Uptake by soils is a relatively small flux in the global budget of atmospheric methane, but CH4 consumption rates in soils could be susceptible to changes in land use and climate. Global estimates of the soil sink for atmospheric CH4 are usually made by multiplying averages of small chamber measurements for various ecosystem types (or other strata) by estimates of the area covered by each stratum. Process-level models driven by gridded databases can also be used to make global flux estimates, to evaluate potential effects of changes in climate and land use, and to identify weaknesses in both data and mechanistic understanding. Methane uptake by soils is an appropriate process to model globally because the probable controls are simple relative to many other microbially mediated soil processes of trace gas production and consumption. Field experience suggests that diffusion of atmospheric CH4 into the soil is the primary factor limiting rates of CH4 oxidation in many soils. We have applied a modified version of Fick's first law based on theoretical computations for diffusivity in aggregated media, together with a soil water balance model run on a 1 degrees global grid, to make independent estimates of CH4 uptake by soils worldwide. Uptake rates were assumed to be zero in very dry desert soils that are mostly devoid of microbial activity, in frozen soils, and in wetlands that are usually CH4 sources. Our mechanistically based model supports a reference case for global net consumption of CH4 in soils of 17-23 Tg yr(-1), which is near the middle of previously reported ranges; and is close to our own mean estimate from extrapolation of flux means across ecosystem strata (21 Tg CH4 yr(-1)). A new inference of our modeling approach is that over 40% of the soil sink for CH4 occurs in warm and relatively dry ecosystems, such as semi-arid steppe, tropical savanna, tropical seasonal forest, and chaparral. This model prediction results from a favorable climate regime, high porosity in coarse-to-medium textured soils, and low moisture content that permits rapid gaseous diffusion in these semi arid and seasonally dry tropical ecosystems. Very few data on CH4 fluxes exist from these areas that can be used to compare with model predictions. Because of this paucity of data where uptake rates may be relatively high, and because humans have altered these landscapes extensively, our results suggest that more study is needed in seasonally dry ecosystems in order to understand the impacts of land-use change on soil sinks for methane. Copyright (C) 1996 Elsevier Science Ltd
