Comparative assessment of CERES and InfoCrop models to evaluate phenology and yield of maize cultivars for the Northeastern Zone of Punjab, India

Abstract

The CERES-Maize and InfoCrop-Maize models were calibrated and validated using two cultivars (JC-4 and PMH-13) grown under four sowing dates (1st June, 15th June, 30th June and 15th July) during 2021-22. The study was conducted at the Punjab Agricultural University Regional Research Station, Ballowal Saunkhri, Punjab. The models were calibrated by repetitive interactions among genetic coefficients. for maize genotypes. The different test criteria viz., mean of observed and simulated values, root mean square error (RMSE), coefficient of determination (R2), mean absolute error (MAE), Nash-Sutcliffe efficiency (NSE) and normalized root mean square error (NRMSE) etc were used to evaluate the performance of models in terms of simulating phenology and yield attributes characters of maize cultivars. The results revealed that CERES-Maize model showed close agreement between observed and simulated data for anthesis date (RMSE= 1.09 days, NSE=1.00 and R2= 0.82), maturity date (RMSE= 0.97 days, NSE=1.00 and R2= 0.97), grain yield (RMSE=431.63 kg/ha, NSE=0.80 and R2= 0.84) and biomass yield (RMSE= 778.29 kg/ha, NSE=0.70 and R2= 0.74). On the other hand, the data simulated by InfoCrop-Maize model showed more error and less efficiency for anthesis date (RMSE= 1.85 days, NSE=0.70 and R2= 0.76), maturity date (RMSE= 2.26 days, NSE=0.80 and R2= 0.77), grain yield (RMSE= 673.95 kg/ha, NSE=0.60 and R2= 0.69) and biomass yield (RMSE= 1197.16 kg/ha, NSE=0.30 and R2= 0.61). The CERES-Maize model demonstrated proficiency in simulating phenological and yield parameters, with simulated values closely matching observed data as compared to InfoCrop-Maize model.