Detecting changes in human mobility and movement patterns associated with wildfires in California
About the Project
The goal of this project is to use movement as a marker to study behavioral responses of people to environmental changes, in particular during natural disasters. As a case study, this project utilizes large and multi-sourced mobility data sets to investigate changes in movement patterns in associations with wildfire events in California. First, the goal is to investigate available data sets and their suitability for representing movement patterns in wildfire-impacted areas. Second, the project focuses on applying machine learning techniques to identify and trace changes in mobility time series and to associate them to the timelines of different wildfire events between 2019-2022 (before, during, and after the fire) compared with a baseline (e.g. a year without wildfire). The outcomes of this project will inform simulation models to assess and estimate wildfire risk on movement flows in Californian communities.
Team
- Ellie Burrell
- Justing Liu
- Lyndsey Umstead
- Piero Trujillo
- Haotian Xia
Mentors
- Dr. Somayeh Dodge, UCSB
- Evgeny Noi, UCSB
- Laura Baracaldo, UCSB
- Enbo Zhou, UCSB
Presentation
About the Sponsor
The MOVE Lab at the University of California Santa Barbara conducts basic and applied research to study movement and spatiotemporal processes such as human mobility, animal movement, migration, disease spread, and natural hazards (e.g. wildfires, hurricanes). Using large arrays of movement and tracking data sets, we develop data analytics, machine learning and knowledge discovery methods, agent-based simulation models, and visualization techniques. We apply spatial data science and computational approaches to advance the knowledge and understanding of how movement patterns are formed in dynamic natural and human systems. Current projects of the MOVE Lab are listed below. If you are interested in joining MOVE@UCSB to get involved in any of these research areas, please contact Dr. Dodge. We are always looking for talented students who are interested in computational solutions for spatiotemporal problems.
