livingstone corers, two metre Kullenberg corers could have been deployed. One aspect which needs attention is the amount of weights that are placed on the system. If these are insufficient then the sediment sequences will be mixed as the corer is unable to penetrate the sediment successfully (Davies, pers. com. 2000). A Percussion corer was also employed in this investigation. This was successful in collecting material from solid ground sites. If this device were to be used in a lagoon, an extremely stable platform would have to be constructed with very good anchors so that the system remained in place throughout the operation. This may help recover a longer core from Honey Camp Lagoon as the extra power provided by the engine may be able to penetrate the hard layer that thwarted hand coring attempts in this investigation. This investigation began as a diatom based study. After the first field season (1999) it became apparent that, as diatom preservation was not a dominant process, this 304 proxy could not be relied upon to provide a continuous record of change. The methodology employed was therefore widened to include the stable isotopes of oxygen and carbon. From the analysis of the results it is clear that the isotopes have provided the most sensitive results of environmental change. The reasons behind this are twofold. Firstly, because of the location of the records, the diatom flora is dominated by littoral species which appear to have been mainly influenced by habitat rather than chemical changes to the lagoon. Secondly, due to the large size of the New River Lagoon the amount of change needed to result in a complete change of habitat would have to be enormous and thus any smaller-scale changes have not been detected by the diatoms. Stable isotopes respond to wider scale changes of climate and catchment which are more likely to show change. The analysis of the isotope data was, however, hindered by the lack of modern samples which would have enabled a better understanding of the modern environment and created comparative reference conditions. The latter is therefore a priority for future work. Of key importance to a paleolimnological study is the creation of a chronology which enables the results to be placed into a meaningful context. Dating was problematic in this study principally due to the lack of terrestrial organic matter available in the sediment. In a carbonate area, such as Belize, it is essential that the dating material is terrestrial so that it is not affected by the hardwater error. The organic matter which was dated in the records was very small and thus its integrity in the core was difficult to ascertain. This is especially true of the two dates from Honey Camp Lagoon 1999 which are modern. Funding was awarded to date freshwater gastropods where they occurred at the same depth as terrestrial organic matter i.e. paired dates. Three paired dates have been produced as part of this study. These had hardwater errors of 1105 i4C years BP (Honey Camp Lagoon), 1660 14C years BP (Lamanai) and 1527 l4C years BP (Hillbank). These are very similar and suggest that these values can be applied as a correction factor to similar lake systems. In order to improve the chronology of the records produced in this study, gastropods should therefore be dated from a number of different horizons and the relevant correction factor applied. 305 9 10 Pb dating was also carried out in this investigation. This only provided a meaningful record for Hillbank 2000. The two other records analysed did not contain a signal. This is because evidence suggests that Outpost 2000 and Honey Camp 2000 do not contain sediments from the present day back and therefore the signal was lost. 210 2 1The unsupported Pb flux from Belize is very low (32 Bq m" yr" ) compared to the global average (185 Bq m 2 yr"1) (Appleby and Oldfield, 1983). The value for Belize compares well with values gained by Davies (2000) for Mexico (Lake Zirahuen: 48 Bq m"2 yr"1, Lake Juanacatlan: 12 Bq m 2 yr"1). The data from Belize therefore adds weight to Davies' suggestion that the global flux of 2l0Pb may be more variable than previously thought. 2l0Pb did not however prove to be a successful dating tool in the New River Lagoon because of the highly mixed nature of the recent Hillbank sediments. This is however useful environmental information which may have been difficult to establish if this technique had not been employed. It is apparent that the environmental signals detected in records are dependent on the proxies used. It therefore could be argued that until a complete range of tools are employed, any environmental reconstructions cannot be regarded as complete. As well as pollen, the amount and types of gastropods present throughout a sequence could be investigated. This would provide ecological information about the lake environment serving as a valuable addition to the diatom data. 9.6 Wider research issues The research which is presented in this thesis does not fall solely into one academic discipline; rather, it draws on botany, archaeology and geography. When data are collected it is often done so with specific goals in mind which can be restricting. In terms of the diatom study it became apparent that much more work is needed on species ecology and the role of geography in influencing speciation. The more work that is undertaken in diverse areas the more knowledge that will be gained. Although this thesis may have been hindered by lack of information on this topic, it has made a contribution which will aid further investigations of diatoms in this part of the 306 Tropics. Most archaeological work in Belize has concentrated on the classification of large temples. While this is very informative in terms of the relative importance of sites, it provides relatively little information about population dynamics and activities. The production of archaeological histories of sites that emphasise human activity rather than the abstract descriptions of archaeological digs would enable a closer link to be developed between lake sediment and human records. The best way in which these goals can be achieved is through collaborative projects. These ensure that maximum information is gleaned from the sites in question. The first priority for further work in Belize would be to increase the number of dates on each core. This is especially important for Honey Camp Lagoon as it would enable this lagoon to be better integrated into the regional record. More sediment cores should be analysed in order to develop a greater understanding of the role of environment and to develop an improved understanding of the spatial dynamics within systems. To develop the remit further sea level could be investigated to constrain the impact of this variable. An ideal site for this would be Cobweb Swamp. The record from Hillbank is very enlightening because it extends beyond the Holocene. The area which is thought to have the most potential to have a long sediment record is the Booth River Wetlands. There are thought to be permanent water bodies located next to the escarpment which would be ideal for coring producing a long record of environmental change (Furley pers.com. 1999). Access to this area is however very difficult. Both the New River Lagoon and Honey Camp Lagoon have documented archaeological sites on their shores. Part of the aim of this thesis was to investigate the human history of the area. One way in which this could be enhanced is through the study of sediments from the actual occupation sites. This would include, for example, pits from raised fields. This strategy was undertaken by Bradbury et al. (1990). They found, however, that by comparison with the lake record, the field and canal records contained less information. These samples did however provide more insight into the human dynamics of the site. 307 9.6 The contribution of this thesis The original focus of this thesis was to address four research questions. The first: to determine the pattern of environmental change in north Belize. This thesis has provided an insight into distinct phases of environmental change, which can be attributed to both climatic and human forces. The second question was to determine whether the records from Belize fit into the wider scale regional patterns. The records from Belize have highlighted most clearly that there are specific difference between various areas of the circum-Caribbean. A greater geographical spread of data is required to develop a better understanding of the regional dynamics of this area. The third aim was to find evidence for the late Holocene dry period. This has been found but its severity is difficult to judge. Evidence from Lamanai suggests that this may have occurred during a period of altered human attitudes to site development, but a causal link between the two cannot be established. This latter point formed part of the fourth aim. The record from Lamanai has not only corroborated known archaeological information it has also added to our appreciation of human impact. This research has also provided evidence of colonial period activity. The principal conclusions drawn from this thesis are summarised below: The present day environment of Belize: 1. The lakes sampled in Belize cover a chemical gradient running from calcium bicarbonate, to calcium sulphate to sodium chloride dominated systems. This provides a great deal of scope for comparative studies in the future. 2. From the data available it is apparent that habitat has the most influence over diatom species distribution. This may change, however, with the collection of more water chemistry data. In terms of fossil diatom records, the use of preservation indices increases the amount of data that can be gained from a sequence. 308 3. Diatom species taxonomy in Belize is a topic in which a great deal of knowledge can be gained concerning the role of environment in determining species characteristics. This was developed in this thesis through the study ofMastogloia smithii var. lacustris. The environmental history of Belize: 1. The Lateglacial was a very variable climatic period in Belize. 2. The transition to the Holocene encompassed the most arid conditions seen in the records produced for Belize. This may be a Younger Dryas signal and equates to the signal found in Lake Miragoane, Haiti. 3. The Holocene was a period of stable, moist conditions in Belize in a similar manner to the rest of the circum-Caribbean. 4. Human disturbance is apparent from c. 4900 years BP in Belize. 5. The transition to the late Holocene dry period was gradual. Pollen evidence indicates that this period began in c. 4800 years BP, whilst isotope evidence places it c. 3000 years BP. 6. The record from Lamanai shows clear evidence of human impact on the environment through the building and modifying of temples. 7. The records from Belize show evidence for the late Holocene dry period from 1200 years BP onwards. This palaeolimnological study presents a picture of dynamic change and spatial variation in northern Belize. The climate of Belize and the surrounding countries of the circum-Caribbean appear to depend on a number of different factors. These factors are related to the strong oceanic influence controlling Belize's climate and the high level of local environmental variation within the country. The findings presented in this thesis have important implications regarding the extent, magnitude, and timing of the climatic changes in the circum-Caribbean during the last -10,000 years. 309 Years BP Hillbank Lamanai Honey Camp 0 Colonial activity Colonial activity 500 3rd building phase Dry 1000 Dry 1500 Very moist ? 2000 2nd building phase 2500 1st building phase (dry) 3000 3500 Moist and stable 4000 4500 Transition to dry conditions Moist and stable 5000 Evidence for humans 5500 Very moist 6000 6500 7000 7500 Moist 8000 8500 9000 9500 Possible dry signal 10000 10500 Dry 11000 >11000 Very variable Figure 9.1 A summary of Holocene environmental change in Belize, inferred from this research Key moist climate dry climate no data available Figure9.2 Reference List: Adams, R.E.W. 1973 The collapse ofMaya civilisation: A review of previous theories. IN Culbert, T.P. (ed) The Classic Maya Collapse. University of New Mexico Press: 21-33. Adams, R.E.W. 1991 Prehistoric Mesoamerica University of Oklahoma Press. Adams, R.E.W. 1994 The Programme for Belize Archaeological project: 1993 Field Season. The University of Texas, San Antonio. Adams, R.E.W., Brown, W.E. and Culbert, T.P. 1981 Radar mapping, archaeology and ancient Maya land use. Science 213: 1457-1464. Adams, R.E.W. and Valdez, F. 1993 The Programme for Belize (PFB) Archaeological Project: Report offield activities, 1992. The University of Texas at San Antonio, San Antonio. Adams, R.E.W. and Valdez, F. 1995 The Programme for Belize Archaeological project: 1994 interim report. The Center for Archaeology and Tropical Studies and The University of Texas at San Antonio. Alcala-Herrera, J.A., Jacob, J.S., Machian Castillo, M.L., and Neck, R.W. 1994 Holocene palaeosalinity in a Maya wetland, Belize, inferred from the microfaunal assemblage. Quaternary Research 41:121-130. Alcala-Herrera, J.A., Machain-Castillo, M.L., Comet, P.A., Jacob, J. and Gio- Argaez, F.R 1995 Stratigraphic and environmental correlation between two sedimentary sequences in northern Belize submitted to the Journal of the Geological Society. Allaby, M. 1994 Concise dictionary ofecology. Oxford University Press. Altschuler, M. 1958 On the environmental limitations ofMaya cultural development. Southwestern Journal ofAnthropology 14: 189-196. Anonymous 1975 Proposals for standardisation of diatom terminology and diagnoses. Nova Hedwigia 19: 323-354. Appleby, P.G. and Oldfield, F. 1983 The assessment of 2l0Pb data from sites with varying sediment accumulation rates Hydrobiologia 103: 29-35. Barker, P. 1992 Differential diatom dissolution in Late Quaternary sediments from Lake Manyara, Tanzania: an experimental approach Journal ofPalaeolimnology 7: 235-251. QfO Barker, P., Gasse, F., Roberts, N. and Taieb, M. 1990 Taphonomy and diagenesis in diatom assemblages; a Late Pleistocene palaeoecological study from Lake Magadi, Kenya. Hydrobiologia, 214: 267-272. Barnhart, E. and Hargrove, C. 1995 Findings of the PFB archaeological project 1994 summer survey IN Adams, R.E.W. and Valdez, F. 1995 The Programmefor Belize Archaeological project: 1994 interim report. The Center for Archaeology and Tropical Studies and The University of Texas at San Antonio. Bartlett, A.S and Barghoorn, E.S. 1973 Phytogeographic history of the Isthmus of Panama during the past 120000 years: a history of vegetation, climate and sea level change. IN Graham, A Vegetation and vegetational history ofnorthern Latin America, Elsevier. Battarbee, R.W. 1986 Diatom analysis. IN Berglund, B.E. (Ed) Handbook of Holocene palaeoecology and palaeohydrology John Wiley and Sons Ltd. Battarbee, R.W. 2000 Palaeolimnological approaches to climate change, with special regard to the biological record. Quaternary Science Reviews 19: 107-124. Beck, J.W., Edwards, R.L., Ito, E., Taylor, F.W., Recy, J., Rougerie, F., Joannot, P. and Henin, C. 1992 Sea surface temperature from coral skeletal Strontium/ Calcium ratios. Science 257: 644-647. Behling, H. 2000 A 2860 year high-resolution pollen and charcoal record from the Cordillera de Talamanca in Panama: a history of human and volcanic forest disturbance. The Holocene 10: 387-393. Bennion, H. 1994 A diatom phosphorus transfer function for shallow, eutrophic ponds in SE England. Hydrobiologia 275/276: 391-410. Berger, A. 1978 Numerical values of caloric insolation from 1,000,000 YBP to 100,000 YAP (astronomical solution of Berger, 1978). Contributions of the Institute ofAstronomy and Geophysics, University Catholique, Loivain-La-Neuve. Belgium 37. Binford, M.W. 1982 Ecological history of Lake Valencia, Venezuela: Interpretation of animal microfossils and some chemical, physical and geological features. Ecological Monographs 52(3): 307-333. Binford, M.W. 1983 Paleolimnology of the Peten Lake District, Guatemala 1: Erosion and deposition of inorganic sediment as inferred from granularity. Hydrobiologia 103:199-203. Binford, M.W., Kolata, A.L., Brenner, M., Janusek, J.W., Seddon, M.T., Abbott, M. and Curtis, J.H. 1997 Climate variation and the rise and fall of an Andean civilization. Quaternary Research 47: 235-248. 313 Birks, H.J.B. 1998 Numerical tools in palaeolimnology - progress, potentialities and problems. Journal ofPaleolimnology 20: 307-332. Birks, H.J.B., Line, J.M., Juggins, S., Stevenson, A.C. and Ter Braak, C.J.F. 1990 Diatoms and pH reconstruction. Philosophical Transactions of the Royal Society of London B 327: 263-278. Blackmore, S. unpublished. The vegetational history ofBelize. Internal report, Natural History Museum, London. Blanton R.E., Kowalewski, S.A., Feinman, G.M. and Finsten, L.M. 1993 Ancient Mesoamerica: a comparison of change in three regions. Cambridge University Press. Bonnefille, R. and Mohammed, U. 1994 Pollen-inferred climatic fluctuations in Ethiopia during the last 3000 years. Palaeogeography, Palaeoclimatology and Plalaeoecology 109: 331-343. Bonnett, P.J.P. and Cambrey, R.S. 1991 The record of deposition of radionuclides in the sediments of Ponsonby Tarn, Cumbria. Hydrobiologia 214: 63-70. Bradbury, J.P. 1979 Quaternary diatom stratigraphy of Lake Valencia, Venezuela. Proceedings of the 4th Latin American Geological Congress, Trinidad and Tobago. Bradbury, J.P. 1999 Continental diatoms as indicators of long-term environmental change. IN Stoermer, E.F. and Smol, J.P. (eds) The diatoms: applicationsfor the environmental and earth sciences. Cambridge University Press: 169-182. Bradbury, J.P., Leyden, B., Salgado-Labouriau, Lewis, W.M., Schubert, C., Binford, M.W., Frey, D.G., Whitehead, D.R., and Weibezahn, F.H. 1981 Late Quaternary environmental history of Lake Valencia, Venezuela. Science 214: 1299-1305. Bradbury, J.P., Forester, R.M., Bryant, W.A. and Covich, A.P. 1990 Paleolimnology of Laguna de Cocos, Albion Island, Rio Hondo, Belize IN Pohl, (ed.) M.D. Ancient Maya wetland agriculture: Excavations on Albion Island, northern Belize. Westview Special studies in Archaeology. Breen, A.M. 1998 A reconstruction ofenvironmental change in North Belize. Unpublished MRes dissertation, The University of Edinburgh. Brenner, M. 1978 Paleolimnological Assessment ofHuman Disturbance in the Drainage Basins of Three Northern Guatemalan Lakes. MS Thesis, University of Florida, Gainsville. Brenner, M. 1983 Paleolimnology of the Maya region. Unpublished PhD thesis, University of Florida, Gainsville. Brenner, M. 1994 Lakes Salpeten and Quexil, Peten, Guatemala, Central America. IN Gierlowski-Kordesch, E. and Kelts, K. (eds.) Global geological record of lake basins, Volume 1. Cambridge University Press. Brenner, M. and Binford, M.W. 1988 A sedimentary record of human disturbance from Lake Miragoane, Haiti. Journal ofPaleolimnology 1: 85-97. Brenner, M., Curtis, J.H., Higuera-Gundy, A., Hodell, D.A., Jones, G.A., Binford., M.W. and Dorsey, K.T. Lake Miragoane, Haiti (Caribbean) IN Gierlowski- Kordesch, E. and Kelts, K. 1994 Global geological record of lake basins volume 1 Cambridge University Press. Brenner, M., Hodell, D.A., Curtis, J.H., Roseinmeier, M.F., Binford, M.W. and Abbott, M.B. 2001 Abrupt climate change and Pre-Columbian cultural collapse. IN Markgraf, V (ed). Interhemispheric climate linkages. Academic Press. Bridgwater, N.D., Holmes, J.A. and O'Hara, S.L. 1999 Complex controls on the trace element chemistry of non marine ostracods: an example from Lake Patzcuaro, Central Mexico. Palaeogeography, Palaeoclimatology, Palaeoecology 148: 117-131. Broecker, W.S., Peteet, D.M. and Rind, D. 1985 Does the ocean-atmosphere system have more than one stable mode of operation? Nature 315: 21-26. Burney, D.A., Burney, L.P. and MacPhee, R.D.E. 1994 Holocene charcoal stratigraphy from Laguna Tortuguero, Puerto Rico and the timing of human arrival on the island. Journal ofArchaeological Science 21: 273-281. Bush, M.B. and Colinvaux, P.A. 1990 A pollen record of a complete glacial cycle from lowland Panama. Journal of Vegetation Science 1: 105-118. Bush, M.B. and Colinvaux, P.A. 1994 Tropical forest disturbance: paleoecological records from Darien, Panama. Ecology 75: 1761-1768. Bush, M.B., Piperno, D.R. and Colinvaux, P.A. 1989 A 6000-year history of Amazonian maize cultivation. Nature 340: 303-305. Bush, M.B., Piperno, D.R, Colinvaux, P.A., De Oliveira, P.E., Krissek, L.A, Miller, M.C. and Rowe, W.E., 1992 A 14300 year palaeoecological profile of a lowland tropical lake in Panama. Ecological Monographs 62: 251-275. Cane, M.A. 1998 A role for the tropical Pacific. Science 282: 59-61. Chase, D.Z. and Chase, A.F. 1982 Yucatec influence in Terminal Classic Northern Belize. American Antiquity 47: 596-614. Chepstow-Lusty, A.J., Bennett, K,D., Switsur, V.R. and Kendall, A. 1996 4000 years of human impact and vegetation change in the central Peruvian Andes - with events paralleling the Maya record? Antiquity 70 : 824-33. Chivas, A.R., Deckker, P., Cali, J.A., Chapman, A., Kiss, E. and Shelly, J.M.G. 1993 Coupled stable isotope and trace-element measurements of lacustrine carbonates as paleoclimatic indicators. IN Swart, P.K., Lohmann, K.C., McKenzie, J. and Savin, S. (eds). Climate change in continental isotopic records. American Geophysical Union Geophysical Monograph78: 113-121. Clark, P.U., Marshall, S., Clarke, G., Hostetler, S., Licciardi, J. and Teller, J. 2001 Freshwater forcing of abrupt climate change during the Last Glaciation. Science 293: 283-287. Coe, M.D. 1966 The Maya. Frederick A. Praeger: New York. Coe, M.D. 1997 The Maya. Thames and Hudson: 5th edition. Coe, W.R. 1957 Environmental limitations on Maya culture: A re-examination. American Anthropologist 59: 325-335. Cook, R.G. and Ranere, A.J. 1992 Prehistoric human adaptations to the seasonally dry forests of Panama. World Archaeology 24: 114-33. Cooper, S.R., Huvane, J., Vaithiyanathan, P., Richardson, C.J. 1999 Calibration of diatoms along a nutrient gradient in Florida Everglades water conservation area-2A, USA. Journal ofPalaeolimnology 22: 413-437. Covich, A.P. 1976 Recent changes in molluscan diversity of a large tropical lake (Lago de Peten, Guatemala). Limnology and Oceanography 21: 51-59. Covich, A.P. 1983 Mollusca: A contrast in species diversity from aquatic and terrestrial habitats. IN Turner, B.L. and Harrison, P.D. (eds) Pulltrouser swamp. Ancient Maya habitat, Agriculture and settlement in northern Belize. University of Texas Press, Austin: 120-139. Covich, A.P. and Stuiver, M 1974 Changes in oxygen-18 as a measure of long term fluctuations in tropical lake levels and molluscan populations. Limnology and Oceanography 19: 682-691. Cowgill, U.M. 1961 Soil fertility and the ancient Maya. Transactions of the Connecticut Academy ofArts and Sciences 42: 1-56. Cowgill, G.L. 1964 The end of the Classic Maya culture: A review of recent evidence. Southwestern Journal ofAnthropology 20: 145-159. Cowgill, U.M. and Hutchinson, G.E. 1963 Ecological and geochemical archaeology in the Southern Maya Lowlands. Southwestern Journal ofAnthropology. 19: 267- 286. Craig, H. 1957 Isotopic standards for carbon and oxygen and correction factors for mass-spectrometric analysis of carbon dioxide. Geochimica et Cosmochimica Acta. 12: 133-149. Craig, H. 1961 Standard for reporting concentrations of deuterium and oxygen-18 in natural waters. Science. 133: 1833-1834. Culbert, T.P. (Ed). 1973 The Classic Maya Collapse. University of New Mexico Press. Culbert, T.P. 1991 Classic Maya political history: hieroglyphic and archaeological evidence. School of American Research: advanced seminar series. Curtis, J.H. and Hodell, D.A. 1993 An isotopic and trace element study of ostracods from lake Miragoane, Haiti: A 10,500 year record of palaeosalinity and palaeotemperature change in the Caribbean IN Swart, P.K., Lohman, K.C., McKenzie, J. and Savin, S (Eds). Climate change in continental isotopic records Geophysical Monographs 78, American Geophysical Union. Curtis, J.H., Hodell, D.A. and Brenner, M. 1996 Climate variability on the Yucatan Peninsula (Mexico) during the past 3500 years, and implications for Maya cultural evolution. Quaternary Research 46: 37-47. Curtis, J.H., Brenner, M., Hodell, D.A., Balser, R.A., Islebe, G.A. and Hooghiemstra, H 1998 A multi-proxy study of Holocene environmental change in the Maya lowlands of Peten, Guatemala. Journal ofPaleolimnology 19: 139-159. Curtis, J.H., Brenner, M. and Hodell, D.A. 1999 Climate change in the lake Valencia Basin, Venezuela, 12600 yr BP to present. The Holocene 9(5): 609-619. Dahlin, B.H. 1977 The initiation of the Albion Island settlement pattern survey. Journal ofBelizian Affairs 5: 44-51. Dahlin, B.H. 1983 Climate and prehistory on the Yucatan Peninsula. Climatic Change 5: 245-261. Dansgaard, W. 1964 Stable isotopes in precipitation. Tellus XVI: 436-468. Davies, S.J. 2000 Environmental change in the west-central Mexican Highlands over the last 1000 years: evidence from lake sediments. Unpublished PhD Thesis, The University of Edinburgh. Davies, S.J., Metcalfe, S.E., Caballero, M.E. and Juggins, S. in press Developing diatom-based transfer functions for Central Mexican lakes. Hydrohiologia. Deevey, E.S. 1978 Holocene forests and Maya disturbance near Quexil Lake, Peten, Guatemala. Polish Archives ofHydrohiology 25: 117-129. Deevey, E.S. and Rice, D.S. 1980 Coluviacion y retencion de nutrientes en el distrito lacustre del Peten central, Guatemala. Biotica 5 129-144. Deevey, E.S. and Stuiver, M. 1964 Distribution of natural isotopes of carbon in Linsley Pond and other New England Lakes. Limnology and Oceanography 9: 1-11. Deevey, E.S., Rice, D.S., Rice, P.M., Vaughan, H.H., Brenner, M., and Flannery, M.S. 1979 Mayan urbanism: Impact on a tropical karst environment. Science 206: 298-306. Deevey, E.S., Brenner, M., and Binford, M.W. 1983 Paleolimnology of the Peten Lake District, Guatemala 3: Late Pleistocene and Gamblian environments of the Maya area. Hydrobiologia 103: 211-216. DeMenocal, P.B. 2001 Cultural responses to climate change during the late Holocene. Science 292: 667-673. Dixit, S.S., Smol, J.P., Kingston, J.C., and Charles, D.F. 1992 Diatoms: powerful indicators of environmental change. Environmental Science Technology 26:23-33. Duncan, K.D. 1966 Aspects of the Hardwood forest industry of British Honduras IN Furley, P.A. (ed). Expedition to British Honduras - Yucatan 1966: General Report. The University of Edinburgh. Dunn, F.L. 1968 Epidemiological Factors: Health and Disease in Hunter-Gatherers. IN Lee, R.B. and DeVore, I. (ed). Man the Hunter. Chicago, Aldine: 221-228. Dunning, N., Rue, D.J., Beach, T., Covich, A and Traverse, A. 1998 Human- environment interactions in a tropical watershed: The palaeoecology of Laguna Tamarindito, El Peten, Guatemala. Journal ofField Archaeology 25: 139-151. Durazzi, J.T. 1977 Stable isotopes in the ostracod shell: a preliminary study. Geochimica et Cosmochimica Acta. 41:1168-1170. Eades, L.J., Farmer, J.G., MacKenzie, A.B., Kirika, A. and Bailey-Watts, A.E. 1998 High-resolution profile of radiocaesium deposition in Loch Lomond sediments. Journal ofEnvironmental Radioactivity. 