Monday, August 4, 2008 - 2:30 PM

COS 15-4: Development of a spatially explicit ecohydrological modeling approach to investigate semi-arid land degradation and its impact on biogeochemical cycling

Laura Turnbull1, Steven J. Del Grosso2, William J. Parton3, Cindy Keough3, John Wainwright1, and Richard E. Brazier4. (1) University of Sheffield, (2) USDA-ARS, (3) Colorado State University, (4) University of Exeter

Background/Question/Methods

A common form of land degradation in semi-arid areas world-wide is the invasion of grasslands by shrubs. The change in the type and distribution of vegetation over the landscape has significant implications for the hydrological and erosion response to rainfall events.  Feedbacks and interactions between biotic and abiotic components of the landscape drive ecosystem dynamics in semi-arid environments. However, these interactions have not yet been well incorporated into ecosystem modelling approaches.  A new spatially explicit ecohydrological modelling approach is outlined, in which Daycent, a daily time-step biogeochemical cycling and plant growth model is coupled with Mahleran, an event-based Model for Assessing Hillslope to Landscape Erosion, Runoff and Nutrients. This coupled modelling approach explicitly incorporates spatial interactions by simulating the spatial redistribution of water, sediment and nutrients that occurs during runoff-generating rainfall events. 
Results/Conclusions

Results from initial simulations show that the redistribution of water greatly affects the spatial variability of soil-moisture content, which in turn affects the spatial variability of ecosystem processes and biogeochemical cycling that is primarily governed by water availability. Thus, these initial results re-enforce the requirement of explicitly including spatial interactions when modelling semi-arid ecosystem processes, since these ecosystems are markedly heterogeneous. Initial results are validated against data collected over a grassland to shrubland transition at the Sevilleta National Wildlife Refuge in New Mexico. Specifically, the spatial properties of vegetation cover, soil nutrient and soil-moisture content and event-based runoff, erosion and nutrient fluxes simulated by the coupled ecohydrological model are validated.