Search for a command to run...
European settlement of North America has involved monumental environmental change. From the late 19th century to the present, agricultural practices in the Great Plains of the United States have dramatically reduced soil organic carbon (C) levels and increased greenhouse gas (GHG) fluxes in this region. This paper details the development of an innovative method to assess these processes. Detailed land-use data sets that specify complete agricultural histories for 21 representative Great Plains counties reflect historical changes in agricultural practices and drive the biogeochemical model, DAYCENT, to simulate 120 years of cropping and related ecosystem consequences. Model outputs include yields of all major crops, soil and system C levels, soil trace-gas fluxes (N2O emissions and CH4 consumption), and soil nitrogen mineralization rates. Comparisons between simulated and observed yields allowed us to adjust and refine model inputs, and then to verify and validate the results. These verification and validation exercises produced measures of model fit that indicated the appropriateness of this approach for estimating historical changes in crop yield. Initial cultivation of native grass and continued farming produced a significant loss of soil C over decades, and declining soil fertility led to reduced crop yields. This process was accompanied by a large GHG release, which subsided as soil fertility decreased. Later, irrigation, nitrogen-fertilizer application, and reduced cultivation intensity restored soil fertility and increased crop yields, but led to increased N2O emissions that reversed the decline in net GHG release. By drawing on both historical evidence of land-use change and scientific models that estimate the environmental consequences of those changes, this paper offers an improved way to understand the short- and long-term ecosystem effects of 120 years of cropping in the Great Plains.