Assessment of the CERES-Rice model for rice production in Ibadan, Nigeria

Abstract

CERES-Rice model was used to simulate growth and yield of a new rice variety in Nigeria with the attention on rates of water application, one major factor that retards crop’s grain yield. The rates were full irrigation (100%ET), medium (75% ET), average (50% ET) and low (25% ET) irrigation intervals. The NEw RIce for AfriCA (acronym for NERICA) 2 variety was grown at the International Institute for Tropical Agriculture (IITA), Ibadan, Nigeria. Experimental agronomic values of parameters such as plant height, root depth, canopy shading, Leaf area index (LAI), biomass and grain yield in relation to water use were obtained and compared with simulated modelled values from CERES and the results were subjected to statistical analysis.

The model predicted slightly higher values in biomass yield (13.74tha^-1), total yield (16.47 tha^-1) and grain yield (2.63 tha^-1) than that observed values of 8.17 tha^-1, 10.58 tha^-1 and 2.41 tha^-1 respectively for NERICA 2 at 100% ET. There was no significant difference between the simulated and observed values of day 0 to flowering in the N2 variety. Highest experimental values of most agronomic parameters were obtained at 100% ET and they were slightly lower than predicted values by the model. At 25% ET, prediction was very low when compared with the observed values which underlined the effect of limited water application rate. The result of ANOVA showed significant differences in biomass and grain yield, LAI, CS, plant height and root depth among treatments (P < 0.05) while RMSE and R² were significant at the same interval. Recalibration, revalidation and localization of CERES-rice model under nitrogen limiting and soil conditions is needed to explore strategic management options for increase resource-use efficiency and productivity.