39: 107-116. Enfield, D.B. and Elfaro, E.J. 1999 The dependence of Caribbean rainfall on the interaction of the tropical Atlantic and Pacific Oceans. Journal of Climate 12: 2093- 2103. Epstein, S., Buchsbaum, R., Loweenstam, H.A. and Urey, H.C. 1953 Revised carbonate-water isotopic temperature scale. Bulletin of the Geological Society of America 64: 1315-1316. « Espindola, J.M., Macias, J.L., Tilling, R.I. and Sheridan, M.F. 2000 Volcanic history of El Chichon volcano (Chiapas, Mexico) during the Holocene, and its impact on human activity. Bulletin ofVolcanology 62: 90-104. Esselman, P.C. and Botes, E. 2001 Status and future needs of limnological research in Belize. Limnology in developing countries 3: 35-68. Eugster, H.P. and Hardie, L.A. 1970 Saline Lakes IN Lerman, A. (ed). Lakes chemistry, geology, physics. Springer, Berlin: 237-294. Fairbanks, R.G. 1989 A 17,000 yr glacio-eustatic level record: influence of glacial melting rates on the Younger Dryas event and deep ocean circulation. Nature 342: 637-642. Flower, R.J. 1993 Diatom preservation: experiments and observations on dissolution and breakage in modern and fossil material. Hydrobiologia 269/270: 473-484. Flower, R. and Likhoshway, Y. 1993 An investigation of diatom preservation in Lake Baikal. IN Grachev, M.A. (ed). Fifth workshop on diatom algae, Russian Botanical Society. Irkutsk. Foged, N. 1984 Freshwater and littoral diatomsfrom Cuba. Bibliotheca Diatomologica. J.Cramer. Folan, W.J. and Hyde, B.H. 1985 Climatic forecasting and recording amongst the ancient and historic Maya. An ethnohistoric approach to epistemological and paleoclimatological patterning. IN Folan, W.J. (ed). Contributions to the archaeology and ethnohistory ofgreater Mesoamerica. Southern Illinois Press. Folan, W.J., Gunn, J., Eaton, J.D. and Patch, R.W. 1983 Paleoclimatological patterning in Southern Mesoamerica. Journal ofField Archaeology 10: 453-468. Fontes, J.C. and Gonfiantini, R. 1967 Comportement isotopique ay cours de l'evaporation de deux basins sahariens. Earth and Planetary Science Letters. 3: 258- 266. Ford, A. and Fedick, S. 1992 Prehistoric Maya settlement patterns in the Upper Belize River area: Initial results of the Belize River archaeological settlement survey. Journal ofField Archaeology 19: 35-49. Freidel, D.A. 1986 Archaeology at Cerros, Belize, Central America: Volume 1. Southern Methodist University Press. 18 13* Fritz, P. and Poplawski, S. 1974 O and C in the shells of freshwater molluscs and their environments. Earth and Planetary Science Letters 24: 91-98. Q Fritz, S.C. 1990 Twentieth-century salinity and water level fluctuations in Devil's Lake, North Dakota: Test of a diatom-based transfer function. Limnology and Oceanography 35: 1771-1781. Fritz, S.C., Juggins, S., Battarbee, R.W., and Engstrom, D.R. 1991 Reconstruction of past changes in salinity and climate using a diatom-based transfer function. Nature 352: 706-708. Furley, P.A., Metcalfe, S.E., Dugmore, A.J., Tudhope, S. and Murray, M.R. unpublished Report to the Leverhulme Trust F/158/BQ. Gann, T. 1928 Maya cities: a record ofexploration and adventure in Middle America. Garber, J.F. 1989 Archaeology at Cerros, Belize, Central America. Volume 2: The Artifacts. Southern Methodist University Press. Gasse, F 1986 East African diatoms: Taxonomy, ecological distribution. J.Cramer. Gasse, F., Techet, R., Durand, A., Gilbert, E. and Fontes, J.C. 1990 The arid-humid transition in the Sahara during the last glaciation. Nature 346: 141-146. Gasse, F., Juggins, S. and Khelifa, L.B. 1995 Diatom-based transfer functions for inferring past hydrochemical characteristics of African lakes. Palaeogeography, Palaeoclimatology, Palaeoecology 117: 31-54. Gasse, F., Barker, P., Gell, P.A., Fritz, S.C. and Chalie, F. 1997 Diatom inferred salinity in palaeolakes: an indirect tracer of climate change. Quaternary Science Reviews 16: 547-563. Gat, J.R. 1980 Isotope hydrology of very saline lakes. IN Nissenbaum, A. (ed). Hypersaline brines and evaporitic environments. Amsterdam: Elsevier: 1-8. Gat, J.R. 1995 Stable isotopes of fresh and saline lakes. IN Ferman, A., Imboden, D.M. and Gat, J.R. Physics and Chemistry ofLakes. Springer, New York: 139-165. Gauch, H.G. Jr. 1982 Multivariate analysis in community ecology. Cambridge University Press. Germain, H 1981 Flore des diatomees eaux douces et saumatres du Massif Armoricain et des contrees voisines d'Europe occidentale. Boubee. Giannini, A., Kushnir, Y. and Cane, M.A. 2000 Interannual variability of Caribbean rainfall, ENSO and the Atlantic Ocean. Journal of Climate 13: 297-3 11. Giannini, A., Cane, M.A. and Kushnir, Y. 2001 Interdecadal changes in the ENSO teleconnection to the Caribbean region and the North Atlantic Oscillation. Journal of Climate 14: 2867-2879. Gifford, J.C., Sharer, R.J., Ball, W., Chase, A.F., Gifford, C.A., Kirkpatrick, M. and Myer, G.H. 1976 Prehistoric Pottery Analysis and the ceramics ofBarton Ramie in the Belize Valley. Memoirs, Peabody Museum ofArchaeology and Ethnology 18. Cambridge University Press. Gill, R.B. 1994 The Great Maya Droughts. Unpublished PhD thesis, Texas University. Global Network for Isotopes in Precipitation 2000. http://isohis.iaea.org/. Godwin, H. 1962 Half-life of radiocarbon. Nature 195: 984. 210 Goldberg, E.D. 1963 Geochronology with Pb: Radioactive dating. Conference proceedings 1962 IAEA, Vienna. Goodfriend, G.A. 1992 The use of land snail shells in paleoenvironmental reconstruction. Quaternary Science Reviews 11: 665-685. Graham, E. 2001 Collapse, conquest and Maya survival at Lamanai, Belize. Archaeology International. Institute of Archaeology, University College London. Graham, E. and Pendergast, D.M. 1989 Excavations at the Marco Gonzalez site, Ambergris Caye, Belize, 1986. Journal ofField Archaeology. 16: 1-16. Grimm, E.C. 1987 CONISS: A FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of instrumental sum of squares. Computers and Geosciences 13: 13-35. Grimm, E.C. 1992 TILIA and TILIAGRAPH, Illinois State Museum, Illinois. Grove, J. 1996 The century time-scale IN Driver, T.S. and Chapman, G.P. (eds). Time-scales and environmental change. Routledge: 39-87. Guderjan, T.H. 1991 New information from La Milpa, The 1990 field season. Mexicon XIII 5:9. Guderjan, T.H. and Garber, J.F. 1995 Maya maritime trade, settlement and populations on Ambergris Caye, Belize. Maya Research Program. Guilderson, T.P., Fairbanks, R.G. and Rubenstone, J.L. 1994 Tropical temperature variations since 20,000 years ago: Modulating interhemispheric climate change. Science 263: 663-665 Guillard, R.R.L. and Kilham, P. 1977 The ecology of marine planktonic diatoms. IN Werner, D. (ed). The biology ofdiatoms. Botanical Monographs 13, Blackwell Scientific Publications: 372-469. 321 Gunn, J. and Adams, R.E.W. 1981 Climatic change, culture, and civilisation in North America. World Archaeology 13(1) 87:100. Haberyan, K.A., Horn, S.P. and Cumming, B.F. 1997 Diatom assemblages from Costa Rican lakes: an initial ecological assessment. Journal ofPalaeolimnology 17: 263-274. Hakansson, H. 1989 A light and electron microscopical investigation of the type species of Cyclotella (Bacillariophyceae) and related forms using original material. Diatom research 4(2): 255-267. Hammond, N. 1981 Settlement patterns in Belize. IN Ashmore, W. (ed). Lowland Maya settlement patterns. University of New Mexico Press: 157-185. Hammond, N. 1982 Ancient Maya civilization. Cambridge University Press. Hammond, N. 1985 Nohmul: A prehistoric Maya community in Belize 1973-1983. BAR International Series. Hammond, N. 1991 Cuello: An early Maya community in Belize. Cambridge University Press. Hammond, N and Bobo, M.R. 1993 Pilgrimage's last mile: late Maya monument veneration at La Milpa, Belize. World Archaeology 26: 19-33. Hammond, N and Miksicek, C.H. 1981 Ecology and economy of a formative Maya site at Cuello, Belize. Journal ofField Archaeology 8: 259-269. Hammond, N., Donaghey, S., Gleason, C., Staneko, J.C., Van Tuerenhout, D. and Kosakowsky, C.J. 1987 Excavations at Nohmul, Belize 1985. Journal ofField Archaeology 14: 257-282. Hansen, J., Lacis, A., Rind, D., Russell, G., Stone, P., Fung, I., Ruedy, R., and Lerner, J. 1984 Climate sensitivity: analysis of feedback mechanisms. IN Hansen, J.E. and Takahashi, T. (eds). Climate processes and climatic sensitivity. Geophysical Monograph series 29: 130-163. Hansen, P.C.S. 1990 Pollen stratigraphy of Laguna de Cocos. IN Pohl, M (ed). Ancient Maya wetland agriculture: excavations on Albion Island, northern Belize. Westview Press. Hartshorn, G., Nicolait, L., Hartshorn, L., Bevier, G., and Brightman, R. 1984 Belize: Country environmental profile: Afield study. Robert Nicolait and Associates Ltd. Hartshorn, G.S. 1988 Tropical and Subtropical vegetation of Meso-America. IN Barber, M.G., and Billings, W.D. North American terrestrial vegetation. Cambridge University Press. ooo Harvey, L.D. 1980 Solar variability as a contributing factor to Holocene climatic change. Progress in Physical Geography 4: 283-304. Hastenrath, S. 1976 Variations in Low latitude circulation and extreme climatic events in the Tropical Americas. Journal of the Atmospheric Sciences 33: 202-215. Hastenrath, S. 1984 Interannual variability and annual cycle: mechanisms of circulation and climate in the tropical Atlantic sector. Monthly Weather Review 112: 1097-1107. Hastenrath, S. 1991 Climate dynamics of the Tropics. Kluwer Academic publishers, Dordrecht: 488pp. Haviland, W.A. 1967 Stature at Tikal: Implications for ancient Maya demography and social organisation. American Antiquity 32: 316-325. Healy, P.F., Lambert, J.D.H., Arnason, J.T. and Hebda, R.J. 1983 Caracol, Belize: Evidence of Ancient Maya agricultural terraces. Journal ofField Archaeology 10: 397-410. Heaton, T.H.E., Holmes, J.A. and Bridgwater, N.D. 1995 Carbon and oxygen isotope variations among lacustrine ostracods: implications for palaeoclimatic studies. The Holocene 5(4): 428-434. Hester, T.R., Shafer, H.J. and Eaton, J.D. (eds). Archaeology at Colha, Belize: The 1981 interim report. Center for Archaeological Research, The University of Texas at San Antonio. High, L.R. 1975 Geomorphology and sedimentology of Holocene coastal deposits, Belize. IN Wantland, K.W. and Pusey, W.C. (eds). Belize Shelf: Carbonate sediments, clastic sediments and ecology. American Association of Petroleum Geologists, Tulsa: 53-96. Higuera-Gundy, A., Brenner, M., Hodell, D.A., Curtis, J.H., Leyden, B.W. and Binford, M.W. 1999 A 10,300 l4C year record of climate and vegetation change from Haiti. Quaternary Research 52: 159-170. Hodell, D.A., Curtis, J.H., Jones, G.A., Higuera-Gundy, A., Brenner, M., Binford, M.W., and Dorsey, K.T. 1991 Reconstruction of Caribbean climate change over the past 10,500 years. Nature 352:790-793. Hodell, D.A., Curtis, J.H., and Brenner, M. 1995 Possible role in climate in the collapse of classic Maya civilization. Nature 375: 391-394. Hodell, D.A., Brenner, M., Curtis, J.H. and Guilderson, T. 2001 Solar forcing of drought frequency in the Maya Lowlands. Science 292: 1367-1370. Holmes, J.A., Street-Perrott, F.A., Ivanovich, M. and Perrott, R.A 1995 A late Quaternary palaeolimnological record from Jamaica based on trace element chemistry of ostracod shells. Chemical Geology 124: 143-160. Holmes, J.A., Street-Perrott, F.A., Allen, M.J., Fothergill, P.A., Harkness, D.D., Kroon, D. and Perrott, R.A. 1997 Holocene palaeolimnology of Kajemarum Oasis, Northern Nigeria: an isotopic study of ostracodes, bulk carbonate and organic carbon. Journal of the Geological Society, London 154: 311-319. Hooghiemstra, H., Cleef, A.M., Noldus, G.W., and Kappelle, M. 1992 Upper Quaternary vegetation dynamics and palaeoclimatology of the La Chonta bog area (Cordillera de Talamanca, Costa Rica). Journal ofQuaternary Science 7: 205-225. Horn, S.P. 1993 Postglacial vegetation and fire history in the Chirripo Paramo of Costa Rica. Quaternary Research 40: 107-116. Horn, S.P. and Sanford, R.L. 1992 Holocene fires in Costa Rica. Biotropica 24: 354- 361. Hustedt, F. 1930a Die Kieselalgen Deutschlands, Okologische und der Schweiz, Volume 1. Reprinted 1977, Otto Koeltz, Koenigstein. Hustedt, F. 1930b Die Kieselalgen Deutschlands, Okologische und der Schweiz, Volume 2. Reprinted 1977, Otto Koeltz, Koenigstein. Hustedt, F. 1930c Die Kieselalgen Deutschlands, Okologische und der Schweiz, Volume 3. Reprinted 1977, Otto Koeltz, Koenigstein. Hustedt, F. 1937-39 Systematische und okologische untersuchungen uber die Diatomeenflora von Java, Bali und Sumatra, nach der material der Deutschen limnologischen Sunda expedition. Teil 1. Hustedt, F. 1956 Kieselalgen (Diatomeen). Franckh, Stuttgart. Hustedt, F, 1959 Die Diatomeen flora des Neusiedlersees in Oesterreichischem Burgenland. Oesterr. BD Zeitr., Bd 106, S. 390-430. Islebe, G.A. and Hoogiemstra, H. 1997 Vegetation and climate history of montane Costa Rica since the last glacial. Quaternary Science Reviews 16: 589-604. Islebe, G.A., Hooghiemstra, H. and van der Borg, K. 1995 A cooling event during the Younger Dryas chronozone in Costa Rica. Palaeogeography, Palaeoclimatology, Palaeoecology 117: 73-80. Islebe, G.A., Hooghiemstra, H., Brenner, M., Curtis, J.H., Hodell, D.A. 1996a A Holocene vegetation history from lowland Guatemala. The Holocene 6:265-271. 0/~» A Islebe, G.A., Hooghiemstra, H and van't Veer, R. 1996b Holocene vegetation and water level history in two bogs of the Cordillera de Talamanca, Costa Rica. Vegetatio 124: 155-171. Jacob, J.S. 1992 The agroecological evolution of Cobweb Swamp, Belize. Ph.D. thesis Texas A+M University. Jacob, J.S. 1995 Ancient Maya wetland agricultural fields in Cobweb Swamp, Belize: construction, chronology and function. Journal ofField Archaeology 22: 175-190. Jacob, J.S., and Hallmark, C.T. 1996 Holocene stratigraphy of Cobweb Swamp, a Maya wetland in North Belize. Geological Society ofAmerica Bulletin 108:883-891. Johnson, W.C. 1983 The physical setting: Northern Belize and Pulltrouser Swamp. IN Turner, B.L. and Harrison, P.D. (eds). Pulltrouser Swamp: Ancient Maya habitat, agriculture and settlement in northern Belize. University of Texas Press: 8-20. Jones, G.D. 1989 Maya resistance to Spanish rule: Time and history on a colonial frontier. University of New Mexico Press, Albuquerque. Jones, J.G. 1991 Pollen evidence ofprehistoric forest modification and Maya cultivation in Belize. Unpublished PhD Thesis, Texas A&M University. Jongman, R.H.G., ter Braak, C.J.F., and Van Tongeren, O.F.R. (eds). 1995 Data analysis in community and landscape ecology. Cambridge University Press, Cambridge. Kaushal, S. and Binford, M.W. 1999 Relationship between C:N ratios of lake sediments, organic matter sources and historical deforestation in Lake Pleasant, Massachusetts, USA. Journal ofPalaeolimnology 22: 439-442. Kent, M. and Coker, P. 1992 Vegetation description and analysis: A practical approach. CRC Press. Kilham, P. 1971 A hypothesis concerning silica and the freshwater planktonic diatoms. Limnology and Oceanography 16: 10-18. Kjellmark, E. 1996 Late Holocene climate change and human disturbance on Andros Island, Bahamas. Journal ofPaleolimnology 15: 133-145. Knudson, B.M. 1953 The diatom genus Tabellaria II. Taxonomy and morphology of the plankton varieties. Annals ofBotany 17: 131-155. Kolbe, R.W. 1927 Pflanzenforchung: zur okologie, morphologie and systematik der brackwasser diatomeen. Die kiesalagen des sperenberger salzgebiets Verlag von Gustav Fischer. Krammer, K. and Lange-Bertalot, H. 1986 Siifiwasserflora von Mitteleuropa: Herausgeben von H.Ettl, J.Gerlof H.Heynig, D.Moltenhauer. Bacillariophyceae 1. Teil Naviculaceae Gustav Fisher Verlag. Krammer, K. and Lange-Bertalot, H. 1988 Siipwasserflora von Mitteleuropa: Herausgeben von H.Ettl, J.Gerlof, H.Heynig, D.Moltenhauer. Bacillariophyceae 2. Teil Bacillariaceae, Epithemiaceae, Surirellaceae Gustav Fisher Verlag. Krammer, K. and Lange-Bertalot, H. 1991a SiijJwasserflora von Mitteleuropa: Herausgeben von H.Ettl, J.Gerlof, H.Heynig, D.Moltenhauer. Bacillariophyceae 3. Teil Centrales, Fragilariaceae, Eunotiaceae Gustav Fisher Verlag. Krammer, K. and Lange-Bertalot, H. 1991b Sii[3wasserflora von Mitteleuropa: Herausgeben von H.Ettl, G.Gartner, J. Gerloff, H.Heynig, D.Moltenhauer. Bacillariophyceae 4. Teil Achnanthaceae Kritische Erganzungen zu Navicula (lineolatoe) und Gomphonema Gustav Fisher Verlag. Krishnaswamy, S. and Lai, D. 1978 Radionuclide limnochronology IN Lerman, A. (ed). Lakes chemistry, geology, physics. Springer Berlin: 153-177. Laird, K.R., Fritz, S.C., Maasch, K.A. and Cumming, B.F. 1996 Greater drought intensity and frequency before AD 1200 in the Northern Great Plains, USA. Nature 384: 552-554. Laird, K.R., Fritz, S.C. and Cumming, B.F. 1998 A diatom-based reconstruction of drought intensity, duration and frequency from Moon lake, North Dakota: a sub decadal record of the last 2300 years. Journal ofPaleo limnology 19: 161-179. Lamb, A.L. 2000 Stable isotope geochemistry of lakes Tilo and Awassa, Ethiopia: a Holocene record of volcanic and climatic change. Unpublished PhD thesis, University of Wales, Aberystwyth. Lamb, H.F., Gasse, F., Benkaddour, A., El Hamouti, N., van der Kaars, S., Perkins, W.T., Pearce, N.J. and Roberts, C.N. 1995 Relation between century scale Holocene arid intervals in tropical and temperate zones. Nature 373: 134-137. Lambert, J.D.H. and Arnason, T. 1978 Distribution of vegetation on Mayan ruins and its relationship to ancient land use at Lamanai, Belize. Turrialba 28 : 33-41. Lambert, J.D.H., Siemens, A.H. and Arnason, J.T. 1984 Ancient Maya drained field agriculture: its possible application today in the New River floodplain, Belize, central America. Agriculture, Ecosystems and Environment. 11: 67-84. Lean, J. and Warrilow, D.A. 1989 Simulation of the regional climatic impact of Amazon deforestation. Nature 342: 411-413. Leng, M.J. Unpublished. Stable isotopes in the lacustrine and marine environment. University College London, Short Course notes, 19-23rd March 2001. Leng, M.J., Heaton, T.H.E., Lamb, H.F. and Naggs, F. 1998 Carbon and oxygen isotope variations within the shell of an African land snail (Limicolaria kambeul chudeaui Germain): a high-resolution record of climate seasonality? The Holocene 8: 407-412. Leyden, B.W. 1984 Guatemalan forest synthesis after Pleistocene aridity. Proceedings of the National Academy ofScience 81:4856-4859. Leyden, B.W. 1985 Late Quaternary aridity and Holocene moisture fluctuations in the Lake Valencia Basin, Venezuela. Ecology 66:1279-1295. Leyden, B.W. 1987 Man and climate in the Maya lowlands. Quaternary Research 28:407-414. Leyden, B.W. 1995 Evidence of the Younger Dryas in Central America. Quaternary Science Reviews 14: 833-839. Leyden, B.W., Brenner, M., Hodell, D.A. and Curtis, J.H. 1993 Late Pleistocene Climate in Central American Lowlands. IN Climate change in continental isotopic records, Geophysical Monograph 78. Leyden, B.W., Brenner, M., Hodell, D.A. and Curtis, J.H. 1994 Orbital and internal forcing of climate on the Yucatan peninsula for the past ca. 36 ka. Palaeogeography, Palaeoclimatology, Palaeoecology 109: 193-210. Leyden, B.W., Brenner, M., Whitmore, T., Curtis, J.H., Piperno, D.R., and Dahlin, B.D. 1996 A record of long- and short-term climatic variation from North West Yucatan: Cenote San Jose Chulchaca. IN Fedick, S.L. The managed mosaic: Ancient Maya agriculture and resource use. University of Utah Press. Leyden, B.W., Brenner., M and Dahlin, B.H. 1998 Cultural and climatic history of Coba, a lowland Maya city in Quintana Roo, Mexico. Quaternary Research 49: 111- 122. Li, H.C. and Ku, T.L. 1997 8I3C / §lhO covariance as a paleohydrological indicator for closed basin lakes Palaeogeography, Palaeoclimatology, Palaeoecology 133: 69- 80. Libby, W.F. 1955 Radiocarbon Dating. University of Chicago Press. Lighty, R.G., Macintyre, I.G. and Stuckenrath, R. 1982 Acropara Palmata reef framework: A reliable indicator of sea level in the western Atlantic for the past 10,000 years. Coral Reefs 1: 125-130. Lister, G.S., Kelts, K., Zao, K., Yu, J. and Niessen, F. 1991 Lake Quinghai, China: closed-basin lake levels and the oxygen isotope record for ostracoda since the latest Pleistocene. Palaeogeography, Palaeoclimatology and Paleoecology 84: 141-162. ■507 Loten, H.S. 1985 Lamanai Postclassic. IN Chase, A.F. and Rice, P.M. (eds). The LowlandMaya Postclassic. University of Texas Press: 85-90. Lowe, J.J. and Walker M.J.C. 1997 Reconstructing Quaternary environments. Longman. Lowe, J.W.G. 1985 The dynamics ofApocalypse: A systems simulation of the Classic Maya collapse University of New Mexico Press. Mann, D.G. 1989 The species concept in diatoms: evidence for morphologically distinct, sympatric gamodemes in four epipelic species. Plant systematics and evolution. 164: 215-237. Mann, D.G. 1994 The origins of shape and form in diatoms: the interplay between morhogenetic studies and systematics. IN Shape andform in plants andfungi. The Linnean Society of London. Mann, D.G. 1999 The species concept in diatoms. Phycologia 38: 437-495. Mann, D.G. and Droop, S.J.M. 1996 Biodiversity, biogeography and conservation of diatoms. Hydrobiologia 336: 19-31. Marchitto, T.M., Curry, W.B. and Oppo, D.W. 1998 Millennial- scale changes in North Atlantic circulation since the last glaciation. Nature 393: 557-560. Masson, M.A. 1993 Changes in Maya community organisation from the Classic to Postealssic periods: a view from Laguna de On, Belize. PhD Thesis, University of Texas at Austin. Masson, M.A. 1995 Understanding the stratigraphic context of the Maya Postclassic in Belize. Geoarchaeology: An International Journal 10: 389-404. Masson, M.A. 1996 Overview of 1996 investigations at Laguna de On Island. IN Masson, M.A. and Rosenwig, R.M. (eds). The Belize Postclassic project: Laguna de On Island excavations 1996. Occasional publication 1, Institute of Mesoamerican Studies, The University at Albany, State University of New York. Masson, M.A. 1997 Cultural transformation at the Maya Postclassic community of Laguna de On, Belize. Latin American Antiquity 8(4): 293-316. Masson, M.A. 1999 Postclassic Maya ritual at Laguna de On Island, Belize. Ancient Mesoamerica. 