Eddy covariance-based methane and CO2 budget of a bog-pine ecosystem in southern Germany

Natural peatlands are commonly known as moderate sinks of atmospheric carbon dioxide (CO2), but they are also reported to be considerable sources of the potent greenhouse gas methane (CH4). However, for many ecosystems comprehensive carbon balances are lacking. In this study, fluxes of methane and CO2 were measured for or one year above a natural bog-pine site in the pre-alpine region of southern Germany. The measurements indicate annual CH4 emissions of +5.3 ±0.33 g C m-2 a-1 and an annual CO2 uptake of -62 ±20 g C m-2 a-1, resulting in a small greenhouse gas sink of -50 ±74 g CO2-C eq. m-2 a-1. Air temperature has been identified as the main controlling factor of methane during the entire annual cycle. Dependences on other factors are indirect or linked to the air temperature variation. We compared three different methods to gap-fill methane fluxes: the mean daily variation approach (MDV), a look-up table (LUT) with various controlling parameters and an Arrhenius-type exponential relation between methane flux and air temperature (non-linear regression, NLR). We found that MDV and LUT provide the best results, while the NLR shows only poor fits and is not able to reproduce the measured variations of the CH4 exchange. The CH4 flux prediction could be considerably improved by including photosynthetic active radiation, water table depth and the distance of the 70% footprint isoline (in addition to air temperature) in the LUT. The choice of gap-filling method affects the annual methane budget estimate by up to ±0.8 g C m-2 a-1, or about 16% of the annual flux.