Conventional flood irrigation is the common irrigation system practiced in many developing countries. Under this regime, a large quantity of water is used to irrigate a variety of crops, such as banana, which consumes about 1,200–2,200 mm of water per year. Using the example of Yemen, the main objective of this study was to explore how to save as much of groundwater as possible and select the most suitable irrigation depth that does not affect the quantity and quality of banana production in the Al Mahwit governorate. Two models were used in this investigation. The BROOK90 (B90) hydrological model was employed to examine the banana field under existing water depth (1,500 mm year−1) and under two irrigation-reduction scenarios (1,000 and 500 mm year−1), and CROPWAT was applied to validate the findings of the B90 model, taking advantage of the monthly climate data in the CROPWAT package. Climate data were collected and field measurements of water consumption rates, plant characteristics, and soil properties were taken. The results show that the current irrigation depth with a quantity of 1,500 mm year−1 is more than the banana water requirements and most of the irrigation water is lost to seepage. The irrigation scenario with 500 mm year−1 appeared to be less than the banana water requirement, and water stress was observed. The scenario of 1,000 mm year−1 was found to be appropriate. Actual evapotranspiration (ET), soil water content (SWC), and seepage (SEE) had standard rates, and there was no stress in the soil. Most importantly, following the suggested irrigation schedule of 1,000 mm year−1 will save as much as 500 mm year−1 of groundwater compared with current water use. Furthermore, evaluation of effective rainfall and the tested irrigation depths to meet the amount of water required by banana indicated a strong correlation between CROPWAT and the B90 model with an average percentage bias of 13.2. Overall, both models yielded identical results, confirming the finding of an optimal irrigation depth.
Agriculture consumes a great amount of freshwater resources, making efficient water management imperative. This research from Yemen presents a universally applicable solution for optimizing irrigation practices and emphasizes the importance of calculating the actual water input for high-demand crops such as banana. The BROOK90 and CROPWAT models can be used in any region of the world to address water scarcity challenges. Furthermore, the strong agreement between the CROPWAT and B90 models in the validation of the proposed optimal irrigation depth makes the study findings more sensible and applicable, with no effect on banana productivity. The research introduces an innovative strategy to improve water-use efficiency in agriculture. In arid and semiarid regions, the irrigation depth of 1,000 mm for banana cultivation is proven to be sufficient, and would result in a considerable saving of groundwater for other uses. From a farmer’s perspective, saving water will result in a great reduction of the fuel required for operating water wells and also will save time and labor. Moreover, conflicts among water users over limited water resources are expected to diminish. Therefore, supplying banana with the actual irrigation input, based on intensive field research, is essential, and promises a significant conservation of groundwater resources on a larger scale.