Evaluation of CSM-CERES-rice in simulating the response of lowland rice cultivars to nitrogen application

Abstract

The objective of this study was to evaluate the ability of CSM-CERES-Rice to simulate responses of lowland rice cultivars to different N-fertilizer applications in a tropical area. Experiments on 4 planting dates (July 6, August 5, and September 9, 2010 and January 14, 2011) were conducted in Khon Kaen province, Thailand. A split-plot randomized complete block design with 3 replications was used. Three N-fertilizer application rates (0, 60 and 120 kg N ha-1) and 2 rice cultivars (TDK8 and TDK11) were assigned to the main-plots and sub-plots, respectively. Soil, plant, management and climatic data were collected from 4 experimental fields. The data from crops planted on July 6, 2010 and January 14, 2011 with application of 120 kg N ha-1 were used for model calibration. The remaining experimental data were used for model evaluation. The results for model calibration showed that the derived genetic coefficients provide simulated values of phenological events, biomass accumulation and grain yields that were in good agreement with their corresponding observed values. The model evaluation results indicated that for 2 rice cultivars grown under 3 rates of N-fertilizer application, the differences between observed and simulated values for time between transplantation and anthesis varied from 0-6 days, and for time between transplantation and maturity varied from 0-14 days. The normalized root mean square error (RMSEn) values for biomass accumulation ranged from 0-34% and for grain yield ranged from 2-354%. In general, however, these results indicate that CSM-CERES-Rice can be used as a tool to support decision-making for rice production in tropical area.