Modeling of soil carbon dynamics, with focus on microbial activity

The representation of soil carbon dynamics is a major source of uncertainty in Earth System Models (ESMs). The terrestrial carbon pool is more than twice the size of the atmospheric pool. Therefore, the role of soil carbon as a source or a sink of atmospheric carbon, and in feedback loops is important to quantify in a changing climate. Decomposition processes of organic matter in soil have often been represented by first order decay equations, which make comparison and validation against observations difficult. Therefore, quantification of the uncertainties  and validation of improved parameterizations are problematic. An emerging approach to tackle these challenges is to represent microbial soil processes explicitly in the models. Following this approach, we have built a process based module that represent the carbon fluxes during soil decomposition, from aboveground litter to soil organic matter (SOM). The role of saprotrophs and mycorrhizal fungi is explicitly represented with separate carbon pools with associated fluxes. On a site level, we compare initial results from the stand alone module with both existing models and observations of carbon pools and fluxes. The observations are from the Norwegian Dovre Mountains, with data from three different alpine communities. These geographic areas are important, because they are subject to changes due to shrubification. In addition, these ecosystems can store large amounts of carbon. By modeling these sites, we gain more insight in the most important processes in soil decomposition, and how different microbial communities affect the carbon dynamics. We will further refine the module by expanding our study with more sites. The long-term objective is to develop an improved module that can be used to represent soil processes in ESMs, and thereby reduce the uncertainty connected to the exchange of carbon between land and atmosphere.