The role of field studies in landscape-scale applications of process models: an example of soil redistribution and soil organic carbon modeling usingCENTURY

Field studies on soil properties and processes in southern Saskatchewan hav e clearly indicated the need to account for both lateral and vertical trans fers of components in the landscape for a better understanding of soil dyna mics at a given point. Extrapolation of these studies requires greater inte gration of the site-specific field results with the current generation of p rocess models. In this paper, we use the results of a field study to assess the ability of the CENTURY model to describe the influence of soil redistr ibution on soil organic carbon (SOC) dynamics. After modifying the erosion input of CENTURY to account for soil deposition, the results from CENTURY w ere compared to measured SOC levels from a chronosequential study of cultiv ation effects on SOC levels in southern Saskatchewan. CENTURY closely simul ated the effects of soil loss on SOC levels in landform segments with domin antly convex profile (i.e., downslope) curvature. CENTURY estimates of SOC changes for landform segments experiencing soil gain are less comparable to the field results; it overestimated SOC loss after 80 years by 16 Mg ha(-1 ) for depressional complexes and 10 Mg ha(-1) for footslope complexes. This leads to a 14% difference in total SOC loss on a landscape-weighted basis (estimated loss based on field data of 36 Mg ha(-1) versus a CENTURY-simula ted loss of 41 Mg ha(-1)). (C) 2001 Elsevier Science B.V. All rights reserv ed.