NITROGEN USE EFFICIENCY, GRAIN YIELD AND QUALITY IN DURUM WHEAT (Triticum turgidum L. Var. Durum) UNDER IRRIGATED AND RAIN-FED ENVIRONMENTS IN CENTRAL ETHIOPIA

Abstract

Wheat has historically been cultivated in Ethiopia using rain-fed agricultural system. However, the practice of irrigated wheat production is a relatively recent development in the country. Despite this shift, there has been a notable lack of research on the recommended improved cultivars, nitrogen application rates, and irrigation frequencies specifically for durum wheat in irrigated systems. Furthermore, the comparative benefits of irrigated versus rain-fed production systems concerning durum wheat grain yield and quality have not been extensively explored. To address these gaps, a series of studies were conducted at the Debre Zeit Agricultural Research Center in Central Ethiopia. The primary objectives of these studies were to assess the performance of durum wheat cultivars under both irrigated and rain-fed production systems, to optimize recommended nitrogen rates for durum wheat production and grain quality within rain-fed systems, to establish nitrogen fertilizer rates and irrigation frequencies for improved durum wheat production and grain quality, to calibrate and validate the DSSAT-CERES-Wheat model for simulating the growth and yield of selected durum wheat cultivars under irrigation, and to evaluate the responsiveness of durum wheat growth and yield to varying nitrogen rates and irrigation frequencies using this model. The first experiment's findings revealed that the combined analysis of variance over multiple years indicated that the cultivars significantly influenced both the yield and grain quality of durum wheat in irrigated and rain-fed systems. In the irrigated environment, the cultivars Tesfaye, Mangudo, Utuba, Tate, and Hitosa achieved the highest grain yields. Conversely, the highest protein content was found in the Bakalcha, Toltu, Bullalla, Fetan and Utuba cultivars. In the rain-fed system, the cultivars Mangudo, Tesfaye, Utuba, Hitosa, and Tate produced the greatest grain yields, while Bakalcha, Toltu, Bullalla, Feta and Utuba demonstrated elevated grain protein content. A pair-wise T-test analysis further illustrated that the irrigated system enhanced grain yield, biomass yield, thousand kernel weight, and hectoliter weight by 40.9%, 36%, 25%, and 39%, respectively, when compared to the rainfed system. However, the rain-fed system exhibited superior values in protein content, gluten content, gluten index, and SDS-sedimentation of durum wheat, relative to the irrigated system. The second experiment focused on the impact of nitrogen rates on durum wheat cultivars. The combined data analysis indicated that varying nitrogen fertilizer rates significantly influenced both the grain yield and quality of the cultivars. In the rain-fed production system, increasing nitrogen rates from 0 to 138 kg N ha–1 led to improvements across all measured parameters. The cost-benefit analysis indicated that a nitrogen application rate of 92 kg N ha–1 is optimal for durum wheat production within the study area and similar agro-climatic conditions. The third experiment examined the interaction between nitrogen rates and irrigation intervals on durum wheat. Results showed that the interaction significantly affected both grain yield and quality. Application rates of nitrogen from 0 to 138 kg N ha–1 improved all studied parameters across all irrigation intervals. Notably, the highest average grain yield (6.9 t ha–1 ) and biomass yield (16.35 t ha–1 ) over two years were achieved with the application of 138 kg N ha–1 at the I1 irrigation interval. The cost-benefit analysis reiterated that a nitrogen application of 138 kg N ha–1 is the recommended rate for durum wheat in irrigated systems in the study area and similar agro-climatic regions. In the fourth experiment, the calibration and evaluation of the DSSAT- model for durum wheat cultivars in an irrigated environment were conducted. This involved calibrating the model with data from two years of field experiments (2018/19 and 2019/20) and evaluating it with data from three additional years (2020/21, 2021/22, and 2022/23). The results indicated that the normalized root mean square error (NRMSE) of the model ranged from 1.4 to 4.6 daysfor physiological maturity and from 5.4 to 9.3 kg ha–1 for grain yield, among other parameters. These findings demonstrate that the model was successfully calibrated and validated for durum wheat in central Ethiopia. The fifth experiment involved utilizing the DSSAT v4.7.5 model to simulate durum wheat yield under varying nitrogen rates and irrigation intervals in central Ethiopia. The evaluation incorporated nitrogen rates of 0, 46, 92, 138, and 184 kg ha–1 and three irrigation intervals (Irrigation-1, Irrigation-2, and Irrigation-3). The evaluation results indicated a strong correlation between the simulated and observed data, with percent NRMSE values ranging from 0.25 to 10%, 1.05 to 14%, and 1.65 to 23.01% under the respective irrigation intervals. Additional metrics, such as error differences (ED) and indexes of agreement (d-stat), further validated the model's performance. In conclusion, the DSSAT model has shown significant promise in accurately simulating durum wheat yield in Central Ethiopia, providing valuable insights into the potential impacts of various management practices and climate variations on durum wheat production