Modeling pastureland and rangeland using APEX and GRASIM models
Abstract
The goal of the this study is to improve the grazing simulation model—GRASIM, for grass-legume mixture study, and develop a comprehensive pastureland and rangeland simulation model based on APEX and GRASIM models. Grass-legume mixture with nitrogen fixation and transfer are incorporated into the GRASIM model to study crop performances. Field testing using 1997 and 1998 data from Logan, Utah showed that the model accurately simulates both orchardgrass and orchardgrass-alfalfa mixture. Sensitivity analysis of the GRASIM model was conducted to investigate the impact of model parameters and initial conditions on output using data from State College, Pennsylvania in 1993 and Bekaa Valley, Lebanon in 1998. Results indicate that the model performs consistently over the two locations with the different base values. The standing biomass is sensitive to photosynthetic fraction partitioned to shoot, leaf photosynthetic efficiency, extinction coefficient, specific leaf area, maximum specific growth rate, soil bulk density, available water, water content at 15 bar, rainfall, temperature, nitrogen mineralization rate, and initial water and nitrogen levels in soil. Nitrogen leaching from top soil layer is sensitive to bulk density, available soil water, mineralization and nitrification rate, leaching coefficient, initial water and nitrogen levels, rainfall, and temperature. The new model developed from APEX and GRASIM is capable of simulating pastureland and rangeland systems with watershed analysis capabilities at a farm or small watershed level. The model evaluates various management strategies essential for economical and environmental sustainability such as crop rotation, grazing management, manure handling, fertilization, irrigation, and sustainable landscape design in a continuous mode over several years using actual or generated weather. It can also simulate non-point source pollution from fertilization (N&P) and pesticide applications in runoff and sediment. Model testing demonstrates that the model predictions of water runoff, nitrate and phosphorus in runoff agree with observations over a three-year period. Comparisons of runoff, sediment, and N and P in runoff and sediment showed that the integrated model improved the overall predictions over the original APEX model.
Degree
Ph.D.
Advisors
Mohtar, Purdue University.
Subject Area
Range management
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.