This is a capstone project for the Master of Environmental Data Science at the Bren School of Environmental Science and Management, University of California, Santa Barbara.
As wildfires become more common in the face of climate change, wildfire preparedness is increasingly important. Maintaining defensible space around structures is key for the mitigation of property damage from wildfires. The Santa Barbara County Fire Department (SBCFD) currently conducts time-consuming field inspections of all structures within their jurisdiction to identify properties not compliant with defensible space requirements. The SBCFD needs a tool to identify compliance so that resources can be allocated more efficiently. Therefore, the capstone team will create a machine learning algorithm that predicts compliance with defensible space requirements based on remote sensing imagery, and an interactive dashboard that presents these results for use by the SBCFD. By allowing for in-time monitoring of defensible space compliance rather than relying on annual field visits, this project intends to increase wildfire preparedness, easing the burden of compliance monitoring on SBCFD.
This organization currently hosts:
- data-preparation: a repository covering data cleaning and training set creation for modeling.
- modeling: a repository covering modeling and model metrics.
- Inspections data: SBCFD
- Remote sensing data: Planet Labs
- Sarah Anderson, Ph.D. | Bren School of Environmental Science & Management
- Mark Buntaine, Ph.D. | Bren School of Environmental Science & Management
- Cesar Martinez-Alvarez, Ph.D | UCSB Department of Political Science
- Santa Barbara County Fire Department
- Faculty Advisor - Mark Buntaine, Ph.D.
- Capstone Advisor - Carmen Galaz-Garcia, Ph.D.
- Cullen Molitor | Center for Effective Global Action | emLab
- Scott Safechuck | SBCFD