Search

Global society has become conscious that efforts towards securing food production will only be successful if agricultural production increases are obtained through mechanisms that ensure active regeneration of the natural resource base. Production options should be targeted in the sense of that their suitability to improve agricultural production and maintain natural resources is evaluated prior to their introduction. Biophysical targeting evaluates production options as a function of the spatial and temporal variability of climate conditions, in interaction with soil, crop characteristics and agronomic management strategies. This thesis contributes to the development of a system-based methodology for biophysical targeting. Cropping system simulation and weather generator tools are interfaced to geographical information systems. Inductive use of two case studies - a green manure cover crop and reduced tillage with residue management - helped to develop the methodology. Insight is gained into the regional potential for and the soil and climate conditions under which successful introduction of these production options may be achieved. The resulting information supports regional stakeholders involved in agriculture in their analysis and discussion, negotiation and decision-making concerning where to implement production systems. This process can improve the supply of appropriate agricultural production practices that enhance production and conserve soil and water resources.

Velvet bean (Mucuna pruriens (L.) DC. cv.-group utilis) is widely promoted in tropical and sub-tropical regions as a green manure cover crop that can reduce weed growth and soil erosion and enhance soil fertility. To provide these benefits, the crop must attain rapid ground cover and develop substantial aboveground biomass. To assist biophysical targeting of the crop to environments that can provide adequate growth conditions, the CROPGRO model was adapted to simulate velvet bean growth and development. This paper evaluates the performance of the model for phenology, growth, senescence and N accumulation for multiple locations that represent a range of environmental and agronomic management scenarios. Vegetative development, as described by main stem leaf appearance rate, varied linearly with thermal time. Time to flowering showed departures from the linear photoperiod response used in the model. Additional research is required to determine whether the crop is influenced by factors besides photoperiod and air temperature, especially water and nutrient deficits. The linear response to photoperiod did, however, provide reasonable values for partitioning to vegetative, reproductive and senesced materials. Simulation of nitrogen concentration for various plant components matched observed data. Sensitivity analyses evaluating the ability of the crop to provide ground cover, intercept light and develop adequate growth for soil protection and weed suppression indicated that a mean temperature of over 22 °C and a soil moisture holding capacity of at least 100 mm are required. The CROPGRO model proved to be a reliable decision support tool for guiding analyses of velvet bean response to crop management and environmental conditions. Further research, however, is warranted to improve its predictive capability, especially for phenology.

Velvet bean (Mucuna pruriens (L.) DC cv. group utilis) is widely promoted as a GMCC for tropical regions. Reports of insufficient biomass production in certain environments and concerns over seed production, however, suggest a need for a more complete description of growth and development of velvet bean under different production scenarios and environments. Process-based simulation models offer the potential for facilitating an assessment of management strategies for different environments, soils and production systems. The objective of this study was to review the physiology of velvet bean and using the generic legume model CROPGRO, to provide a structured and quantitative framework for describing crop response to management and environment. Model coefficients used to describe growth and development of soybean (Glycine max (L.) Merr.) served as initial reference values. Information on velvet bean from published sources was then used to revise the functions and parameters of the model. Phenology, canopy development, growth and partitioning were calibrated for two velvet bean varieties using experimental data from three sites in Mexico. Compared to soybean, velvet bean has a much longer growth cycle, allowing a very large numbers of nodes to form. Velvet bean has larger, thinner leaves than soybean, resulting in more rapid leaf area development, and larger seeds, which affects germination, early season growth and pod development. A modification to CROPGRO to track senesced tissues was incorporated. Overall, the physiological processes underlying growth and development of velvet bean appear to be similar to other tropically adapted legumes. The new model, incorporated as part of the DSSAT, version 3.5 suite of crop simulation models, has potential for evaluating management strategies in specific environments and to identify potential regions for introduction of velvet bean as a green manure cover crop.