10: 51-68. Masson, M.A. Unpublished. Postclassic Maya communities at Progresso Lagoon and Laguna Seca, Northern Belize. R9R McKenzie, J.A. 1985 Carbon isotopes and productivity in the lacustrine and marine environment. IN: Stumm, W. (ed). Chemical Processes in Lakes. Wiley: 99-118. Meggers, B.J. 1954 Environmental limitation on the development of culture. American Anthropologist 56: 801-824. Messerli, B., Grosjean, M., Hofer, T., Nunez, L. and Pfister, C. 2000 From nature- dominated to human-dominated environmental changes. Quaternary Science Reviews 19: 459-479. Messenger, L.C. 1990 Climatic settings and prehistoric social complexity in Mesoamerica. Ancient Mesoamerica 1:21 -40. Metcalfe, S.E. 1988 Modern diatom assemblages in Central Mexico: the role of water chemistry and other environmental factors as indicated by TWINSPAN and DECORANA. Freshwater Biology 19: 217-233. Metcalfe, S.E. 1995 Holocene environmental change in the Zacapu Basin, Mexico: a diatom-based record. The Holocene 5:196-208. Metcalfe, S.E., Street-Perrott, F.A., O'Hara, S.I., Hales, P.E., and Perrott, R.A. 1994 The paleolimnological record of environmental change: Examples from the arid frontier of Mesoamerica. IN Millington, A.C., and Pye, K. Environmental change in drylands: Biogeographical and Geomorphological perspectives. John Wiley and Sons Ltd. Meyers, P.A. 1994 Preservation of elemental and isotopic source identification of sedimentary organic matter. Chemical Geology 114: 289-302. Mook, W.G. 1986 Recommendations/ resolutions adopted by the Twelfth International Radiocarbon Conference. Radiocarbon 28: 2A 799. Mook, W.G. and Vogel, J.C. 1968 Isotopic equilibrium between shells and their environment. Science: 874-875. Morely, S.G. 1920 The inscriptions at Copan. Carnegie Institution of Washington, Publication No. 219. Morely, S.G. 1938 The inscriptions ofPeten. Carnegie Institution of Washington. Publication no 437, 5 vols. Murray, M.R. 1995 The environmental effects ofmangrove clearance in Belize, Central America. Unpublished PhD thesis, The University of Edinburgh. Murray, M.R. in press Is laser particle size determination possible for carbonate-rich lake sediments? Journal ofPaleolimnology. Murray, M.R. unpublished. Internal report, Department of Geography, The University of Edinburgh. Neale, J.W. 1988 Ostracods and paleosalinity reconstruction. EN Decker, P.D. (ed). Ostracoda in Earth Sciences. Elsevier: 125-155. Nieuwenhuize, J., Maas, Y.E.M. and Middelburg, J.J. 1994 Rapid analysis of organic carbon and nitrogen in particulate materials. Marine Chemistry 45: 217-224. Northrup, L.A. and Horn, S.P. 1996 Pre-Columbian agriculture and forest disturbance in Costa Rica: palaeoecological evidence from two lowland rainforest lakes. The Holocene 6: 289-299. O'Hara, S.L., Street-Perrott, F.A. and Burt, T.P. 1993 Accelerated soil erosion around a Mexican highland lake caused by pre-hispanic agriculture. Nature 362:48- 51. Olsson, I.U. 1986 Radiometric dating. IN Berglund, B.E. (ed). Handbook of Holocene palaeoecology andpalaeohydrology. John Wiley and Sons: 298-312. Patrick, R and Reimer, C.W. 1966 The diatoms of the United States - volumes 1 and 2. Monographs of the Academy of Natural Sciences of Philadelphia. Patrick, R. 1977 Ecology of freshwater diatoms - Diatom communities. IN Werner, D. (ed). The Biology ofDiatoms. Botanical Monographs 13. Blackwell Scientific Publications: 284-332. Pearson, F.J. and Coplen, T.B. 1978 Stable isotope studies of lakes. IN Lerman, A. (ed). Lakes chemistry, geology, physics. Springer Berlin:325-339. Pendergast, D.M. 1971 Evidence of early Teotihucan - Lowland Maya contact at Altun Ha. American Antiquity 36: 455-460. Pendergast, D.M. 1975 The church in the jungle: The ROM's first season at Lamanai. Rotunda. Pendergast, D.M. 1981 Lamanai, Belize: Summary of excavation results 1974-1980. Journal ofField Archaeology 8: 29-53. Pendergast, D.M. 1984 Excavations at Lamanai Belize, 1983. Mexicon VI: 5-11. Pendergast, D.M. 1985 Lamanai, Belize: An updated view. IN Chase, A.F. and Rice, P.M. (eds). The Lowland Maya Postclassic. University of Texas Press: 91-103. Pendergast, D.M. 1986 Historic Lamanay: Royal Ontario Museum 1985 excavations at Lamanai Belize. Mexicon VIII: 9-13. OCA Pendergast, D.M. 1987 The Southern Maya Lowlands contact experience: The view from Lamanai, Belize. IN Thomas, D.H. (ed). Columbian consequences volume 3: The Spanish borderlands in Pan-American perspective. Smithsonian Institution Press: 336-354. Pendergast, D.M. 1990 Up from the dust: the Central Lowlands Postclassic as seen from Lamanai and Marco Gonsalez, Belize. IN Clancy, F.S. and Harrison, P.D. (eds). Vision and revision in Maya studies. University of New Mexico Press, Albuquerque: 169-177. Pendergast, D.M., Jones, G.D. and Graham, E. 1993 Locating Maya Lowlands Spanish Colonial towns: a case study from Belize. Latin American Antiquity 4(1) 59:73. Peterson, L.C., Overpeck, J.T., Kipp, N.G. and Imbrie, J. 1991 A high resolution late Quaternary upwelling record from the anoxic Cariaco Basin, Venezuela. Paleoceanography 6: 99-119. Piperno, D.R., Bush, M.B. and Colinvaux, P.A. 1990 Paleoenvironments and human occupation in Late Glacial Panama. Quaternary Research 33: 108-116. Piperno, D.R., Bush, M.B. and Colinvaux, P.A. 1991 Paleoecological perspectives on human adaption in Central Panama: The Holocene. Geoarchaeology 6: 227-250. Podzorski, A.C. 1984 The reaction of epiphytic diatoms (Bacillariophyta) to environmental change in Broad River, Western Jamaica. Nova Hedwigia 40: 487- 509. Pohl, M. (Ed) 1985 Prehistoric lowlandMaya environment and subsistence economy. Harvard University Press. Pohl, M. (Ed) 1990 Ancient Maya wetland agriculture: excavations on Albion Island, northern Belize. Westview Press. Pohl, M., Bloom, P. Pope, K. 1990 Interpretation of wetland farming in Northern Belize: excavations at San Antonio, Rio Hondo. EN Pohl, M (Ed). AncientMaya wetland agriculture: excavations on Albion Island, northern Belize Westview Press: 187-254. Pohl, M.D., Pope, K.O., Jones, J.G., Jacob, J.S., Piperno, D.R., de France, S.D., Lentz, D.L., Gifford, J.A., Danforth, M.E. and Josserand, J.K. 1996 Early agriculture in the Maya Lowlands. Latin American Antiquity 7(4): 355-372. Pope, K.O. and Dahlin, B.H. 1989 Ancient Maya wetland agriculture: New insights from ecological and remote sensing research. Journal of Field Archaeology 16: 87- 106. OOH Pope, K.O., Rejmankova, E. and Paris, J.F. 2001 Spaceborne imaging radar-C (SIR- C) observations of groundwater discharge and wetlands associated with the Chicxulub impact crater, northwestern Yucatan Peninsula, Mexico. Geological Society ofAmerica Bulletin 113: 403-416. Pyburn, K.A., Dixon, B., Cook, P. and McNair, A. 1998 The Albion Island settlement pattern project: Domination and resistance in Early Classic Northern Belize. Journal ofField Archaeology 25: 37-62. Rathje, W.L. 1971 The origin and development of Lowland Classic Maya civilisation American Antiquity 36(3) 275:285. Reed, J.M. 1995 The potential of diatoms and other palaeolimnological indicators for Holocene palaeoclimate reconstruction from Spanish salt lakes, with special reference to Laguna de Medina (Cadiz, southwest Spain). Unpublished PhD Thesis, University College, London. Reed, J.M. 1998a A diatom-conductivity transfer function for Spanish salt lakes. Journal ofPalaeolimnology 19: 399-416. Reed, J.M. 1998b Diatom preservation in the recent sediment record of Spanish saline lakes: implications for palaeoclimate study. Journal ofPaleolimnology 19: 129-137. Rejmankova, E., Pope, K.O., Pohl, M.D. and Rey-Benayas, J.M. 1995 Freshwater wetland plant communities of northern Belize: Implications for paleoecological studies of Maya wetland agriculture. Biotropica 27: 28-36. Rice, D.S. 1976 The historical ecology of Lakes Yaxha and Sacnab, El Peten, Guatemala. Unpublished PhD, Pennsylvania State University. Rice, D.S. and Culbert, T.P. 1990 Historical contexts for population reconstruction in the Maya Lowlands. IN Culbert, T.P. and Rice, D.S (eds). Pre-columbian population history in the Maya Lowlands. University of New Mexico Press, Albuquerque: 1-36. Rice, P. 1986 The Peten Postclassic: A settlement perspective from the Central Peten lakes. IN Sabloff, J.A. and Andrews, E.W. (eds). Late Lowland Maya Civilisation: Classic to Postclassic. University of New Mexico Press, Albuquerque: 251-299. Ricketson, O.G. and Ricketson, E.B. 1937 Uaxactun, Guatemala: Group E 1926- 1931. Washington D.C.: Carnegie Institution. Roberts, N. 1996 Long-term environmental stability and instability in the Tropics and Subtropics. IN Driver, T.S. and Chapman, G.P. (eds). Time-scales and environmental change. Routledge: 25-38. T39 Robichaux, H.R. 1995 Survey in the peripheral zones of the La Milpa and Dos Hombres ancient Maya sites in northwestern Belize: The 1994 season. IN Adams, R.E.W. and Valdez, F. 1995 The Programme for Belize Archaeological project: 1994 interim report. The Center for Archaeology and Tropical Studies and The University of Texas at San Antonio. Roseinmeier, M.F., Hodell, D.A., Brenner, M., Curtis, J.H. and Guilderson, T.P. in press A 3500-year record of climate change and human disturbance from the Southern Maya Lowlands, Peten, Guatemala. Quaternary Research. Round, F.E., Crawford, R.M. and Mann, D.G. 1990 The Diatoms: Biology and morphology of the genera. Cambridge University Press. Roys, R.L. 1957 The political geography of the Yucatan Maya. Publication 613. Carnegie Institution of Washington, Washington D.C. Rozanski, K., Araguas-Araguas, L. and Gonfiantinin, R. 1993 Isotopic patterns in modern global precipitation Climate Change in continental isotopic records, Geophysical Monograph 78. Rubinson, M. and Clayton, R.N. 1969 Carbon-13 fractionation between aragonite and calcite. Geochimica et Cosmochimica Acta 33: 997-1002. Rue, D.J. 1987 Early agriculture and early postclassic Maya occupation in Western Honduras. Nature 326: 285-286. Rull, V. 1996 Late Pleistocene and Holocene climates of Venezuala. Quaternary International 31: 85-94. Ryves, D.B. 1994 Diatom dissolution in saline lake sediments: an experimental study in the Great Plains ofNorth America. Unpublished PhD Thesis, University College, London. Ryves, D.B. and Batterbee, R.W. unpublished. Assessing the impact of diatom dissolution in biasing quantitative salinity reconstructions from saline lake sediments. NERC Research Grant Report GR9/02033. Ryves, D.B., Juggins, S., Fritz, S.C. and Battarbee, R.W. 2001 Experimental dissolution and the quantification of microfossil preservation in sediments. Palaeogeography, Palaeoclimatology, Palaeoecology 172: 99-1 13. Sabloff, J.A. 1991 The new archaeology and the ancient Maya. New York. Sabloff, J.A.and Wiley G.R., 1967 The collapse of the Maya civilisation in the Southern Lowlands: A consideration of history and process. Southwestern Journal of Anthropology 23: 311-336. Sanders, W.T. 1962 Cultural ecology of the Maya Lowlands, Part 1. Estudios de Cultura Maya 2: 79-121. Sanders, W.T. 1963 Cultural ecology of the Maya Lowlands, Part 2. Estudios de Cultura Maya 2: 203-241. Sanders, W.T. 1973 The cultural ecology of the Lowland Maya: A re-evaluation. IN Culbert, T.P. (ed). The Classic Maya Collapse. University of New Mexico Press, Albuquerque: 325-366. Santley, R.S., Killion, T.W., and Lycett, M.T. 1986 On the Maya collapse. Journal ofAnthropological Research 123-159. Scarborough, V.L. 1998 Ecology and ritual: water management and the Maya. Latin American Antiquity 9(2) 135:159. Scarborough, V.L., Becher, M.E., Baker, J.L., Harris, G and Valdez, F. 1995 Water and land at the ancient Maya community of La Milpa. Latin American Antiquity 6(2) 98:119. Scott, R.F. 1980 Further comments on faunal analysis and ancient subsistence activities at Colha. IN Hester, T.R., Eaton, J.D. and Shafer, H.J. (eds). The Colha project, second season interim report. Center for Archaeological Research, University of Texas at San Antonio. Scott, R.F. 1982 Notes on continuing faunal analysis for the site of Colha, Belize: data from the Early Postclassic. IN Hester, T.R., Shafer, H.J. and Eaton, J.D. (eds). Archaeology at Colha, Belize: The 1981 interim report. Center for Archaeological Research, The University of Texas at San Antonio. Sharer, R.J. 1973 The Maya collapse revisited: internal and external perspectives. IN Culbert, T.P. (ed). 1973 The Classic Maya Collapse. University of New Mexico Press. Sharer, R.J. 1994 The Ancient Maya. Stanford University Press, Stanford, California: Fifth Edition. Sheets, P. 2000 The effects of explosive volcanism on simple to complex societies in ancient middle America. IN Markgraf, V. (ed). Interhemispheric Climate Linkages. San Diego: Academic Press. Siemens, A.H. 1978 Karst and the Pre-Hispanic Maya in the Southern Lowlands. IN Harrison, P.D. and Turner, B.L. (eds). Pre-Hispanic Maya Agriculture. University of New Mexico Press, Albuquerque: 117-143. Stoermer, E.F. 2001 Diatom taxonomy for paleolimnologists. Journal of Paleolimnology 25: 393-398. Stoermer, E.F. and Smol, J.P. (eds). 1999 The diatoms: applications for the environmental and earth sciences. Cambridge University Press. Street-Perrott, F.A. and Harrison, S.P. 1985 Fake levels and climate. IN Hecht, A.D. (ed). Paleoclimate analysis and modeling. Wiley, New York: 291-340. Street-Perrott, F.A., Hales, P.E., Perrott, R.A., Ch.Fontes, J., Switsur, V.R. and Pearson, A. 1993 Fate Quaternary palaeolimnology of a tropical marl lake: Wallywash Great Pond, Jamaica. Journal ofPalaeolimnology 9: 3-22. Street-Perrott, F.A., Holmes, J.A., Waller, M.P., Allen, M.J., Barber, N.G.H., Fothergill, P.A., Harkness, D.D., Ivanovich, M., Kroon, D and Perrott, R.A. 2000 Drought and dust deposition in the West Africal Sahel: A 5500 year record from Kajemarum Oasis, northeastern Nigeria. Holocene 10(3): 293-302. Stuiver, M. 1970 Oxygen and carbon isotope ratios of freshwater carbonates as climatic indicators. Journal ofGeophysical Research 75: 5247-5257 Stuiver, M. and Reimer, P.J. 1993 Extended l4C database and revised CALIB 3.0 age calibration. Radiocarbon 35: 215-230. Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B., McCormac, F.G., v.d. Plicht, J. and Spurk, M. 1998 Radiocarbon 40: 1041-1083. Stumm, W. and Morgan, J.J. 1970 Aquatic Chemistry. Wiley and Sons. Sutton, R.T., Jewson, S.P. and Rowell, D.P. 2000 The elements of climate variability in the tropical Atlantic region. Journal of Climate 13: 3261-3284. Tainter, J.A. 1988 The collapse of complex societies. Cambridge University Press. Talbot, M.R. 1990 A review of the palaeohydrological interpretation of carbon and oxygen isotopic ratios in primary lacustrine carbonates. Chemical Geology 80: 261 - 279. Tarutani, T., Clayton, R.N. and Mayeda, T.K. 1969 The effect of polymorphism and magnesium substitution on oxygen isotope fractionation between calcium carbonate and water. Geochimica et Cosmochimica Acta 33: 987-996. Ter Braak, C.J.F. 1987-1992 CANOCO- a FORTRAN program for canonical community ordination by [partial] [detrended] [canonical] correspondence analysis, principal components analysis and redundancy analysis. Agricultural Mathematics Group, Wageningen. Thompson, J.E.S. 1954 The Rise and Fall of the Maya Civilisation. University of Oklahoma Press, Norman. Thompson, J.E.S. 1966 The Rise and Fall ofMaya Civilisation. University of Oklahoma Press: Second Edition. Thompson, R. 1984 A global review of palaeomagnetic results from wet lake sediments. IN Elowarth, E. and Lund, J. (eds). Lake sediments and environmental history. University of Minneapolis Press, Minneapolis: 154-164. Thompson, L.G., Davis, M.E., Mosley-Thompson, E. and Liu, K-B 1988 Pre-Incan agricultural activity recorded in dust layers in two tropical ice cores. Nature 336: 763-765. Tilman, D., Kilham, S.S. and Kilham, P. 1982 Phytoplankton community ecology: the role of limiting nutrients. Annual Review ofEcology and Systematics 13:349-372. Tourtellot, G., Clarke, A. and Hammond, N. 1993 Mapping La Milpa: a Maya city in northwestern Belize. Antiquity 67: 96-108. Tucker, M.E. and Wright, V.P. 1990 Carbonate sedimentology. Blackwell Scientific publications. Turner, B.L. and Harrison, P.D. 1981 Prehistoric raised field agriculture in the Maya lowlands Science 213: 399-405. Turner, B.L., and Harrison, P.D. 1983 Pulltrouser Swamp: Ancient Maya habitat, agriculture and settlement in Northern Belize University of Texas Press. University of Columbia, 2000. http://ingrid.ldgo.columbia.edu. Urey, H.C. 1947 The thermodynamic properties of isotopic substances. Journal of Chemical Society, London. 108: 562. Van Geel, B., Van der Plicht, J., Kilian, M.R., Klaver, E.R., Kouwenberg, J.H.M., Renssen, H., Reynaud-Farrera, I. And Waterbolk, H.T. 1998 The sharp rise of l4C c. 800 cal BC: possible causes, related climatic teleconnections and the impact on human environments. Radiocarbon 40: 535-550. Van Geel, B., Heusser, C.J. and Renssen, H. 2000 Climatic change in Chile at around 2700 BP and global evidence for solar forcing: A hypothesis. Holocene 10: 659-664. Vaughan, H.H. 1979 Prehistoric disturbance of vegetation in the area ofLake Yaxha, Peten, Guatemala. Unpublished PhD thesis, University of Florida, Gainsville. Vaughan, H.H., Deevey, E.S. and Garrett-Jones, S. 1985 Pollen stratigraphy of two cores from the Peten Lake District. IN Pohl, M. (ed). Prehistoric Lowland Maya environment and subsistence economy. Harvard University, Papers of the Peabody Museum: 73-89. Walker, S.H. 1973 Summary ofclimatic records for Belize. Foreign and Commonwealth Office, Supplement Report 3. Walling, D. 1995 Biogeochemistry of small catchments - a tool for environmental research Applied Geography 15(3): 300-301. Watts, W.A. 1975 A late Quaternary record of vegetation from Lake Annie, south- central Florida. Geology 3: 344-346. Watts, W.A. and Hansen, B.C.S. 1994 Pre-Holocene and Holocene pollen records of vegetation history from the Florida peninsula and their climatic implications Palaeogeography, Palaeoclimatology and' Palaeoecology 109: 163-176. Webb, M.C. 1964 The post-classic decline of the Peten Maya: an interpretation in the light ofgeneral theory ofstate society. Unpublished PhD University of Michigan. Webb, M.C. 1973 The Maya Peten decline viewed in the perspective of state formation. IN Culbert, T.P. (ed). The Classic Maya Collapse. University of New Mexico Press, Albuquerque: 367-404. Webb, R.S., Rind, D.H., Lehman, S.J., Healy, R.J and Sigman, D. 1997 Influence of ocean heat transport on the climate of the Last Glacial Maximum. Nature 385: 695- 699. White, C.D., Healy, P.F. and Schwarcz, H.P. 1993 Intensive agriculture, social status and Maya diet at Pacbitun, Belize. Journal ofAnthropological Research 49: 347-375. Whitmore, T.J., Brenner, M., Curtis, J.H., Dahlin, B.H., Leyden, B.W. 1996 Holocene climatic and human influences on lakes of the Yucatan Peninsula, Mexico: an interdisciplinary, palaeolimnological approach. The Holocene 6: 273-287. Willey, G.R. 1974 The Classic Maya hiatus: a rehearsal for the collapse? Hammond, D. (ed). Mesoamerican Archaeology: New Approaches. University of Texas Press, Austin. Willey, G.R. 1982 Maya archaeology. Science 2\5: 260-261. Wiseman, F.M. 1983 Analysis of pollen from fields at Pulltrouser Swamp. IN Turner, B.L. and Harrison, P.D. (eds). Pulltrouser swamp. Ancient Maya habitat, Agriculture and settlement in northern Belize. University of Texas Press, Austin: 105-119. Wiseman, F.M. 1985 Agriculture and vegetation dynamics of the Maya collapse in central Peten, Guatemala. IN Pohl, M. (ed). Prehistoric Lowland Maya environment and subsistence economy. Peabody Museum of Archaeology and Ethnology, Harvard University: 63-71. Wiseman, F.M. 1990 San Antonio: A Late Holocene record of agricultural activity in the Maya Lowlands. IN Pohl, M. (ed). 1990 Ancient Maya wetland agriculture: excavations on Albion Island, northern Belize Westview Press 313-322. Wolin, J.A. and Duthie, H.C. 1999 Diatoms as indicators of water level change in freshwater lakes. IN Stoermer, E.F. and Smol, J.P. (eds). 1999 The diatoms: applications for the environmental and earth sciences. Cambridge University Press. Wright, A.C.S, Romney, D.H., Arbuckle, R.H. and Vial, V.E. 1959 Land in British Honduras: Report of the British Honduras Land Use Survey Team. Colonial Research Publications No. 24. QQfi Appendix One: The following plates are pictures of diatoms species and a selection of the dissolution stages that were encountered in the study of Honey Camp 1999. The species shown cover a variety of forms to provide an overview of the dissolution process in diatoms. The numbers in brackets correspond to the dissolution stage shown. Mastogloia smithii var lacustris Perfect (1) Both ends missing (3) One end missing (2) Central area (4) Central area (4) Faint Brachysira neoexilis Perfect (1) maaii r i! I % i : i Centre (4) Encyonema carina Ends missing (3) Edges missing (2) Centre (4) Perfect (1) Diploneis elliptica *jt rsp«M- \ I : V -2£*r2 ' Edges degraded (2) Edges and End Missing (3) yy. 7/ Centre (4) Centre (4) Navicula radiosa Ends Missing (3) Plate One: A: Achnanthes exigua B: Achnanthes minutissima C: Brachysira neoexilis var large D: Brachysira neoexilis var capitate E: Brachysira neoexilis var large capitate F: Brachysira neoexilis G: Cymbella microcephala H: Cymbella mesiana I: Cocconeis placentula var euglypta Plate Two: J: Denticula elegans K: Denticula tenuis L: Encyonema carina M: Fragilaria fasciculata N: Fragilaria construens O: Gomphonema gracile Plate Three: P: Mastogloia smithii Q: Mastogloia elliptica var dansei R: Gyrosigma acuminatum S: Species 15 T: Navicula florinae U: Unidentified species (Almond Hill Lagoon) Plate Four: V: Navicula radiosa W: Navicula radiosa var tenella X: Navicula cuspidata Plate Five: Y: Nitzschia palea Z: Nitzschia amphibia var rostrata AA: Nitzschia amphibia AB: Schistauron crucicula AC: Nitzschia gracilis Plate Five Appendix Three This section contains a complete list of species encountered in the modern samples collected from sediments, surface water and vegetation from 1998-2000. The sites are listed across the top of each table and the figure in brackets relates to the year of sampling. The figures are in percentages. SEDIMENT AC008D AC134A AD009A/XA009X/AC013A AC009A AC031A AC160A ACCX)8A/SS001 C AP001A AM009C AM006A XM007X XM009X XM006X AM008A AN009A XU009X AU003A XU008X BR010A BR010C/XB020C BR010D BR010B/XB010B C0009D C0001C/XCP001Z C0001B C0001A CY004A CM009M/EY01 OA CM004A CM023A EY018A CM110A DE002A DE001A/MUC005 DP010A DP007A DP001A DP065A/MUC007 DP061A EC001A/EY016A/EY011A/XE009X EU009C/XN009X EU017A EU009I/EU047A XS005X EU108A EU002A/EU009S PS001A SR001A/FR002A FR057A SS002A/FR001A FR060A FG(X)1U XS014X G0004A G0025B/G0014A G0025H/MUC009 GP009S GY005A MA002B XF009X MA001B/C/D XF007X NA066B NA008C/NA056A NA365A SP012S XH009X NA102A NA058A NA009P SL001A/NA014A FA001A/NA010A NA003A NA751A SP009H UN009S XN006X/NA650A XN005X/NA009S NA144A NE001A NI014A/XNA002Z XI009X NI065A NI008A NI017A NI009V XS007X NI009A NI005S XI008X PI009S/MUC013 PI008A PI005A SV009P RH001A SS001C NA001P NI011S XP014X SV020C XS006X ST009J SR103A SY003A SY001A XU001X SV005C XM008X Total Counted Species Achnanthes exigua var elliptica Achnanthes helvetica Achnanthes minutissima Achnanthes minutissima variety Achnanthes sp Achnanthes thermalis Achnanthes exigua Amphipleura pellucida Amphora clevi Amphora coffeaeformis Amphora sp Amphora sp 1 Amphora subturgida var thin Amphora thumensis Anomoeoneis sphaerophora Aulacoseira ambigua Aulacoseira granulata Aulacoseira sp Brachysira neoexilis Brachysira neoexilis var capitate Brachysira neoexilis var large capitate Brachysira neoexilis var small Cocconeis disculus Cocconeis placentula var incisa Cocconeis placentula var euglypta Cocconeis placentula var placentula Cyclotella stelligera Cymbella mesiana Cymbella microcephala Cymbella pusilla Cymbella turgida Cymbella turgidula Denticula elegans Denticula tenuis Diploneis finnica Diploneis oblongella Diploneis ovalis Diploneis parma Diploneis subovalis Encyonema carina Eunotia camelus Eunotia flexuosa Eunotia incisa Eunotia incisa variety Eunotia intermedia Eunotia solerolii Fragilaria brevistriata Fragilaria construens Fragilaria fasciculata Fragilaria pinnata Fragilaria tenera Fragilaria ulna Fragillaria constmens var javanica Gomphonema gracile Gomphonema intricatum Gomphonema intricatum var vibrio Gomphonema sp Gyrosigma acuminatum Mastogloia elliptica var dansei Mastogloia smithii var fine Mastogloia smithii var lacustris Mastogloia sp Navicula capitata var hungarica Navicula cuspidata Navicula florinae Navicula florinae variety Navicula heimansii Navicula laevissima Navicula phyllepta Navicula pseudosigma Navicula pupula Navicula pygmea Navicula radiosa Navicula radiosa var tenella Navicula sp Navicula sp Navicula stroemii Navicula subtillissima Navicula utermoehlii Neidium iridis Nitzschia amphibia Nitzschia amphibia var rostrata Nitzschia archibaldii Nitzschia frustulum Nitzschia gracilis Nitzschia levidensis var vidorae Nitzschia liebetruthii Nitzschia palea Nitzschia palea variety Nitzschia pellucida Pinnularia appendiculata Pinnularia divergens Pinnularia major Pinnularia side view (unid) Rhopalodia gibba Schistauron crudcula Species 1 Progresso Species 11 South Species 14 ahsm Sp 20 Chiwa (side view) Species 6 Stephanodiscus minultus Striatella unipunctata Synedra acus Synedra ulna unid side view unid side view (sp 5) Unid spedes Authority Hustedt (Hustedt) Lange-Bertalot and Krammer 1989 Kutzing 1833 (Rabenhorst) Schoenfeld 1907 Grunow in Cleve and Grunow 1880 Kutzing Grunow (Agardh) Kutzing 1844 / / / (A.Mayer) Cleve-Euler 1932 (Ehrenberg) Pfitzer 1871 (Grunow) Simonsen 1979 (Ehrenberg) Simonsen 1979 / Lange-Bertalot 1994 / / / (Schumann) Cleve in Cleve and Jentzsch 1882 Ehrenberg (Ehrenberg 1854) Grunow 1884 Ehrenberg 1838 Cleve and Grunow (in Van Heurck) 1882 Cholnoky 1955 Grunow in Van Heurck 1880 Grunow in A. Schmidt 1875 Gregory 1856 Grunow in A. Schmidt 1875 Kutzing 1844 Kutzing 1844 (Ehrenberg) Cleve 1891 (Naegeli) Cleve-Euler 1922 (Hilse) Cleve 1891 Cleve 1891 Cleve 1894 Lange-Bertalot and Krammer nov.spec. Ehrenberg (Brebisson) Kutzing 1849 Gregory 1854 / (Krasske ex Hustedt) Norpel and Lange-Bertalot 1991 (Kutzing) Rabenhorst 1864 Grunow in Van Heurck 1885 (Ehrenberg) Grunow 1862 (C.Agardh) Lange-Bertalot 1980 Ehrenberg 1843 (W.Smith) Lange-Bertalot 1980 (Nitzsch) Lange-Bertalot 1980 Hustedt 1942 Ehrenberg 1838 Kutzing 1844 (Ehrenberg) Cleve / (Kutzing) Rabenhorst 1853 (Thwaites) Cleve 1895 / Grunow 1878 / (Grunow) Ross 1947 (Kutzing) Kutzing 1844 Moller 1950 Van Dam and Kooyman 1982 Kutzing 1844 Kutzing 1844 Kutzing 1844 Kutzing 1849 Kutzing 1844 (Brebisson ex Kutzing) Van Heurck 1885 / / Hustedt 1931 Cleve 1891 Hustedt 1943 (Ehrenberg) Cleve 1894 Grunow 1862 Hustedt 1959 Lange-Bertalot 1980 (Kutzing) Grunow in Cleve and Grunow 1880 Hantzsch 1860 (Grunow) Cholnoky 1966 Rabenhorst 1864 (Kutzing) W.Smith 1856 / Grunow in Cleve and Grunow 1880 (Agardh) Cleve 1895 W.Smith 1853 (Kutzing) Rabenhorst 1853 / (Ehrenberg) O.Muller (Grunow ex Cleve) Ross / / (Kutzing) Cleve and Moller 1878 (Lyngbye) Agardh Kutzing 1844 (Nitzsch) Ehrenberg 1832 / Almond Hill (00) Almond F 0.00 0.00 4.88 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.35 0.00 3.83 0.00 0.00 16.72 2.79 2.09 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.35 0.00 0.00 0.00 0.00 0.00 0.00 2.44 0.00 0.00 0.00 0.00 9.06 0.35 0.00 0.70 0.00 0.00 0.00 0.00 0.00 0.70 1.74 0.35 0.00 0.00 4.88 0.00 0.00 0.00 0.00 11.85 9.41 9.76 10.45 0.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.70 0.00 0.00 0.00 0.00 0.35 0.00 0.00 2.79 2.44 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.35 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 287 I 2 (00) 0.00 0.00 26.04 0.00 0.00 0.00 0.26 0.00 0.00 0.00 2.86 0.00 0.00 0.00 0.00 0.00 0.00 0.78 2.34 3.91 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.04 0.00 0.00 0.00 0.00 31.77 0.00 0.00 2.34 0.00 5.21 0.52 0.26 0.00 0.00 2.34 0.00 0.00 0.00 0.00 0.26 0.26 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.52 1.04 0.00 0.00 0.00 0.00 2.86 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.69 0.00 2.34 0.00 0.00 0.00 0.00 0.00 0.00 0.00 384 Booth (98) 0.00 0.75 0.00 12.25 0.00 0.00 8.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.50 0.00 1.50 4.25 0.00 16.00 0.00 0.00 0.00 0.00 2.75 0.00 1.00 0.00 0.75 0.00 0.00 0.00 0.00 0.00 0.25 0.00 0.00 16.00 3.75 0.00 0.00 0.00 7.50 3.50 0.00 0.00 0.00 0.00 0.00 0.25 0.00 0.00 0.00 2.00 0.00 0.00 0.00 0.00 4.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.75 3.25 0.00 0.00 0.00 0.75 0.00 0.00 4.00 0.00 0.00 0.00 0.00 0.00 0.00 1.