Assessing the photoreactivity of peatland derived carbon in aquatic systems

Abstract

Northern peatlands are a globally important soil carbon (C) store, and aquatic systems draining peatland catchments receive a high loading of dissolved and particulate forms of C from the surrounding terrestrial environment. Once incorporated into the aquatic environment, internal processes occur to modify the C pool. Of these, photo-processing preferentially targets terrestrially derived C and therefore might have a significant effect on the C budget of peatland draining aquatic systems. The overarching aim of this study was to investigate photochemical processing of C in Scottish peatland draining aquatic systems in order to determine the importance of this pathway in aquatic biogeochemical cycles. For initial laboratory experiments, water samples from a peatland headwater stream (Auchencorth Moss, SE Scotland) were collected. Laboratory based irradiation experiments were conducted at a range of temperatures, and different filtration treatments, including unfiltered samples, were employed to understand the fraction of C most susceptible to photo-processing. UV irradiation and temperature had a significant effect on DOC and gas headspace concentrations, with Q10 values of ~1.42 and ~1.65 derived for CO2 and CO photoproduction in unfiltered samples, respectively. However, filtration treatment did not induce significant changes in gaseous C production between light and dark samples, indicating that the experimental conditions favoured breakdown of DOC rather than POC to CO2 and CO. In all light treatments a small but significant increase in CH4 concentration was detected. These data were compared to results from experiments conducted in ambient light and temperature conditions. DOC normalised CO2 photoproduction was an order of magnitude lower than in laboratory conditions, although relative abundances of C species within overall budgets were similar and these experiments demonstrated that ambient exposure is sufficient to generate photo-processing of aquatic peatland C. Overall these data show that peatland C, particularly the < 0.2 μm fraction, is highly photoreactive and that this process is temperature sensitive. Further laboratory irradiation experiments were conducted on filtered water samples collected over a 13-month period from two contrasting aquatic systems. The first was the headwater stream draining Auchencorth Moss peatland with high DOC concentrations. The second was a low DOC reservoir (Loch Katrine, C Scotland) situated in a catchment with a high percentage peat cover. Samples were collected monthly from May 2014 to May 2015 and from the stream system during two rainfall events. Significant variation was seen in the photochemical reactivity of DOC between the two systems, with total irradiation induced change typically two orders of magnitude greater and DOC normalised CO2 production a factor of two higher in the headwater stream samples. This is attributed to longer water residence times in the reservoir rendering a higher proportion of the DOC recalcitrant to photo-processing. Overall the magnitude of photo-induced C losses was significantly positively correlated with DOC concentration in the headwater stream, which varied seasonally with highest concentrations detected in late autumn and winter. Rainfall events were identified as important in replenishing the stream system with photoreactive material, with lignin phenol data indicating mobilisation of fresh DOC from woody vegetation in the upper catchment during a winter rainfall event. Whilst these data clearly demonstrate that peatland catchments generate significant volumes of photoreactive DOC, the degree to which it is processed in the aquatic environment is unclear. Field investigations were undertaken to address this uncertainty. In-situ experiments with unfiltered water samples in light and dark conditions were conducted in two contrasting open water peatland pool systems. At the high DOC site (Red Moss of Balerno, SE Scotland), DOC concentrations in surface light exposed samples decreased by 18% compared to dark controls over 9 days and light treatments were enriched in CO2 and CH4. Photochemical processing was evident in δ13C-DOC and δ13C-DIC signatures of light exposed samples, which were enriched and depleted, respectively, relative to dark controls (+0.23 ‰ and -0.38 ‰) after 9 days of surface exposure. At the low DOC site (Cross Lochs, Forsinard, N Scotland) net production of DOC occurred in both light and dark samples over the experiment duration, in part due to POC breakdown. δ13C-DIC signatures indicated photolysis had occurred in light exposed samples (-1.98 ‰), whilst δ13C-DOC data suggest an absence of photo-processing, as the signatures in both treatments were similar. Accounting for light attenuation through the water column, 46 ± 4.9 and 8.7 ± 0.5 g C-CO2 eq m−2 yr−1 was processed by photochemical and microbial activity in peatland pools within the catchments at the high and low DOC sites, respectively. At both sites, light driven processing was responsible for a considerable percentage (34 and 51%) of gaseous C production when compared to equivalent estimates of microbial C processing and thus should be considered a key driver of peatland pool biogeochemical cycles. It is clear from this study that temperature, seasonal cycles, rainfall events and water residence time provide strong controls on the photoreactivity of aquatic C in Scottish peatland systems. The photo-processing pathway has the potential to alter the C balance of peatland catchments with a high percentage coverage of aquatic systems. Under climate change scenarios where light, temperature and rainfall conditions are expected to change, this process may become increasingly important in aquatic C cycling, particularly if the upward trend in DOC concentrations in northern aquatic systems continues.