Nutrient controls on decomposition rates: A global modeling approach

Background/Question/Methods

From their geophysical origins, global climate models have increasingly become more ecologically relevant.  For example, modern climate models include explicit representations of land, vegetation cover, hydrology, land use change, atmospheric chemistry, glacial retreat, and biogeochemistry to predict future climate scenarios.  This presents enormous opportunities and challenges for ecological and climate-modeling communities to both learn from and inform each other.  Foremost, processes newly included in climate models need to be evaluated to assess if model simulations seem realistic.  We used observational data from the Long-term Intersite Decomposition Experiment Team (LIDET) to evaluate soil carbon and nitrogen biogeochemistry routines in the Community Land Model (CLM-CN 4.0) and DayCENT.  To match LIDET observations in our modeling experiment we prescribed soil temperature and soil moisture effects on decomposition through the climatic decomposition index at each of the 26 LIDET sites and simulated litter decomposition over a decade for 9 different litter types that varied in litter quality. 

Results/Conclusions

Across biomes DayCENT predictions of litter decomposition rates and N dynamics were in reasonably good agreement with LIDET observations.  Comparatively, in CLM-CN simulations initial litter mass loss was too rapid, N immobilization was unrealistically high, and litter took too long to mineralize N.  Rates of litter decomposition and N immobilization in both models showed sensitivity to assumptions made about the availability of inorganic N in soils.  Collectively, these data suggest that a more faithful implementation of “CENTURY-like” soil biogeochemistry should improve soil C-N simulations in global climate models.  More broadly, our work highlights utility of well-coordinated, long-term observations and experiments that are replicated over broad geographic scales for ecological and climate-modeling communities.  Moreover, we hope that global climate models can be used by ecologists to justify and scale-up observational measurements into regional and global models that simulate how the terrestrial biosphere responds to and affect environmental change.