فهرست مطالب s. r. amiri

Maize (Zea mays L.) is one of the most important cereals after wheat and rice in the tropical and temperate regions of the world. Also, its mean production is 8 ton ha-1. Moreover, the total area of under cultivation is 132572 hectares in Iran. Crop simulation models can play an important role in improving agricultural production systems in many developing countries. Crop models can simulate plant growth processes and grain yield instead of conducting several years of field experiments. On the other hands, crop simulation models should be calibrated and evaluated with independent data sets under different climatic conditions. Therefore, the purpose of this research was evaluation of the APSIM model for simulation of growth, development and yield of maize hybrids in Kerman province under different amounts of nitrogen.

The APSIM model was calibrated and validated using measured data from a two-year field experiment conducted in the 2014 and 2015 growing seasons. The experiment was a factorial arrangement based on a randomized complete block design (RCBD) with three replications conducted at Kerman province in Iran. Four nitrogen rates (0 (control), 92, 220 and 368 kg ha-1) and two maize hybrids (KSC 704 and Maxima) were included in the study. Moreover, inputs of APSIM model were climatic, soil, plant and management data. In order to calibrate the APSIM model, the data of field experiment in the first year (2014) (including flowering date, physiological maturity date, leaf area index, biological yield and grain yield) were included. Moreover, Data from the second experiment (2015) were used to validate the model.

Our results showed that APSIM model accurately predicted phenology (nRMSE=4.5%). But the APSIM model did not capture the effect of nitrogen stress on phenology. At the evaluation step, the model couldn’t accurately predict the maximum leaf area index (nRMSE=26 and 18% for SC 704 and Maxima hybrids, respectively) which led to overestimate of the results. The nRMSE values for the biological yield of SC 704 and Maxima hybrids were 13.9% and 5.7%, respectively. Furthermore, the values of Wilmot agreement index (d) for these SC 704 (0.95) and Maxima (0.99) indicated a close agreement between the field-measured and simulated values. Furthermore, the nRMSE for grain yield simulation of SC 704 and Maxima hybrids were 13.2 and 11.9 percentage, respectively, revealed that the model accurately simulated the grain yield of maize hybrids.

The evaluation of the APSIM model with the experimental data revealed that the model predicted grain yield, biological yield, days to flowering and maturity of maize hybrids reasonably well. This indicates that the model could be applied for assessing various management practices in maize agro-ecosystems under all parts of the semi-arid regions which has the similar characteristics to the study location. On the other hands, the APSIM model couldn’t predict the effect of different nitrogen levels on phenology.