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.50 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 400 E-888888888888888888S88S88888888 S8S888S888888 8 SS88888S888S8 8 88K888888888888S8S888 8 88 8 88888888888888 8.8888 8 888885 Soddddo'dddooddoododNNdsddddddddo'drjddddo'oiDoddooNOoddo'do'dooddo'doddodddoddddio'uiododdoVdddodo'ododdddododddd £c£~8 8E888888888888888!qo8888888:S88 8.838888 8 8 8 8S!58838888888e888888S888888K88S8K888888!85S888 8 88888K88888888888888888oddddddddddddo'do'ddoo--dddo'dddoddo'dcDddddo'ddddddd-M:do'do'dddo'o'dddddgd--ddo'dddd---ddo'ddo'o'dd--ddddddddo'o'ddddddddddddddddd" COc II Q88S888S88888888§882S8ddddd 00000000 oindddo'1'sr8F:8K88o88888888888s£83!88888888K888888888o8888s888885!88888888888888888888S!f:8888!3!8;s?888888S8888888888888g8888888 £d-dciddrddddddddddd-ddNddddddddNddddddddc|j-dddddddddd-dddddddd-ddddddddddd--dddwnddddddddddddddddddddddddddddd< <0 t »- T- 88888888888o83888S!S8&8888888288r;8888Kr28SS888S88S8SS88888888888888:8888888888K88!f888888888888S8888888; Soodvaioritdddddo'dodddridiriddsudonddo'ddddrddiiiddodoiNddrdddddddo'doddddddodddddddNOo'o'do'o'ddo'dddoNddddddddddd >5moir S-88S888S!888888888888S888888888888a888888Si88888g!8Sa88888888888s8S8888888aSS8888S8S?S88888888888888888888888888; d^dddddy-ddddddddddddd^ddddddddddddddddddridd^-dd^-dddddddciddddy-ddddddddddddddddddddd^rridddddddddddddddddddddddddd'' ^ n cm £3 =-88888888888888888 8 8S8a88S"88§!8888888888888 8 888 8 8E8 8, 888™83888S88288888888E?888888E888888888888888888888888888-ddddddddddddddddddddt11 c1 o88S888?888888888888a8S88888S8888?888888Si8888888888a888888888888888aa88t88:8888888S8a888888888888888888888?888 Boob 00000000 00 dddddddddddddddodd do 000 000 Nodddoddd 0000000 00 000 00000 o •- 0 o odd d dddcMo'ddddddddo'dddddddddddd odd 00-00' "Ea>1®8 8. 8888888888888 8 888S8S8888 8.8$8888£888888S 8 8888 8 8888888888 8, 8, 88S>88888888888S8888§!8888888888;!i8888888888888888888t 2-dddddo o'dddddddddddtdcvi^d dodo 0000010000 00 ddddddddddddddddddddddNddddddddoddoidddd'-ddddddddddifld oddddddd 0000000000^ 0) cm co a)2 s883888888888888888S!88888888388883888888588888S888888858888888888888888S8S;88888888888888888888838888888888888§SddNddo'ddddddddddd-'-od'-dddo'dfflddddioddddddrddddddddddddddddddddoddddddddd'-dd^dddrddddddddddddddddioddddddddddddd" cd2 0888888888888888888888S888888"8888888S88888888"88888S888S8888S!88888888888S8888888E8888"888888888a888888888888E 2- o *- o *- co odd o* dddddoddod o'o'»-ddddddcodd dddddddo'v dodo dw>ddo'do'»-ddd^:o'r-' od--ddddd odd ddo'ddicidddo'di'ddddod cm 00" 00* o'o* odd 000' dddod o'do" 101 &S88^88w8888?®88S8wwS8w8888888w88w88?888cS8888. 8S8w88888®88888^°8?888?888S®S8888S8oio28ScM8888w8w8888888888888^88o 2-d 00* dddvoddd'-r odd o'd^-cd 0-00 o'dddddddddo'r^ddddddd 00 o'ddddd o'o'*-dddo'dcoco o o o o o o'o' o *- v o'o' dodduinoido'dddddd o'o' o'o o'ddodd 00000000'* f838a88?88888888888SS8S8883883838a88888888S8 888888 88888838888:8888888888832888288 SS8888S838888888SS888S8888888S o^oocmoooooooooooooooo)ocooooooooooo»-oooooooooooooooooooooo 00 odd*-ddddd 000 odddiodddciddNddddddddddddd ddood odd 000000^ .poppopoooor -ooooooooroooooivjooooooooooroooooo-'wooooooooc '88SS888888888288888888S82S8SS8S888SS8S88S tuooooooooooooc >ooooro-»oc JOOOOOOOOOOOO O CDOOOOOO O O WC >888888S88 8™838 888SS8 88iiiS2s8 8iS88888888$8888823;88-s PLANKTON SS001C/AC008A AD009A/XA009X/AC013A AC009A AC160A XU009X XU008X BR010A BR010C/XB020C BR010E XB010B CO001B CO001A EY010A CM004A DE002A DE001A EC001A/EY016A/EY011A/XE009X EU009I/EU047A XN008X FG001U GO004A GO025H MA002B MA001B/C/D NA008C XH009X NA123A NA003A NA751A UN009S XN005X NA144A NI014A XI009X NA267A NI017A NI009A NI009S PI009S PI008A PI005A RH001A SU024B SY001A Species Achnanthes exigua Achnanthes minutissima Achnanthes minutissima variety Achnanthes thermalis Aulacoseira ambigua Aulacoseira sp Brachysira neoexilis Brachysira neoexilis var capitate Brachysira neoexilis var large Brachysira neoexilis var small Cocconeis placentula var euglypta Cocconeis placentula var placentula Cymbella mesiana Cymbella microcephala Denticula elegans Denticula tenuis Encyonema carina Eunotia incisa Eunotia side view Fragilaria ulna Gomphonema gracile Gomphonema intricatum var vibrio Mastogloia elliptica var dansei Mastogloia smithii var lacustris Navicula cuspidata Navicula heimansii Navicula modica Navicula radiosa Navicula radiosa var tenella Navicula sp Navicula subtillissima Navicula utermoehlii Nitzschia amphibia Nitzschia amphibia var rostrata Nitzschia calida Nitzschia gracilis Nitzschia palea Nitzschia silqua Pinnularia appendiculata Pinnularia divergens Pinnularia major Rhopaiodia gibba Surirella capronii Synedra ulna Authority Aguacaliente Booth Botes Crooked Tree Grunow in Cleve and Grunow 1880 14.76 0.82 0.00 0.61 Kutzing 1833 0.00 0.00 0.00 0.00 0.00 4.95 0.00 2.45 (Rabenhorst) Schoenfeld 1907 0.00 0.00 0.00 0.00 (Grunow) Simonsen 1979 0.00 0.00 0.54 0.00 0.00 0.00 0.54 0.00 Lange-Bertalot 1994 0.00 29.40 3.26 65.03 0.56 0.00 0.00 0.00 0.00 10.71 0.00 1.84 0.00 0.00 0.82 0.00 (Ehrenberg 1854) Grunow 1884 0.00 0.00 0.00 0.00 Ehrenberg 1838 0.00 0.00 0.00 0.00 Cholnoky 1955 0.00 0.00 5.43 0.00 Grunow in Van Heurck 1880 0.00 4.95 12.23 0.00 Kutzing 1844 0.00 0.00 0.00 0.00 Kutzing 1844 0.00 2.20 0.00 1.23 Lange-Bertalot and Krammer nov.spe< 0.00 14.01 23.37 0.00 Gregory 1854 0.00 0.00 0.82 0.00 / 0.00 0.00 2.99 0.00 (Nitzsch) Lange-Bertalot 1980 0.00 0.00 0.00 3.07 Ehrenberg 1838 0.00 0.00 1.63 0.00 (Ehrenberg) Cleve 0.00 0.00 0.00 0.00 (Thwaites) Cleve 1895 0.00 0.00 0.00 0.00 Grunow 1878 0.00 4.40 15.22 0.00 (Kutzing) Kutzing 1844 2.79 0.27 0.00 0.61 Van Dam and Kooyman 1982 0.00 0.00 13.32 0.00 Husetedt 1945 0.00 0.00 0.00 0.00 Kutzing 1844 0.00 4.95 0.00 0.61 (Brebisson ex Kutzing) Van Heurck 16 0.56 9.34 0.00 3.68 / 7.52 0.00 0.00 0.00 Cleve 1891 0.00 0.00 10.33 0.00 Hustedt 1943 0.00 0.00 0.00 0.00 Grunow 1862 0.00 0.00 0.27 0.00 Hustedt 1959 0.00 0.00 0.00 0.00 Grunow in Cleve and Grunow 1880 8.36 0.00 0.00 0.00 Hantzsch 1860 0.00 3.30 0.00 5.52 (Kutzing) W.Smith 1856 56.27 2.20 0.00 13.50 Archibald 5.85 0.00 0.00 0.00 (Agardh) Cleve 1895 0.84 8.52 0.00 1.84 W.Smith 1853 0.00 0.00 7.61 0.00 (Kutzing) Rabenhorst 1853 0.00 0.00 1.36 0.00 (Ehrenberg) O.Muller 0.00 0.00 0.27 0.00 de Brebisson 2.51 0.00 0.00 0.00 (Nitzsch) Ehrenberg 1832 0.00 0.00 0.00 0.00 359 364 368 163 Doubloon Harry Jones Hillbank Centre Hillbank Far Irish Creek Lemonal Monkey Tail 0.00 0.51 1.03 0.00 0.35 0.36 0.54 16.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.65 2.58 1.22 12.89 15.41 1.09 0.00 0.00 0.00 0.00 0.00 0.00 17.39 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.22 38.52 16.28 3.67 15.33 7.17 22.83 19.11 0.51 37.73 51.99 10.10 11.47 5.43 0.00 13.27 12.40 0.00 5.92 4.30 19.57 4.44 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.36 1.09 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.33 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.78 0.00 0.70 1.08 3.80 0.44 0.26 0.52 3.98 0.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 24.44 15.31 11.89 8.26 8.71 7.53 8.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.34 2.07 2.45 3.48 13.26 4.35 2.22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 6.89 1.29 1.22 0.70 1.43 1.63 0.44 0.00 0.00 0.00 0.00 0.00 0.00 24.00 4.08 4.65 18.65 4.18 1.43 3.26 0.00 0.26 0.26 0.00 0.00 0.00 0.00 0.89 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.30 0.00 0.00 0.00 0.00 0.00 0.89 1.79 1.29 2.75 1.05 0.72 0.00 0.00 1.02 0.78 2.75 1.39 1.79 5.98 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.44 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3.36 0.00 2.79 0.36 0.54 0.00 0.00 2.33 1.22 4.88 0.00 1.63 0.89 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.44 2.04 0.26 0.00 3.83 19.35 1.63 0.44 1.28 0.52 1.83 23.00 13.98 0.54 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.33 0.00 0.00 0.00 0.00 0.00 0.00 225 392 387 327 287 279 184 Ramgoat Rio Bravo 0.00 0.00 0.00 0.00 8.66 14.14 0.00 0.00 0.00 0.00 0.00 0.00 16.27 7.07 19.42 18.94 1.84 1.26 0.00 0.00 0.00 6.82 0.00 6.06 0.00 0.00 1.05 12.63 1.31 0.00 0.00 0.00 34.65 22.47 0.00 0.00 0.00 0.00 1.84 0.51 0.00 2.27 2.89 2.27 0.00 0.00 2.89 1.26 0.00 0.00 0.00 0.00 0.00 0.00 1.57 0.51 4.46 1.26 0.00 0.00 0.00 0.00 0.00 0.00 0.26 0.00 0.00 0.00 0.00 0.00 0.79 0.00 2.10 2.53 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 381 396 EPIPHYTE AD009A/AC013A AC008A/SS001C XM007X XM006X XU009X XU008X BR010A BR010C BR010E XB010B XY009X CY004A XM009A EY010A CM004A CM023A DE002A DE001A/MUC005 DP007A DP061A EC001A XN009X EU017A EU047A XS005X XN008X EU002A SR001A/FR002A FR057A FR060A XS014X FU002A G0004A MA002B MA001A XF009X MA001B XN019X NA066B NA001P/NA022A XH009X NA102A SL001A/NA014A NA003A NA751A XO032X XN006X/NA650A XN005X NI014A/XNA002Z XI009X NI003A/XND001Z NI008A NI017A XS007X NI008N NI009A XI008X XI006X PI009S/MUC013 PI008A PI005A RH001A XS006X ST021A SY001A XU001X XV009X XM008X XS017X Species Achnanthes minutissima Achnathes exigua Amphora sp Amphora subturgida var thin Aulacoseira ambigua Aulacoseira sp Brachysira neoexilis Brachysira neoexilis var capitate Brachysira neoexilis var large Brachysira neoexilis var small Cyclotella sp Cyclotella stelligera Cymbella cesatii Cymbella mesiana Cymbella microcephala Cymbella pusilla Denticula elegans Denticula tenuis Diploneis oblongella Diploneis subovalis Encyonema carina Eunotia camelus Eunotia flexuosa Eunotia incisa Eunotia incisa variety Eunotia side view Eunotia solerolii Fragilaria construens Fragilaria tasciculata Fragilaria tenera Fragillaria construens var javanica Frustulia rhomboides Gomphonema gracile Mastogloia elliptica var dansei Mastogloia smithii Mastogloia smithii var fine Mastogloia smithii var lacustris Nacicula sp 15 ahsm Navicula capitata var hungarica Navicula halophila Navicula heimansii Navicula laevissima Navicula pupula Navicula radiosa Navicula radiosa var tenella Navicula sp 32 ahsm Navicula stroemii Navicula subtillissima Nitzschia amphibia Nitzschia amphibia var rostrata Nitzschia denticula Nitzschia frustulum Nitzschia gracilis Nitzschia liebetruthii Nitzschia nana Nitzschia palea Nitzschia pellucida Nitzschia pusilla Pinnularia appendiculata Pinnularia divergens Pinnularia major Rhopalodia gibba Species 6 Stephanodiscus minultus Synedra ulna unid side view unid side view unidentified sp unidentified sp Authority Kutzing 1833 Grunow in Cleve and Grunow 1880 Grunow) Simonsen 1979 Lange-Bertalot 1994 Cleve and Grunow (in Van Heurck) 1882 (Rabenhorst) Grunow 1881 Cholnoky 1955 Grunow in Van Heurck 1880 Grunow in A. Schmidt 1875 Kutzing 1844 Kutzing 1844 (Naegeli) Cleve-Euler 1922 Cleve 1894 Lange-Bertalot and Krammer nov.spec. Ehrenberg (Brebisson) Kutzing 1849 Gregory 1854 / / (Kutzing) Rabenhorst 1864 (Ehrenberg) Grunow 1862 (C.Agardh) Lange-Bertalot 1980 (W.Smith) Lange-Bertalot 1980 Hustedt 1942 (Ehrenberg) de Toni Ehrenberg 1838 (Thwaites) Cleve 1895 Thwaites 1856 / Grunow 1878 / (Grunow) Ross 1947 (Grunow) Cleve 1894 Van Dam and Kooyman 1982 Kutzing 1844 Kutzing 1844 Kutzing 1844 (Brebisson ex Kutzing) Van Heurck 1885 / Hustedt 1931 Cleve 1891 Grunow 1862 Hustedt 1959 Grunow in Van Heurck 1880-1885 (Kutzing) Grunow in Cleve and Grunow 186 Hantzsch 1860 Rabenhorst 1864 Grunow in Van Heurck 1881 (Kutzing) W.Smith 1856 (Agardh) Cleve 1895 Grunow 1862 emend. Lange-Bertalot 1976 (Agardh) Cleve 1895 W.Smith 1853 (Kutzing) Rabenhorst 1853 (Ehrenberg) O.Muller / (Kutzing) Cleve and Moller 1878 (Nitzsch) Ehrenberg 1832 / / / / Almond Hill (00) 1.74 0.00 1.52 0.00 0.00 0.00 3.70 2.61 0.00 0.43 1.30 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.22 0.00 0.00 0.00 0.00 0.00 0.87 0.00 0.00 0.00 71.52 1.09 0.00 0.00 2.17 0.00 0.00 0.43 1.09 2.39 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.52 0.00 0.00 0.00 0.22 0.00 0.00 0.00 2.39 0.00 0.00 0.00 1.74 0.00 0.00 0.00 0.00 0.22 0.00 0.00 0.00 0.65 0.00 0.43 460 -CDOOOO-^-^-LOO O 00 O O O O LO 1 O CM o o o o o •ddddddcbddodoo OOOCDOOOCOOOOO^OCOOOOOOOCDOOOO^OOCOOOLO o o o o ooo^rooooooo o oo o OOOOOO^tOOOOOOOOCMO^OCOOOOOOOO' O O O O h- O O O O LO o o o o o OOOOOOOOOOCMOOOOO — r>, o o o o o ■ O h- O O O O O I CM O C\J O CO O O CD O CD O •»- O O-LOOOOOOCOCOOOOOOOOOOOOO^-OOOOOOOOOOO CMOOO-M-OOOCDCOOOOOOOOCO0000^000 OOOOOOCD(OCOOOOOOOOOOOO(0-M-CMOOOOOdddddddddddodd •<3-00000000100000CMOOOOOOOOLodddddodd oooododo o o o o o (O CO O o o o o o CO o CO o CO o CM o o o o o o CO CM o o o o o Is- CM o CM o CM o LQ o o d o o o Is- d o o o o o o o o o o o LO o o o o o o o o CO o o o CM o o o o o o o o CD o o o o o o o o o o o o o o o o o (OOOOOOOOOOOOOOCD(OOOOOOOOON-OOOOCM o o o o o oooooooo^oooooooooooo o0 ^-OOOOOOT-g)oqqqqqc\jSodddddco CO T~ 1jco OOOOOOOOOCDOOOOOOOOOOOOOOOOOOOOOh-OOOOOOOOOOOOOO r- o o o o o o o o o o OOOOOOOLOCOOOOCMOOOCOT- OOOOCMOOOOOOOOOCD 00000000000001s-OOOOOOOOOOOOOOCDCMOOOOOOOOOOOOO 0,00000 < CO o o o o o LOOLOOOOOOOlOlOlOCMOCMOOOOOOCMh-CMcMd-r^dddddioddcM OOOOOOOOlOqqqqqqqoh-odddddddioddd OOOOOOOOOIOOOOOOOOOOOOOOOOOOIOOOOIOOOOOOOOOIOOOOOOCMOOOOIOOOOOOOOOOO --~~---~0010^0000010^000000000 CMOOOCMOOOOOT-T- 0000000 .OOOOOCDCMCOO-^-OOOKOOOOLnOLOOOOOOOOCOO'OLOoqqqqN-D>.qr^qqqqqqqqcMS-ddddddcocModddddoddddo ooo(b oqqoqqqh-O'dddddddcviodd CDOOOCOOOCDOOOCOCDOOOLOOOOOOOOOOOOLOOOOOCOCOOOO(OOOCDOOOh~O^OOCMOOLOOOOOOOOOC\JOOOOOOOOO^OOCDOOOOOOCM oooooooooooooooooooo o o o o ^ o - Is- O O O O O t 0 0)00 OOCMO^OOOOCOOOOCOOOCO^tOOOOOCM O T- O CD O O COOOOOOOOOh-OO oooooooo OOOOOOLOOO CMOOOCOOOOOOOOOOOOOOLOCOOOOOOOOOOOOOOOCOOOCO'ddcModddddddddi-do00 COOOOOLOCOCOOOOOOoo^ooqocM-^qqh-qqqqOcodddddcbNodddd OOCOOCMOCDOOOOOOO-^-LnOOCDOOlO-^lOOOOOOOOi-OOO o o o o o o odd r^oooD-ooo^tcMOdddddddd-^od h- o d OOCMOCMOOOOOh-OOOOOLOOOOlO OOOCMOOOOOOOOLOOOOOOr^O ooo-3-oooooooo o o o o o o o CM CD doddddddddddddcM Doubloon A (00) Doubloon B (00) Hillbank 1A (00) Hillbank 1B (00) Hillbank 1C (00) Hillbank 2A (00) Hillbank 2C (00) 0.24 1.96 17.69 3.30 12.36 18.51 16.52 0.00 0.00 0.00 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3.53 7.82 0.87 1.18 6.52 1.58 1.58 5.18 1.71 19.65 11.56 17.75 25.51 16.06 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.41 2.20 7.86 6.60 12.36 8.58 6.33 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3.06 2.20 0.44 0.00 0.45 0.23 1.13 0.00 0.00 0.00 0.47 0.00 0.00 0.00 2.82 2.20 0.00 0.00 0.00 0.00 0.00 3.53 5.62 0.00 1.65 2.70 0.00 0.68 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.49 0.00 0.24 0.22 0.45 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 22.12 27.14 30.79 28.54 15.96 18.06 20.59 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.45 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 000 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.24 6.11 1.18 3.15 2.71 8.60 0.00 0.24 0.22 0.00 0.00 0.00 0.23 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.73 0.00 0.00 0.00 0.00 0.00 26.59 19.07 9.39 29.48 4.27 14.67 20.59 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.20 2.18 3.30 2.70 3.16 2.04 0.00 0.73 0.00 0.00 0.00 0.00 0.00 0.00 0.24 0.22 0.00 0.22 0.00 0.00 4.00 2.44 0.00 1.65 0.90 0.45 0.45 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.90 0.00 1.88 1.71 0.22 5.42 0.67 1.35 0.23 0.24 1.71 0.00 0.00 0.22 0.00 0.45 0.00 0.49 0.00 0.71 0.22 0.00 0.23 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.98 0.00 0.00 0.00 0.00 0.00 15.76 9.29 0.00 0.00 0.22 0.00 0.23 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.47 3.18 0.44 1.18 0.00 0.23 0.00 0.00 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3.93 3.07 18.88 3.16 4.07 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3.29 4.16 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 425 409 458 424 445 443 442 OCOOOOO'sl-CMOOCMOOir>OOOLOOOOr--OOCM -OJOOOOOOCVJOCOO^J-CDOOOCDOOOO^ CM CM ci o o c\i o o o ooo'tfooocooooo^o CM O O O O O O O O O CM O O o o o o CMOOOOOOOOOOOOOOOOCOcoooenooooooooooooo^t6060^000000006000006^ 00 00 00 00 00 cdcd 00cm OO OO OO OO cm 00 00 coco 00 00 00 cocm 0 0 d 0 0 cm cd O O O O O co 0 0 co 0 0 0 cdin 0 0 0 0 0 0 0 0 0 ■m" o o o 00 o o o o o o o o o 0 0 0 0 0 0 0 co 0 0 0 0 0 0 0 0 0 cm o o o t— o o o o o o cm o o o en 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o o o o CT> o o o o o o o o o o 0 0 0 0 0 0 0 0 0000000000000^OOOOOOOOOOOOOT-6666066666666^ ^-.00000000000 ■ COCM .00000000 Tt CD oooooo-^-ooocoooooooo'^-oooloooor^oooocmqcmooo-^OOOOOOOr^opocmoooct)000000i^000k0000000cm0c)cd00c>0ir) cd _wwwOOOLOOOOOOOCOOOOOOOCOOOOOOOOOOOOOOOOOCMOOOOOO-tfOOOOOOCnOOOOO" -~~'6066666666666666 ' o I o i cd 00^000000000 .OOOOOCOOCDOr^OOOOOO^tOOOCMOOOOOOOOOOOOOOOCDOOOOOOCDOOOCDOOOOOOO o o o o o o CM £6 66666cowo OOO-^tOOLOOOOCT)0666066660 0000 0000000000000 ooooooooooooococooooooocMoqo-»-qqoqqoqcqqqqqqqqqqqqqqcM666666666666666666666666066666666- O co O O 0 0 0 0 en 0 cm O q 0 0 0 0 in 0 co O d d 0 d d in 0 LOOOOCOOOCMOOO^tOcnooo-si-qqqqqqcoqcbddddddddddr^dodoooooooor^-ooocoooo 00000000000000000^0000000ooooooom-^ooooocmooooooooddddddddodddddddddddddd o o o o o n- cm o o o o t- t~ D^ C) D^ C) co o o o o o o h- 0 co 0 0 0 00 0 0 0 O O 0 00 0 0 0 0 0 0 0 0 0 0 0 t— O O O cm 0 0 0 00 0 cm 0 0 0 co ■<3- co 0 0 0 0 0 0 0 0 q 0 co 0 0 0 0 0 cm O O 0 co 0 0 0 0 0 0 0 0 0 0 0 cd O O O q 0 0 0 co 0 cm q 0 0 0 cm 0 0 0 0 0 cm 0 0 co 0 encm 0 0 0 0 0 0 T~ O d 0 en 0 0 0 0 0 0 0 0 0 0 0 co O O O T~ 0 0 0 cm 0 O •*" 0 0 0 0 0 0 0 0 0 0 0 T_ 0 0 oooooor-oooocoOOOOOONOOOOr-00000060000^ ^ o o o o c\iddddddcddddoodddcMod CDOOCOOCOOCDOOOOOCOOOOCOOI0^-00000do O O O O O LD O o o o o o r- o OOOCDOCOOOOOOCOOOOOOOOOCMOOOOCDOOOO^CMOOOOOCMOOOOOOOOlDOOOOCO CM O O O CM O 1 ddddddcoodOLOOoocMooooooooooooooooooooooocMoooo \ (00) Laguna Verde (00) Outpost (00) Outpost B (00) Outpost C (00) Progresso (99) Progresso 2 (99) 15.72 44.58 18.64 24.19 28.43 0.00 0.00 0.00 0.00 0.00 0.22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.24 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 8.90 1.18 0.00 0.22 0.00 1.40 0.00 1.70 0.47 21.79 27.43 35.66 1.40 1.41 0.00 0.00 0.00 0.00 0.00 0.00 0.47 2.65 4.95 2.18 6.70 6.48 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.33 0.00 0.00 0.00 0.00 0.00 0.00 14.77 0.00 0.00 0.00 0.00 0.00 0.00 0.95 2.36 1.94 0.65 1.00 0.00 0.00 1.14 0.00 0.24 0.00 0.00 0.00 1.88 0.00 0.00 0.00 0.00 0.00 0.00 0.00 9.47 0.47 3.63 0.00 0.00 1.68 64.08 0.00 0.00 0.00 0.00 0.00 3.64 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 10.98 34.67 17.43 14.90 8.48 0.84 6.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.38 0.00 0.00 0.00 0.00 0.00 0.00 0.19 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 85.71 16.20 3.41 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.27 1.65 3.15 1.51 2.74 0.28 1.88 0.00 0.00 0.73 0.22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 3.79 0.24 18.16 12.10 2.49 0.56 0.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 11.55 1.65 0.00 0.43 1.75 0.00 0.00 0.00 0.00 0.24 0.00 0.00 0.00 0.00 0.19 0.00 0.00 0.00 0.00 0.00 0.00 3.03 0.47 0.48 0.00 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.92 0.47 0.00 0.22 0.00 0.00 0.00 0.76 0.00 5.33 5.40 6.23 0.00 0.00 1.33 0.00 4.84 3.67 3.74 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.28 0.00 0.24 0.00 0.43 0.00 0.28 0.00 0.57 0.00 0.00 0.22 0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.24 0.00 0.00 0.25 4.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.21 1.30 2.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5.90 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 528 424 413 463 401 357 426 Appendix Four: List of diatom samples collected from each lagoon visited (1998- 2000) Code Water Body Type Year Diatoms Present PS Aguacaliente Swamp S 1999 a/ PP Aguacaliente Swamp P 1999 V AY Almond Hill Lagoon S 2000 V AZ Almond Hill Lagoon s 2000 V BA Almond Hill Lagoon E 2000 V BB Almond Hill Lagoon S 2000 V BF Almond Hill Lagoon P 2000 X AS Bluewater River E 2000 V BP Bluewater River P 2000 X BQ Bluewater River S 2000 X OS Booth River S 1999 V OP Booth River P 1999 V XG Booth River s 1998 V X Botes Lagoon s 2000 V Y Botes Lagoon E 2000 V Z Botes Lagoon P 2000 V AA Botes Lagoon E 2000 V BK Caledonia, New River Lagoon P 2000 X LS Chiwa Lagoon S 1999 V LP Chiwa Lagoon P 1999 V LE Chiwa Lagoon E 1999 V BM Cobweb Swamp P 2000 X AL Cobweb Swamp E 2000 V AM Cobweb Swamp E 2000 A/ AN Cobweb Swamp S 2000 a/ OD Crooked Tree Lagoon P 1999 A/ AF Doubloon Lagoon E 2000 A/ AG Doubloon Lagoon E 2000 A/ AH Doubloon Lagoon S 2000 A/ AI Doubloon Lagoon St/E 2000 V AJ Doubloon Lagoon P 2000 V BE Fabers Lagoon S 1999 X XE Harry Jones Creek S 1998 V XEP Harry Jones Creek P 1998 V XF Harry Jones Creek side S 1998 a/ BI Hillbank P 1998 X XPP Hillbank centre P 1998 V A Hillbank 1 s 2000 A/ B Hillbank 1 E 2000 V C Hillbank 1 E 2000 V D Hillbank 1 E 2000 V F Hillbank 2 S 2000 V G Hillbank 2 E/S 2000 V H Hillbank 2 E 2000 V I Hillbank 2B S 2000 V J Hillbank 3 S 2000 V K Hillbank 3 E 2000 V L Hillbank 3 E 2000 V M Hillbank 3 E 2000 V BG Hillbank 3 P 2000 X XC Hillbank far side S 1998 V XDP Hillbank far side P 1998 V XD Hillbank this side s 1998 V OA Honey Camp Lagoon s 1999 V U Honey Camp Lagoon E 2000 V V Honey Camp Lagoon S 2000 V BJ Honey Camp Lagoon P 1999 X XH Irish Creek s 1998 A/ XHP Irish Creek p 1998 V BD Jones Lagoon s 1999 X BS Kates Lagoon s 1999 V BE Kates Lagoon E 1999 V OC Kates Lagoon s 1999 A/ AC Kates Lagoon s 2000 V AD Kates Lagoon s 2000 V AE Kates Lagoon E 2000 A/ AC Kates Lagoon s 2000 V AD Kates Lagoon s 2000 V AE Kates Lagoon E 2000 V BL Kates Lagoon P 1999 X AP Laguna Seca E 2000 V AQ Laguna Seca E 2000 V BN Laguna Seca S 2000 X BO Laguna Seca P 2000 X AU Laguna Verde E 2000 V AX Laguna Verde S 2000 V BR Laguna Verde S 2000 X BS Laguna Verde p 2000 X OE Lamanai s 1999 V BH Lamanai p 2000 X XP Lemonal Creek p 1998 V XA Monkey Tail River s 1998 A/ XI Monkey Tail River St 1998 A/ XAP Monkey Tail River p 1998 A1 BC Northern Lagoon S 1999 X 0 Outpost E 2000 V P Outpost S 2000 V Q Outpost E 2000 V R Outpost E 2000 V MS Progresso Lagoon S 1999 V ME Progresso Lagoon E 1999 V ON Progresso Lagoon E 1999 V XRP Ramgoat Creek P 1998 V XB Rio Bravo S 1998 V XBP Rio Bravo P 1998 V OF Small Croc Lagoon s 1999 V FP Small Croc Lagoon p 1999 a/ OJ Southern Lagoon s 1999 a/ OK Wagner Lagoon s 1999 X E: Epiphyte S: Sediment P: Plankton St: Stone scraping V: Full Diatom Count X: No Diatoms The Modern data set (Figure 5.3 and 5.4) Species NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.7916 0.606 0.4988 0.295 1 AC00 8A 1.8429 1.5619 0.4001 0.6639 116 7.58 2 ACOO 9A 1.5421 1.7047 2.2212 1.1612 450 12.57 3 AC031A 0.7432 1.7398 0.2473 1.5258 47 1.77 4 AC134A 2.817 1.631 1.165 0.3911 34 3.66 5 AC16OA 3.4951 2.3254 4.6941 0.5718 33 1.06 6 ADOO 9A 4.2695 2.1839 3.5533 1.4773 1772 17.85 7 AMOO 6A -0.4581 1.6581 2.261 1.4296 10 1 8 ANOO 9A 5.0024 -1.1694 1.973 4.2024 15 1.72 9 APOO1A 0.5519 1.707 0.3598 1.5406 18 1 10 AUOO3A 2.8981 1.4652 1.0466 1.7439 73 2.86 11 BR01 OA 3.9243 1.4656 1.9238 0.8395 1516 26.26 12 BR01 OB 2.7767 1.4637 1.5292 0.2284 516 5.61 13 BR01 OC 3.8505 1.5653 2.485 0.015 1610 19.21 14 BR01 OE 4.4215 2.0852 3.1076 1.3281 1825 25.56 15 CMOO 4A 2.4864 1.3708 4.5697 2.7377 424 10.24 16 CM02 3A 4.3948 1.4077 0.6533 3.0271 48 4.24 17 CM11 OA 2.2106 1.3181 5.0446 1.5076 74 1.65 18 COOO 1B -0.0343 1.6552 3.1206 1.4579 181 2.78 19 COOO 1C -0.4581 1.6581 2.261 1.4296 17 1 20 COOO 9D -0.4581 1.6581 2.261 1.4296 11 1 21 CYOO4A 3.5384 1.6165 2.3091 0.3177 41 2.92 22 DEOO 1A 4.7157 0.7566 1.7999 1.3576 27 2.88 23 DEOO2A 4.6677 -0.2755 1.6164 -0.5642 1461 11.61 24 DPOO1A 2.0526 1.5516 0.3771 1.4214 59 2.94 25 DPOO7A 4.6801 0.6431 1.1652 0.8586 66 4.07 26 DP065A -0.4492 1.644 2.2093 1.4263 31 1.07 27 ECOO1A 3.5152 1.4806 2.2891 1.8523 4855 41.38 28 EUOO2A 4.2194 0.5367 1.2457 2.5428 127 5.01 29 EUOO9C 2.7322 1.5925 1.2693 1.0347 27 2.69 30 EUOO 9D 3.3331 1.5665 1.6749 2.8262 38 2.24 31 EUOO 91 3.397 1.1523 1.0015 0.8491 11 1.75 32 EUOO 9S 1.6634 1.5906 0.5196 0.5788 32 4.74 33 EU01 7A 4.7161 0.6096 1.3436 3.9307 30 1.23 34 EU04 7A 4.7131 0.1594 1.9244 3.7093 25 5.17 35 EU10 8A 1.3562 1.6835 0.9649 1.3434 40 1.6 36 EY01 OA 3.7257 1.2112 4.1887 2.3927 538 12.4 37 EY01 8A 2.9023 1.6252 1.2999 0.3045 19 3.44 38 FGOO 1U 3.2139 1.9014 1.3551 0.6976 124 7.19 39 FR057A 5.6078 0.9255 2.9844 0.8244 965 3.9 40 FR06OA 4.7098 1.8935 1.4687 0.0981 71 3.98 41 FUOO2A 3.8162 1.8666 3.6127 3.2253 13 1 42 GOOO 4A 4.2295 1.8889 2.9322 2.8091 434 27.51 43 GO02 5B 1.2475 1.04 5.2689 1.1205 12 1 44 GO02 5H 2.8781 1.8074 1.4683 0.6672 132 8.17 45 GPOO 9S 1.2917 1.0606 5.2518 1.149 36 1.12 46 GYOO 5A 1.5474 1.8001 0.0994 1.4986 31 1.21 47 MAOO 1A 4.5321 1.2311 0.6839 3.4455 21 1.21 48 MA00 1B 4.3222 1.0002 1.6665 2.7151 3366 32.51 49 MAOO 2B 5.0375 -0.7369 2.4367 3.8869 65 3.09 50 NAOO1P -0.4581 1.6581 2.261 1.4296 17 1 51 NAOO3A 4.1812 0.8787 1.8281 0.1031 439 25.21 52 NAOO 8C 1.287 1.1042 5.2296 1.1355 216 1.14 53 NAOO 9P 3.022 1.4214 0.835 2.3698 21 1 54 NAOO 9S 3.5362 1.4772 2.3911 -0.1877 50 3.2 55 NA123A 3.7333 2.0631 2.4008 1.2952 9 1 56 NA144A 3.4099 1.879 1.6055 0.2352 37 3.18 57 NA267A 1.8911 2.2038 -0.473 1.6004 30 1 58 NA365A -0.4581 1.6581 2.261 1.4296 82 1 59 NA751A 2.0049 1.6581 2.4795 0.6537 304 15.7 60 NEOO1A 4.7009 0.291 1.0593 -0.0603 28 1.72 61 NIOO 3A 5.3407 -2.2509 2.2653 -0.2026 31 1 62 NIOO 5S 4.331 0.5366 1.7881 -0.7567 26 1 63 NIOO 8A 4.4889 0.5426 1.815 0.4407 59 5.72 64 NIOO 8N 3.8162 1.8666 3.6127 3.2253 37 1 65 NIOO 9A 2.1232 1.7861 -0.0947 1.9423 743 9.11 66 NIOO 9S 1.8911 2.2038 -0.473 1.6004 21 1 67 NIOO 9V 1.9213 1.8104 0.0786 1.494 13 1 68 NI01 1S -0.4581 1.6581 2.261 1.4296 9 1 69 NI01 4A 3.8813 1.2887 2.8842 0.7949 392 21.93 70 NI01 7A 3.7674 1.7397 1.2058 1.7231 285 7.52 71 NI06 5A -0.4581 1.6581 2.261 1.4296 8 1 72 PIOO 8A 4.3852 0.9193 2.0436 3.3304 38 1.77 73 PIOO 9S 3.6371 1.6257 1.149 1.696 92 3.46 74 PSOO 1A 0.1719 4.9598 2.2286 0.3879 11 1.42 75 RHOO 1A 4.6641 0.5132 1.482 2.9006 35 1.88 76 SLOO 1A 4.5301 0.3675 1.7895 -0.0485 73 6.46 77 SPOO 9H 1.9213 1.8104 0.0786 1.494 25 1 78 SP01 2S -0.4581 1.6581 2.261 1.4296 75 1 79 SROO1A 4.9041 -0.1278 2.514 2.6488 44 1.39 80 SR10 3A 3.022 1.4214 0.835 2.3698 9 1 81 SSOO1C 2.0275 1.9617 -0.1782 1.4422 134 2.4 82 STOO 9J 3.022 1.4214 0.835 2.3698 9 1 83 ST02 1A 4.5091 2.4451 3.9365 3.4835 46 1 84 SU024B 1.8911 2.2038 -0.473 