Use of Soil-moisture Retrievals from SMAP to Refine Global Land Emissions of Trace Gases

CH4 and N2O become increasingly important for accurate climate predictions due to their strong direct global warming potentials. However, the specific sources and magnitudes of these trace-gas emissions as well as the involved mechanisms are still not well understood. The state and amount of water in the soil is known as a key player in modulating these trace gas emissions. In this study, we use the unprecedented global high-resolution land surface soil moisture and freeze/thaw state from SMAP together with other land-based observations to constrain our recently developed soil N2O and CH4 emission modules integrated in the CLM-CN version 3.5 (CLMCN-CH4-N2O) across a wide range of field sites. We employ the Gaussian process emulator to assess the sensitivity of the model estimates to various uncertain parameters so that the key parameterizations associated with these trace-gas emissions can be better understood and enhanced. Our ultimate goal is to have an enhanced CLMCN-CH4-N2O model that can provide current trace gas emission estimates with reduced uncertainty and support greater confidence in predicting their fate and assessing their relative contributions to the climate feedbacks under various climate-change scenarios.