Simulating the impact of sowing methods and water deficit levels on wheat yield under a semi-arid environment
Academic Article
Overview
Additional Document Info
View All
Overview
abstract
The present study was undertaken to investigate the effect of deficit irrigation (DI) on wheat grain yield under different sowing methods. The field experiment was layout using randomized complete block design (RCBD) with split plot arrangements by keeping four DI regimes (I1, I2, I3, I4 as 0%, 20%, 35% and 50% deficit respectively) and two sowing methods (bed sowing and drill sowing) in sub plots and main plots respectively with three repeats. Research also evaluated the ability of AquaCrop model to simulate wheat gain yield and biological yield under full and deficit water conditions in a semi-arid environment. The model was calibrated for the full irrigation treatments under both sowing method and remaining six treatments were used for validation purpose. The results showed that the maximum grain yield was recorded 5.724 tons/ha for 0% DI treatment under drill sowing method. The DI (I1, I2, I3, and I4) levels under drill sowing method gave 13.12%, 14.28%, 16.38%, and 19.59% more grain yield than the corresponding DI levels (I1, I2, I3, and I4) under bed sowing method respectively, whereas the average crop water use efficiency for I1, I2, I3, and I4 treatments under bed sowing was found higher than the corresponding treatments under drill sowing method by 17.94%, 14.34%, 9.96%, and 5.36% respectively. AquaCrop model simulated gain yields and biological yield showed a good agreement with measured values of both gain yield (RMSE=0.25ton ha-1, NRMSE=5.41, d=0.96 and NSE=0.79) and biological yield (RMSE=0.59 ton ha-1, NRMSE=5.36, d=0.94 and NSE=0.74). The high values of the statistical indicators confirmed that the AquaCrop model (v3.0) can simulate wheat yield under no to mild water stress conditions which makes it very useful for evaluating the deficit irrigation strategies under different cultural and management practices with minimal input data requirements and ease of use.
