Evaluating a crop-weed simulation model as a tool for weed management in irrigated transplanted rice (Oryza sativa L.)
1995
Abamu, F.J.
Results showed that grain yield responded to nitrogen, weed density and time of weed emergence. Highest yields were obtained at 150 kg N/ha except when the crop competed with a high weed density that emerged early when the yield at 90 kg N/ha was highest. Yield loss due to weed competition was greatest when rice and weeds competed at 30 kg N/ha. Weeds reduced yield by reducing the number of panicles produced. Harvest index and 1,000-grain weight were not affected by competition. Predictive ability of the model was strongly affected by the amount of nitrogen the crop received. Simulated grain yield and dry matter production did not differ significantly from the observed when 150 kg N/ha was applied but not at lower nitrogen rates. The trend of simulated LAI [leaf area index] development was similar to the observed. Simulated LAI did not differ from the observed at early growth stages when canopy LAI was about 1. Differences occurred when maximum canopy LAI was reached. To accurately simulate plant height, model parameters were recalibrated using experimental data. A logistic height growth parameters (HS) of 0.00365 was best to simulate the time course of plant height in both seasons. Parameters for maximum plant height (HMAX) were, however, season specific. Because weed competition is usually more important in ecosystems with limited nutrient supply capacity more precise assessments of rice growth patterns as affected by weed competition at low nutrient conditions are needed to improve the predictive ability of the model
显示更多 [+] 显示较少 [-]