1.6004 9 1 85 SVOO 9P 0.6616 1.2978 0.9819 1.4859 24 1.71 86 SV02OC 3.1352 1.365 0.8733 2.0496 16 1.75 87 SYOO1A 4.5596 1.928 3.4741 0.9046 233 6.04 88 SYOO3A 2.847 1.6718 1.7998 0.7337 19 1 89 UNOO 9S 1.8911 2.2038 -0.473 1.6004 27 1 90 XAOO 9X 4.9158 4.9706 2.2654 1.1173 93 1.19 91 XFOO 7X 5.1853 -1.7991 2.7368 4.4712 16 1 92 XFOO 8X 5.1853 -1.7991 2.7368 4.4712 14 1 93 XFOO 9X 4.868 -0.2149 1.3519 3.8131 69 3.45 94 XHOO9X 4.4855 1.4142 2.366 2.6259 386 14.13 95 XIOO 6X 6.2057 3.5593 2.7244 2.1689 8 1 96 XIOO 9X 4.8421 0.1625 1.8355 0.6357 502 7.32 97 XMOO 6X 4.6609 0.2776 0.9539 2.9681 10 1.52 98 XMOO 7X 5.3726 4.9367 2.4177 0.7959 18 1.91 99 XMOO 8X 4.342 1.4693 0.6043 1.9467 66 2.55 100 XMOO 9A 4.7057 1.1969 1.9687 0.6296 115 1.92 101 XNOO5X 4.4365 1.1822 1.4867 3.0212 294 13.58 102 XN00 8X 4.4478 1.6849 3.0205 2.377 25 2.73 103 XNOO9X 4.872 0.2374 1.298 3.8063 13 1.37 104 XN01 9X 5.8214 4.8525 2.5527 1.3582 16 1.86 105 XP01 4X 4.8433 4.9818 2.2213 0.2527 18 1 106 XSOO 5X 4.9002 4.9751 2.231 0.2695 129 1.12 107 XSOO 6X 5.027 4.9591 2.2437 0.2894 11 1.46 108 XSOO 7X 5.521 4.9052 2.3642 0.5758 28 2.82 109 XS01 5X 4.8356 0.0285 1.5438 0.1156 72 4.39 110 XS01 6X 4.9663 4.9712 2.2673 0.3518 11 1.2 111 XS01 7X 5.4664 4.7228 2.1261 0.0552 14 2.09 112 XUOO 8X 4.8325 0.1352 1.7614 3.9386 222 3.88 113 XVOO 9X 4.5521 2.4571 4.2845 1.4224 35 1.85 s NAME AX1 AX2 AX3 AX4 WEIGH1 N2 EIG 0.7916 0.606 0.4988 0.295 1 BS 3.9676 0.8675 2.0949 0.9149 407 4.2 2 BE 3.7341 1.426 2.2104 1.6446 402 2.13 3 PP 2.0476 1.8832 0 1.728 359 2.85 4 LS 3.2748 1.3319 1.5906 1.7607 380 8.21 5 LE 3.6208 1.4913 2.5434 2.0894 385 7.44 6 OS 3.6438 1.2256 1.8907 1.5807 371 11.07 7 OP 3.5076 1.5269 2.1383 1.3157 364 6.9 8 ME 5.3332 0.9586 2.7504 0.88 357 1.35 9 ON 4.7343 0 2.026 0 426 2.24 10 OJ 0 1.6732 2.299 1.4531 336 7.24 11 OE 4.0515 0.6933 1.8775 0.4998 265 5.34 12 OD 3.509 1.5532 1.6349 1.066 163 2.23 13 XA 3.057 1.3657 3.646 1.8246 394 6.23 14 XB 1.7235 1.5972 2.0852 1.377 364 6.83 15 XC 3.8814 1.0148 1.9728 0.5774 401 5.79 16 XD 3.5306 1.4867 2.2769 0.5521 410 4.86 17 XE 3.1182 1.463 1.7388 0.8562 396 4.49 18 XF 3.1572 1.4772 1.8239 0.9791 377 9.4 19 XG 2.7217 1.5276 1.7043 1.0638 400 10.83 20 XH 2.2369 1.7116 0.8218 1.3562 379 11.67 21 XI 1.946 1.1841 4.6491 1.4116 403 3.54 22 XDP 3.8173 1.3733 2.157 0.7813 327 3.15 23 XHP 3.116 1.6152 1.649 1.2508 287 8.17 24 XAP 3.6078 1.7732 2.7889 1.075 184 7.13 25 XBP 2.9277 1.5526 2.569 1.3915 396 6.06 26 XP 3.0828 1.6977 1.639 1.2366 279 7.99 27 XPP 3.765 1.5803 2.312 0.79 387 4.9 28 XEP 3.5983 1.6044 2.0673 1.1661 392 4.87 29 XRP 3.4132 1.5106 2.1774 1.1253 381 5.07 30 A 4.3048 1.0384 2.1226 1.575 395 7.78 31 B 4.0407 1.7186 2.7102 1.6545 458 5.39 32 C 4.0504 1.3905 2.2069 1.8944 423 5.15 33 D 4.1726 1.6966 2.7867 1.2449 444 7.55 34 F 4.2154 1.0223 2.1292 0.9824 408 8.55 35 G 4.1496 1.712 2.6886 1.6286 439 5.99 36 H 4.1269 1.6254 2.6255 1.792 442 6.59 37 I 4.404 0.3931 1.8159 0.0713 397 3.12 38 J 4.5556 0.4834 1.799 1.4141 389 4.33 39 K 4.1136 1.5133 2.4601 1.8052 435 5.62 40 L 4.1225 1.7443 2.7484 1.0603 416 4.75 41 M 4.0679 1.6334 2.6077 1.7848 416 5.58 42 O 4.1854 1.5141 2.6563 1.6188 412 6.45 43 P 4.4487 0.7789 2.0217 1.8314 399 4.7 44 Q 4.1746 1.7037 2.799 1.5062 463 5.56 45 R 4.2289 1.8406 3.01 1.3735 401 4.41 46 U 4.136 1.0682 1.7825 2.4262 438 3.5 47 V 4.228 1.0873 2.0418 1.2906 372 14.17 48 X 4.3486 0.9093 1.826 2.6921 400 2.44 49 Y 3.7377 1.3291 2.8329 2.2293 403 6.77 50 Z 3.8981 1.2851 2.5049 2.4791 361 7.28 51 AA 4.2845 0.9926 2.1088 2.7461 395 8.58 52 AC 4.2497 0.9101 2.0762 1.8284 388 4.93 53 AD 3.9345 1.019 1.9662 1.3262 420 5.42 54 AE 3.52 1.4536 1.9578 1.8225 407 2.73 55 AF 4.0954 1.7424 2.8473 2.0017 357 5.51 56 AG 3.9284 1.3352 1.7696 1.8972 418 6.18 57 AH 4.0097 1.2134 1.9561 1.7447 435 6.07 58 Al 3.9506 1.2704 1.8642 1.747 402 7.17 59 AJ 4.0966 1.5741 2.5256 1.7947 225 5.46 60 AL 3.7919 1.5281 2.289 1.899 429 3.16 61 AM 3.7011 1.4807 1.9209 1.9603 433 4.64 62 AN 4.1661 1.3057 2.1031 1.9253 422 6.37 63 AP 4.1939 1.7214 2.7664 1.2454 445 3.74 64 AQ 4.2639 1.3361 2.3182 1.2943 528 10.31 65 AS 4.0729 1.2431 2.2048 0.8862 372 9.14 66 AU 3.9891 1.8401 3.0188 1.5757 423 3.05 67 AX 3.6991 1.3155 2.4773 1.5034 395 3.58 68 AY 4.6085 0.3231 2.1778 3.0674 272 10.94 69 AZ 4.7736 4.1841 2.3807 0.7673 363 5.49 70 BA 5.4036 1.4542 2.8567 0.9444 444 1.8 71 BB 4.865 1.3725 2.597 1.0277 392 4 The Modern Data set: Habitat (Figure 5.5) Species NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.7916 0.606 0.4988 0.295 1 AC00 8A 1.8429 1.5619 0.4001 0.6639 116 7.58 2 ACOO 9A 1.5421 1.7047 2.2212 1.1612 450 12.57 3 AC03 1A 0.7432 1.7398 0.2473 1.5258 47 1.77 4 AC134A 2.817 1.631 1.165 0.3911 34 3.66 5 AC16 0A 3.4951 2.3254 4.6941 0.5718 33 1.06 6 ADOO 9A 4.2695 2.1839 3.5533 1.4773 1772 17.85 7 AMOO 6A -0.4581 1.6581 2.261 1.4296 10 1 8 ANOO 9A 5.0024 -1.1694 1.973 4.2024 15 1.72 9 APOO1A 0.5519 1.707 0.3598 1.5406 18 1 10 AUOO3A 2.8981 1.4652 1.0466 1.7439 73 2.86 11 BR01 OA 3.9243 1.4656 1.9238 0.8395 1516 26.26 12 BR01 OB 2.7767 1.4637 1.5292 0.2284 516 5.61 13 BR01 OC 3.8505 1.5653 2.485 0.015 1610 19.21 14 BR01 OE 4.4215 2.0852 3.1076 1.3281 1825 25.56 15 CMOO 4A 2.4864 1.3708 4.5697 2.7377 424 10.24 16 CM02 3A 4.3948 1.4077 0.6533 3.0271 48 4.24 17 CM11 OA 2.2106 1.3181 5.0446 1.5076 74 1.65 18 COOO 1B -0.0343 1.6552 3.1206 1.4579 181 2.78 19 COOO 1C -0.4581 1.6581 2.261 1.4296 17 1 20 COOO 9D -0.4581 1.6581 2.261 1.4296 11 1 21 CYOO4A 3.5384 1.6165 2.3091 0.3177 41 2.92 22 DEOO1A 4.7157 0.7566 1.7999 1.3576 27 2.88 23 DEOO2A 4.6677 -0.2755 1.6164 -0.5642 1461 11.61 24 DPOO 1A 2.0526 1.5516 0.3771 1.4214 59 2.94 25 DPOO7A 4.6801 0.6431 1.1652 0.8586 66 4.07 26 DP065A -0.4492 1.644 2.2093 1.4263 31 1.07 27 ECOO1A 3.5152 1.4806 2.2891 1.8523 4855 41.38 28 EUOO2A 4.2194 0.5367 1.2457 2.5428 127 5.01 29 EUOO9C 2.7322 1.5925 1.2693 1.0347 27 2.69 30 EUOO 9D 3.3331 1.5665 1.6749 2.8262 38 2.24 31 EUOO 91 3.397 1.1523 1.0015 0.8491 11 1.75 32 EUOO 9S 1.6634 1.5906 0.5196 0.5788 32 4.74 33 EU01 7A 4.7161 0.6096 1.3436 3.9307 30 1.23 34 EU047A 4.7131 0.1594 1.9244 3.7093 25 5.17 35 EU10 8A 1.3562 1.6835 0.9649 1.3434 40 1.6 36 EY01 OA 3.7257 1.2112 4.1887 2.3927 538 12.4 37 EY01 8A 2.9023 1.6252 1.2999 0.3045 19 3.44 38 FGOO 1U 3.2139 1.9014 1.3551 0.6976 124 7.19 39 FR057A 5.6078 0.9255 2.9844 0.8244 965 3.9 40 FR06OA 4.7098 1.8935 1.4687 0.0981 71 3.98 41 FUOO2A 3.8162 1.8666 3.6127 3.2253 13 1 42 GOOO 4A 4.2295 1.8889 2.9322 2.8091 434 27.51 43 GO02 5B 1.2475 1.04 5.2689 1.1205 12 1 44 GO02 5H 2.8781 1.8074 1.4683 0.6672 132 8.17 45 GPOO 9S 1.2917 1.0606 5.2518 1.149 36 1.12 46 GYOO 5A 1.5474 1.8001 0.0994 1.4986 31 1.21 47 MAOO 1A 4.5321 1.2311 0.6839 3.4455 21 1.21 48 MA00 1B 4.3222 1.0002 1.6665 2.7151 3366 32.51 49 MAOO 2B 5.0375 -0.7369 2.4367 3.8869 65 3.09 50 NAOO 1P -0.4581 1.6581 2.261 1.4296 17 1 51 NAOO 3A 4.1812 0.8787 1.8281 0.1031 439 25.21 52 NAOO 8C 1.287 1.1042 5.2296 1.1355 216 1.14 53 NAOO 9P 3.022 1.4214 0.835 2.3698 21 1 54 NAOO 9S 3.5362 1.4772 2.3911 -0.1877 50 3.2 55 NA123A 3.7333 2.0631 2.4008 1.2952 9 1 56 NA144A 3.4099 1.879 1.6055 0.2352 37 3.18 57 NA267A 1.8911 2.2038 -0.473 1.6004 30 1 58 NA36 5A -0.4581 1.6581 2.261 1.4296 82 1 59 NA75 1A 2.0049 1.6581 2.4795 0.6537 304 15.7 60 NEOO 1A 4.7009 0.291 1.0593 -0.0603 28 1.72 61 NIOO 3A 5.3407 -2.2509 2.2653 -0.2026 31 1 62 NIOO 5S 4.331 0.5366 1.7881 -0.7567 26 1 63 NIOO 8A 4.4889 0.5426 1.815 0.4407 59 5.72 64 NIOO 8N 3.8162 1.8666 3.6127 3.2253 37 1 65 NIOO 9A 2.1232 1.7861 -0.0947 1.9423 743 9.11 66 NIOO 9S 1.8911 2.2038 -0.473 1.6004 21 1 67 NIOO 9V 1.9213 1.8104 0.0786 1.494 13 1 68 NI01 1S -0.4581 1.6581 2.261 1.4296 9 1 69 NI01 4A 3.8813 1.2887 2.8842 0.7949 392 21.93 70 NI01 7A 3.7674 1.7397 1.2058 1.7231 285 7.52 71 NI06 5A -0.4581 1.6581 2.261 1.4296 8 1 72 PIOO 8A 4.3852 0.9193 2.0436 3.3304 38 1.77 73 PIOO 9S 3.6371 1.6257 1.149 1.696 92 3.46 74 PSOO 1A 0.1719 4.9598 2.2286 0.3879 11 1.42 75 RHOO 1A 4.6641 0.5132 1.482 2.9006 35 1.88 76 SLOO 1A 4.5301 0.3675 1.7895 -0.0485 73 6.46 77 SPOO 9H 1.9213 1.8104 0.0786 1.494 25 1 78 SP01 2S -0.4581 1.6581 2.261 1.4296 75 1 79 SROO 1A 4.9041 -0.1278 2.514 2.6488 44 1.39 00o SR10 3A 3.022 1.4214 0.835 2.3698 9 1 81 SSOO1C 2.0275 1.9617 -0.1782 1.4422 134 2.4 82 STOO 9J 3.022 1.4214 0.835 2.3698 9 1 83 ST02 1A 4.5091 2.4451 3.9365 3.4835 46 1 84 SU024B 1.8911 2.2038 -0.473 1.6004 9 1 85 SVOO 9P 0.6616 1.2978 0.9819 1.4859 24 1.71 86 SV02OC 3.1352 1.365 0.8733 2.0496 16 1.75 87 SYOO 1A 4.5596 1.928 3.4741 0.9046 233 6.04 88 SYOO 3A 2.847 1.6718 1.7998 0.7337 19 1 89 UNOO 9S 1.8911 2.2038 -0.473 1.6004 27 1 90 XAOO 9X 4.9158 4.9706 2.2654 1.1173 93 1.19 91 XFOO 7X 5.1853 -1.7991 2.7368 4.4712 16 1 92 XFOO 8X 5.1853 -1.7991 2.7368 4.4712 14 1 93 XFOO 9X 4.868 -0.2149 1.3519 3.8131 69 3.45 94 XHOO 9X 4.4855 1.4142 2.366 2.6259 386 14.13 95 XIOO 6X 6.2057 3.5593 2.7244 2.1689 8 1 96 XIOO 9X 4.8421 0.1625 1.8355 0.6357 502 7.32 97 XMOO 6X 4.6609 0.2776 0.9539 2.9681 10 1.52 98 XMOO 7X 5.3726 4.9367 2.4177 0.7959 18 1.91 99 XMOO 8X 4.342 1.4693 0.6043 1.9467 66 2.55 100 XMOO 9A 4.7057 1.1969 1.9687 0.6296 115 1.92 101 XNOO5X 4.4365 1.1822 1.4867 3.0212 294 13.58 102 XN008X 4.4478 1.6849 3.0205 2.377 25 2.73 103 XNOO 9X 4.872 0.2374 1.298 3.8063 13 1.37 104 XN01 9X 5.8214 4.8525 2.5527 1.3582 16 1.86 105 XP01 4X 4.8433 4.9818 2.2213 0.2527 18 1 106 XSOO 5X 4.9002 4.9751 2.231 0.2695 129 1.12 107 XSOO 6X 5.027 4.9591 2.2437 0.2894 11 1.46 108 XSOO 7X 5.521 4.9052 2.3642 0.5758 28 2.82 109 XS01 5X 4.8356 0.0285 1.5438 0.1156 72 4.39 110 XS01 6X 4.9663 4.9712 2.2673 0.3518 11 1.2 111 XS01 7X 5.4664 4.7228 2.1261 0.0552 14 2.09 112 XUOO 8X 4.8325 0.1352 1.7614 3.9386 222 3.88 113 XVOO 9X 4.5521 2.4571 4.2845 1.4224 35 1.85 Samples N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.7916 0.606 0.4988 0.295 1 BS 3.9676 0.8675 2.0949 0.9149 407 4.2 2 BE 3.7341 1.426 2.2104 1.6446 402 2.13 3 PP 2.0476 1.8832 0 1.728 359 2.85 4 LS 3.2748 1.3319 1.5906 1.7607 380 8.21 5 LE 3.6208 1.4913 2.5434 2.0894 385 7.44 6 OS 3.6438 1.2256 1.8907 1.5807 371 11.07 7 OP 3.5076 1.5269 2.1383 1.3157 364 6.9 8 ME 5.3332 0.9586 2.7504 0.88 357 1.35 9 ON 4.7343 0 2.026 0 426 2.24 10 OJ 0 1.6732 2.299 1.4531 336 7.24 11 OE 4.0515 0.6933 1.8775 0.4998 265 5.34 12 OD 3.509 1.5532 1.6349 1.066 163 2.23 13 XA 3.057 1.3657 3.646 1.8246 394 6.23 14 XB 1.7235 1.5972 2.0852 1.377 364 6.83 15 XC 3.8814 1.0148 1.9728 0.5774 401 5.79 16 XD 3.5306 1.4867 2.2769 0.5521 410 4.86 17 XE 3.1182 1.463 1.7388 0.8562 396 4.49 18 XF 3.1572 1.4772 1.8239 0.9791 377 9.4 19 XG 2.7217 1.5276 1.7043 1.0638 400 10.83 20 XH 2.2369 1.7116 0.8218 1.3562 379 11.67 21 XI 1.946 1.1841 4.6491 1.4116 403 3.54 22 XDP 3.8173 1.3733 2.157 0.7813 327 3.15 23 XHP 3.116 1.6152 1.649 1.2508 287 8.17 24 XAP 3.6078 1.7732 2.7889 1.075 184 7.13 25 XBP 2.9277 1.5526 2.569 1.3915 396 6.06 26 XP 3.0828 1.6977 1.639 1.2366 279 7.99 27 XPP 3.765 1.5803 2.312 0.79 387 4.9 28 XEP 3.5983 1.6044 2.0673 1.1661 392 4.87 29 XRP 3.4132 1.5106 2.1774 1.1253 381 5.07 30 A 4.3048 1.0384 2.1226 1.575 395 7.78 31 B 4.0407 1.7186 2.7102 1.6545 458 5.39 32 C 4.0504 1.3905 2.2069 1.8944 423 5.15 33 D 4.1726 1.6966 2.7867 1.2449 444 7.55 34 F 4.2154 1.0223 2.1292 0.9824 408 8.55 35 G 4.1496 1.712 2.6886 1.6286 439 5.99 36 H 4.1269 1.6254 2.6255 1.792 442 6.59 37 I 4.404 0.3931 1.8159 0.0713 397 3.12 38 J 4.5556 0.4834 1.799 1.4141 389 4.33 39 K 4.1136 1.5133 2.4601 1.8052 435 5.62 40 L 4.1225 1.7443 2.7484 1.0603 416 4.75 41 M 4.0679 1.6334 2.6077 1.7848 416 5.58 42 0 4.1854 1.5141 2.6563 1.6188 412 6.45 43 P 4.4487 0.7789 2.0217 1.8314 399 4.7 44 Q 4.1746 1.7037 2.799 1.5062 463 5.56 45 R 4.2289 1.8406 3.01 1.3735 401 4.41 46 U 4.136 1.0682 1.7825 2.4262 438 3.5 47 V 4.228 1.0873 2.0418 1.2906 372 14.17 48 X 4.3486 0.9093 1.826 2.6921 400 2.44 49 Y 3.7377 1.3291 2.8329 2.2293 403 6.77 50 Z 3.8981 1.2851 2.5049 2.4791 361 7.28 51 AA 4.2845 0.9926 2.1088 2.7461 395 8.58 52 AC 4.2497 0.9101 2.0762 1.8284 388 4.93 53 AD 3.9345 1.019 1.9662 1.3262 420 5.42 54 AE 3.52 1.4536 1.9578 1.8225 407 2.73 55 AF 4.0954 1.7424 2.8473 2.0017 357 5.51 56 AG 3.9284 1.3352 1.7696 1.8972 418 6.18 57 AH 4.0097 1.2134 1.9561 1.7447 435 6.07 58 Al 3.9506 1.2704 1.8642 1.747 402 7.17 59 AJ 4.0966 1.5741 2.5256 1.7947 225 5.46 60 AL 3.7919 1.5281 2.289 1.899 429 3.16 61 AM 3.7011 1.4807 1.9209 1.9603 433 4.64 62 AN 4.1661 1.3057 2.1031 1.9253 422 6.37 63 AP 4.1939 1.7214 2.7664 1.2454 445 3.74 64 AQ 4.2639 1.3361 2.3182 1.2943 528 10.31 65 AS 4.0729 1.2431 2.2048 0.8862 372 9.14 66 AU 3.9891 1.8401 3.0188 1.5757 423 3.05 67 AX 3.6991 1.3155 2.4773 1.5034 395 3.58 68 AY 4.6085 0.3231 2.1778 3.0674 272 10.94 69 AZ 4.7736 4.1841 2.3807 0.7673 363 5.49 70 BA 5.4036 1.4542 2.8567 0.9444 444 1.8 71 BB 4.865 1.3725 2.597 1.0277 392 4 1999 Water Chemistry (Figures 5.6 and 5.7) Species NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.8685 0.4345 0.1873 0.0572 1 AC00 8A 6.0505 0.907 1.7536 1.6074 9 1.59 2 ACOO 9A 4.0815 1.7814 1.9277 2.5507 71 2.83 3 AC16OA 1.7378 2.3539 2.9474 2.8895 1 1 4 ADOO 9A 4.7594 2.2492 2.8499 2.7834 32 1.28 5 AMOO 6A 6.6995 1.0279 1.5963 1.6639 10 1 6 AUOO3A 1.2366 3.5978 2.3431 3.5019 47 1.86 7 BR01 OA 1.5072 1.0324 -0.3089 0.6992 339 4.34 8 BR01 OB 0.8667 3.0786 1.3666 2.6497 107 2.28 9 BR01 OC 0.8795 1.3978 2.309 2.5641 88 5.52 10 BR01 OE 1.3366 0.0951 2.7561 2.1563 89 3.04 11 CMOO 4A 1.4579 1.2123 3.5304 2.1803 78 2.88 12 COOO 1B 6.6995 1.0279 1.5963 1.6639 32 1 13 COOO 1C 6.6995 1.0279 1.5963 1.6639 17 1 14 COOO 9D 6.6995 1.0279 1.5963 1.6639 11 1 15 DEOO 1A 0.4728 1.7254 0.3392 0.1047 27 2.88 16 DEOO2A 0.3048 -0.1341 1.9637 2.6356 523 2.78 17 DPOO 1A 0.897 4.059 1.5029 3.6927 25 1.17 18 DP065A 6.6951 1.1019 1.6904 1.7603 31 1.07 19 ECOO 1A 1.5667 -0.414 2.2115 0.8685 674 4.14 20 EUOO2A 1.6804 2.5234 2.8802 2.9662 50 1.17 21 EUOO9C 1.7378 2.3539 2.9474 2.8895 5 1 22 EUOO 9D 1.1106 3.1897 3.095 2.8721 38 2.24 23 EUOO 91 1.5036 2.6293 2.7584 2.9737 11 1.75 24 EY01 OA 0.8006 1.8399 0.4898 -0.0667 51 2.35 25 FGOO 1U 1.739 1.3722 -1.9703 0.0583 6 1.38 26 FR057A -0.2627 1.8171 0.9259 -0.202 386 1.51 27 FUOO2A 1.2394 1.0848 4.609 0.3128 13 1 28 GOOO 4A 1.3303 1.6021 2.8759 2.7368 34 5.12 29 MAOO 1B 1.2004 2.2912 2.8116 2.3038 330 5.21 30 NAOO 1P 6.6995 1.0279 1.5963 1.6639 17 1 31 NAOO3A 1.1815 0.4143 1.9494 1.083 62 3.88 32 NAOO 8C 3.1073 0.9237 0.3416 1.4438 12 1.41 33 NAOO 9P 0.9262 4.2929 1.3876 3.8489 21 1 34 NAOO 9S 0.9229 -0.484 0.3142 -1.2371 10 1.47 35 NA267A 3.6204 1.0279 1.5963 1.6639 30 1 36 NA365A 6.6995 1.0279 1.5963 1.6639 82 1 37 NA75 1A 2.0417 0.5522 2.4474 1.4897 70 2.83 38 NIOO 3A 0.0101 -0.0731 2.4881 4.535 31 1 39 NIOO 5S 0.7131 -0.9933 0.023 -1.9288 26 1 40 NIOO 8A 0.8473 -0.477 2.0489 2.3881 24 2.07 41 NIOO 8N 1.2394 1.0848 4.609 0.3128 37 1 42 NIOO 9A 3.4127 1.1809 0.472 1.065 270 1.75 43 NIOO 9S 3.6204 1.0279 1.5963 1.6639 21 1 44 NI01 1S 6.6995 1.0279 1.5963 1.6639 9 1 45 NI01 4A 0.5762 0.2181 2.5458 1.2135 16 1 46 NI01 7A 1.6196 0.4253 -0.8272 0.9294 21 1.96 47 NI06 5A 6.6995 1.0279 1.5963 1.6639 8 1 48 PI00 9S 1.5121 0.088 1.6026 2.6488 92 3.46 49 PSOO 1A 6.6995 1.0279 1.5963 1.6639 2 1 50 SP01 2S 6.6995 1.0279 1.5963 1.6639 75 1 51 SR103A 0.9262 4.2929 1.3876 3.8489 9 1 52 SSOO1C 3.3551 0.9539 1.2158 1.6757 50 1.18 53 STOO 9J 0.9262 4.2929 1.3876 3.8489 9 1 54 SU02 4B 3.6204 1.0279 1.5963 1.6639 9 1 55 SVOO 9P 2.5839 3.6725 1.7855 3.4556 24 1.71 56 SV02OC 1.1949 3.6923 2.2585 3.5491 16 1.75 57 UNOO 9S 3.6204 1.0279 1.5963 1.6639 27 1 Samples N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.8685 0.4345 0.1873 0.0572 1 PP 3.4609 1.0908 0.8918 1.3334 359 2.85 2 BS 1.0255 0.2329 1.7476 1.2738 407 4.2 3 BE 1.3958 0 2.1898 1.3151 402 2.13 4 OD 1.8639 0.999 0 0.8845 163 2.23 5 OE 0.9015 0.6521 1.9636 2.1129 265 5.34 6 LS 1.1759 2.8419 1.9575 2.6491 380 8.21 7 LE 1.4483 1.0169 2.529 1.4771 385 7.44 8 ON 0.3286 0.2669 1.91 2.1941 426 2.24 9 ME 0 1.7142 0.935 0 357 1.35 10 OJ 6.1985 1.0244 1.6457 1.6722 336 7.24 11 OS 1.8767 1.5025 2.1577 2.1842 371 11.07 12 OP 1.6853 0.6723 1.3708 1.3859 364 6.9 Water Chemistry 1999 (Figure 5.8) Species NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.868 0.814 0.6771 0.3569 1 AC00 8A 0.4763 2.4166 0.2708 -0.3616 9 1.59 2 ACOO 9A 0.1763 -0.1954 -0.6717 0.9082 71 2.83 3 AC16 0A -0.2194 -0.1828 -0.8319 1.0085 1 1 4 ADOO 9A 0.2262 -0.2259 -0.6953 0.8814 32 1.28 5 AMOO 6A 3.3452 -0.5276 0.2607 -0.0088 10 1 6 AUOO3A -0.2725 -0.2313 -0.9454 0.3004 47 1.86 7 BR01 OA -0.2287 -0.0332 -0.7905 1.1373 339 4.34 8 BR01 OB -0.3177 -0.2888 -0.6514 -0.3588 107 2.28 9 BR01 OC -0.3365 -0.2505 -0.1248 -0.5309 88 5.52 10 BR01 OE -0.2726 -0.2329 -0.7109 0.2221 89 3.04 11 CMOO 4A -0.2539 -0.2564 -0.6394 0.2601 78 2.88 12 COOO 1B 3.3452 -0.5276 0.2607 -0.0088 32 1 13 COOO 1C 3.3452 -0.5276 0.2607 -0.0088 17 1 14 COOO 9D 3.3452 -0.5276 0.2607 -0.0088 11 1 15 DEOO 1A -0.3561 -0.2402 0.4528 0.9077 27 2.88 16 DEOO2A -0.4313 -0.3706 0.9515 -0.1219 523 2.78 17 DPOO 1A -0.309 -0.2639 -0.9097 -0.1383 25 1.17 18 DP065A 3.2302 -0.5165 0.2254 0.024 31 1.07 19 ECOO1A -0.2398 -0.3494 -0.2797 -0.818 674 4.14 20 EUOO2A -0.226 -0.1889 -0.8461 0.9198 50 1.17 21 EUOO9C -0.2194 -0.1828 -0.8319 1.0085 5 1 22 EUOO 9D -0.2961 -0.2529 -0.9956 -0.0133 38 2.24 23 EUOO 91 -0.2493 -0.2093 -0.7668 0.6635 11 1.75 24 EY01 OA -0.3235 -0.3365 -0.4813 -0.8714 51 2.35 25 FGOO 1U -0.1889 0.3079 -0.9165 2.4368 6 1.38 26 FR057A -0.5039 -0.3929 1.7934 0.4766 386 1.51 27 FUOO2A -0.3026 -0.2589 -1.0096 -0.1009 13 1 28 GOOO 4A -0.267 -0.3186 0.15 -0.0302 34 5.12 29 MAOO 1B -0.2868 -0.2729 -0.7073 -0.1209 330 5.21 30 NAOO 1P 3.3452 -0.5276 0.2607 -0.0088 17 1 31 NAOO 3A -0.2852 -0.2474 -0.414 0.1847 62 3.88 32 NAOO 8C 0.1241 2.709 0.0835 -0.0078 12 1.41 33 NAOO 9P -0.3026 -0.2589 -1.0096 -0.1009 21 1 34 NAOO 9S -0.3105 -0.3557 -0.3291 -1.0207 10 1.47 35 NA26 7A 0.1891 3.2284 0.2767 -0.3825 30 1 36 NA36 5A 3.3452 -0.5276 0.2607 -0.0088 82 1 37 NA751A -0.1352 -0.0759 -0.8144 0.6475 70 2.83 38 NIOO 3A -0.5089 -0.3927 1.852 0.5354 31 1 39 NIOO 5S -0.3333 -0.3989 -0.2034 -1.528 26 1 40 NIOO 8A -0.3589 -0.3699 0.0484 -1.0821 24 2.07 41 NIOO 8N -0.3026 -0.2589 -1.0096 -0.1009 37 1 42 NIOO 9A 0.1598 2.3915 0.1753 -0.0137 270 1.75 43 NIOO 9S 0.1891 3.2284 0.2767 -0.3825 21 1 44 NI01 1S 3.3452 -0.5276 0.2607 -0.0088 9 1 45 NI01 4A -0.3821 -0.3222 0.2395 -0.5683 16 1 46 NI01 7A -0.2037 0.0696 -0.8754 1.7431 21 1.96 47 NI06 5A 3.3452 -0.5276 0.2607 -0.0088 8 1 48 PI00 9S -0.248 -0.1305 -0.5571 0.1025 92 3.46 49 PSOO 1A 3.3452 -0.5276 0.2607 -0.0088 2 1 50 SP01 2S 3.3452 -0.5276 0.2607 -0.0088 75 1 51 SR10 3A -0.3026 -0.2589 -1.0096 -0.1009 9 1 52 SSOO1C 0.1572 2.9673 0.186 -0.2369 50 1.18 53 STOO 9J -0.3026 -0.2589 -1.0096 -0.1009 9 1 54 SU02 4B 0.1891 3.2284 0.2767 -0.3825 9 1 55 SVOO 9P 0.0017 -0.2802 -0.7849 -0.0397 24 1.71 56 SV02 OC -0.2766 -0.2351 -0.9541 0.2458 16 1.75 57 UNOO 9S 0.1891 3.2284 0.2767 -0.3825 27 1 Samples N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.868 0.814 0.6771 0.3569 1 PP 0.1888 3.2284 0.2766 -0.3825 359 2.85 2 BS -0.3616 -0.4077 -0.0348 -1.3838 407 4.2 3 BE -0.3047 -0.3898 -0.374 -1.674 402 2.13 4 OD -0.183 0.4061 -0.9334 2.7225 163 2.23 5 OE -0.3821 -0.3222 0.2396 -0.5683 265 534 6 LS -0.3175 -0.2915 -1.082 -0.231 380 8.21 7 LE -0.2875 -0.2266 -0.9382 0.0275 385 7.44 8 ON -0.4926 -0.4525 1.475 -0.0416 426 2.24 9 ME -0.5285 -0.3214 2.3018 1.2238 357 1.35 10 OJ 3.3455 -0.528 0.2606 -0.0089 336 7.24 11 OS -0.2044 -0.2812 -0.7773 0.7624 371 11.07 12 OP -0.2347 -0.0823 -0.8875 1.2594 364 6.9 Environmental Variables N NAME AX1 AX2 AX3 AX4 0.9999 0.9986 0.9813 0.9571 1 Calcium 0.5069 0.6025 0.3421 0.2257 2 Magnesiu 0.9675 -0.0691 0.2097 0.0737 3 Sodium 0.9679 -0.179 0.1688 0.0242 4 Potassiu 0.9672 -0.1788 0.1721 0.0199 5 Chloride 0.9576 -0.1862 0.2106 0.0359 6 Sulphate 0.719 0.4556 0.4223 0.1814 7 Bicarbon 0.3096 -0.3986 0.0766 0.6254 8 Ph 0.2961 0.1261 0.5779 -0.413 9 Conducti 0.8648 -0.181 0.4426 0.121 New River Lagoon (Figures 5.9 and 5.10) Species N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.5048 0.2908 0.115 0.056 1 AC00 8A 1.0759 -0.2189 2.1657 1.8686 28 7 2 ACOO 9A -0.4622 3.1573 0.668 1.8313 166 5.95 3 AC13 4A -0.7413 -0.541 0.5779 -0.3085 26 2.4 4 ADOO 9A 3.0478 1.2811 1.8535 0.735 835 10.83 5 AMOO 8A 0.646 0.2853 2.3132 0.4103 1 1 6 BR01 OA 0.1744 2.7093 0.1877 1.7177 564 8.08 7 BR01 OB -0.686 -0.7645 1.1304 -0.1908 337 2.97 8 BR01 OC 0.587 1.3219 1.6977 -0.1487 1237 12.33 9 BR01 OD 0.646 0.2853 2.3132 0.4103 5 1 10 BR01 OE 2.6029 2.001 1.8119 0.7055 1253 13.64 11 CMOO 4A -0.0838 3.0839 1.1335 2.3717 16 5.82 12 COOO 1A -0.5161 -0.2905 1.6801 -0.8198 2 2 13 COOO 1B -0.8475 2.6172 0.4846 -0.0867 6 1.38 14 DEOO 2A 1.2704 0.4957 -0.3452 -0.4155 641 5.4 15 DPOO 1A -0.2453 -0.1818 1.3292 -0.389 6 2.57 16 DPOO7A 2.6754 0.9553 -0.2603 1.6283 9 4.76 17 DP01 OA 4.3982 1.3349 -3.4057 -4.0495 1 1 18 ECOO 1A 1.3286 0.9609 1.2477 1.6602 1561 18.71 19 EUOO2A 3.7031 0.8675 -1.3444 -1.2929 41 3.79 20 EUOO9C -0.9288 -0.929 0.9508 0.7659 6 1.8 21 EU00 9I 0.646 0.2853 2.3132 0.4103 1 1 22 EUOO 9S -0.4368 -0.3967 0.0174 -0.7099 10 2.38 23 EU01 7A 4.3982 1.3349 -3.4057 -4.0495 1 1 24 EU047A 4.0282 0.9613 4.0939 3.0699 7 2.58 25 EY01 OA 2.8971 0.9774 0.2888 -0.033 43 10.45 26 EY01 8A -0.8093 -0.9175 0.9572 0.0156 14 2.13 27 FAOO 1A 4.2986 0.9879 7.8757 5.7516 1 1 28 FGOO 1U -0.4704 3.4151 1.2894 1.6338 90 4.27 29 FR057A 4.2986 0.9879 7.8757 5.7516 1 1 30 GOOO 4A 2.9582 1.1254 1.9922 1.5465 232 11.28 31 GO02 5H -0.4617 -0.0184 0.6831 1.7224 92 5.03 32 MAOO 1A 2.9637 0.6144 0.1145 0.4662 1 1 33 MAOO 1B 2.3482 0.3526 1.2836 1.4918 1405 16.3 34 MAOO 2B 3.8815 1.0306 -1.2185 -1.3245 28 3.77 35 NAOO 3A 0.7644 0.3826 -0.0756 -0.2277 175 11.96 36 NAOO 8C -0.0279 2.9389 1.4989 2.6251 2 2 37 NAOO 9S -0.4423 -0.525 1.9179 -0.9206 37 1.94 38 NA066B 4.1833 1.1995 -1.9268 -2.2824 7 1.81 39 NA10 2A 3.1946 1.0177 -1.199 -1.1111 4 4 40 NA12 3A -0.16 3.0879 0.8383 1.7575 9 1 41 NA14 4A -0.4651 -0.4877 1.6003 2.458 28 2.14 42 NA75 1A -0.2075 0.0282 0.5818 -0.4075 108 6.41 43 NEOO 1A 2.546 1.3984 -0.1843 0.8637 7 3.27 44 NIOO 8A 3.1791 0.7129 5.6287 3.9382 27 2.14 45 NIOO 9A -0.2928 3.4709 1.5886 1.5248 88 4.4 46 NI01 4A 1.2307 -0.0533 2.2195 0.4999 207 11.19 47 NI01 7A -0.7234 3.7532 1.0428 1.4726 85 2.24 48 PIOO 5A 2.6096 1.9536 -0.1536 -0.0178 1 1 49 PI00 8A 3.3358 1.5949 -2.1379 -2.4044 5 1.92 50 PI00 9S 0.646 0.2853 2.3132 0.4103 2 1 51 SLOO 1A 2.9744 1.0638 -1.0414 -0.933 36 3.9 52 SROO1A 4.2742 1.0096 7.722 5.6111 38 1.05 53 SSOO1C -0.1176 3.1546 1.5821 2.721 7 2.33 54 SVOO 9P 0.646 0.2853 2.3132 0.4103 2 1 55 SYOO1A 2.8013 1.5789 1.0289 0.3059 214 5.13 56 SYOO3A -0.8234 -0.2464 0.3854 -0.2211 19 1 57 XHOO9X 2.8079 1.1351 0.0935 0.5802 134 11.29 58 XIOO 9X 3.8823 1.0622 -0.5185 -0.5554 362 4.27 59 XNOO5X 2.4794 1.0003 1.9535 2.3701 51 4.02 60 XNOO6X 1.9521 1.7284 0.0515 1.7114 10 2.38 61 XS01 4X 2.1456 0.7331 -0.3044 1.6675 8 1.68 62 XS01 5X 3.7622 0.9718 -1.9612 -2.0992 50 2.82 63 XUOO 9X 1.9305 0.7896 -0.4684 1.1027 4 1 Samples N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.5048 0.2908 0.115 0.056 1 OE 1.1643 0.6152 0.7867 0.4036 271 5.58 2 XC 0.915 0.6708 0.86 0.2487 401 5.79 3 XD 0.4697 0.95 1.3383 0.3876 410 4.88 4 XE 0.114 0 1.1534 0.6366 396 4.49 5 XF 0.1896 0.6167 1.0354 0.6144 377 9.46 6 XDP 0.912 1.1477 1.3365 0.4846 327 3.15 7 XP 0 2.7155 1.1413 1.3394 279 7.99 8 XPP 0.8347 1.5523 1.3035 0.832 387 4.9 9 XEP 0.6184 2.0895 0.7952 1.5003 392 4.87 10 XRP 0.6663 1.5616 1.0437 1.1628 381 5.07 11 A 2.404 0.9045 0.7059 0.6225 396 7.82 12 B 2.1117 1.2581 1.4954 1.0491 458 5.39 13 C 1.9328 1.0054 1.3008 1.232 424 5.18 14 D 2.0318 1.4335 1.2935 0.7676 445 7.58 15 F 1.7296 1.1325 0.6193 0.66 419 9 16 G 2.2124 1.2944 1.4805 0.9704 443 6.09 17 H 2.216 1.1457 1.4574 1.0611 442 6.59 18 I 1.3188 0.8733 0.0964 0.0324 403 3.21 19 J 2.9813 0.7833 0 0 396 4.48 20 K 2.1492 1.0653 1.3498 1.0749 436 5.65 21 L 1.8547 1.485 1.5302 0.5454 416 4.75 22 M 2.1687 1.1146 1.488 1.091 417 5.61 23 O 2.3286 1.0946 1.38 0.8766 413 6.49 24 P 2.9385 0.7569 1.6872 1.3019 401 4.74 25 Q 2.3308 1.241 1.5888 0.8623 463 5.56 26 R 2.4509 1.4036 1.6452 0.7185 401 4.41 Epiphyte (Figures 5.11 and 5.12) Species NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.7327 0.3048 0.1754 0.091 27 ST02 1A -0.7922 2.7189 3.6053 4.4918 46 1 28 SYOO 1A -0.5366 0.0519 -1.1523 0.9877 217 5.26 00CO XNOO8X -0.3141 2.6259 2.0796 2.5558 14 1.81 29 XB01 OB -0.2289 0.3172 -0.6772 1.0896 549 10.84 25 NI01 7A -0.2184 0.9383 3.454 2.1091 152 3.24 1 ADOO 9A -0.2108 1.5368 0.6892 2.0062 1510 13.45 10 EU01 7A -0.1917 3.4994 0.1419 0.3884 29 1.15 24 NI01 4A -0.1714 0.4848 -0.7344 0.3571 129 6.88 5 CM02 3A -0.1528 -0.1064 4.0327 2.2651 22 2.14 11 EU047A -0.1518 3.314 -0.2742 -0.7177 12 1.67 6 CYOO4A -0.1452 2.6295 1.5926 1.7852 27 1.62 43 XUOO 8X -0.1212 3.3805 1.29 2.5168 110 1.54 35 XMOO 8X -0.0752 -0.2021 3.8879 1.8714 31 1.98 31 XHOO 9X -0.0743 2.6385 0.0647 -0.5391 266 9.52 9 ECOO 1A 0.0484 0.2328 1.6369 0.7558 2836 20.58 37 XNOO 5X 0.077 2.7299 1.849 -0.3135 161 8.81 17 MAOO 1B 0.0816 0.4683 1.7684 0.6548 1408 18.12 15 FUOO2A 0.263 -2.1867 0.3181 2.7444 13 1 22 NIOO 8N 0.263 -2.1867 0.3181 2.7444 37 1 36 XMOO 9A 0.2722 2.7898 1.2283 1.9323 115 1.92 19 NA75 1A 0.306 -2.0652 0.2779 2.6978 25 1.18 30 XB02 OC 0.3135 -0.0427 -0.2647 0.2728 1323 15.46 26 PIOO 9S 0.3229 -1.7162 1.1352 -1.5356 32 1 33 XIOO 7X 0.3982 0.0306 3.4551 0.0653 17 3.04 12 EY01 OA 0.4932 2.5824 0.9909 -0.9819 185 9.3 18 NAOO 3A 0.5222 1.5417 2.2222 0.5306 110 11.2 34 XIOO 9X 0.661 0.5413 -1.0362 1.9572 118 5.05 3 BR01 OE 0.6971 -1.9059 0.6088 2.345 47 2.05 4 CMOO 4A 0.9418 1.6005 -1.2942 -1.8938 145 2.88 21 NIOO 8A 1.1954 -0.6348 1.293 0.1225 30 3.36 23 NIOO 9A 1.4232 -0.3082 2.1911 1.8024 265 6.83 16 GOOO 4A 1.447 2.1189 -0.0184 1.6744 309 16.62 44 XVOO 9X 1.6826 1.3039 1.5029 -0.7718 33 1.65 14 FR06OA 2.0066 2.3966 -0.088 2.352 61 3.06 2 BR01 OA 2.1602 2.1984 1.2133 2.2519 486 12.82 8 DEOO 2A 2.3075 1.5053 1.6599 0.3954 622 4.29 20 NIOO 3A 2.5525 1.2183 1.1495 -0.1369 31 1 42 XS01 7X 2.7401 1.45 0.7314 2.0598 11 1.42 40 XO03 2X 2.8351 1.5243 0.4959 2.014 11 1.86 41 XSOO 7X 2.8373 1.526 0.4907 2.0129 17 1.84 13 FR057A 2.8397 