Modeling mixed plant species pasture growth and tree effect on forage growth in a silvopastoral system
Abstract
The main objectives of this research were to: (1) Estimate plant growth parameters for GRASIM using numerical optimization procedures while enforcing plant physiological constraints; (2) Simulate the competition between coexisting forage species for light, water and nutrients and their utilization for dry matter production; and (3) Simulate tree effects on understorey crop/forage growth in terms of light and rainfall interception, soil water and nutrient uptake. A numerical optimization procedure was designed to systematically search for the best combination of the growth parameters used by GRASIM. Physiologically-based ranges for these parameters derived from literature were used as upper and lower limits for the parameters during optimization. The procedure was utilized to estimate these growth parameters for a barley crop. The optimized parameter set allowed GRASIM to simulate closely its growth from a two-year experiment conducted at the American University of Beirut research center in the Bekaa Valley, Lebanon. In the second part of the research, competition among coexisting forage species on a naturalized pasture was modeled on the basis of leaf area distribution in the mixture for the partitioning of light interception and soil water evapotranspiration demand and on the basis of dry mass distribution for the soil nitrogen uptake. Two-year grazing experiments on a mixed naturalized pasture at the Pennsylvania State University Beef Research Farm, State College, PA provided forage biomass data for model evaluation. The multispecies GRASIM reasonably simulated the growth dynamics of the mixed pasture across two growing seasons. In the last part of the research, the tree effects on understorey pasture growth in a silvopastoral system were evaluated by explicit simulation of tree canopy light and rainfall interception, water and nutrient uptake. The algorithms for modeling these effects were incorporated into the multispecies GRASIM to form a simple agroforestry model. The new model was evaluated using forage biomass and soil moisture data collected from a silvopasture field experiment that was conducted on naturalized grassland in a forest plantation, West Lafayette, Indiana. The model closely simulated the growth of multiple forage species under a walnut tree canopy.
Degree
Ph.D.
Advisors
Mohtar, Purdue University.
Subject Area
Agricultural engineering|Agriculture|Environmental science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.