Determination of sowing window for kharif maize in Punjab, India using sensitized, calibrated and validated CERES-Maize model

Abstract

Crop models help in optimizing the farming practices under climate change scenarios. The CERES-Maize was sensitized for genetic coefficients (P1, P2, P5, G2, G3 and PHINT) using sensitivity index (SI) through mathematical
and graphical approach. The sensitized range was used for calibrating the model for maize hybrids Punjab Maize Hybrid1 (PMH1) and Punjab Maize Hybrid2 (PMH2) for the year 2018 and further validated for the year 2019 using statistical indices. A good coefficient of determination (R ²) for PMH1 and PMH2 was obtained for anthesis (0.82, 0.80), maturity (0.67, 0.94), yield (0.95, 0.95) and Leaf Area Index (LAI) (0.85, 0.82) respectively. The Normalized Root Mean Square Error (NRMSE) was found to be excellent (<10%) for all the parameters except LAI where it was good. The model simulated 20^th May to 7^th June as the optimum sowing window for maize with grain yield / LAI for PMH1 being 5200-6000 kg ha^-1 / 2.9-3.2 and for PMH2 being 4200-5400 kg ha^-1/ 2.8-3.0. With delay in sowing from June 8^th to 18^th the grain yield/LAI varied between 5000 - 5400 kg ha^-1/3.1-3.4 for PMH1 and 4000 - 5000 kg ha^-1/ 2.7-3.2 for PMH2. Delay in sowing after June 7^th reduces the grain yield at the expense of profuse vegetative
growth, i.e. the LAI increases upto June 18^th and 24^th for PMH1 and PMH2, respectively. The deviation of grain yield and Harvest Index (HI) from their mean for the sowing window, respectively showed depreciation after June
9^th (-0.31%, -2.31%) for PMH1 and after June 12^th (-6.49%, -0.13%) for PMH2. The HI and grain yield decreased while LAI and biomass increased with delayed sowing. The calibrated CERES-Maize model can further be used for analysing the climate change impact on maize in Punjab, India.