Evaluation of ORYZA2000 model in yield simulation and production productivity of rice under crop managements

Iran is a semi-arid region with an average annual rainfall of 240 mm and 0.57 million hectares paddy fields. The unceasing growth in demand for water in the industrial sector, drinking water and reduction in the amount of water available for agricultural sector has led to a reduction of water usage in rice, which threatens its production. Crops simulation models can be used to carry out various studies such as selection of suitable cultivar and plant, determining the best agricultural management and production capacity of the area. The purpose of this study was to investigate the ORYZA2000 accuracy in simulating grain and biomass yields, and studying water balance and productivity of rice affected by irrigation and planting dates.

In order to evaluate the ORYZA2000 model and investigate the productivity of rice production under irrigation management and planting date, a split plot experiment based on a randomized complete block design with three replications was carried out on a local (Hashemi) cultivar in the years of 2016 and 2017 in the Rice Research Institute of Iran, Rasht. Irrigation interval was considered as the main factor at 4 levels including full flooding, 5, 10 and 15 days irrigation intervals, and transplanting date was assigned to subplot at three levels (April 21th, May 11th and May 31th). Evaluation of simulated and observed values of grain yield and biological yield was conducted based on coefficient of determination, T-test, root mean square error (RMSE) and normalized root mean square error (RMSEn). In this research, the water balance equation throughout the growing season was considered which its components included irrigation, rainfall, actual evaporation, actual transpiration, leakage and deep penetration, and changes in the water stored in the root development zone. Irrigation amount was measured for each plot, rainfall was also obtained from Rasht's meteorological station and other components of the water balance equation were calculated using the ORYZA2000 model. Potential evapotranspiration in ORYZA2000 model was calculated using Priestley-Taylor equation. Water productivity was investigated based on the grain yield of rice for transpiration, evapotranspiration, irrigation and total precipitation and irrigation.

The results of this study revealed that normalized root means square error of the grain yield and biological yield were determined 8% and 6%, respectively. Also, the results showed that among water managements, flooding irrigation and 15-day irrigation interval had the highest water productivity regarding transpiration and evapotranspiration, and the amount of input water and irrigation, respectively. Among the planting dates, the planting date of May, 11th had the highest water productivity based on transpiration, evapotranspiration and planting date of April, 21th, had the highest water productivity based on irrigation, and irrigation and rainfall. In these conditions, the planting date of April, 11th and May, 21th, with an average of 136 and 116 millimeters, had the highest and lowest water reserves, respectively. The highest amount of water saving during the two years of experiment was observed in irrigation intervals of 10 and 15 days (145 and 143 mm, respectively) and the lowest was recorded in the flood treatment (92 mm).

By considering paddy and biomass yield of rice, water productivity and water consumption, five days irrigation treatment had the best paddy and biomass yield in April, 21st planting date. This treatment was the best treatment in terms of productivity and rice production, with 9% reduction in water use and 6% reduction in paddy yield of rice. According to the present study, the ORYZA2000 model can be used to support the results of experiments under irrigation management conditions and planting dates.