A spatial agent-based model for hydraulic fracturing water distribution

Agent-based modeling (ABM) has been employed to understand and capture the complexity of the coupled human-nature processes in water resource systems. One of the challenges is to model human decisions in the coupled human and natural systems. Hydraulic fracturing water uses were distributed through a depot-based water allocation system in the Bakken region of western North Dakota, United States. In this study, a spatially explicit ABM was developed to simulate this unique water allocation system. In the ABM, institution theory was used to model the State's regulatory policies and procedures, while evolutionary programming was employed to allow the water-depot owners (or agents) to select appropriate strategies when applying for water permits. Cognitive maps simulated the water-depot agents' ability and willingness to compete for more water sales. All agents had their influence boundaries that restricted their competitive behavior toward their neighbors, but not for non-neighboring agents. The decision-making process was constructed and parameterized with both quantitative and qualitative information. The ABM was calibrated against real-world water-use data, and the calibration results showed that the spatial ABM performed well in simulating the total number of water depots as well as the locations and water uses of water depots at the county level. By linking institution theory, evolutionary programming, and cognitive maps, our study exhibited a new exploration of modeling the highly complex dynamics of the decision-making process involved in coupled human-nature water resource systems.