Effect of water table on greenhouse gas emissions from peatland mesocosms

Peatland landscapes typically exhibit large variations in greenhouse gas (GHG) emissions due to microtopographic and vegetation heterogeneity. As many peatland budgets are extrapolated from small-scale chamber measurements it is important to both quantify and understand the processes underlying this spatial variability. Here we carried out a mesocosm study which allowed a comparison to be made between different microtopographic features and vegetation communities, in response to conditions of both static and changing water table. Three mesocosm types (hummocks + Juncus effusus, hummocks + Eriophorum vaginatum, and hollows dominated by moss) were subjected to two water table treatments (0–5 cm and 30–35 cm depth). Measurements were made of soil-atmosphere GHG exchange, GHG concentration within the peat profile and soil water solute concentrations. After 14 weeks the high water table group was drained and the low water table group flooded. Measurement intensity was then increased to examine the immediate response to change in water table position. Mean CO2, CH4 and N2O exchange across all chambers was 39.8 μg m−2 s−1, 54.7 μg m−2 h−1 and −2.9 μg m−2 h−1, respectively. Hence the GHG budget was dominated in this case by CO2 exchange. CO2 and N2O emissions were highest in the low water table treatment group; CH4 emissions were highest in the saturated mesocosms. We observed a strong interaction between mesocosm type and water table for CH4 emissions. In contrast to many previous studies, we found that the presence of aerenchyma-containing vegetation reduced CH4 emissions. A significant pulse in both CH4 and N2O emissions occurred within 1–2 days of switching the water table treatments. This pulsing could potentially lead to significant underestimation of landscape annual GHG budgets when widely spaced chamber measurements are upscaled.