Simulation of Durum Wheat (Triticum turgidum L.) Phenology, Growth and Yield Influenced by Varying Nitrogen Rates and Irrigation Intervals in Central Ethiopia.

Abstract

Calibration and evaluation of crop model is the first step to use appropriate crop model simulation for researchers to forecast how various factors such as weather, soil and agronomic management practices can affect the crop growth and yield performances. The present study aimed to calibrate and evaluate the CERES model for predicting growth and yield of wheat under nitrogen rates and irrigation intervals in central Ethiopia. The treatments of the experiment were included five nitrogen fertilizer rates (0 kg ha^–1, 46 kg ha^–1, 92 kg ha^–1, 138 kg ha^–1, and 184 kg ha^–1), three irrigation intervals (I₁: application of irrigation water every seven days, I₂: every ten days, and I₃: every thirteen days), and one durum wheat cultivar “Utuba”. The experiments were conducted during the 2021/22, 2022/23, and 2023/24 cropping seasons at Debre Zeit Agricultural Research Center (DZARC) in central Ethiopia. The calibration of the CERES model used one year of field data collected during the 2021/22 cropping season. The data used for model evaluation were collected from two years of field experiments conducted during the 2022/23 and 2023/24 cropping seasons.  The result of study indicated that the calibrated genetic coefficient of the Utuba cultivar were 10, 20, 380, 11, 79, 0.8, 80 for P1V, P1D, P5, G1, G2, G3 and PHINT, respectively. On the other hand,  the model evaluation showed that the strong agreement between the simulated and observed Utuba grain yields, with the percent normalized root-mean-square error (NRMSE %) values ranging from 2.89% to 6.14%, 2.26% to 14%, and 11.6% to 23.01% under seven days(I₁), ten days(I₂), and thirteen days(I₃) irrigation intervals during 2022/23 cropping season and 3.09% to 8.89%, 5.85% to 8.57%, and 7.20% to 22.5% under seven days(I₁), ten days(I₂), and thirteen days(I₃) the respective irrigation water applications during 2023/24 cropping season. Additionally, the error differences (ED) and index of agreement (d-stat) further supported the model's performance. Overall, the evaluation of CERES model demonstrated good accuracy in simulating the growth and yield of the Utuba cultivar in central Ethiopia, highlighting its potential for studying the impacts of various management practices and climate change scenarios.