1.2772 0.9492 1.0611 902 3.42 39 XN01 9X 2.8936 1.5682 0.3642 1.9858 12 1.18 32 XIOO 6X 2.92 1.5812 0.3254 1.9774 8 1 7 DEOO 1A 2.9383 0.9382 1.8578 -0.1601 13 1 NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.7327 0.3048 0.1754 0.091 1 BE 0.2676 0 1.3931 0.5856 402 2.13 2 LE 0.5873 0.0832 0.8954 1.1795 371 6.98 3 ME 2.6825 1.1802 1.0417 1.0365 357 1.35 4 ON 2.1805 1.338 1.388 0.4768 426 2.24 5 B 0.1226 0.631 0.6731 0.9443 456 5.34 6 C 0.1535 0.6214 1.1014 0.6664 421 5.1 7 D 0.1669 0.6629 0.241 0.9764 442 7.48 8 G 0.1057 0.6495 0.5944 0.8779 435 5.88 9 H 0.1875 0.744 0.7421 0.9497 441 6.56 10 K 0.1521 0.6071 0.9189 0.8098 428 5.45 11 L 0.1173 0.4817 0 0.8148 415 4.73 12 M 0.1478 0.6845 0.817 0.9549 416 5.58 13 O 0.2094 0.6515 0.6522 0.8883 409 6.36 14 Q 0.0791 0.5992 0.4414 0.9573 461 5.51 15 R 0.0944 0.6622 0.1635 0.9741 400 4.39 16 Y 0.4433 1.2889 0.7093 0 402 6.74 17 AA 0.2989 2.1554 1.0881 0.9783 385 8.17 18 AD 0.8504 0.9247 1.56 0.643 420 5.41 19 AE 0.4263 0.3787 1.5733 0.8493 408 2.74 20 AF 0 1.3025 1.4014 1.7633 351 5.33 21 AG 0.2692 0.6323 2.0103 0.9971 424 6.34 22 Al 0.3955 0.7433 1.7867 1.0022 397 7 23 AL 0.1708 0.5145 1.3823 0.8993 428 3.14 24 AM 0.3974 0.4223 1.4145 0.9596 429 4.56 25 AP 0.1247 1.5119 1.0458 1.6291 443 3.71 26 AQ 0.5585 1.7691 0.9694 1.084 526 10.24 27 AS 0.8625 1.0349 0.5864 1.2748 368 8.95 28 AU 0.0925 1.0546 1.0112 1.2029 422 3.04 29 BA 2.5867 1.323 0.8411 1.2026 431 1.7 30 BB 2.0307 1.347 0.9087 1.2675 383 3.82 Plankton (Figures 5.13 and 5.14) Species NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.6919 0.1993 0.1147 0.0343 21 NA26 7A -1.1557 1.0424 0.3169 0.6285 30 1 24 NI00 9S -1.1557 1.0424 0.3169 0.6285 21 1 30 SU024B -1.1557 1.0424 0.3169 0.6285 9 1 31 UNOO 9S -1.1557 1.0424 0.3169 0.6285 27 1 29 SSOO1C -0.2894 0.9045 0.2659 0.7619 59 1.62 18 NAOO 8C -0.0365 1.091 1.2927 0.4738 14 1.88 23 NIOO 9A 0.475 0.9527 0.6057 -0.3945 370 2.88 28 PIOO 9S 2.0747 2.1315 0.8255 0.7092 37 1.4 26 NI01 7A 2.106 1.2517 0.0862 2.6429 102 3.11 13 FGOO 1U 2.3349 0.5028 -0.1638 2.0748 102 5.13 25 NI01 4A 2.3699 -0.5926 -0.5662 -4.3804 32 3.37 20 NA14 4A 2.3829 -0.6561 -0.1158 -2.5777 23 2.25 1 ACOO 9A 2.5285 2.1174 1.6317 0.9437 237 6.38 22 NA751A 2.5779 1.48 -0.5959 0.6821 100 5.62 4 BR01 OA 2.6296 1.6854 1.3858 -0.3726 652 7.27 6 BR01 OE 2.7655 1.0454 -0.5369 0.8982 219 5.75 17 NAOO 3A 2.9709 0.4633 1.8755 0.0889 55 5.63 19 NA123A 2.99 2.0961 7.6117 -0.5699 9 1 15 GO02 5H 3.0863 1.8206 2.9696 0.3663 65 4.22 2 AC16 0A 3.3384 1.8757 -1.9533 1.9322 32 1 5 BR01 OC 3.4515 -0.4569 0.1605 1.049 583 5.22 10 DEOO 2A 3.5828 -1.5657 1.2616 -0.7583 24 2.82 9 COOO 1B 3.8433 4.2807 -0.6194 1.4779 30 1.23 11 ECOO 1A 3.9575 1.7453 1.2844 -0.1016 604 7.98 8 COOO 1A 4.0271 4.5676 -0.2191 1.3148 24 1 16 MAOO 1B 4.5467 -0.1627 0.4355 0.3452 259 6.13 7 CMOO 4A 4.6277 2.7069 -0.5034 1.1785 132 3.53 14 GOOO 4A 5.0779 2.848 -0.0099 1.0937 20 2.82 3 ADOO 9A 5.2551 1.0424 0.3169 0.6285 36 1 32 XB01 OB 5.4655 1.0424 0.3169 0.6285 13 1.55 12 EY01 OA 6.048 1.0424 0.3169 0.6285 23 1.29 33 XHOO9X 6.1312 1.0424 0.3169 0.6285 51 1.08 34 XNOO5X 6.1436 1.0424 0.3169 0.6285 39 1.05 27 PIOO 8A 6.1669 1.0424 0.3169 0.6285 28 1 35 XNOO 8X 6.1669 1.0424 0.3169 0.6285 11 1 N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.6919 0.1993 0.1147 0.0343 1 PP 0 0.9776 0.502 0.0579 352 2.74 3 OD 2.252 1.5076 1.04 0 161 2.18 5 XHP 2.3746 0.979 0.6816 0.0452 287 8.17 8 XP 2.4225 1.1121 0.5669 1.0361 279 7.99 2 OP 2.8704 1.53 0.7445 0.3188 352 6.61 10 XEP 2.9007 1.4679 1.2196 0.2145 392 4.87 6 XAP 3.0401 1.3669 0 0.6539 184 7.13 9 XPP 3.1613 0.535 0.4843 0.3596 387 4.9 11 XRP 3.2927 1.1594 0.9546 0.3001 381 5.07 4 XDP 3.5245 0 0.4634 0.5848 327 3.15 7 XBP 3.5263 1.7997 0.6111 0.6707 396 7.25 13 AJ 4.3043 0.6667 0.5942 0.4389 219 5.18 12 Z 5.0122 1.2812 0.4941 0.4366 355 7.04 Sediment (Figures 5.15 and 5.16) Species N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.8185 0.6791 0.297 0.1982 6 AM00 6A -0.3826 2.2025 1.37 1.4805 10 1 20 COOO 1C -0.3826 2.2025 1.37 1.4805 17 1 21 COOO 9D -0.3826 2.2025 1.37 1.4805 11 1 47 NAOO 1P -0.3826 2.2025 1.37 1.4805 17 1 54 NA36 5A -0.3826 2.2025 1.37 1.4805 82 1 61 NI01 1S -0.3826 2.2025 1.37 1.4805 9 1 64 NI06 5A -0.3826 2.2025 1.37 1.4805 8 1 70 SP01 2S -0.3826 2.2025 1.37 1.4805 75 1 26 DP06 5A -0.3462 2.1634 1.4201 1.6136 31 1.07 19 COOO 1B 0.4065 0.7034 3.5496 1.2797 151 2.06 18 COOO 1A 0.5274 0.7601 3.554 1.3767 83 1.9 8 APOO 1A 0.7777 0.1861 4.0731 1.5177 18 1 3 AC031A 0.9521 0.5781 3.6777 1.6615 47 1.77 41 GO02 5B 0.9876 3.5242 -0.5866 1.4469 12 1 49 NAOO 8C 0.9876 3.5242 -0.5866 1.4469 202 1 43 GPOO 9S 1.019 3.4823 -0.5689 1.4219 36 1.12 66 PSOO1A 1.0459 4.9964 2.1475 2.1524 11 1.42 59 NIOO 9A 1.2352 1.17 2.5977 1.8068 96 3.45 14 CMOO 4A 1.3749 2.3804 -0.1837 1.8098 144 3.91 44 GYOO 5A 1.4771 1.6864 2.5869 1.8949 31 1.21 17 CM11 OA 1.5651 2.8601 -0.2933 1.0985 74 1.65 38 FR05 7A 1.6082 4.7017 -0.114 1.6632 63 3.79 60 NIOO 9V 1.6621 1.8905 1.2018 1.9293 13 1 69 SPOO 9H 1.6621 1.8905 1.2018 1.9293 25 1 73 SSOO 1C 1.6819 1.9037 1.1871 1.954 75 1.05 33 EU10 8A 1.6853 1.7355 3.0809 2.4048 40 1.6 2 ACOO 9A 1.7916 0.6475 2.5229 1.7737 213 6.21 75 SVOO 9P 1.8277 -0.402 0.1476 3.5531 24 1.71 32 EUOO9S 1.8804 0.6064 2.859 0.7966 32 4.74 55 NA751A 1.9202 0.4661 2.5788 0.2351 179 8.96 22 CYOO 4A 2.2866 2.1236 1.8828 2.4567 14 1.56 24 DPOO 1A 2.2904 0.1256 1.499 2.9274 59 2.94 1 ACOO 8A 2.4424 1.8018 2.6773 1.4924 104 6.29 42 GO02 5H 2.7827 -0.757 1.8845 0.498 67 3.97 34 EY01 OA 2.9311 1.8762 0.9722 1.6482 309 5.06 37 FGOO 1U 2.9479 -0.1766 0.7296 0.9633 22 4.32 4 AC13 4A 3.1123 -0.7938 1.3874 -0.7196 34 3.66 63 NI01 7A 3.209 -0.1338 2.2387 -0.4909 29 2.56 35 EY01 8A 3.2678 -1.0747 1.5766 0.5999 19 3.44 77 SYOO3A 3.3485 -1.3035 1.8314 -3.0379 19 1 11 BR01 OB 3.5161 -0.6182 1.5237 0.7745 595 7.13 62 NI01 4A 3.5593 1.4104 0.8893 0.7482 231 12.5 9 AUOO 3A 3.579 0.2599 0.6756 3.2559 73 2.86 53 NA14 4A 3.7801 -2.5642 2.5133 1.6096 14 1 85 XMOO 7X 3.8521 5.0806 2.3112 2.2598 11 1 89 XP01 4X 3.8521 5.0806 2.3112 2.2598 18 1 90 XSOO 5X 3.8521 5.0806 2.3112 2.2598 122 1 91 XSOO 6X 3.8521 5.0806 2.3112 2.2598 9 1 N92 XSOO 7X 3.8521 5.0806 2.3112 2.2598 11 1 94 XS01 6X 3.8521 5.0806 2.3112 2.2598 10 1 78 XAOO9X 3.9317 5.0612 2.6723 2.2312 89 1.09 27 ECOO1A 3.9325 1.0176 2.2574 0.7904 1217 13 51 NAOO9S 3.9951 -0.3325 0.0562 -0.5046 50 3.2 15 CMOO 9M 4.0487 -0.8517 -0.0445 3.925 11 1 29 EUOO5S 4.0487 -0.8517 -0.0445 3.925 1 1 50 NAOO 9P 4.0487 -0.8517 -0.0445 3.925 21 1 72 SR103A 4.0487 -0.8517 -0.0445 3.925 9 1 74 STOO 9J 4.0487 -0.8517 -0.0445 3.925 9 1 30 EUOO 9D 4.0668 -0.6664 0.112 3.8136 24 1.28 5 ADOO 9A 4.0848 3.4177 2.5595 2.0444 223 9.67 76 SV02OC 4.0938 -0.3885 0.53 3.6466 16 1.75 31 EUOO 91 4.2274 0.5118 2.0066 3.0964 11 1.75 65 PIOO 9S 4.2538 1.0459 2.555 2.9388 23 1.19 12 BR01 OC 4.2659 1.5137 0.4475 0.3917 614 7.97 40 GOOO 4A 4.3342 3.6828 2.5735 1.9344 104 11.3 10 BR01 OA 4.5812 1.6869 1.3595 0.3346 377 13.4 48 NAOO 3A 4.6718 1.3217 1.6192 0.3591 267 11.9 13 BR01 OD 4.7583 1.6098 1.2905 0.9206 21 2.74 CDCO XMOO 8X 4.9926 2.0931 2.6763 2.6733 35 1 57 NIOO 5S 5.0197 1.8107 0.9087 -0.389 26 1 16 CM02 3A 5.0268 2.1705 2.6933 2.6228 12 1.18 39 FR06OA 5.0746 3.8557 2.9525 2.2549 10 1.92 87 XNOO5X 5.1122 2.0064 2.5247 1.9241 89 4.06 28 EUOO 2A 5.1647 1.4056 1.195 2.8432 116 4.21 68 SLOO 1A 5.177 1.3254 1.4404 -0.0541 65 5.17 45 MAOO 1B 5.2201 1.3041 0.9842 2.6622 1486 12.1 23 DEOO 2A 5.3111 1.8636 0.3691 -0.6516 815 7.11 82 XHOO 9X 5.3401 2.1379 1.9095 0.8517 65 6.08 25 DPOO 7A 5.3827 2.7643 2.7323 1.4951 43 1.9 67 RHOO 1A 5.4684 2.1205 1.568 3.3058 27 1.16 84 XMOO 6X 5.5151 2.9635 3.0751 1.3094 9 1.25 56 NEOO 1A 5.5167 2.6824 2.3427 0.3234 28 1.72 58 NIOO 8A 5.6268 2.0725 -0.0739 0.7984 29 2.47 52 NA066B 5.8358 2.5833 1.8363 1.2221 17 3.48 93 XS01 5X 5.903 2.1979 -0.1141 0.5088 67 3.82 81 XFOO9X 5.9308 3.5577 3.5704 1.3949 42 1.9 95 XUOO 8X 6.0269 3.2322 3.3662 2.8468 110 2.49 COCO XNOO 9X 6.055 2.0147 2.9848 3.2292 12 1.18 83 XIOO 9X 6.0621 2.3995 -0.225 0.8406 361 4.18 71 SROO 1A 6.1358 2.1696 -0.2749 1.6151 38 1.05 7 ANOO 9A 6.1761 3.3775 3.8085 1.5867 12 1.18 46 MAOO 2B 6.2321 3.2928 3.3369 1.1058 56 2.32 79 XFOO 7X 6.284 3.4521 3.8851 1.4026 16 1 80 XFOO 8X 6.284 3.4521 3.8851 1.4026 14 1 3S NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.8185 0.6791 0.297 0.1982 12 OJ 0 1.9404 1.7414 1.4953 336 7.28 14 XI 1.3282 2.989 0 1.4277 403 3.56 13 XB 1.3894 0.8123 2.8942 1.2921 364 7.81 15 XH 2.0521 1.1558 1.8241 1.5511 378 11.6 16 XA 2.4668 1.8942 1.0106 1.3853 394 6.23 17 XG 3.0725 0.7752 1.8912 1.3518 385 10.2 4 XF 3.5812 0.3941 1.6065 0.685 375 9.36 3 XE 3.6476 0 1.6213 0.9812 392 4.4 11 AZ 3.7855 4.6228 2.1532 2.0447 356 5.28 2 XD 3.9008 1.0493 1.1151 0.5056 410 4.88 10 AX 3.946 1.3411 1.9101 0.8483 395 3.58 5 LS 3.9646 0.3679 0.8698 2.4306 381 8.34 7 OS 4.2084 1.3046 1.556 1.7646 367 10.8 1 XC 4.4161 1.2443 0.9946 0.5986 401 5.79 6 BS 4.4557 1.3337 1.32 0.4736 417 4.4 9 AH 4.4727 1.4481 1.734 1.2251 436 6.09 8 OE 4.571 1.4724 1.0133 0.4537 270 5.53 27 AN 4.6575 1.6789 1.5189 1.5263 420 6.31 23 F 4.6756 1.7081 1.2356 0.6666 406 8.47 25 V 4.7714 2.0951 1.6605 1.1158 375 14.3 26 AC 4.786 1.5794 1.21 1.584 388 4.93 22 A 4.9104 1.7413 1.1014 1.3172 391 7.63 24 I 4.9439 1.5052 0.8534 0 395 3.09 21 X 5.1667 1.6499 1.4026 2.4599 397 2.4 20 P 5.272 1.8629 0.6856 1.6683 394 4.58 19 AY 5.3879 2.7699 2.6718 1.6791 272 10.9 18 J 5.5078 1.9138 0.6072 1.3338 388 4.3 Additional species codes for fossil/modern data set MUC001 Achnanthes delicatula var haukiana (Grunow) Lange-Bertalot in Lange Bertalot and Ruppel 1980 MUC003 Achnanthes hungarica (Grunow) Grunow in Cleve and Grunow 1880 AC001A Achnanthes lanceolata (Brebisson) Grunow in Cleve and Grunow 1880 AC049A Achnanthes ploenensis Hustedt 1930 AM001A Amphora ovalis (Kutzing) Kutzing 1844 AM094B Amphora proteus Greg. AM9990 Amphora subcapitata (Kisselev) Hustedt 1959 CA006A Caloneis amphisbaena (Bory) Cleve 1894 XCIOOIZ Caloneis incognita Hustdet CH009X Chaetoceros sp XCHOOIZ Chiwa Lagoon sp 5 CI004A Craticula cuspidata Kutz (Mann, 1990) MUC004 Cyclostephanous dubius (Fricke) Round 1982 CY028A Cyclotella distinguenda Hustedt 1927 CY003A Cyclotella meneghiana Kutzing 1844 CY035A Cyclotella plitviscensis Hustedt 1945 XCTOOIZ Cyclotella stelligera var tenuis CM011A Cymbella parva (W.Smith) Cleve DE002B/3A Denticula elegans varieties MUC006 Diatoma tenuis Agardh 1812 DP009A Diploneis elliptica (Kutzing) Cleve 1891 DPOOIB Diploneis ovalis variety EYOIOA Encyonema mesianum (Cholnoky) Mann 1990 Ey012A Encyonema muelleri (Hustedt) Mann 1990 EY016A Encyonema silesiacum (Bleisch in Rabenhorst) Mann 1990 EP003A Epithemia argus (Ehrenberg) Kutzing 1844 EP004A Epithemia turgida (Ehrenberg) Kutzing 1844 EU009D Eunotia exigua (Brebisson ex Kutzing) Rabenhorst 1864 EU048A Eunotia naegelii Migula in Thome 1907 EU005S Eunotia sp FAOOIA Fallacia pygmaea (Kutz) Stickle and Mann 1990 FR005A Fragilaria virescens Ralfs 1843 MUC008 Gomphonema intricatum var lunata nov. var. Germain 1981 SP009H Irish Creek Navicula sp XM005X Kates Lagoon Cymbella sp MUCOIO Mastogloia muradi Voigt MUCOll Mastogloia recta Hustedt MUC012 Navicula bremensis Hustedt 1957 XNBOOIZ Navicula brasiliana variety NA051A Navicula cari Ehrenberg 1836 NA021A Navicula cincta (Ehrenberg) Ralfs in Pritchard 1861 NA050A Navicula dementis Grunow 1882 NA067A Navicula crucicula (W.Smith) Donkin 1872 NA039A Navicula festiva Krasske 1925 SP012S Navicula florinae (dissolved) XNG001Z Navicula gaweniensis Gasse 1986 NA418A Navicula helvetica Brun 1895 NA047A Navicula protracta (Grunow) Cleve 1894 NA003B Navicula radiosa variety XNV008Z Navicula sp 8 XNS001Z Navicula subatomoides Hustedt Unid 17 Neidium sp (L) XNF001Z Nitzschia amphibia f.frauenfeldii (Grunow) Lange-Bertalot 1987 XNM001Z Nitzschia amphibiodes Hustedt 1942 XNM003Z Nitzschia angustatula Lange-Bertalot 1987 NI023A Nitzschia scalaris (Ehrenberg) W.Smith 1853 NIO11S Nitzschia sp NI005S Nizschia sp Unid 23 Northern Lagoon species 14 PI170A Pinnularia braunii (Grunow) Cleve 1895 XPN001Z Pinnularia sp PI007A Pinnularia viridis Hustedt 1930 XPL001Z Plagiotropis lepidoptera (Pfitzer) Cleve PS001A Pseudostaursira brevistriata (Grun. In Van Heurck) Williams and Round 1987 XSC001Z Small Croc Lagoon sp 15 Unid 24 Small Croc Lagoon species 15 (large variety) XSS001Z Small side view (L) Unid 33 Species A (L) SS002A Staurosirella pinnata (Ehrenb) D.M.Williams and Round ST021A Stephanodiscus minultus (Kutzing) Cleve and Moller 1878 SR012S Striatella sp Unid 1 Aulacoseira sp (L) Unid 2 Centric sp 15 (L) Unid 3 Chiwa Lagoon species 20 Unid 4 Cyclotella sp 16 (L) Unid 5 Cyclotella sp 9 (L) Unid 6 Gomphonema sp 8 (L) Unid 7 Mastogloia species small (L) Unid 8 Mastogloia species small dissolution stage 4 (L) Unid 9 Navicula sp 16 (L) Modern/Fossil Data Set (Complete) (figure 5.17) Species N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.4992 0.3915 0.2973 0.2528 1 AC 001A 1.7785 1.4619 -0.2647 -1.7767 1 1 2 AC 008A 1.7012 1.6926 -0.1539 -1.2711 1280.41 19.07 3 AC 008D -0.4288 0.0974 -0.7686 -0.2565 12.01 3.01 4 AC 009A -0.7121 0.6381 0.6529 -0.0041 1345.6 4.34 5 AC 013A 1.7356 0.495 0.1581 -0.7571 81 8.9 6 AC 031A -0.8008 1.0123 1.3355 -0.0738 141.17 3.52 7 AC 049A 0.5361 -1.7094 0.559 0.3609 1 1 8 AC 134A -0.6227 0.5827 0.573 -0.0587 102.12 3.97 9 AC 160A 0.0363 0.775 0.4 -0.2122 4 4.01 10 AD 009A -0.5216 0.0136 -0.6792 -0.1599 5383.4 4.48 11 AM 001A 3.0143 1.0853 0.5764 -2.1605 3 1 12 AM 006A -1.1089 1.4686 3.6555 0.4376 30.04 3.01 13 AM 008A -0.9524 1.1796 2.6521 0.3425 9.01 3.53 14 AM 094B 2.3835 1.6477 -0.0642 -0.906 71 7.49 15 AM 099X 1.1107 -1.2724 0.7458 0.1811 2 1 16 AM 9990 3.0143 1.0853 0.5764 -2.1605 5 1 17 AN 009A 0.4669 0.2059 -0.5782 0.0994 68.05 7.35 18 AP 001A -0.8604 1.0191 1.4805 -0.0331 54.06 3.01 19 AU 003A -0.7043 0.6603 0.6664 -0.0059 219.26 3.84 20 BR 01 OA -0.3276 0.0343 0.0026 -0.0582 19797.41 8.67 21 CA 006A 0.4499 -1.396 0.4617 0.2047 2 2 22 CH 009X 2.5398 2.2431 -0.4063 -1.731 15 2.65 23 CI004A -0.4793 -0.0317 -0.5298 -0.1801 15.02 3.37 24 CM 004A -0.6566 0.5383 0.3379 -0.0307 1271.49 5.65 25 CM 009M -0.8007 0.6552 0.8177 0.0541 33.04 3.01 26 CM 011A -0.7701 0.8855 0.8652 0.0924 224.26 3.52 27 CM 023A -0.3835 -0.1157 -0.5215 -0.0971 153.18 4.38 28 CO 001A -0.6747 1.0279 1.4318 -0.0565 344.38 4.37 29 CO 001B -0.8987 1.0319 1.7258 0.0764 537.64 3.81 30 CO 001C -0.2375 1.4906 2.608 -0.067 70.06 5.36 31 CO 009D -1.1089 1.4686 3.6555 0.4376 33.04 3.01 32 CY 003A 2.0379 1.6988 -0.2154 -1.8757 4 2.67 33 CY 004A -0.3638 0.2468 -0.2681 -0.2699 136.15 5.76 34 CY 028A 1.8394 0.3112 0.2151 -0.7487 1762 33.6 35 CY 035A 0.42 -1.3235 0.4195 0.2278 57 6.81 36 DE 001A 2.3076 2.2783 -1.094 6.8958 473.1 5.07 37 DE 002A 0.6773 -1.026 0.4535 0.1317 13721.64 52.04 38 DE 002B -0.8308 0.7097 0.7467 0.3255 1333.59 3.73 39 DE 003A 2.0409 1.6731 -0.3158 -1.496 2 1 40 DP 001A -0.3406 0.4552 0.8322 -0.0345 224.21 6.09 41 DP 001B 0.2138 -1.0619 -0.0557 0.2384 1 1 42 DP 007A -0.2435 0.0744 -0.5927 -0.1635 222.24 5.03 43 DP 009A 1.7115 0.6267 -0.023 0.9695 3 3 44 DP 01 OA -0.3186 -0.0847 -0.8287 -0.1243 30.03 4.14 45 DP 061A -0.5691 0.1073 -1.1838 -0.0496 33.04 3.56 46 DP 065A -1.0968 1.4389 3.5541 0.4282 93.11 3.07 47 EC 001A -0.2397 -0.1374 -0.1233 -0.0784 18745 9.49 48 EP 003A 2.1628 1.1575 -0.0862 -0.7546 3 3 49 EP 004A 1.0404 -2.3492 1.1901 0.5223 1 1 50 EU 002A -0.5295 0.1702 -0.1847 -0.0043 381.45 5.24 51 EU 005S -0.8007 0.6552 0.8177 0.0541 3 3.01 52 EU 009C -0.6193 0.6619 0.5665 -0.0824 81.1 3.8 53 EU 009D -0.788 0.6238 0.7367 0.052 114.14 3.69 54 EU 009I -0.738 0.5717 0.5796 0.1289 33.04 3.51 55 EU 009S 0.6805 0.9813 0.3921 -0.6782 188.11 13.14 56 EU 017A -0.5515 0.0128 -0.8691 -0.2045 90.11 3.2 57 EU 047A -0.4893 0.0132 -0.7902 -0.162 75.09 4.11 58 EU 048A 0.7852 -1.3713 0.6266 0.2414 1 1 59 EU 108A -0.6451 0.81 0.8147 -0.1385 120.14 3.44 60 EU 999X 1.1107 -1.2724 0.7458 0.1811 2 1 61 EY 01 OA -0.3655 0.2983 -0.1234 -0.139 1712.88 6.44 62 EY012A 0.6174 -1.3909 0.4907 0.2601 273 30.63 63 EY016A -0.7396 0.6285 0.5987 0.034 198.24 3.75 64 EY018A -0.612 0.5745 0.5433 -0.0618 57.07 3.94 65 FA 001A -0.7879 0.824 1.5901 0.1218 45.05 4.62 66 FG 001U -0.6626 0.5563 0.5383 -0.0425 355.41 4.36 67 FR 001A 2.4683 1.9751 -0.3209 -1.8172 20 5.13 68 FR 002A 2.3956 2.0492 -0.3677 -1.7906 58 7.28 69 FR 005A 1.0928 -2.1918 1.0756 0.5205 1 1 70 FR 057A -0.8389 0.5749 -0.4264 0.6227 2900.45 5.12 71 FR060A -0.5073 0.0133 -0.7454 -0.1092 213.25 4.01 72 FU 002A -0.7804 0.5932 0.6634 0.0289 39.05 3.01 73 GO 004A -0.2058 0.27 -0.4952 -0.265 1514.56 6.54 74 GO 014A 0.8242 -1.5903 0.695 0.3593 24 7.38 75 GO 025B -1.0279 1.1357 1.5819 0.2653 36.04 3.01 76 GO 025H -0.6699 0.574 0.5776 -0.0519 387.46 4.24 77 GP 009S -1.0095 1.1187 1.5271 0.2518 108.13 3.12 78 GY 005A -0.6916 0.9998 1.0701 -0.1482 93.11 3.19 79 MA 001A 0.8665 -0.9768 0.3903 0.2152 1457.1 54.89 80 MA 001B 0.1394 -0.2507 -0.1742 0.0759 17748.36 14.19 81 MA 001C -0.5945 0.3758 0.6917 0.1058 283.3 4.73 82 MA 001D -0.7544 0.553 0.6258 0.0439 189.23 3.97 83 MA 002B -0.443 -0.0353 -0.8168 -0.1179 282.34 4.01 84 MA 002C 1.4717 -0.3983 0.2855 0.6429 667 29.73 85 MU C001 0.4557 -1.4456 0.532 0.2173 1 1 86 MU C003 0.9324 -1.5583 0.578 0.5261 3 1.8 87 MU C004 0.6423 -1.7357 0.6348 0.3615 75 4.8 88 MU C005 0.748 -1.6788 0.8754 0.2553 1 1 89 MU C006 0.2138 -1.0619 -0.0557 0.2384 1 1 90 MU C007 0.7862 -1.4958 0.6976 0.2273 14 8.91 91 MU C008 1.0404 -2.3492 1.1901 0.5223 1 1 92 MU C009 0.9413 -2.0801 1.0399 0.4152 6 1.8 93 MU C010 0.9901 -2.0264 0.9511 0.5718 59 2.19 94 MU C011 0.9684 -2.2578 1.1 0.4992 7 1.32 95 MU C012 0.7882 -2.0293 0.8745 0.4416 2 2 96 MU C013 1.0404 -2.3492 1.1901 0.5223 1 1 97 NA 001P -1.1089 1.4686 3.6555 0.4376 51.06 3.01 98 NA 003A 0.1356 -0.5813 0.1654 0.0382 2912.59 23.24 99 NA 003B 2.002 1.4379 -0.193 -1.5474 40 8.79 100 NA 008C -1.0155 1.1173 1.5366 0.2444 648.77 3.14 101 NA009P -0.8007 0.6552 0.8177 0.0541 63.08 3.01 102 NA 009S -0.6111 0.4986 0.4679 -0.0108 150.18 3.9 103 NA 01 OA 2.5373 2.1895 -0.2922 -2.1083 9 3.52 104 NA 014A 2.1166 -0.0041 0.4946 -0.6969 137 11.21 105 NA 021A 0.5361 -1.7094 0.559 0.3609 3 1 106 NA 022A 0.5361 -1.7094 0.559 0.3609 8 1 107 NA 039A 0.5361 -1.7094 0.559 0.3609 1 1 108 NA 047A 0.1308 -1.0483 0.4001 0.1075 1 1 109 NA 050A 0.5939 -1.5765 0.6207 0.2998 35 6.35 110 NA 051A 0.9396 -2.2212 1.0639 0.49 5 1.47 111 NA 056A 2.0485 0.7332 0.0251 0.3027 59 18.61 112 NA 058A -0.4288 0.0974 -0.7686 -0.2565 12.01 3.01 113 NA 066B -0.092 0.1962 -0.6854 -0.3449 61.06 4.97 114 NA 067A 2.7136 1.9025 -0.1187 -1.9364 25 4.31 115 NA 102A 0.0076 -0.0793 -0.3934 0.0486 80.07 6.95 116 NA 123A -0.7146 0.4689 0.4843 -0.0263 27.03 3.01 117 NA 144A -0.6551 0.5085 0.4646 -0.0251 108.13 3.87 118 NA 267A -0.8954 0.9985 1.0517 -0.1029 90.11 3.01 119 NA 365A -1.1089 1.4686 3.6555 0.4376 246.29 3.01 120 NA 418A 1.0404 -2.3492 1.1901 0.5223 1 1 121 NA 650A 0.7936 -1.1391 0.5725 0.144 20 6.67 122 NA 751A -0.4511 0.2154 0.6071 0.0281 1089.05 7.06 123 NE 001A -0.3816 -0.0077 -0.6672 -0.2042 85.1 3.58 124 Nl003A -0.953 0.8501 0.855 0.4665 93.11 3.01 125 Nl005S -0.6569 0.3799 0.5462 0.0825 78.09 3.01 126 Nl008A -0.0798 0.3847 -0.21 -0.2127 212.21 7.53 127 Nl008N -0.7804 0.5932 0.6634 0.0289 111.13 3.01 128 Nl009A -0.6644 0.5251 0.376 -0.0658 2250.66 5.55 129 Nl009S -0.8954 0.9985 1.0517 -0.1029 63.08 3.01 130 Nl009V -0.6735 0.9977 1.0261 -0.1606 39.05 3.01 131 Nl 011S -1.1089 1.4686 3.6555 0.4376 27.03 3.01 132 Nl 014A 1.7156 1.2619 -0.2268 0.0322 4925.39 36.06 133 Nl017A -0.5947 0.2149 -0.1203 -0.0789 847.01 5.46 134 Nl 023A 2.3519 1.4597 0.064 -1.2841 1 1 135 Nl 065A -1.1089 1.4686 3.6555 0.4376 24.03 3.01 136 PI 005A 0.59 0.4983 -0.5435 -0.0159 60.04 9.29 137 PI 007A 0.9064 -1.9214 0.815 0.4424 3 1.8 138 PI 008A -0.0518 -0.212 -0.3962 -0.1169 152.14 6.13 139 PI 009S -0.6946 0.4854 0.4583 0.1014 276.33 3.94 140 PI 170A 3.0143 1.0853 0.5764 -2.1605 12 1 141 PS 001A -0.8816 0.4782 -1.1518 0.1757 33.04 3.75 142 RH 001A -0.3617 -0.1697 -0.519 -0.0738 105.13 3.56 143 SL 001A -0.0029 -0.628 -0.0936 -0.002 361.26 10.58 144 SP 009H -0.6735 0.9977 1.0261 -0.1606 75.09 3.01 145 SP012S -1.1089 1.4686 3.6555 0.4376 225.27 3.01 146 SR 001A -0.399 0.0258 -0.7778 -0.1448 132.16 3.32 147 SR 103A -0.8007 0.6552 0.8177 0.0541 27.03 3.01 148 SS 001C -0.3812 1.0464 0.8891 -0.2057 448.48 4.57 149 SS 002A -0.6117 0.1161 -1.405 -0.0958 21.03 3.48 150 ST 009J -0.8007 0.6552 0.8177 0.0541 27.03 3.01 151 ST 021A -0.5918 0.0177 -0.8275 -0.2177 138.16 3.01 152 SU 024B -0.8954 0.9985 1.0517 -0.1029 27.03 3.01 153 SV 005C -0.8007 0.6552 0.8177 0.0541 12.01 3.01 154 SV 009P -0.8074 0.7095 1.0144 0.1018 72.09 3.49 155 SV 020C -0.78 0.6214 0.7222 0.0824 48.06 3.51 156 SY 001A -0.5121 -0.0046 -0.6125 -0.1598 711.85 4.18 157 SY 003A -0.4531 0.623 0.4593 -0.185 61.07 3.39 158 UN 009S -0.8954 0.9985 1.0517 -0.1029 81.1 3.01 159 U NID1 2.8027 1.6488 0.1039 -2.0119 6 3.6 160 UN ID17 2.2525 0.8857 0.2064 -1.415 8 5.33 161 U NID2 2.201 -0.7226 0.84 -0.1884 1 1 162 UN ID23 3.0143 1.0853 0.5764 -2.1605 1 1 163 UN ID24 2.274 1.8168 -0.3047 -1.2279 2 1 164 U NID3 1.1107 -1.2724 0.7458 0.1811 2 1 165 UN ID33 0.3993 -1.2383 0.4317 0.1659 493 24.14 166 U NID4 3.0143 1.0853 0.5764 -2.1605 5 1 167 U NID5 1.6663 -1.2728 0.9904 -0.3454 1 1 168 U NID6 1.6663 -1.2728 0.9904 -0.3454 1 1 169 U NID7 1.9656 -0.1543 0.4146 0.6607 4 2.67 170 U NID9 2.3894 1.3054 -0.18 0.7021 11 4.17 171 XA 009X -0.8187 0.2568 -2.117 0.1242 279.33 3.18 172 XC 008X -0.8405 0.352 -1.5235 0.369 27.03 3.6 174 XCH 001Z 2.2595 1.7348 -0.1893 -1.7249 74 6.49 175 XCI 001Z 2.0221 1.4441 -0.1618 -1.6893 3 1.8 176 XCP001Z 1.3685 0.2609 -0.2499 -0.3095 1 1 177 XCT 001Z 2.7313 2.379 -0.3239 -2.2926 3 1 178 XEU 001Z 2.1968 1.6006 -0.2314 -1.5403 113 13.2 179 XF 007X -0.5239 0.0594 -1.079 -0.1811 48.06 3.01 180 XF 008X -0.5239 0.0594 -1.079 -0.1811 42.05 3.01 181 XF 009X -0.4613 0.0107 -0.8265 -0.1171 228.27 4.01 182 XH 009X -0.4965 -0.0242 -0.659 -0.1616 1168.39 4.36 183 XI006X -0.8481 0.4272 -0.9662 0.5701 24.03 3.01 184 XI007X -0.4231 -0.1075 -0.4575 -0.0971 63.08 4.05 185 XI008X -0.4202 -0.0772 -0.586 -0.1446 18.02 3.61 186 XI009X -0.3917 -0.0634 -0.6804 -0.1184 1538.8 4.38 187 XM 005X -0.5109 -0.0089 -0.3988 -0.1356 18.02 3.01 188 XM 006X -0.4398 0.0783 -0.7751 -0.2347 30.04 3.39 189 XM 007X -0.9647 0.3607 -2.5398 0.4576 36 2.62 190 XM 008X -0.3543 -0.2654 -0.7377 -0.1572 132 2.87 191 XM 009A -0.513 -0.2145 -1.1818 -0.3001 234.01 2.72 192 XM 009X -0.3039 -0.1528 -0.9456 -0.2099 26 2.66 193 XN 005X -0.4104 -0.215 -0.8439 -0.1934 597.03 3.84 194 XN 006X -0.4285 -0.0556 -0.2677 -0.0573 50.11 2.41 195 XN 008X -0.5359 -0.1255 -1.1117 -0.2862 50 2.93 196 XN 009X -0.5627 0.2136 -0.1949 -0.0376 320.05 1.19 197 XN 019X -0.7746 0.344 -0.9919 0.3263 63.11 2.94 198 XNA 002Z 0.5123 -1.2746 0.5398 0.2381 337.21 16.14 199 XNB001Z -0.4281 0.3246 -0.1311 -0.076 786.66 1.12 200 XND001Z -0.1361 0.4203 -0.1334 -0.2281 16.05 1.45 201 XNF001Z -0.5781 0.2474 -0.1208 -0.0352 392.39 1.01 202 XNG 001Z 1.3885 -1.2726 0.8681 -0.0822 2 2 203 XNM 001Z 0.2494 0.5998 -0.1024 -0.5669 18.04 2.09 204 XNM 003Z -0.5136 0.2255 -0.0988 -0.0375 39.14 1.06 205 XNS 001Z 3.0143 1.0853 0.5764 -2.1605 1 1 206 XNV 008Z 1.7652 1.7022 -0.3242 -1.4701 45.04 5.19 207 XO 032X -0.6945 0.2849 -0.754 0.1466 109.26 2.07 208 XO 037X -0.6368 0.2636 -0.4701 0.0288 87.27 1.34 209 XP 014X -0.8806 0.2596 -2.6406 0.1478 45.03 2.78 210 XPL 001Z 2.8496 2.4255 -0.7274 3.065 7 2.33 211 XPN 001Z 2.1193 1.6533 -0.2681 -1.4469 141 10.5 212 XS 005X -0.8284 0.2629 -2.1461 0.1354 387.46 3.12 213 XS 006X -0.82 0.272 -1.9691 0.1738 33.04 3.36 214 XS 007X -0.8078 0.3211 -1.4047 0.331 84.1 3.83 215 XS 014X -0.3686 -0.143 -0.4648 -0.1187 24.03 3.48 216 XS 015X -0.4047 -0.0608 -0.6885 -0.1487 216.26 4.05 217 XS 016X -0.8327 0.2734 -2.1061 0.1587 33.04 3.18 218 XS017X -0.7435 0.2719 -1.0726 0.2892 42.05 3.64 219 XS 026X -0.4421 -0.0661 -0.6475 -0.1258 21.03 3.27 220 XSC015Z 2.7008 1.353 0.2007 -1.3864 247 4.59 221 XSS 001Z 2.3968 1.8691 -0.2997 -1.6661 255 13.21 222 XU 001X -0.4919 -0.0639 -0.4701 -0.161 18.02 3.01 223 XU 008X -0.5069 0.0028 -0.8687 -0.1539 666.79 4 224 XU 009X -0.4294 -0.0487 -0.602 -0.1722 27.03 3.91 225 XV 009X -0.5866 0.0383 -0.7864 -0.1342 105.13 3.55 226 XY 009X -0.8481 0.4272 -0.9662 0.5701 18.02 3.01 Samples N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.4992 0.3915 0.2973 0.2528 1 A -0.1018 -0.3167 -0.7929 -0.1438 396 7.44 2 B -0.5927 -0.0925 -0.8246 -0.3537 458 4.52 3 C -0.3253 -0.3318 -0.646 -0.1702 424 4.7 4 D -0.6327 -0.1056 -0.8633 -0.3641 445 4.65 5 F 0.0023 -0.6027 -0.2336 -0.1799 419 6.47 6 G -0.5776 -0.1112 -0.8148 -0.2964 443 4.57 7 H -0.4594 -0.1217 -0.8675 -0.2966 442 5.58 8 I 0.5555 -1.5482 0.6147 0.1824 403 3 9 J -0.1905 -0.4695 -1.2122 -0.1417 396 4.48 10 K -0.3845 -0.2726 -0.7081 -0.205 436 5.1 11 L -0.6323 -0.0054 -0.4592 -0.3128 416 2.46 12 M -0.4641 -0.1358 -0.8366 -0.298 417 5 13 O -0.2386 -0.05 -0.8017 -0.2235 413 6.11 14 P -0.1073 -0.0187 -1.2936 -0.2316 401 4.73 15 Q -0.3894 0.1053 -0.8945 -0.3019 463 4.59 16 R -0.4866 0.2224 -0.9613 -0.3658 401 3.66 17 S -0.0814 -0.4772 -0.7316 0.025 444 2.83 18 U -0.0947 -0.4496 -0.7121 -0.0071 446 3.62 19 V -0.2678 0.1947 -1.2572 -0.5953 403 14.34 20 X -0.1186 -0.2963 -0.9245 0.045 400 2.43 21 Y -0.6096 0.2316 -0.4702 -0.1897 403 6.64 22 Z -0.5559 -0.0463 -0.9781 -0.3447 368 7.52 23 AA -0.6549 -0.0658 -1.6524 -0.4691 400 8.75 24 AC 0.0591 -0.7235 -0.3507 0.0376 389 4.9 25 AD -0.0935 -0.7005 0.0476 -0.0731 426 5.3 26 AE -0.5142 -0.1248 -0.1464 -0.2328 414 2.8 27 AF -0.7574 -0.0447 -1.434 -0.4797 357 5.26 28 AG -0.3741 -0.2649 -0.4046 -0.1166 425 6.13 29 AH -0.2179 -0.4768 -0.6187 -0.198 442 6.17 30 Al -0.3123 -0.266 -0.5192 -0.1679 409 7.06 31 AJ -0.4455 -0.2073 -0.7246 -0.2176 225 4.75 32 AL -0.474 -0.23 -0.5934 -0.2457 429 3.13 33 AM -0.4563 -0.0554 -0.3205 -0.1735 433 4.51 34 AN -0.238 -0.328 -0.7772 -0.1515 436 6.13 35 AP -0.7262 -0.1074 -1.591 -0.5121 446 3.72 36 AQ -0.5367 -0.3841 -1.4282 -0.4563 528 9.47 37 AS -0.2394 -0.0342 -0.7257 -0.2613 385 7.24 38 AU -0.7884 -0.085 -1.4053 -0.4821 424 3.05 39 AX -0.4572 -0.29 -0.3603 -0.3087 406 3.77 40 AY -0.5535 