Soil models that couple carbon (C) and nitrogen (N) cycles use ecological stoichiometry to regulate flows of C and N between stabilized pools. These models simulate emergent relationships between C and N pools and climate, litter, and edaphic variables that can be evaluated against data. Relative to C-only configurations that are typically evaluated against measurements of C pools, coupled C-N models can also use pool stoichiometry to discern where models cleave to or diverge from measured reality. We used a model that explicitly represents microbial biomass, coupled C and N cycles, and multiple pools of stabilized organic matter (MIMICS-CN) to simulate a continent-scale range of sites and assess relationships between soil C/N and edaphic and climate variables. Next, we evaluated these model-generated patterns against a synthesis dataset of C and N in soil size fractions and the novel Soil Data Harmonization (SoDaH) dataset. We also used the SoDaH dataset to explore depth relationships between C and N in soil pits across the US to further explore the drivers of C-N relationships.
Results/Conclusions
Our model simulations indicated strong relationships between soil clay content, mineral-associated organic matter, and decreasing whole soil C/N. While these trends were reflected in both the synthesis dataset of soil fractions and the larger SoDaH database, climate variables (mean annual temperature and precipitation) had stronger impacts on soil C/N in the database than in our model simulations. In addition, pH was an important driver of soil C/N relationships in the database that was not represented in the model. In the soil pits included in the SoDaH database, C/N declined modestly with depth, but none of the included climate or edaphic variables modified the relationships between C and N with depth as we expected. We will discuss the implications of our work for interpreting model results and guiding model development, and we will highlight important future data synthesis activities and soil measurements that would benefit coupled C-N soil models.