Evaluation of CERES-Rice Model (V. 4.0) Under Temperate Conditions of Kashmir Valley, India

The CERES-rice model (version 4.0) was calibrated and validated using the data from a field experiment carried out during the rainy season of 2004 and 2005 at Shalimar, Srinagar (35° 5′ N latitude and 74° 89′ E longitude, 1587 m above the mean sea level, India. The experiment included six rice cultivars each transplanted on 25 May, 10 June and 25 June. Data of 25 May transplanting was used for model calibration and development of the genetic coefficients of the rice cultivars. The predicted and observed dates of phenological events were in close agreement with root mean square error (RMSE), mean absolute error (MAE) and D-index of 5.0 days, 4.3 days and 0.91, respectively, for anthesis and 3.7 days, 3.1 days and 0.91, respectively, for physiological maturity of the crop. The predicted and observed grain yields were also very close with a RMSE of 0.63 Mg ha−1, MAE of 0.58 Mg ha−1 and D-index of 0.89, respectively. Corresponding values for above ground biomass was 1.17 Mg ha−1, 1.01 Mg ha−1 and 0.82. Sensitivity test showed that simulated yield responded to temperature and atmospheric CO 2 concentration. Nitrogen 240 kg ha−1 at 25 May transplanting, recorded highest simulated grain yield (9.71 Mg ha−1). Further, 3 seedlings hill−1 produced highest simulated grain yield. The results suggest that the model can be applied in the temperate Kashmir to estimate crop productivity and optimize the management practices.