B13G-0271:
Methane Emissions from Semi-natural, Drained and Re-wetted Peatlands in Germany
Abstract:
Drained peatlands contribute around 5% to the total German greenhouse gas emissions. While these areas are hotspots for carbon dioxide (CO2) and nitrous oxide (N2O) emissions, some re-wetted peatlands may emit large amounts of methane (CH4). To quantify the GHG emission reductions achieved by the re-wetting of peatlands, the reduced CO2 emissions and the potential CH4fluxes need to be balanced.We synthesized methane flux data from 14 peatlands with 122 sites. At each site, methane fluxes were measured for one to three years with static chambers. The sites comprise arable land, intensive and extensive grassland, forest and peat mining areas as well as semi-natural and re-wetted peatlands on both bog peat, fen peat and other soils rich in organic carbon. Besides the groundwater table we consider further potential drivers for the CH4fluxes such as soil properties (carbon, nitrogen, pH, and physical properties), climatic parameters, land use, and vegetation composition.
Annual methane fluxes ranged from low uptake rates (around -1 g CH4-C m² a-1) to very high emissions (> 200 g CH4-C m² a-1). Intensively drained sites showed very low emissions, while for annual mean water levels higher than 5-10 cm below ground, elevated emissions of more than 20 g CH4-C may occur. At some re-wetted sites CH4 emissions of more than 100 g CH4-C m² a-1 were measured, which roughly equal the Global Warming Potential of the CO2-emissions from intensively drained agricultural sites. These high fluxes were probably caused by a combination of nutrient-rich conditions, the dieback of poorly adapted plants and a fast accumulation of organic sediments. However, this was the exception and not the rule even for very wet re-wetted sites. Achieving a model efficiency of 0.72 during cross-validation, a boosted regression tree (BRT) model was well able to describe logarithmic CH4-fluxes. Groundwater level, biotope type, soil nitrogen content, and ponding duration during summer were the most important controls. Combining the BRT model with soil, land use, and groundwater table maps as well as weather data, methane fluxes were upscaled for Germany.