Refinement, testing, and application of the DAYCENT model to investigate ecological impacts of agriculture

Abstract

Reliable models are necessary to assess the ecological impacts of different land management options. The DAYCENT terrestrial ecosystem model was improved by reworking the soil organic matter decomposition submodel, revising the plant growth routines to make net primary productivity (NPP) simulations more mechanistic, and adding features to increase model versatility. DAYCENT was extensively tested using data from various global sites representing a range of vegetation types and land management alternatives. The model reliably simulated differences in mean values among sites/treatments for soil organic matter (SOM), grain yields, CH4 uptake, and N2O emissions. DAYCENT was also able to mimic the observed within site annual and seasonal patterns for grain yields, soil water content, soil temperature, soil mineral N levels, and trace gas fluxes. Sequestration of CO2 in SOM has been suggested as a means to compensate for anthropogenic greenhouse gas (GHG) emissions. However, changes in land management that affect SOM levels also will affect fluxes of two important non-CO2 GHG's, N2O and CH4. To accomplish a full GHG accounting, the model was used to compare the effects of climate, vegetation cover, land management, and soil type on GHGnet. GHGnet was defined as the integrated GHG effect of CO2 fluxes, N2O emissions, CH4 uptake, and the CO2 costs of N fertilizer production. Model results suggest that GHGnet in dryland systems can be substantially reduced by conversion to no till cultivation and by reducing the summer fallow period. GHGnet from rainfed or irrigated intensive cropping systems can be reduced by conversion to no till and by utilizing precision fertilizer application. The land management changes recommended to reduce GHGnet will maintain or increase overall crop yields. DAYCENT should be one of many tools used to assess the effects of human activity on C and N cycling, regional and global climate, and biodiversity.