Oxidation of atmospheric methane in Northern European soils, comparison with other ecosystems, and uncertainties in the global terrestrial sink

Abstract

Summary This paper reports the range and statistical distribution of oxidation rates of atmospheric CH 4 in soils found in Northern Europe in an international study, and compares them with published data for various other ecosystems. It reassesses the size, and the uncertainty in, the global terrestrial CH 4 sink, and examines the effect of land‐use change and other factors on the oxidation rate. Only soils with a very high water table were sources of CH 4 ; all others were sinks. Oxidation rates varied from 1 to nearly 200 μg CH 4 m −2 h −1 ; annual rates for sites measured for ≥1 y were 0.1–9.1 kg CH 4 ha −1 y −1 , with a log‐normal distribution (log‐mean ≈ 1.6 kg CH 4 ha −1 y −1 ). Conversion of natural soils to agriculture reduced oxidation rates by two‐thirds –‐ closely similar to results reported for other regions. N inputs also decreased oxidation rates. Full recovery of rates after these disturbances takes > 100 y. Soil bulk density, water content and gas diffusivity had major impacts on oxidation rates. Trends were similar to those derived from other published work. Increasing acidity reduced oxidation, partially but not wholly explained by poor diffusion through litter layers which did not themselves contribute to the oxidation. The effect of temperature was small, attributed to substrate limitation and low atmospheric concentration. Analysis of all available data for CH 4 oxidation rates in situ showed similar log‐normal distributions to those obtained for our results, with generally little difference between different natural ecosystems, or between short‐and longer‐term studies. The overall global terrestrial sink was estimated at 29 Tg CH 4 y −1 , close to the current IPCC assessment, but with a much wider uncertainty range (7 to > 100 Tg CH 4 y −1 ). Little or no information is available for many major ecosystems; these should receive high priority in future research.