A plant process economic model for weed management decisions in irrigated onion

A simulation model was built as a decision aid for management of five weed species in direct seeded irrigated onion (Allium cepa L.), The model uses t he state variable approach and simulations are driven by temperature and su nlight as photosynthetically active radiation (PAR), It predicts yield redu ction caused by competition for PAR according to the ratio of crop leaf are a index (LAI) to weed LAI and respective light extinction coefficients (k), Input variables are plant density by species and average number of leaves by species. Number of leaves per plant is used by the model to provide an e stimate of initial leaf area per plant. The model calculates initial specie s LAIs by multiplying species density times average leaf area per plant. Th e model accurately describes competitive interactions, taking into account respective plant densities, time of emergence, and time of weed removal, It permits economic evaluation of management factors such as handweeding, che mical weed control, herbicide phytotoxicity due to early application, and c ontrol of weed flushes during the season. The model is also used to evaluat e mechanisms of plant competition for sunlight. In a sensitivity analysis, onion yield loss was more sensitive to weed PAR interception than to PAR us e efficiency, the latter a species-dependent constant in the model.