Effects of broadleaf woodland cover on streamwater chemistry and risk assessments of streamwater acidification in acid-sensitive catchments in the UK

Abstract

Acidification of surface waters has been recognised as the major water quality problem in the UK uplands. The adverse effects of conifer afforestation on streamwater chemistry and ecology are well documented in acid-sensitive catchments and have mainly been attributed to the enhanced deposition of atmospheric pollutants onto conifer canopies (the “scavenging effect”). Currently, international and national policies promote the expansion of native broadleaf woodland in the UK. Pollutant deposition onto broadleaf canopies is considered less than onto the more aerodynamically rough conifers, but there is concern that largescale broadleaf planting could delay the recovery of acidified waters or lead to further acidification in most sensitive areas. However, there has been limited investigation of the influence of broadleaf woodland cover on streamwater chemistry in the UK. To investigate the effect of woodland cover 14 catchments with different (0-78%) percentages of broadleaf woodland cover were identified in representative acidsensitive areas in north-western and central Scotland (Glen Arnisdale and Loch Katrine area) and northern and south-western England (Ullswater area and Devon) using spatial datasets in a GIS. Streamwater was sampled at high flow from the catchment outlets in winter and spring 2005 and 2006 and was analysed for major cations, anions and trace metals using standard methods. The number of samples ranged from two in the Glen Arnisdale catchments to 10 in the Loch Katrine area catchments which were sampled more intensively. Significant positive correlations were found between percentage broadleaf woodland cover and streamwater NO3 (rs = 0.51) and soluble Al (rs = 0.64) concentrations. The greater NO3 leaching to streamwater in the three most forested catchments (> 50%) was probably due to enhanced N deposition onto woodland canopies and nitrification by alder in the Ullswater area forested catchments. Streamwater NO3 concentrations equalled or exceeded non-marine SO4 in the above catchments indicating that NO3 was the principal excess acidifying ion in catchments with greater woodland cover.The woodland effect on streamwater chemistry in the study catchments was masked to some extent by variability in acid deposition climate and soil type composition. Seasalt inputs were found to be a more important control than woodland cover for streamwater chemistry in the maritime Glen Arnisdale catchments. A risk assessment of acid-sensitivity in the study catchments was conducted by calculating streamwater critical load exceedances using the Steady-State Water Chemistry (SSWC) and First-order Acidity Balance (FAB) models and modelled pollutant deposition for 1995-97 and 2002. Critical loads were exceeded by 0.01 to 1.74 keq H ha-1 yr-1 in two catchments which had woodland covers > 50% and in the Devon control catchment. The remaining 11 study catchments were assessed to be not at risk of acidification, probably due to significantly reduced non-marine S deposition from 1986 to 2001, but seasalt inputs to the Glen Arnisdale catchments might cause acidic streamwater episodes. Acid-sensitivity was also assessed using macroinvertebrates sampled in 11 of the study catchments and the results generally agreed with the critical load assessments. More detailed estimates of the enhancement of dry S and N deposition onto birchwoods in the Loch Katrine area catchments using calculated roughness length within FRAME showed that it posed no risk for streamwater acidification in these catchments because of the high rainfall environment. However, in acid-sensitive areas of the UK with lower rainfall and closer to major pollution sources, enhanced pollutant scavenging by broadleaf woodland canopies could pose a greater risk of acidification to freshwaters. The finding that almost all study catchments with woodland covers less than 30% are well protected from acidification suggests that this is a sensible threshold value for use in risk assessments of the effects of broadleaf woodland planting conducted within the Forests and Water Guidelines. The results of a sensitivity analysis of the Guidelines’ methodology, conducted using parameters such as numbers and timing of streamwater sampling, different runoff estimates and critical acid neutralising capacity values, showed that the Guidelines should be able to protect sensitive freshwaters from acidification in areas where broadleaf woodland is expanding.