COMPUTATIONAL SOCIAL SCIENCE

CSS SEMINAR - FUHS

Friday, November 14, 3:00 p.m.
Center for Social Complexity Suite
Research Hall, Third Floor

Agent-based modeling of human migration in response to climate change in northern latitudes

Brendon Fuhs, PhD Student
Computational Social Science
George Mason University

ABSTRACT: Migration is one of the primary geospatially measurable human responses to effects of climate change. Such migration-causing effects represent situational changes for humans and societies, because they include impacts to the economy, changes in the availability of natural biological resources, and changes in the appeal of living in a particular location. Understanding how climate change affects patterns of human migration requires knowledge of how these components interact with human decision-making processes as part of a complex socio-natural system.
Instructed by anthropological understandings of human migration, an agent-based approach to modeling migration is being taken by the Mason-Smithsonian Joint Project on Climate and Societal Modeling: Cyber-enabled Understanding of Complexity in Socio-Ecological Systems via Computational Modeling. We focus on Arctic and Subarctic (boreal) latitudes, where climate change is occurring most rapidly, and where a large portion of residents lives lifestyles dependent on natural systems affected by climate change.
In this talk, I’ll present two agent-based migration models, a simple empirical model of Canadian migration over the last century, and an abstract toy model designed to explore the general contribution of different agent-level behaviors to emergent spatio-temporal patterns of migration. I’ll explain some of the inherent difficulties in modeling the social effects of climate change, and present some preliminary results from both of the models, including the contribution of various climate factors to migration in Canada, and the emergent effects of kin and affinity based migration dynamics on macro-level migration patterns.

This material is based upon work supported by the National Science Foundation under Grant No. 1125171. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the NSF.