Agent-based modelling to understand irrigated farmland dynamics and farmer decision-making

Abstract

An Advanced Irrigation-Related Agent-Based Model (AIRABM) of farmers' decision-making mechanism and feedback among farmers is developed. The model explores the interactions among human and non-human agents in the irrigation system. In this paper, we discuss harvest patterns as they result from more equal or unequal water distribution in the system. In a baseline model run, farmers are not restricted in their water use. For those situations that yields are low on the system or farmer level, we allow gate settings to be adjusted to improve poor harvest situations. Our model results show that 1) in the baseline scenario, upstream farmers generally receive more water and gain higher yields compared to downstream farmers; 2) gate capacity adjustments of upstream and middle stream farmers can push more water to downstream farmers, but those specific variations are considerable. We observe unexpected emerging system performance. The AIRABM model offers options for how combinations of individual farmers' decisions on water use and farming create (un)equal yield patterns in irrigation systems.