Studying coupled human and natural systems from a decentralized perspective: the case of agent-based and decentralized modeling

Agent-based modeling;decentralized modeling;Integrated Modeling;coupled human and natural systems;water resources;High Plains Aquifer;irrigation;individual behavior;individual heterogeneity

Noel, Paul Henri Charles

关键词: Agent-based modeling;    decentralized modeling;    Integrated Modeling;    coupled human and natural systems;    water resources;    High Plains Aquifer;    irrigation;    individual behavior;    individual heterogeneity;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/78533/NOEL-THESIS-2015.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

The science of coupled human and natural systems deals with the interactions between humans and their environment. This science focuses on the complex dynamics and patterns that emerge through these interactions. One of the most insightful ways to study coupled human and natural systems consists in developing models to reproduce the patterns seen in these systems. Models of coupled human and natural systems are particular in the sense that they require the integration of knowledge from various and differing fields such as economics, social sciences, ecology, hydrology, biology, climate sciences and many others. In this thesis, we claim that most coupled human and natural systems are decentralized and would better be modeled from a decentralized perspective. Agent-based models, especially can be very useful to model human systems. A review of the literature shows that agent-based modeling is a commonly used tool in all the fields related to coupled human and natural systems such as socio-ecology – or social and ecological systems, hydro-economic systems, socio-hydrology or integrated environmental modeling. While agent-based models present a lot of challenges, they appear as promising tools for the representation of humans in models of coupled human and natural systems.Using an agent-based model of farmers’ decision-making on irrigation, coupled with a model of groundwater flow and aquifer/stream interactions, we studied the role of individuals in a coupled agricultural and hydrologic system. The model was designed to simulate the interactions between farmers pumping groundwater to irrigate their corn fields and the water levels within a portion of the aquifer below the Republican River Basin in the High Plains region in Nebraska. A set of simulations show that incorporating behavioral heterogeneity of individuals in the model leads to the formation of spatial and temporal patterns. In other words, some of the patterns found in the real system could be partially explained by behavioral heterogeneity of farmers. Additionally, we find that model results are more accurate when accounting for individual heterogeneity. Including individuals in the model also helps understand how these individuals are impacted by system dynamics such as new policies or environmental change. This can prove useful for policy making when knowing the differences between individuals can help devise better policies. The challenge in modeling individuals and their behavior is to decide how complex these models should be. We suggest that individual behavior should be considered as another source of uncertainty rather than a source of unnecessary complexity.