An agent-based model of shared irrigation resources using the Theory of Planned Behaviour in arid and semi regions

The expansion of irrigated agriculture and recurrent drought periods poses a serious threat to the renewability and sustainability of common water resources in arid and semi-arid regions. These shared resources can take the form of dam water which is shared between farmers according to a predefined schedule or groundwater which the farmers independently extract. The dam water is less expensive to use but this source can be limited in drought years risking crop productivity. Groundwater is a more reliable resource but is more expensive to extract and can cause soil salinity. Simulating agricultural management systems requires understanding and quantifying how biophysical and socio-economical constraints influence the decisions of farmers. Therefore, this research aimed to develop an agent-based modelling (ABM) approach to simulate farmer behaviour in irrigation management. The Theory of Planned Behaviour was used as a theoretical framework to simulate decision models that were integrated with a biophysical model describing the interaction of farmers with water resources and how limitations of water resources and salinity impact crop yield. Through modelling, we explore various strategies to improve sustainable water use. The methodology is applied to an irrigated perimeter of Al Haouz Basin, Morocco, as a case study, where there are different stakeholders and water user associations with conflicting objectives. The ABMs were parameterised using data collected by surveying 70 farmers. The findings indicate that the existing irrigation scheduling was usually satisfactory. However, with the exacerbation of drought periods, the use of dam water resources is unreliable. Farmers responded by seeking alternative water resources and changing their irrigation systems and cropping patterns which led to the potential of overexploitation of groundwater and increased accumulated salt content.