0.0804 -2.19 -0.3692 287 11.87 41 AZ -1.4761 0.6772 -5.6171 0.5286 384 6.05 42 BA -1.5269 1.185 -1.8503 1.8869 464 1.96 43 BB -1.0554 0.629 -1.0921 1.0274 396 3.61 44 5L 1.1108 -1.2724 0.7458 0.1812 401 3.23 45 46L 1.9408 -1.4603 1.1434 -0.1223 435 2.28 46 51L 2.2009 -0.7226 0.8401 -0.1884 401 3.62 47 56L 2.0409 -0.8927 0.9367 -0.7072 424 3.17 48 61L 1.6663 -1.2728 0.9905 -0.3454 420 3.1 49 67L 3.0144 1.0853 0.5766 -2.162 429 7.9 50 71L 1.3382 -2.0673 1.2587 0.3123 406 1.5 51 76L 2.352 1.4597 0.0641 -1.2842 422 8.06 52 81L 1.8154 1.1598 -0.1766 -1.0743 437 8.66 53 86L 2.279 2.0869 -0.2099 -2.3256 401 9.63 54 91L 1.7785 1.4619 -0.2644 -1.7762 412 10.42 55 96L 3.0375 2.5258 -0.293 -2.4282 433 6.53 56 101L 2.7314 2.3791 -0.3238 -2.2924 406 8.1 57 106L 2.5093 1.4086 0.0441 -1.5145 400 9.36 58 111L 1.3686 0.2609 -0.2497 -0.3098 430 3.84 59 116L 2.2741 1.8168 -0.3046 -1.2277 426 6.76 60 121L 2.2739 2.0486 -0.2679 -1.5458 410 6.88 61 126L 2.0409 1.6731 -0.3157 -1.4958 402 8.37 62 132L 2.6009 2.2173 -0.365 -1.9751 408 6.97 63 139L 3.1817 2.7706 -0.3514 -2.5238 404 5.15 64 144L 2.6475 2.7087 -0.5504 -2.3539 447 5.86 65 149L 2.8411 2.5429 -0.5047 -1.3311 418 3.63 66 154L 1.3905 1.0272 -0.3105 -1.2227 417 9.15 67 159L 1.9661 1.5465 -0.4248 -1.3344 411 7.28 68 164L 2.0065 0.746 -0.1896 -0.8861 398 4.37 69 168L 1.6917 0.537 -0.154 -0.8902 415 4.16 70 189L 1.311 -0.8929 0.3044 0.3707 411 3.84 71 194L 2.5867 1.3296 -0.1697 0.2513 279 2.66 72 199L 1.9624 -0.1981 0.3645 0.2977 353 3.39 73 204L 1.3094 -0.6723 0.189 0.1042 451 3.73 74 209L 1.9828 -0.2999 0.5334 0.0508 415 3.28 75 215L 2.0813 0.3887 0.3183 -0.2929 427 5 76 219L 2.5897 2.1453 0.0626 -0.424 280 3.99 77 228L 2.8023 2.1699 -0.5614 0.2942 267 1.89 78 238L 3.3803 3.2928 -1.7454 11.1812 401 3.45 79 248L 1.6598 -0.8449 0.6499 0.1612 412 2.86 80 253L 2.6494 2.0618 -0.4162 -0.1209 303 3.22 81 263L 2.0606 -0.0806 0.345 1.3874 217 3.79 82 269L 3.6516 3.71 -1.5209 8.064 242 2.15 83 289L 2.2386 1.8488 -1.3191 7.1507 389 3.1 84 294L 2.0791 1.0779 -0.6439 3.0725 322 3.94 85 10H 1.0928 -2.1918 1.0756 0.5204 315 2.53 86 90H 1.3255 -1.7599 1.0332 0.7559 400 3.49 87 113H 0.9269 -1.7459 0.8451 0.3875 400 2.99 88 123H 1.0403 -2.3492 1.1901 0.5223 437 3.2 89 132H 0.8803 -1.6263 0.806 0.1956 437 3.24 90 143H 0.7851 -1.3713 0.6267 0.2412 425 5 91 203H 1.2044 -1.9126 1.0326 0.5263 411 2.74 92 270H 1.2676 -2.3489 1.2385 0.7289 400 2.69 93 290H 0.672 -1.3087 0.4272 0.0399 419 5.18 94 31 OH 0.454 -1.1198 0.1913 0.1416 494 4.13 95 338.5H 0.536 -1.7094 0.559 0.3611 402 5.96 96 358.5H 1.0425 -1.4372 0.4795 0.6251 446 4.72 97 378.5H 0.6332 -1.506 0.6003 0.228 396 4.54 98 398.5H 0.6763 -1.6661 0.6399 0.3021 401 4.18 99 418.5H 0.8237 -1.4276 0.5933 0.0971 405 5.23 100 420H 0.5086 -1.3386 0.2987 0.2713 390 3.96 101 440H 0.5752 -1.375 0.471 0.2146 399 4.27 102 460H 0.3499 -1.122 0.15 0.2092 442 4.29 103 480H 0.709 -1.5798 0.803 0.1679 400 3.53 104 500H 0.7431 -1.5418 0.7396 0.2009 396 5.32 105 524H 0.7118 -1.8005 0.775 0.3275 400 3.73 106 544H 0.4557 -1.4456 0.532 0.2173 409 5.39 107 564H 0.5944 -1.3408 0.3595 0.3289 432 5.63 108 584H 0.8553 -1.5125 0.5063 0.371 407 4.53 109 604H 0.3262 -0.9493 0.0993 0.1999 396 4.72 110 630H 0.5335 -1.3805 0.2939 0.286 435 4.31 111 650H 0.1478 -0.9226 0.1365 -0.038 416 3.46 112 670H 0.2899 -0.7574 -0.1817 0.1069 459 2.68 113 690H 0.1058 -0.7618 -0.0614 0.1491 397 3.66 114 71 OH 0.668 -1.0908 0.594 -0.1347 430 7.01 115 733H 0.4932 -1.3687 0.5771 0.2657 398 5.05 116 753H 0.4811 -1.4275 0.7975 0.2574 450 4.5 117 773H 0.453 -1.378 0.5264 0.249 440 5.86 118 793H 0.4583 -1.3282 0.4902 0.1344 396 6.61 119 801,5H 0.574 -1.6345 0.8328 0.2285 410 3.22 120 821,5H 0.3971 -1.2632 0.3741 0.2067 434 5.38 121 841,5H 0.5443 -1.3265 0.6011 0.2071 420 4.9 122 861,5H 0.4099 -1.2678 0.775 0.113 391 3.83 123 881.5H 0.2081 -1.106 0.4893 0.0272 404 5.76 124 920H 0.4955 -1.0364 0.1284 0.0793 470 3.99 125 931,5H 0.2973 -1.0422 0.0664 0.1622 403 3.61 126 941,5H 0.2137 -1.0619 -0.0558 0.2384 405 3.32 127 961,5H 0.8778 -1.7888 0.979 0.2478 406 2.78 128 981,5H 0.1308 -1.0483 0.4 0.1076 406 4.85 129 1001.5H 0.188 -0.9916 0.1484 0.0823 408 5.56 130 1021.5H 0.5796 -1.4654 0.7593 0.188 424 4.31 131 1048H 0.8624 -1.9587 1.0599 0.3857 427 2.35 132 1068H 0.4156 -1.1819 0.4726 0.1787 461 5.51 133 1088H 0.4773 -1.3117 0.5209 0.162 419 4.49 134 1108H 0.8515 -1.58 0.6465 0.3735 422 3.88 135 1128H 0.7479 -1.6788 0.8752 0.2557 398 3.23 136 1340H 0.3991 -1.3069 0.3483 0.2293 400 5.16 137 1345H 0.4451 -1.385 0.5226 0.2422 457 5.22 138 1350H 0.4432 -1.3369 0.4384 0.2193 401 4.34 139 1355H 0.339 -1.1481 0.1968 0.2069 453 4.71 140 1360H 0.1767 -1.0574 0.2617 0.1261 437 5.88 141 1365H 0.4358 -1.348 0.423 0.2063 400 4.51 142 1370H 0.4658 -1.3431 0.512 0.26 422 4.93 143 1375H 0.6429 -1.6102 0.6363 0.3014 433 3.7 144 BS -0.5755 0.0352 0.5921 0.1402 417 4.99 145 BE -0.6091 -0.0114 -0.086 -0.1356 402 2.22 146 PP -1.292 1.8932 2.1102 -0.4177 359 2.85 147 LS -1.0074 0.862 1.4077 0.0549 385 8.72 148 LE -0.9464 0.6757 0.9441 -0.0208 385 9 149 OS -0.8089 0.5372 0.4889 0.3264 377 12.14 150 OP -0.6332 0.4169 0.4331 0.5069 364 4.88 151 ME -1.3278 1.4837 -1.2977 3.0897 357 1.35 152 ON -1.4648 1.4474 1.5194 1.2937 426 2.42 153 OJ -1.9331 3.3052 9.9313 1.2077 346 7.68 154 OE -0.5233 0.7007 0.889 0.3493 271 5.72 155 OD -0.7777 0.4929 0.3755 0.136 163 2.12 156 XA -0.6975 1.3895 0.745 -0.0388 394 6.45 157 XB -1.1868 1.9551 3.3986 -0.2072 364 7.62 158 XC 0.0459 -0.6452 0.413 0.0408 401 4.27 159 XD -0.3515 0.397 0.223 -0.1808 410 2.72 160 XE -0.4692 0.3908 0.3074 -0.1924 396 4.2 161 XF -0.5383 0.4572 0.4719 -0.213 377 5.29 162 XG -0.3242 1.1178 0.7318 -0.6297 400 8.93 163 XH -0.6254 1.8908 2.0334 -0.5905 379 11.65 164 XI -1.6897 2.3052 3.7033 0.6889 403 3.56 165 XDP -0.365 -0.0929 0.1023 -0.08 327 2.81 166 XHP -0.7083 0.7558 0.6468 -0.1614 287 5.45 167 XBP -0.9461 0.7695 1.1195 -0.173 396 5.84 168 XP -1.0568 0.7767 0.7572 -0.1968 279 6.35 169 XPP -0.5555 0.2213 0.1728 -0.1863 387 2.17 170 XEP -0.7487 0.3025 0.406 -0.1869 392 3.2 171 XRP -0.6517 0.1246 0.2755 -0.1903 381 3.66 New River Lagoon (Figure 5.18) Species N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.3169 0.2328 0.1212 0.1159 1 AC 001A 1.0639 0.4796 -0.0366 0.0691 3.01 3.02 2 AC 008A 1.2864 0.4965 -0.0261 0.0919 2887.98 4.5 3 AC 009A -1.0027 1.5623 0.0788 1.9238 499.92 4.18 4 AC 013A 0.7403 0.0566 0.0126 0.0983 243.62 5.2 5 AC 049A -0.3976 -0.7944 0.0371 -0.0085 3.01 3.01 6 AC 134A -0.8768 1.3106 -0.1057 1.1438 78.3 3.75 7 AC 160A 1.2247 0.3732 -0.0025 0.1064 3.01 3.02 8 AD 009A -0.879 1.3148 -0.3419 -0.4829 2535.71 4.37 9 AM 001A 1.2492 0.1685 0.0019 0.1778 9.02 3.02 10 AM 008A -0.832 1.7124 7.644 -1.0983 3.01 3.02 11 AM 094B 1.2858 0.3713 -0.0002 0.0984 213.58 4.24 12 AM 099X 0.5721 -0.1552 0.046 0.0942 6.02 3.02 13 AM 9990 1.2492 0.1685 0.0019 0.1778 15.04 3.02 14 AN 009A 1.0993 0.1005 0.0039 0.03 69.19 4.32 15 BR 01 OA -0.477 0.4035 -0.0389 0.1461 20732.05 6.33 16 CA 006A -0.3913 -0.6686 0.029 0.0276 6.01 3.61 17 CH 009X 1.3562 0.4915 -0.0277 0.1146 45.12 3.81 18 CM 004A 0.1672 0.8703 -0.0594 0.6771 105.34 5.88 19 CM 011A -0.2718 -0.9652 0.071 -0.0205 6.01 3.01 20 CM 023A -0.1724 -0.4213 -0.0089 0.007 18.04 4.4 21 CO 001A 1.0421 0.495 -0.0268 0.1795 75.21 4.63 22 CO 001B -0.9545 1.5589 -0.0783 1.9166 18.07 3.33 23 CO 001C 1.1613 0.4093 -0.0195 0.083 57.16 4.24 24 CY 003A 1.1593 0.4861 -0.0285 0.1449 12.03 3.81 25 CY 004A 0.6422 -0.0389 0.0048 0.1055 39.1 5.08 26 CY 028A 0.7931 0.0183 0.0061 0.0686 5299.72 5.28 27 CY 035A -0.3866 -0.6151 0.0105 -0.0123 171.34 4.21 28 DE 001A 1.4709 0.3715 -0.0967 -0.2939 1179.22 3.98 29 DE 002A -0.0667 -0.4897 0.0354 0.0247 33706.76 5.75 30 DE 002B -0.832 1.7124 7.644 -1.0983 277.07 3.02 31 DE 003A 1.1423 0.4701 -0.0106 0.0789 6.02 3.02 32 DP 001A -0.0568 -0.2691 0.3081 0.0748 159.38 5.87 33 DP 001B -0.3847 -0.6165 -0.0137 -0.1263 3.01 3.01 34 DP 007A 0.324 0.3372 -0.0879 -0.1288 99.29 5.98 35 DP 009A 0.7355 -0.0048 -0.0065 -0.0227 9.02 4.79 36 DP 01 OA 0.0639 0.1066 -0.1737 -0.722 12.03 5.17 37 EC 001A -0.3538 -0.0603 -0.0194 0.0213 18118.25 6.01 38 EP 003A 1.1197 0.2736 -0.0033 0.0711 9.02 3.88 39 EP 004A -0.3334 -0.9763 0.0843 0.047 3.01 3.01 40 EU 002A -0.824 1.2663 -0.558 -1.9564 123.47 4.01 41 EU 009C -0.8906 1.344 -0.1206 1.1086 18.07 3.55 42 EU 009I -0.832 1.7124 7.644 -1.0983 3.01 3.02 43 EU 009S 0.8406 0.3449 -0.0185 0.1723 306.85 5.33 44 EU 017A -0.8808 1.4029 -0.804 -3.0253 3.01 3.02 45 EU 047A -0.7917 1.3978 -0.5503 -2.2111 21.08 3.8 46 EU 048A -0.2933 -0.7051 0.0419 0.0435 3.01 3.01 47 EU 999X 0.5721 -0.1552 0.046 0.0942 6.02 3.02 48 EY 01 OA 0.5753 0.5351 0.0889 -0.1067 520.55 5.62 49 EY012A 50 EY018A 51 FA 001A 52 FG 001U 53 FR 001A 54 FR 002A 55 FR 005A 56 FR 057A 57 GO 004A 58 GO014A 59 GO 025H 60 MA 001A 61 MA 001B 62 MA 001C 63 MA 002B 64 MA 002C 65 MU C001 66 MU C003 67 MU C004 68 MU C005 69 MU C006 70 MU C007 71 MU C008 72 MU C009 73 MU C010 74 MU C011 75 MU C012 76 MU C013 77 NA 003A 78 NA 003B 79 NA 008C 80 NA 009S 81 NA 01 OA 82 NA 014A 83 NA 021A 84 NA 022A 85 NA 039A 86 NA 047A 87 NA 050A 88 NA 051A 89 NA 056A 90 NA 066B 91 NA 067A 92 NA 102A 93 NA 123A 94 NA 144A 95 NA 418A 96 NA 650A 97 NA 751A 98 NE 001A 99 Nl008A 100 NI 009A 101 Nl014A 102 Nl017A -0.3398 -0.8725 -0.7525 -0.8921 1.2799 1.2852 -0.2718 -0.533 0.0372 -0.3032 -0.9397 -0.022 -0.0519 -0.2509 -0.8362 0.3481 -0.3986 -0.2638 -0.3646 -0.3483 -0.3847 -0.3109 -0.3334 -0.3346 -0.2988 -0.3426 -0.3655 -0.3334 -0.2866 1.1131 -0.9478 -0.8399 1.3376 0.9789 -0.3976 -0.3976 -0.3976 -0.502 -0.369 -0.3462 0.9303 0.2163 1.3457 0.1343 -1.0121 -0.8966 -0.3334 -0.1691 -0.5522 -0.7028 0.3098 -0.6997 1.1443 -1.1182 -0.6894 1.2781 1.4555 1.568 0.3996 0.4541 -0.9652 0.071 0.8579 -0.7703 1.3588 -0.5057 -0.1298 -0.7886 1.3313 -0.3297 -0.6959 -0.7731 -0.7996 -0.7283 -0.6165 -0.7279 -0.9763 -0.9019 -0.8614 -0.9503 -0.8853 -0.9763 -0.3999 0.3976 1.3374 1.2397 0.5176 0.0037 -0.7944 -0.7944 -0.7944 -0.436 -0.7174 -0.9399 0.0947 0.8955 0.4084 0.0096 1.4438 1.3088 -0.9763 -0.5236 -0.0113 0.7017 0.8642 1.519 0.3847 1.9684 0.0291 -0.117 -0.5603 -0.039 -0.0299 -0.0313 0.071 -0.1594 -0.0248 0.045 -0.096 0.022 -0.0312 0.0486 -0.59 0.0133 0.0482 0.0351 0.0436 0.0567 -0.0137 0.0509 0.0843 0.0756 0.06 0.0776 0.0607 0.0843 0.0924 -0.0238 -0.1059 -0.1143 -0.0218 0.0207 0.0371 0.0371 0.0371 0.0113 0.04 0.0749 -0.0025 -0.3531 -0.0142 -0.0692 -0.0968 -0.1175 0.0843 0.0393 0.0179 -0.2645 0.9517 0.2395 -0.0008 -0.0269 0.0077 0.9756 -2.6471 2.1961 0.1017 0.1011 -0.0205 -0.7712 -0.2758 0.0122 1.2859 0.0138 -0.1338 -0.0163 -2.0679 -0.0152 0.0495 -0.0524 -0.0078 0.1139 -0.1263 0.0515 0.047 0.054 -0.0037 0.0391 0.0192 0.047 0.0317 0.0961 1.0934 0.7788 0.1543 0.0861 -0.0085 -0.0085 -0.0085 0.1802 0.0333 0.0359 0.038 -1.2379 0.1299 -0.2063 1.4754 0.9602 0.047 0.0921 0.371 -0.9332 -0.8139 1.8708 0.0083 3.1829 820.65 42.16 3.01 292.1 60.16 174.47 3.01 9.02 1315.35 72.14 277.07 4131.04 26384.86 84.17 84.32 2005.67 3.01 9.02 225.45 3.01 3.01 42.08 3.01 18.04 177.36 21.04 6.01 3.01 5264.89 120.33 6.02 111.43 27.07 412.13 9.02 24.05 3.01 3.01 105.21 15.03 177.47 42.14 75.21 72.19 27.1 84.32 3.01 60.13 953.51 24.09 186.6 313.15 11916.25 255.98 4.45 3.67 3.02 4.38 4.31 4.3 3.01 4.29 5.95 4.23 4.12 5.5 6.3 4.03 4 5.87 3.01 3.54 4.09 3.01 3.01 4.28 3.01 3.54 3.68 3.28 3.61 3.01 5.36 4.28 3.63 3.61 3.96 4.33 3.01 3.01 3.01 3.01 4.19 3.37 5.04 5.22 4.05 6.25 3.02 3.68 3.01 4.77 5.68 4.46 5.59 4.76 4.69 3.71 103 Nl023A 104 PI005A 105 PI007A 106 PI008A 107 PI009S 108 PI170A 109 SL 001A 110 SR 001A 111 SS001C 112 SV 009P 113 SY 001A 114 SY 003A 115 U NID1 116 UN ID17 117 U NID2 118 UN ID23 119 UN ID24 120 U NID3 121 UN ID33 122 U NID4 123 U NID5 124 U NID6 125 U NID7 126 U NID9 127 XCH 001Z 128 XCI 001Z 129 XCP 001Z 130 XCT 001Z 131 XEU001Z 132 XH 009X 133 XI 009X 134 XN 005X 135 XN 006X 136 XNA 002Z 137 XNB001Z 138 XND 001Z 139 XNF001Z 140 XNG 001Z 141 XNM001Z 142 XNM 003Z 143 XNS 001Z 144 XNV 008Z 145 XPL 001Z 146 XPN 001Z 147 XS 014X 148 XS 015X 149 XSC 015Z 150 XSS 001Z 151 XU 009X 1.2247 1.051 -0.291 0.1534 -0.832 1.2492 -0.4418 -0.7573 1.0982 -0.832 -0.9344 -0.5259 1.3136 1.1125 0.9112 1.2492 1.2426 0.5721 -0.3962 1.2492 0.6886 0.6886 0.9324 1.2047 1.2194 1.1215 0.8306 1.4102 1.1901 -0.8675 -0.7886 -0.8316 -0.7965 -0.3283 1.3087 1.0593 0.8529 0.6303 0.9536 0.9112 1.2492 1.3593 1.4649 1.1727 -0.7901 -0.8322 1.2604 1.2753 -0.7789 0.3732 0.2945 -0.8832 -0.2153 1.7124 0.1685 -0.3362 1.4599 0.5163 1.7124 1.3664 1.2611 0.3333 0.2353 -0.171 0.1685 0.4382 -0.1552 -0.5987 0.1685 -0.224 -0.224 -0.0298 0.2977 0.4449 0.4214 0.1239 0.5347 0.4042 1.2077 1.1939 1.0459 0.8579 -0.7574 0.3513 0.3714 -0.1887 -0.1896 0.3002 -0.171 0.1685 0.4688 0.4499 0.4344 0.9976 1.2599 0.2676 0.426 0.9881 -0.0025 -0.0251 0.0555 -0.0406 7.644 0.0019 -0.0472 -0.5571 -0.0007 7.644 -0.3464 -0.0879 -0.0112 0.0014 0.0436 0.0019 -0.0216 0.046 0.0204 0.0019 0.0424 0.0424 0.0462 -0.0009 -0.0214 -0.0289 -0.0443 -0.0222 -0.0241 -0.3106 -0.5329 -0.2492 -0.1993 0.0444 -0.0193 -0.0238 0.0499 0.0442 -0.0141 0.0436 0.0019 -0.0343 -0.0341 -0.0282 -0.2519 -0.5983 -0.0036 -0.0259 -0.2552 0.1064 0.0295 -0.0419 -0.2161 -1.0983 0.1778 -0.2033 -2.5917 0.2876 -1.0983 -0.3762 1.2184 0.1483 0.1397 0.0971 0.1778 0.0417 0.0942 0.0624 0.1778 0.1321 0.1321 0.0516 0.0467 0.1194 0.0835 -0.1144 0.1733 0.0872 -0.4531 -1.9593 -0.3369 -0.3004 0.0224 0.0568 0.0721 0.1303 0.1132 0.1413 0.0971 0.1778 0.1083 -0.0404 0.0866 -0.4523 -2.1294 0.1263 0.0862 -0.5031 3.01 75.21 9.02 129.33 6.02 36.1 535.27 114.44 168.48 6.02 644.48 69.25 18.05 24.07 3.01 3.01 6.02 6.02 1481.98 15.04 3.01 3.01 12.03 33.09 222.61 9.02 3.01 9.02 339.93 403.55 1174.36 153.59 30.12 835.68 120.33 9.02 3.01 6.02 18.05 3.01 3.01 102.28 21.06 424.16 24.09 150.58 743.03 767.1 12.05 3.02 4.69 3.54 6.22 3.02 3.02 5.27 3.08 4.19 3.02 4.13 3.74 3.97 4.14 3.02 3.02 3.02 3.02 4.45 3.02 3.02 3.02 3.81 4.04 4.2 3.54 3.02 3.02 4.36 4.34 4.38 4.03 3.75 4.42 4.27 3.02 3.02 3.62 4.4 3.02 3.02 4.05 3.73 4.32 3.5 3.85 4.08 4.36 3.02 Samples N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.3169 0.2328 0.1212 0.1159 1 A -1.0199 1.5139 -1.2091 -3.7808 396 7.44 2 B -1.4617 1.9854 -0.8485 -0.7719 458 4.52 3 C -1.1249 1.008 -0.5629 -0.4996 424 4.7 4 D -1.7283 2.6367 -1.1272 -0.8637 445 4.65 5 F -1.0292 1.0664 -0.6825 -1.3288 419 6.47 6 G -1.5229 2.1214 -0.9375 -0.8567 443 4.57 7 H -1.2426 1.9068 -0.8276 -1.0324 442 5.58 8 I -0.7404 -0.5617 -0.1422 -0.4857 403 3 9 J -1.3363 2.3158 -2.3322 -8.9247 396 4.48 10 K -1.202 1.3546 -0.7221 -0.9339 436 5.1 11 L -1.5701 2.2002 -0.6823 0.1486 416 2.46 12 M -1.251 1.7419 -0.7949 -1.0022 417 5 13 O -1.0563 1.8557 -0.9415 -1.7097 413 6.11 14 P -0.9499 2.4742 -1.5993 -7.7858 401 4.73 15 Q -1.3185 2.3756 -1.0479 -1.4602 463 4.59 16 R -1.4976 2.9751 -1.2276 -1.448 401 3.66 17 OE -1.1892 3.248 23.107 -3.1216 271 5.72 18 XC -0.8767 0.1281 -0.0648 0.8328 401 4.27 19 XD -1.1395 1.6734 -0.2492 2.0548 410 2.72 20 XE -1.3498 2.0371 -0.3065 3.0434 396 4.2 21 XF -1.3963 2.341 -0.2158 4.4888 377 5.29 22 XDP -1.1821 1.165 -0.1765 1.7049 327 2.81 23 XP -2.3675 5.0084 0.1872 13.3069 279 6.35 24 XPP -1.3441 1.7982 -0.2608 2.3283 387 2.17 25 XEP -1.7314 2.4389 -0.2055 4.6296 392 3.2 26 XRP -1.5219 1.6083 -0.0987 3.376 381 3.66 27 5L 0.0428 -0.7935 0.1525 0.2224 401 3.23 28 46L 0.6321 -1.3093 0.1953 0.2381 435 2.28 29 51L 1.063 -0.8411 0.1455 0.2313 401 3.62 30 56L 0.8878 -0.8943 0.164 0.3312 424 3.17 31 61L 0.3934 -1.0006 0.1413 0.3366 420 3.1 32 67L 2.0798 0.1804 0.0199 0.4735 429 7.9 33 71L -0.0764 -1.63 0.2245 0.2558 406 1.5 34 76L 2.0062 0.7959 0.0063 0.2591 422 8.06 35 81L 1.5086 0.7905 -0.0577 0.1562 437 8.66 36 86L 2.103 1.318 -0.0661 0.599 401 9.63 37 91L 1.5223 1.1159 -0.096 0.1473 412 10.4 38 96L 2.8322 1.2304 -0.0521 0.3494 433 6.53 39 101L 2.5639 1.2818 -0.0529 0.4607 406 8.1 40 106L 2.0427 0.5915 -0.0257 0.2775 400 9.36 41 111L 0.8204 0.046 -0.1194 -0.4049 430 3.84 42 116L 2.0598 0.9916 -0.051 0.0647 426 6.76 43 121L 2.178 1.2355 -0.0152 0.32 410 6.88 44 126L 1.7582 1.0875 -0.0179 0.1768 402 8.37 45 132L 2.5612 1.1127 -0.0635 0.2962 408 6.97 46 139L 3.1389 1.2694 -0.0816 0.4202 404 5.15 47 144L 2.6043 1.4257 -0.1158 0.3617 447 5.86 48 149L 2.8774 1.2185 -0.0579 0.1335 418 3.63 49 154L 0.9806 0.7268 -0.1222 0.1221 417 9.15 50 159L 1.7023 0.7865 -0.1186 0.0552 411 7.28 51 164L 1.4843 0.2127 -0.0995 -0.1771 398 4.37 52 168L 1.1369 0.3299 -0.0342 -0.1879 415 4.16 53 189L 0.4647 -0.8921 0.0131 -0.3624 411 3.84 54 194L 2.2324 0.4638 0.0247 -0.0283 279 2.66 55 199L 1.1875 -0.5043 0.0884 -0.0483 353 3.39 56 204L 0.5054 -0.6988 -0.0062 -0.3932 451 3.73 57 209L 1.1518 -0.5302 0.1302 0.1206 415 3.28 58 215L 1.3497 0.1365 0.1173 0.3554 427 5 59 219L 2.7644 1.0171 0.0678 0.2669 280 3.99 60 228L 2.7241 1.0257 -0.0342 -0.105 267 1.89 61 238L 3.2409 1.0125 -0.3515 -1.1382 401 3.45 62 248L 0.6413 -0.8655 0.1223 0.0463 412 2.86 63 253L 2.4634 1.0362 0.0021 0.1825 303 3.22 64 263L 1.2581 -0.4683 0.1858 -0.0125 217 3.79 65 269L 3.71 1.4594 -0.2375 -0.7022 242 2.15 66 289L 1.9636 0.438 -0.3241 -1.1532 389 3.1 67 294L 1.6302 0.1601 -0.1786 -0.6817 322 3.94 68 10H -0.2826 -1.8837 0.1685 -0.1005 315 2.53 69 90H -0.0096 -1.6789 0.1715 -0.0879 400 3.49 70 113H -0.314 -1.3785 0.1336 0.0272 400 2.99 71 123H -0.4677 -1.9172 0.2091 0.1026 437 3.2 72 132H -0.3708 -1.2957 0.1138 0.1807 437 3.24 73 143H -0.3473 -1.1018 0.0804 0.0915 425 5 74 203H -0.1563 -1.6567 0.1715 0.0235 411 2.74 75 270H -0.2474 -1.8711 0.1946 -0.025 400 2.69 76 290H -0.4297 -1.0799 0.0359 -0.0546 419 5.18 77 31 OH -0.4562 -0.8107 -0.0426 -0.2065 494 4.13 78 338.5H -0.6606 -1.3702 0.0673 -0.0644 402 5.96 79 358.5H -0.1221 -1.2431 0.0187 -0.3178 446 4.72 80 378.5H -0.5285 -1.1478 0.0788 0.1117 396 4.54 81 398.5H -0.5171 -1.257 0.0505 0.0123 401 4.18 82 418.5H -0.3456 -1.1508 0.0508 0.0081 405 5.23 83 420H -0.4578 -1.0963 0.0126 -0.2561 390 3.96 84 440H -0.4835 -1.0639 0.0591 0.0036 399 4.27 85 460H -0.5544 -0.8368 -0.037 -0.192 442 4.29 86 480H -0.5253 -1.1286 0.109 0.2912 400 3.53 87 500H -0.4792 -1.2461 0.1292 0.1653 396 5.32 88 524H -0.532 -1.4329 0.1445 0.0458 400 3.73 89 544H -0.664 -1.0743 0.0985 0.1096 409 5.39 90 564H -0.4856 -1.0625 0.0207 -0.1241 432 5.63 91 584H -0.2682 -1.2425 0.0354 -0.2311 407 4.53 92 604H -0.5816 -0.7687 -0.0266 -0.1 396 4.72 93 630H -0.456 -1.1446 0.0281 -0.294 435 4.31 94 650H -0.7768 -0.6629 -0.0096 0.1675 416 3.46 95 670H -0.4079 -0.644 -0.1214 -0.5235 459 2.68 96 690H -0.6659 -0.3791 -0.1248 -0.1619 397 3.66 97 71 OH -0.4773 -0.8747 0.1514 0.3602 430 7.01 98 733H -0.6508 -0.7719 0.0341 0.2437 398 5.05 99 753H -0.7373 -0.8536 0.0791 0.4674 450 4.5 100 773H -0.6829 -0.8465 0.0682 0.1691 440 5.86 101 793H -0.6592 -0.994 0.1067 0.1734 396 6.61 102 801.5H -0.6879 -1.0028 0.1107 0.3962 410 3.22 103 821,5H -0.6323 -0.8925 0.0031 0.0339 434 5.38 104 841,5H -0.6319 -0.6674 0.0234 0.2999 420 4.9 105 861.5H -0.7694 -0.6224 0.0797 0.6302 391 3.83 106 881.5H -0.922 -0.6375 0.0147 0.5469 404 5.76 107 920H -0.3819 -0.8293 -0.0508 -0.2541 470 3.99 108 931,5H -0.5423 -0.7933 -0.071 -0.2453 403 3.61 109 941,5H -0.622 -0.8355 -0.086 -0.4186 405 3.32 110 961.5H -0.4337 -1.3229 0.19 0.2725 406 2.78 111 981.5H -0.9749 -0.293 -0.0119 0.5025 406 4.85 112 1001.5H -0.7518 -0.5517 -0.0609 0.0422 408 5.56 113 1021.5H -0.6121 -1.0253 0.0992 0.3491 424 4.31 114 1048H -0.5009 -1.3167 0.1716 0.2875 427 2.35 115 1068H -0.6481 -0.7072 0.0107 0.2526 461 5.51 116 1088H -0.6297 -0.792 0.0167 0.214 419 4.49 117 1108H -0.3092 -1.2344 0.0735 -0.1027 422 3.88 118 1128H -0.5125 -1.1717 0.126 0.3036 398 3.23 119 1340H -0.6226 -0.8993 -0.0282 -0.0889 400 5.16 120 1345H -0.6592 -0.9098 0.0264 0.0178 457 5.22 121 1350H -0.6008 -0.8067 -0.016 -0.0629 401 4.34 122 1355H -0.6117 -0.6816 -0.087 -0.3899 453 4.71 123 1360H -0.7808 -0.4614 -0.0646 0.1236 437 5.88 124 1365H -0.6359 -0.7665 -0.053 -0.2355 400 4.51 125 1370H -0.5804 -0.856 0.0345 0.0487 422 4.93 126 1375H -0.5209 -1.0947 0.045 -0.112 433 3.7 Dominant Data Set (Complete) (Figures 5.19 and 5.20) Species NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.506 0.3661 0.2932 0.2002 1 AC 008A 1.8726 0.4905 -1.527 -0.1968 1048 16.81 2 AD 009A -0.8467 1.9275 -0.1324 1.4534 1797 18.32 3 BR 01 OA -0.4174 0.449 -0.0643 -0.0515 8922 86.4 4 CY 028A 0.9246 -0.3877 -0.6789 0.188 1762 33.6 5 DE 001A 3.0497 1.133 5.095 0.4601 419 4.11 6 DE 002A -0.1634 -0.8392 0.0757 0.4768 11680 73.23 7 EC 001A -0.4126 0.2601 -0.0036 -0.3295 9218 93.07 8 MA 001B -0.0346 -0.0809 0.1835 -0.4921 10824 94.03 9 NA 003A -0.3214 -0.3197 0.0754 -0.104 2021 73.32 10 Nl 014A 1.7162 0.2778 -0.3657 0.129 4144 38.71 JS NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 0.506 0.3661 0.2932 0.2002 1 A -0.2912 0.4818 0.0935 -0.4732 278 4.28 2 B -0.9131 1.7129 -0.1012 0.5489 395 3.45 3 C -0.5638 0.5933 0.1423 -1.1589 357 3.42 4 D -0.8883 1.5554 -0.1438 0.6797 325 3 5 F -0.5087 0.2427 -0.0871 0.6217 343 4.43 6 G -0.8746 1.7216 -0.0815 0.6773 387 3.54 7 H -0.7757 1.4994 -0.0045 0.3712 368 4.09 8 I -0.4465 -1.1018 0.1259 1.198 358 2.39 9 J -0.1176 -0.4386 0.3987 -1.198 182 2.09 10 K -0.7109 1.1247 0.0639 -0.1858 383 4 11 L -0.865 1.4269 -0.1852 0.2974 353 1.8 12 M -0.8097 1.5636 -0.0253 0.4267 367 4 13 0 -0.5686 1.4972 -0.0926 0.7721 362 4.81 14 P 0.1102 0.3469 0.1636 -1.3963 216 2.04 15 Q -0.6876 1.9945 -0.2078 1.2701 422 3.84 16 R -0.7773 2.401 -0.3221 2.037 353 2.87 17 S -0.3699 0.3333 0.356 -1.5988 388 2.19 18 U -0.3084 0.1149 0.3921 -2.0368 320 1.93 19 V -0.205 1.4917 -0.5402 0.6503 216 5.99 20 X -0.1228 0.057 0.4894 -1.9926 290 1.34 21 Y -0.4855 0.534 0.1043 -1.5023 204 2.72 22 Z -0.5249 0.4296 0.1863 -1.787 158 2.32 23 AA -0.5818 0.568 0.1402 -1.5037 142 2.9 24 AC -0.3619 -0.2156 0.3313 -0.7579 307 3.25 25 AD -0.5818 -0.2345 0.1183 -0.3374 339 3.56 26 AE -0.7185 0.583 0.0453 -1.298 327 1.87 27 AF -1.0837 2.5641 -0.1915 2.1509 260 3.11 28 AG -0.471 0.1808 0.2347 -1.4389 284 3.36 29 AH -0.5019 0.195 0.114 -1.0118 317 3.42 30 Al -0.4872 0.3736 0.116 -0.9633 285 3.79 31 AJ -0.768 1.3776 0.0237 0.119 206 4 32 AL -0.7851 1.1554 0.033 -0.4696 390 2.61 33 AM -0.6127 0.7962 0.1206 -0.9694 341 3.13 34 AN -0.4749 0.5609 0.1757 -0.7636 330 3.73 35 AP -1.2642 3.569 -0.3162 5.0333 320 2.01 36 AQ -0.8572 1.4422 -0.0965 1.8378 301 4.94 37 AS -0.4411 0.984 -0.2372 0.6292 267 4.04 38 AU -1.2499 3.0552 -0.2486 3.0461 369 2.34 39 AX -0.793 0.6939 -0.0392 -0.8761 274 1.9 40 AY -0.2901 0.9821 0.0016 -0.4375 88 4.25 41 AZ -0.8193 2.6493 -0.4702 2.4526 42 2.19 42 BA -0.8931 1.7963 -0.1657 0.8592 44 1.84 43 BB -0.772 1.0377 -0.0914 0.1393 164 2.24 44 5L -0.151 -1.1972 0.1049 1.1213 346 2.42 45 46L 0.2393 -1.7678 -0.1257 1.7959 384 1.8 46 51L 0.7361 -1.3162 -0.3161 1.4433 353 2.84 47 56L 0.5378 -1.4683 -0.6267 1.4101 387 2.65 48 61L 0.1937 -1.3812 -0.381 1.3791 387 2.65 49 67L 0.9716 -0.4893 -0.5566 0.3229 222 4.84 50 71L -0.2087 -1.8495 0.1275 1.9323 389 1.38 51 76L 1.6766 0.0732 -1.293 0.3159 321 4.91 52 81L 1.3198 0.1724 -1.0427 -0.2443 356 5.92 53 86L 1.7971 0.4923 -2.0102 -0.2561 294 5.51 54 91L 1.0563 0.296 -1.0417 -0.4245 274 5.53 55 96L 2.4633 0.407 -2.2027 0.0867 331 4.01 56 101L 2.2594 0.4617 -2.0289 -0.055 293 4.4 57 106L 1.6007 -0.1126 -1.1273 0.1146 282 5 58 111L 0.67 0.0004 -0.1266 -1.3298 332 2.38 59 116L 1.7669 0.3551 -0.9239 -0.3531 312 3.76 60 121L 2.0038 0.4994 -1.6026 -0.3286 328 4.5 61 126L 1.4442 0.3878 -1.1022 -0.3322 304 5.04 62 132L 2.1809 0.4018 -1.8661 -0.197 324 4.53 63 139L 2.7688 0.5371 -2.5436 -0.0458 328 3.46 64 144L 2.2841 0.7962 -2.1235 -0.4132 354 3.81 65 149L 2.6484 0.6171 -1.6704 -0.1178 348 2.54 66 154L 0.58 0.3893 -0.5801 -0.9454 308 5.26 67 159L 1.3196 0.3578 -0.8574 -0.6998 324 4.64 68 164L 1.2143 -0.1573 -0.7297 -0.6986 342 3.27 69 168L 0.9339 -0.1779 -0.6756 -0.6927 374 3.39 70 189L 0.4053 -0.8036 0.1449 -0.2435 333 2.7 71 194L 2.1248 -0.0113 -0.6629 0.5677 256 2.25 72 199L 1.0569 -0.8293 -0.2764 0.889 324 2.87 73 204L 0.423 -0.6189 0.0663 -0.6075 382 2.78 74 209L 0.9686 -0.963 -0.3896 1.212 387 2.86 75 215L 1.1802 -0.4966 -0.6411 0.9417 372 3.83 76 219L 2.4675 0.3367 -0.8188 0.1986 186 1.94 77 228L 2.6832 0.5217 -0.7971 0.1877 247 1.62 78 238L 3.9791 1.6197 7.4278 0.9006 357 2.76 79 248L 0.3722 -1.268 0.014 1.2366 375 2.38 80 253L 2.2296 0.4397 -0.9073 0.2993 253 2.27 81 263L 1.1082 -0.8269 0.584 1.3171 201 3.26 82 269L 3.9956 1.3323 4.3519 1.137 239 2.1 83 289L 2.3688 0.8884 4.7043 -0.446 367 2.77 84 294L 1.9101 0.5112 1.8505 -0.7226 270 2.91 85 10H -0.2573 -1.6973 0.3551 1.0355 252 1.69 86 90H -0.226 -1.6179 0.2856 0.9815 276 1.86 87 113H -0.3292 -1.2885 0.2258 0.7629 360 2.44 88 123H -0.4144 -1.481 0.2315 1.1469 331 1.91 89 132H -0.3541 -1.1016 0.0735 0.7107 397 2.69 90 143H -0.3719 -0.835 0.132 0.2092 362 3.7 91 203H -0.2787 -1.5851 0.2061 1.1396 340 1.92 92 270H -0.2929 -1.708 0.3054 1.2273 313 1.71 93 290H -0.3141 -0.6431 0.0345 -0.185 357 3.85 94 31 OH -0.3039 -0.4926 0.2454 -0.7205 464 3.66 95 338.5H -0.4763 -0.4622 0.2328 -0.5581 304 3.64 96 358.5H -0.2153 -0.8137 0.3317 -0.5482 321 2.77 97 378.5H -0.4444 -0.7537 0.1517 0.1429 345 3.54 98 398.5H -0.4056 -0.9305 0.2218 0.3022 338 3.04 99 418.5H -0.254 -0.8149 0.0328 0.131 340 3.8 100 420H -0.3687 -0.6418 0.3226 -0.6134 354 3.29 101 440H -0.4086 -0.776 0.2103 -0.0408 373 3.74 102 460H -0.4129 -0.3683 0.289 -0.875 408 3.67 103 480H -0.4104 -1.0027 0.081 0.7598 362 2.91 104 500H -0.3674 -0.904 0.0915 0.35 330 3.77 105 524H -0.4268 -1.0969 0.2269 0.5042 352 2.91 106 544H -0.4575 -0.7156 0.221 -0.051 361 4.29 107 564H -0.4077 -0.2596 0.272 -1.0344 311 3.85 108 584H -0.2497 -0.8954 0.2832 -0.2442 328 3.13 109 604H -0.4896 -0.0255 0.2029 -1.2232 334 3.49 110 630H -0.329 -0.6058 0.3348 -0.8087 383 3.39 111 650H -0.5873 0.0004 0.0551 -0.8118 380 2.91 112 670H -0.2245 -0.0845 0.3197 -1.8081 420 2.25 113 690H -0.3978 0.0845 0.3015 -1.4904 354 2.95 114 710H -0.3804 -0.6349 -0.0192 0.0894 357 4.99 115 733H -0.4952 -0.4978 0.1588 0.1142 342 3.87 116 753H -0.5165 -0.7313 0.115 0.6455 372 3.18 117 773H -0.4682 -0.4932 0.1935 -0.1154 372 4.33 118 793H -0.4545 -0.5174 0.1295 -0.2491 339 5.01 119 801,5H -0.5137 -0.8783 0.1023 0.8538 379 2.76 120 821.5H -0.4237 -0.4639 0.2247 -0.4697 379 4.2 121 841,5H -0.4421 -0.4875 0.0913 0.2385 355 3.63 122 861,5H -0.4943 -0.6617 