Simulating phenology and yield of rice using ceres-rice model in north western Himalayas

A rice crop model CERES-Rice was parameterized and validated in Himachal Pradesh using experimental results. Cultivar specific genotypic coefficients were derived for four varieties during calibration. Validation based on several independent sets of yield data, including different locations, years, nitrogen and irrigation water treatments showed good agreement (R²=0.7785) between observed and simulated grain yield. The model predicted phenology of varieties reasonably well. The model, however, fails to simulate single grain weight. Significant association between simulated and observed grain yield were supported by the tests of significance for intercept and slope of the regression line (R²=0.7289). Likewise harvest index was also simulated fairly well by the model. The model was used to design management practices of four varieties.