Representation of Soil Processes in Climate Models: We Need Your Help.

The soil-plant-atmosphere continuum is a crucial determinant of the climate people experience, and biophysical and biogeochemical feedback processes at this interface are one of the biggest sources of uncertainty in projections of climate change.  Modern climate models include representations of physical, hydraulic, physiological, and biogeochemical processes that are familiar to soil scientists, but typically resolve exchanges only on scales on the order of billions of square meters. Soil properties and processes are usually measured and best understood on scales that occur on scales smaller than a single square meter. Soil scientists and climate scientists must work very hard to cross these nine orders of magnitude in spatial scale, and creative new approaches will be required. These include meaningful sampling of probability density functions of soil properties at large scales, and the recognition of extreme heterogeneity in the soil-plant-atmosphere environment that cannot be resolved. We must work together to build model representations that solve strongly nonlinear functions of soil texture, moisture, composition, and plant water relations on scales a billion times larger than those of local measurements. We review recent progress in multiscale representations of land-atmosphere interactions in global climate models, and chart some potential paths forward in this work.