Performance assessment of the FAO Aquacrop model for irrigated cabbage in Uganda

Abstract

This study assesses the FAO AquaCrop model's performance in predicting cabbage yield and water use efficiency under drip irrigation in Uganda's tropical environment. The research addresses the critical need for sustainable agricultural practices due to climate change and population growth, the importance of irrigation, specifically drip irrigation, for enhancing crop productivity and food security in Uganda. Additionally, the FAO AquaCrop model's role in simulating crop yield under varying irrigation conditions is emphasized, filling a research gap in the Ugandan context resulting from the scarcity of research on smart irrigation for cabbage cultivation in the region. The study investigates water application effects on cabbage yield, models yield under different water treatments, and calibrates the AquaCrop model as the specific objectives. The research aims to develop a practical tool for farmers, leveraging the AquaCrop model to optimize cabbage production amidst climate uncertainty and resource constraints. This research was conducted at Makerere University Agricultural Research Institute- Kabanyolo in Uganda, at the FoodLAND project site. The study focused on the Gloria F1 variety and diverse irrigation scenarios during the growing season from 30th December, 2023 to 1st April, 2024. A randomized complete block design with three treatments; Recommended level (100% of I.R), Substantial water reduction (50% of 100% of I.R) and a rain-fed control, formed the experimental framework. For the canopy assessment between the observed and simulated data, the Pearson Correlation Coefficient (r) value was 0.96, RMSE was 4.6%, CV(RMSE) was 17.8%, model efficiency (EF) was 0.89 and the index of agreement was 0.98. For the above ground biomass and final yield, the deviation between the observed and simulated results were 0.438% and 0.052% for the 100% of I.R, 0.454% and 0.186% for 50% of I.R, and 0.482% and 0.115% for the rain-fed control, respectively. The reference harvest index obtained was 70%. The model best simulated canopy cover, followed by the harvestable yield but over-estimated the biomass. Therefore, the model can be used as a decision support tool by district agricultural production managers, consultants and irrigation engineers in helping irrigation practicing farmers make informed decisions. However, the model should be tested with supplementary setups of research, if possible, including irrigation and cumulative biomass and yields such that the comparison can be viable enough to verify the validity of the model for application in Uganda.