0.0415 0.676 357 3.23 123 881,5H -0.5888 0.0754 -0.0512 -0.4109 306 3.81 124 920H -0.2807 -0.4318 0.2099 -0.8985 439 3.51 125 931,5H -0.4092 -0.3323 0.2943 -0.9793 382 3.26 126 941,5H -0.3838 -0.0045 0.3491 -1.767 353 2.55 127 961,5H -0.3549 -1.3582 0.131 1.1442 379 2.43 128 981,5H -0.6217 0.2241 0.0258 -0.3878 328 3.38 129 1001.5H -0.4593 -0.0437 0.1609 -0.9072 357 4.4 130 1021.5H -0.4487 -0.8528 0.1046 0.5528 368 3.29 131 1048H -0.4157 -1.3471 0.1891 1.2736 402 2.09 132 1068H -0.4725 -0.4343 0.1407 -0.1221 404 4.36 133 1088H -0.4878 -0.5768 0.1242 0.1908 381 3.76 134 1108H -0.2732 -1.0572 0.2693 0.1899 366 2.97 135 1128H -0.4473 -1.1068 0.1407 0.9261 356 2.62 136 1340H -0.4233 -0.5069 0.2596 -0.4258 345 3.93 137 1345H -0.4419 -0.688 0.2326 0.0038 401 4.08 138 1350H -0.4278 -0.6992 0.2329 0.0261 373 3.77 139 1355H -0.3922 -0.4626 0.2931 -0.6486 411 3.91 140 1360H -0.5187 -0.1533 0.1624 -0.3972 379 4.52 141 1365H -0.4476 -0.7107 0.2169 0.1353 367 3.82 142 1370H -0.4223 -0.7278 0.2391 0.0453 368 3.79 143 1375H -0.4002 -1.0186 0.2508 0.4254 396 3.12 144 BS -0.6825 0.2317 0.0467 -0.8139 273 2.64 145 BE -0.788 0.7068 -0.0181 -1.4189 320 1.44 146 PP 2.695 1.3146 -4.0994 -0.8218 9 1.53 147 LS -0.6275 0.8488 0.001 -0.8312 138 1.63 148 LE -0.6471 0.7558 0.0307 -1.1311 164 2.64 149 OS -0.6236 1.2074 0.4135 0.172 180 4.83 150 OP -0.5236 0.9275 0.515 -0.6178 235 2.28 151 ME 1.9289 1.1893 6.6609 0.9333 34 3.64 152 ON -0.6506 0.0023 0.0591 -0.4824 49 2.69 153 OJ -0.6332 2.5538 -0.6646 2.0399 10 3.57 154 OE -0.1353 0.3133 -0.1089 -0.5536 141 4.78 155 OD -0.7008 1.2411 0.099 -0.214 112 1.06 156 XA 1.6704 0.713 -0.7367 -0.2758 61 1.99 157 XB -0.0315 1.0743 -1.0215 -0.8467 63 2.61 158 XC -0.505 -0.0762 0.0881 -0.1395 365 3.59 159 XD -0.4005 1.0055 -0.2776 -0.4228 340 1.91 160 XE -0.4462 0.8476 -0.1569 -0.822 300 2.5 161 XF -0.4938 0.8066 -0.0969 -0.8617 250 2.48 162 XG 0.2086 0.9169 -0.8908 -0.8776 219 3.62 163 XH 1.4148 0.8224 -1.4312 -0.7101 62 4.41 164 XI -0.0437 0.4602 -0.1609 -0.9671 18 3.12 165 XDP -0.5833 0.6564 -0.0048 -0.7172 296 2.32 166 XHP -0.3459 0.8829 -0.1968 -0.5586 146 2.35 167 XBP -0.8003 0.9453 -0.1038 -0.9194 204 2.12 168 XP -0.7852 0.9976 -0.1241 -0.6801 91 1.82 169 XPP -0.6647 1.0142 -0.1636 -0.5284 337 1.66 170 XEP -0.7746 0.9762 -0.1165 -0.6645 289 1.82 171 XRP -0.7663 0.8416 -0.0824 -0.9114 298 2.33 New River Lagoon Dominant (Figures 5.21 and 5.22) Species N NAME AX1 AX2 AX3 AX4 WEIGHT N2 EIG 1 AC008A 2 AD 009A 3 BR 01 OA 4 CY 028A 5 DE 001A 6 DE 002A 7 EC 001A 8 MA 001B 9 NA 003A 10 Nl 014A Samples N NAME EIG 1 A 2 B 3 C 4 D 5 F 6 G 7 H 8 I 9 J 10 K 11 L 12 M 13 O 14 P 15 Q 16 R 17 OE 18 XC 19 XD 20 XE 21 XF 22 XDP 23 XP 24 XPP 25 XEP 26 XRP 27 5L 28 46L 29 51L 30 56L 31 61L 32 67L 0.5131 0.332 0.2989 0.1623 1.8099 0.2624 -1.4841 -0.0122 960 14.5 -0.691 2.2333 -0.0327 -0.4428 842 11 -0.4456 0.7091 -0.0585 -0.3685 6890 61.2 0.7245 -0.3697 -0.6345 -0.0039 1762 33.6 3.079 0.6139 4.8064 -0.5733 392 3.62 -0.3094 -0.7001 0.0779 -0.3392 11210 68.3 -0.3858 0.2448 0.0002 0.2632 6024 74.5 -0.042 -0.031 0.1673 0.6645 8773 79 -0.4028 -0.2353 0.0818 0.0758 1751 60.6 1.5899 0.1136 -0.3526 -0.1489 3961 35.6 X1 AX2 AX3 AX4 WEIGHT N2 0.5131 0.332 0.2989 0.1623 -0.3158 0.8234 0.1082 1.0373 278 4.28 -0.8394 2.3356 -0.0257 -0.2822 395 3.45 -0.5532 0.9187 0.1435 1.3225 357 3.42 -0.855 2.2799 -0.0744 -1.0811 325 3 -0.5822 0.7074 -0.051 -0.8851 343 4.43 -0.8133 2.4306 -0.0075 -0.4799 387 3.54 -0.7159 2.0899 0.0577 0.1986 368 4.09 -0.634 -0.7602 0.1349 -1.5185 358 2.39 -0.1837 -0.2857 0.3639 2.299 182 2.09 -0.6702 1.6096 0.1025 0.6311 383 4 -0.867 2.2435 -0.1342 -1.672 353 1.8 -0.7404 2.1341 0.0421 0.2052 367 4 -0.5316 2.072 -0.0201 -0.1703 362 4.81 0.0871 0.5447 0.1499 2.4783 216 2.04 -0.6266 2.7206 -0.1101 -0.8257 422 3.84 -0.7083 3.2858 -0.1959 -1.7596 353 2.87 -0.2182 0.6383 -0.0964 -0.1081 141 4.78 -0.602 0.3816 0.0895 -0.5578 365 3.59 -0.4525 1.6103 -0.2526 -1.3359 340 1.91 -0.4811 1.3792 -0.144 -0.4339 300 2.5 -0.5275 1.3504 -0.0879 -0.3437 250 2.48 -0.628 1.2518 -0.001 -0.4945 296 2.32 -0.8052 1.6528 -0.1068 -1.0324 91 1.82 -0.7019 1.7107 -0.1451 -1.3215 337 1.66 -0.7979 1.6386 -0.1001 -1.0432 289 1.82 -0.7684 1.2997 -0.067 -0.2489 298 2.33 -0.3517 -0.9618 0.1091 -1.1608 346 2.42 -0.0446 -1.6531 -0.0998 -1.4226 384 1.8 0.4479 -1.2694 -0.2798 -1.1074 353 2.84 0.2417 -1.3742 -0.5638 -0.9818 387 2.65 -0.0701 -1.1906 -0.336 -1.2268 387 2.65 0.7624 -0.4827 -0.5203 -0.0504 222 4.84 33 71L -0.4617 -1.6565 0.1356 -1.802 389 1.38 34 76L 1.4484 -0.0219 -1.213 -0.4796 321 4.91 35 81L 1.1353 0.1597 -0.981 0.2016 356 5.92 36 86L 1.6112 0.3781 -1.8997 -0.1625 294 5.51 37 91L 0.8943 0.3139 -0.9728 0.3465 274 5.53 38 96L 2.2243 0.1472 -2.083 -0.3566 331 4.01 39 101L 2.0385 0.2433 -1.9177 -0.2192 293 4.4 40 106L 1.37 -0.2156 -1.0561 0.108 282 5 41 111L 0.5641 0.0772 -0.1324 2.1385 332 2.38 42 116L 1.5824 0.2435 -0.8797 0.3987 312 3.76 43 121L 1.8101 0.3478 -1.519 0.1436 328 4.5 44 126L 1.2666 0.349 -1.0362 0.2116 304 5.04 45 132L 1.9704 0.1742 -1.7665 0.0978 324 4.53 46 139L 2.5287 0.2076 -2.4103 -0.2336 328 3.46 47 144L 2.1103 0.5542 -2.0192 -0.0665 354 3.81 48 149L 2.4296 0.3035 -1.5897 -0.037 348 2.54 49 154L 0.4856 0.4749 -0.5487 0.9307 308 5.26 50 159L 1.1757 0.274 -0.8135 0.8704 324 4.64 51 164L 1.0262 -0.1855 -0.6851 1.4916 342 3.27 52 168L 0.755 -0.1082 -0.6321 1.3385 374 3.39 53 189L 0.2492 -0.7435 0.1286 1.0974 333 2.7 54 194L 1.8716 -0.2225 -0.6268 -0.4545 256 2.25 55 199L 0.8158 -0.8639 -0.2575 -0.5567 324 2.87 56 204L 0.2824 -0.5478 0.0538 1.5034 382 2.78 57 209L 0.7088 -0.9745 -0.3577 -1.0375 387 2.86 58 215L 0.9274 -0.4714 -0.5927 -1.05 372 3.83 59 219L 2.2205 0.0678 -0.7758 -0.0552 186 1.94 60 228L 2.4419 0.2136 -0.7611 -0.1626 247 1.62 61 238L 3.8529 0.923 6.8601 -1.3033 357 2.76 62 248L 0.1374 -1.1707 0.0204 -0.9953 375 2.38 63 253L 1.9901 0.2735 -0.8553 -0.5908 253 2.27 64 263L 0.86 -0.8765 0.5452 -1.2643 201 3.26 65 269L 3.7888 0.6956 4.0092 -1.661 239 2.1 66 289L 2.2861 0.5228 4.3376 0.8595 367 2.77 67 294L 1.7924 0.2765 1.6904 1.2649 270 2.91 68 10H -0.4625 -1.5233 0.339 -0.3982 252 1.69 69 90H -0.4303 -1.4433 0.2749 -0.3813 276 1.86 70 113H -0.5054 -1.0858 0.221 -0.4344 360 2.44 71 123H -0.6013 -1.3288 0.2325 -0.8144 331 1.91 72 132H -0.5182 -0.9348 0.0835 -0.5124 397 2.69 73 143H -0.5056 -0.6461 0.1351 0.0445 362 3.7 74 203H -0.4837 -1.4244 0.2051 -0.6918 340 1.92 75 270H -0.506 -1.521 0.2959 -0.751 313 1.71 76 290H -0.427 -0.5176 0.0431 0.5611 357 3.85 77 310H -0.3908 -0.2977 0.2295 1.2464 464 3.66 78 338.5H -0.5494 -0.3155 0.2238 0.9223 304 3.64 79 358.5H -0.3263 -0.6477 0.3075 1.3626 321 2.77 80 378.5H -0.5651 -0.5636 0.1529 0.0073 345 3.54 81 398.5H -0.5401 -0.7297 0.2156 -0.0429 338 3.04 82 418.5H -0.3938 -0.6436 0.0407 0.1981 340 3.8 83 420H -0.4538 -0.5015 0.3025 1.2637 354 3.29 84 440H -0.5258 -0.5973 0.205 0.3525 373 3.74 85 460H -0.4762 -0.1932 0.2708 1.357 408 3.67 86 480H -0.5741 -0.7632 0.0908 -0.8154 362 2.91 87 500H -0.5157 -0.7058 0.0992 -0.2087 330 3.77 88 524H -0.5775 -0.9178 0.2241 -0.2456 352 2.91 89 544H -0.576 -0.4905 0.2166 0.1818 361 4.29 90 564H -0.4649 -0.0671 0.2548 1.4683 311 3.85 91 584H -0.3757 -0.7191 0.2663 0.8984 328 3.13 92 604H -0.5193 0.1301 0.1929 1.5622 334 3.49 93 630H -0.4134 -0.4575 0.3129 1.5169 383 3.39 94 650H -0.6249 0.15 0.0633 0.8132 380 2.91 95 670H -0.2548 0.0679 0.2895 2.7276 420 2.25 96 690H -0.4264 0.3587 0.2759 1.8224 354 2.95 97 71 OH -0.5216 -0.3932 -0.0002 -0.2086 357 4.99 98 733H -0.6111 -0.1509 0.157 -0.4018 342 3.87 99 753H -0.6657 -0.3673 0.1215 -1.1074 372 3.18 100 773H -0.5836 -0.1518 0.1891 -0.077 372 4.33 101 793H -0.5623 -0.2885 0.1331 0.2728 339 5.01 102 801,5H -0.672 -0.5589 0.1113 -1.2313 379 2.76 103 821,5H -0.5115 -0.2407 0.2144 0.7082 379 4.2 104 841,5H -0.5746 -0.0845 0.0947 -0.6806 355 3.63 105 861,5H -0.6483 -0.272 0.0542 -1.2618 357 3.23 106 881,5H -0.6543 0.3804 -0.0336 -0.1278 306 3.81 107 920H -0.356 -0.29 0.1964 1.5586 439 3.51 108 931,5H -0.4619 -0.1865 0.2745 1.5589 382 3.26 109 941.5H -0.3951 0.1272 0.3202 2.5404 353 2.55 110 961.5H -0.56 -1.099 0.1391 -1.1517 379 2.43 111 981,5H -0.6957 0.7246 0.0344 -0.5839 328 3.38 112 1001.5H -0.5144 0.2283 0.1536 0.9346 357 4.4 113 1021.5H -0.596 -0.5946 0.1117 -0.6642 368 3.29 114 1048H -0.6185 -1.0385 0.1915 -1.4177 402 2.09 115 1068H -0.5751 -0.1429 0.1398 -0.0037 404 4.36 116 1088H -0.6053 -0.2917 0.1267 -0.3647 381 3.76 117 1108H -0.4263 -0.8319 0.2563 0.257 366 2.97 118 1128H -0.62 -0.8428 0.1467 -1.0366 356 2.62 119 1340H -0.5113 -0.2871 0.2468 0.7449 345 3.93 120 1345H -0.5615 -0.4231 0.2259 0.0971 401 4.08 121 1350H -0.5482 -0.429 0.2241 0.0628 373 3.77 122 1355H -0.4783 -0.2191 0.2751 0.976 411 3.91 123 1360H -0.5975 0.1988 0.1593 0.1605 379 4.52 124 1365H -0.5709 -0.4389 0.2116 -0.0804 367 3.82 125 1370H -0.5468 -0.453 0.2302 0.043 368 3.79 126 1375H -0.5514 -0.7717 0.2432 -0.2549 396 3.12 Stratigraphy Key '///. vss. sss, sss. *///. '///* X-X- X-Xxx XX VCyC ^ES v^v t 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1. Gastropods 2. Fine silt with gastropods 3. Coarse silt 4. Silt and Clay 5. Fine silt and organics 6. Banded silt/clay with gastropods 7. Silt/clay and organics 8. Clay, organics and gastropods 9. Sandy clay, organics and gastropods 10. Fine sand 11. Sandy clay and roots 12. Sand/silt with carbonate clasts 13. Mottled sandy/silt 14. Sand, gastropods and organics 15. Sand and gastropods 16.Clay Hillbank 1998 Stratigraphy Depth (cm) Description 0-40 Clay with organics 40-183 Clay, sand and gastropods 183-310 Sand, gastropods and scattered organics 310-358.5 Clay and sand 358.5-440 Sand and gastropods 440-544 Clay and sand 544-630 Sand and gastropods 630-811.5 Sand and clay 811.5-941.5 Sand, clay and gastropods 941.5-1128 Clay and sand 1128-1198 Sediment missing 1198-1222 Fine clay 1222-1234 Banded silty clay 1234-1241.5 Silty clay with gastropods 1241.5-1290 Banded silty clay 1290-1340 Sediment missing 1340-1386 Silty clay Stratigraphy of Hillbank 2000 Depth cm Munsell Code Colour Description 0-9 10 YR 3/3 7.5 YR 5/2 Dark brown Brown Smearing of darker section down to 18.5cm. Moussy sediment with large organic fragments (notably at 4-6, 9-10cm) 6-7cm - small shell new species. Mid section has lighter patches 9-12/14 7.5 YR 3/2 Dark Brown Darker section, more uniform in colour. Same moussy texture. Diffuse boundary to next layer. 12/14- 23.5 10 YR 5/2 Greyish brown Lighter section with organic fragments (18.5- 19.5, 21.5-22.5), same moussy, moist texture as above. 23.5-30 10 YR 4/1 Dark grey 4cm wide section on far side of core, sticky plastic clay with a few shells present at the base. 23.5-70 2.5 Y 6/2 2.5 Y 7/3 Light brownish grey Pale yellow Central area darker and more moist than the edges, few scattered shells, some roots at the bottom of the record. Lamanai 1999 2/1 0-42 cm Depth (cm) Munsell code Colour Comments 0-9 10Y 6/1 Greenish grey Moist. Few coarse grains which may have been washed in. Rootlets. V Fine texture 9-21 10 Y 7/1 Light greenish grey Less moist than above. Abundant gastropods, husk of a brown root at 19- 20cm 21-25.5 2.5Y 7/3 Pale yellow Gastropod layer, sediment has the same texture as above layer 25.5-27.5 2.5 Y 7/2 Light grey Same texture, no visible gastropods 27.5-29.5 2.5Y 6/2 Light brownish grey Distinct dark band with abundant gastropods 29.5-42 Top: 10YR 7/2 Base 2.5 Y 6/2 Top: light grey Base: light brownish grey Colour and texture grades through getting coarser and darker. Gastropods throughout. Lamanai 1999 2/2 0-89 cm Depth (cm) Munsell Code Colour Comment 0-30 2.5 Y 6/2 Light brownish grey Fine silt: similar to above 26-27cm - shell layer 30-33 2.5 Y 6/2 Light brownish grey Distinct shell layer 33-67 2.5 Y 4/2 Dark brownish grey Darker section, slightly sticky silt/clay 41-42cm - organic traces 67-79 2.5 Y 6/3 Light yellow brown Distinct lighter band, coarse silt. 70.5 -71.5cm - large shell. 78.5cm - organic trace: 10 YR 3/3 dark brown 79-89 2.5 Y 5/2 Greyish brown Silt with a coarser fraction Lamanai 1999 2/3 0-9 lcm Depth (cm) Munsell Code Colour Comment 0-31 10YR 5/2 Greyish brown Fine silt, moist, shells visible from 18cm onwards, distinct band 22- 23cm. 31-35 2.5Y 5/2 Greyish brown Same texture, shells present. 31-32- darker lens (2.5Y 3/2 very dark greyish brown, same texture) does not extend full width. Very fine roots below this lens. 35-37 2.5 Y 5/3 Light olive brown Same texture. 37-40.5 2.5 Y 5/2 Greyish brown This is a darker band. Shells visible, same texture. 40.5-43.5 2.5 Y 6/3 Light yellowish brown This is a light band, few visible shells, same texture. 41cm - very fine darker band 43.5-45.7 2.5Y 5/2 Greyish brown This is a darker band, moister than the level above but the same basic texture. 45.7-47 2.5 Y 6/3 Light yellowish brown This is a lighter band but moist with same texture as the above layer. 47-49.5 2.5 Y 5/2 Greyish brown Distinct shell layer, quite a diffuse boundary. Texture difficult to define as the coarser fragment could be broken shells. 49.5-57.5 2.5 Y 6/2 Light brownish grey Very greasy feel. 57.5-58 2.5 Y 5/3 Light olive brown This darker band is discontinuous 2.5Y 4/2 Greyish brown between the two colours. Same texture as above. 58-61 2.5 Y 6/2 Light brownish grey Very greasy feel 61-62 2.5 Y 5/3 Light olive brown Slightly coarser 62-83.5 2.5 Y 6/2 Light brownish grey 63cm - shells and organic Dark staining 70.5 / 69.5cm Organics - 80.5,82, 82.5 Coarser fraction lost. 83.5-84.5 2.5Y 4/2 Dark greyish brown Very dark organic layer, lots of root hairs. Munsell code is the background colour 84.5-91 2.5 Y 5/2 Greyish brown 85.7-86 - organics Lamanai 1999 2/4 0-97cm Depth (cm) Munsell Code Colour Comment 0-10.5/13 2.5 Y 6/2 Light brownish grey Coarse silt, 8-11.5 - shell and root zone 10.5/13 -15 2.5Y 5/2 Greyish brown Very fine texture 15-15.5 2.5 Y 4/2 Dark greyish brown Darker band, same texture - silt/clay 15.5-16.5 2.5 Y 6/3 Light yellowish brown Lighter band, texture as above 16.5-20.5 2.5 Y 6/2 Light brownish grey 19.5 - organic, texture as above, shells below the organic and occasional fine root hairs 20.5-22 2.5 Y 4/2 Dark greyish brown Dark stain, fine root hairs 22-32 2.5 Y 6/2 Light brownish grey Same texture, organic - 31cm, 32cm-3mm wide root running along width 32-35.5 2.5 Y 6/2 Light brownish grey Same texture 35.5-46 2.5 Y 6/2 Light brownish grey Clear tough organic layer - 39cm Thinner organic layer - 46cm 46-52 2.5 Y 5/2 Greyish brown Darker band, few shells, same texture. 52-63 2.5 Y 7/2 Light grey Light band, light yellow root 3mm wide running along width, same texture. 63-64.5 2.5 Y 6/2 Light brownish grey Shell fragments, same silty/clay texture 64.5-67 2.5 Y 5/2 2.5 Y 4/1 Greyish brown Dark grey Darker patch does not extend across, no textural difference 76.5-89 2.5 Y 5/1 Grey 80,82cm - organics 89-90 2.5 Y 5/2 Greyish brown Lighter band, same texture 90-93 2.5 Y 5/1 Grey Darker band, same texture 93-94 2.5 Y 6/2 Light brownish grey Light band, same texture. 94-97 2.5 Y 5/2 Greyish brown Contains a fine light layer Outpost 2000 Stratigraphy Depth (cm) Munsell Code Colour Description 0-6 10 YR 6/2 Light brownish grey Moussy clay, organic fragments, few shells present. 6-13.5 2.5Y 5/3 Light olive brown Large organic fragments, large shell, same moussy clay but smaller silt fraction present. 13.5-19 10 YR 6/3 Pale brown Few small shells present, same moussy texture, silt fraction same as above. 19-21.5 10 YR 6/2 Light brownish grey Darker band, moussy texture with no silt fraction. 21.5-23.5 2.5 Y 5/2 Greyish brown Distinct darker band, few very small organic fragments. 23.5-74 10 YR 6/3 Pale brown Uniform sediment, organic fragments, moussy sediment with a silty fraction, lots of small shells. Honey Camp Lagoon 1999 2/1 0-100cm Depth Munsell Code Colour Description 0-4/9 2.5 y 6/2 Light brownish grey Litter fragments at the top, shells and shell fragments. Moist Clay. Fine root material common 4/9-15 10 YR 4/2 Dark greyish brown Sharp boundary, more organic layer, lots of partly decayed organic matter (some quite coarse). Shells present (more than above)Quartz clast 5mm wide at 6.5cm More silt 15-23 10YR 6/2 Light brownish grey Diffuse boundary, very sandy clay, gastropods, occasional organic matter 23-30 10YR 7/2 Light grey Diffuse boundary, sandy clay, organic fragments, gastropods, medium root fragments common, occasional very fine roots quartz clast 2mm at 29cm (well rounded) 30-40 2.5 Y 7/2 Light grey Fairly abrupt boundary, moist sand, coarse organic fragments common, gastropods, large shell - 36-39cm 40-73 Gley 10Y7/1 2.5Y 8/2 Light greenish grey Pale yellow Fine sand, small sedimentary clast at 64.5cm 2mm wide, root fragment 60- 62cm, occasional scattered medium/fine root fragments. Pale yellow found in slight bands - 10% of section 73- 77/80 2.5 Y 8/2 Pale yellow Abrupt boundary, clayey sand with a high fine fraction, occasional root hairs (medium and fine) 75.5cm sedimentary clast - taken out for analysis 77/80- 100 Gley 10Y 7/1 10YR 8/2 Light greenish grey Very pale brown Similar to 40-73cm, much more sand than 73-80cm Coarse roots, channel with fragments remaining through the section, lighter patches in discrete zones Honey Camp Lagoon 1999 2/2 0-100cm Depth Munsell Code Colour Description 0-2 2.5 Y 8/2 Pale yellow Large carbonate clast 30x25x20 (reacts with 10% HCL) within a sand/silt mix 2-35 5 Y 7/1 2.5 Y 8/2 5 PB 2.5/1 Light grey (88%) Pale yellow (7%) Bluish black (5%) (old root channel) Abrupt boundary, more clasts up to 10mm across 3/4cm, 45cm, 32.5-33cm (5mm) banded clay especially at the top (up to 10- 11cm) occasional fine root fragments up to 2cm long mixture of sand and silt. 35-41 2.5 Y 8/2 10 YR 7/4 Pale yellow Very pale brown (10%) Abrupt boundary coarser than above, several clasts up to 20mm across, rare medium roots 41-56 10 YR 7/4 5Y7/1 2.5 Y 8/2 5 PB 2.5 /1 Very pale brown Light grey Pale yellow Bluish black Abrupt boundary, back to 2-35cm unit, occasional roots with black staining around pale brown mottles more obvious than in the original unit, silt/sand - more silty than original 56- 68/73 10YR 8/2 5 Y 6/3 Very pale brown Pale olive (10%) Massive silty clay, very fine 68/73- 100 2.5 Y 6/6 10 YR 8/2 10 Y 8/1 Olive yellow (7%) Very pale brown Light greenish grey (main) Massive clay, very fine flecks of black, orange mottles at the base (87-89cm), white mottles (5%), 75.5-79 - pale section (vpb) Honey Camp Lagoon 1999 2/3 0-100cm Depth Munsell Code Colour Description 0-1 Tube empty 0-10.5 5Y7/1 Light Grey Moist sand/silt mixture 10.5-21 10 YR 8/2 5 Y 7/2 Very pale brown Light grey Moist, sticky silty clay very soft 21- 34/41 Moist sand silt mixture, tube is not filled, very watery at the bottom, occasional sedimentary clasts 34/41- 47/49 Gap 47/49- 52/54 10 YR 6/8 Brownish yellow Moist sand silt mixture with occasional sedimentary clasts and orange mottles 52/54- 100 10 Y 8/1 10 GY 7/1 5G 4/2 10 YR 6/8 2.5 YR 4/3 2.5 YR 4/6 Light greenish grey Light greenish grey Greyish green Brownish yellow Reddish brown (92- 94cm) Red (97-98cm) Massive mottled clay, very small holes throughout, white clasts, dark green mottles in a pale grey matrix which grades into a green matrix at 58cm, orange mottles and red mottles 87-88 - fine black line carbonates run below this. 40% mottles, quartz clast 99-100 cm 4mm across Honey Camp Lagoon 1/1 Depth Munsell Code Colour Description 0-3 2.5Y 7/1 10 YR6/2 Light Grey Light brownish grey Moussy texture, few scattered shells, light brown patches 3-5.5 10 YR 6/2 Light brownish grey Light brown band with scattered shells, ends in black decomposing leaves in a layer 5.5-17 10 YR 7/2 2.5 Y 5/2 Light Grey Greyish brown Few scattered shells, very fine bands at 9 and 11 cm 17-18.5 5 Y 5/2 Olive Grey Very fine organic bands, could indicate shallowing 18-5- 21.5 10 YR 6/2 Light brownish grey Very fine, faintly laminated, occasional shells 21.5-30 10YR 5/2 7.5 YR 3/3 2.5Y 3/1 Greyish brown Dark brown (f) Very dark grey (f) Increase in shells, distinct organic flecks up to 1cm across - mini layers? Moussy texture 30-32 10 YR5/2 Greyish brown Brown band with occasional shells, much more uniform than layer above 32-35 10YR 6/2 10 YR 7/2 Light brownish grey Light grey At 32mm 2mm thin band 35-36 7.5 YR 5/2 Brown Brown band, shells 36-37.5 7.5 YR 6/2 Pinkish Grey Shells Honey Camp Lagoon 4/1 Depth Munsell Code Colour Description 0-1 5Y 8/1 White Very fine moussy sediment 5 YR 5/4 Reddish Brown 1-9 5Y 7/1 Light grey Uniform, very fine 9-10 5Y6/1 Grey Band 10-12.5 2.5Y 6/1 Grey Band with shells, darker than one above 12.5-13 2.5Y 7/1 Light grey Pale band 13-13.5 2.5Y 5/1 Grey Darker band 13.5-14 10Y 7/1 Light grey Pale band 14-15 2.5Y 6/1 Grey 15-17 2.5Y 7/1 Light grey 17-18.5 5Y6/1 Grey Darker band 18.5- 5Y6/2 Light olive grey Pale band 19.5 19.5-21 5Y6/1 Grey Darker layer, few shells 21-21.2 5Y 7/2 Light grey Very fine pale lens 21.2- 5Y 4/1 Dark grey 22.5 22.5-23 2.5Y 7/1 Light grey Pale layer with organic layers/flecks 2.5Y 3/1 Very dark grey 23-24 5 Y 4/1 Dark grey Very dark layer with shells 24-26 2.5Y7/1 Light grey 26-27 2.5 Y 7/2 Light grey Slightly pinker layer 27-28 5Y6/2 Light olive grey Darker layer 28-29.5 2.5Y 7/1 Light grey Pale layer 29.5-30 2.5Y 5/1 Grey Darker layer 30-34 5Y6/2 Light olive grey Few shells, uniform 34-36 5Y 5/2 Olive grey Increase in organic flecks within 5Y 2.5/2 Black matrix 36-38.5 5Y6/1 Grey Uniform 38.5- 5Y4/2 Olive grey Increase in organic flecks and number 45.5 5Y 3/2 Dark olive grey and size of shells 45.5- 5Y6/1 Grey Possible contamination at the bottom 46.5 Depth Munsell Code Colour Description 0-4 2.5Y7/1 Light Grey Very fine silt/clay, moussy texture with scattered gastropods 4-7.5/11.5 7.5YR 6/2 Pinkish Grey Very fine sediment with scattered gastropods 7.5/11.5-8/12 2.5 Y 7/1 Light Grey Thin diagonal band 8/12-18 10 YR 7/1 5YR 6/2 7.5YR 4/1 10 YR 7/1 10YR 6/1 Light Grey Pinkish Grey Dark Grey Light Grey Grey Colour at top of section 12cm: dark band 2-3mm wide Colour in between the two dark bands 15cm: dark band 2-3mm wide Zone in between From 17cm dark grey discontinuous band 18-19.5 7.5 YR 6/2 Pinkish Grey Same texture as above with scattered gastropods 19.5-20 7.5 YR 3/1 Very dark Grey Dark band with woody fragments < 1mm long 20-21.5 10YR 6/1 Grey Slightly darker band which is discontinuous at base of layer. Scattered gastropods 21.5-26.5 2.5Y 6/1 5YR 5/2 Grey Reddish grey Scattered gastropods Discontinuous fibrous band at 23.5cm 26.5-28.5 5YR 5/3 Reddish Brown Dark organic band full of gastropods Same texture as above 28.5-30 2.5Y7/1 Light Grey Coarser than above 30-79 2.5Y7/1 Light Grey Continuous grey sediment with high levels of gastropods throughout Stratigraphy for Honey Camp 2000 Appendix 7 One of the issues that has been raised in this thesis is that the taxonomy of tropical species may not be exactly the same as those detailed in European floras. Mastogloia smithii var. lacustris was investigated in Chapter 5 and was found to have some important distinctions to both the published and type material. Two further species are also of interest: Cyclotella distinguenda and Cyclotella plitvicensis. According to Hakansson (1989) the key feature which distinguishes these two species is the lack of undulation on the central part of the valve face in C. plitvicensis. Under the light microscope although it was apparent that there were two different forms of Cyclotella in the samples, the valve margins could not be categorised in this way. As a result the samples were examined under a Scanning Electron Microscope (SEM). Cyclotella distinguenda: Characteristics: Hustedt (1927) (in Hakansson, 1989) described C.distinguenda as being 10- 35pm in diameter, having an undulate centre to the valve face with 12-14 striae in 10pm. The central zone was described as smooth or irregularly punctate. The reference material with which the Belizian examples were compared against was Krammer and Lange-Bertalot (1991) plate 43 and the written description. The key characteristics that match are (when looking under the SEM): 1. The central area is punctate with occasional fultoportulae. 2. The central area is formed from a convex raised undulation (this is only noticeable under SEM) 3. The structure of the striae match with rows of areolae increasing in number on the mantle. There are some discrepancies though: 1. The published description of Cyclotella distinguenda has a fultoportulae every third costae. The Belizian form is at a greater frequency to this. 2. Samples from Spain analysed by Jane Reed have several fine areolae between costae (Reed, pers.com.). In the Belizian samples these are found at a much lower frequency. This may be due to dissolution. 3. It is difficult to find evidence for the openings of the fultoportulae on the external view of the Belizian samples. This may also be related to dissolution. 4. The Belizian examples do not show a smooth boundary between the central area and the striae. From this it is apparent that the Belizian samples are very close but not exactly the same as the published form. Cyclotella plitvicensis: This was described by Hustedt in 1945 (Hakansson, 1989). This species was described as having a diameter of 12-40pm, with a flat valve centre and more widely spaced striae in the marginal zones (8-10 in 10pm). Despite the differences between this species and C.distinguenda they are thought to be found in the same ecological conditions. The examples from Belize do match the published description. The significance of the difference is therefore difficult to ascertain especially as the species were not found in the modern Belize samples and therefore their ecological tolerances could not be judged. The following plates illustrate the points raised above. Cyclotella distingeunda (external and internal view SEM) from Hakansson (1989) Examples of Cyclotella plitvicensis from Belize External view showing undulate central area Examples of undulated central area in Belizian Cyclotella distinguenda (SEM). This feature is much harder to differentiate under LM, An example of the Belizian Cyclotella distinguenda showing that the central area boundary is not smooth.