Katie Kolodner ’24


Operations Research and Financial Engineering

Project Title

Local Damages From Hurricanes: Application of Machine Learning With Satellite Data

Presentation Link

View Katie's Presentation

Certificate(s): Applications of Computing, Statistics and Machine Learning

Estimates of the damages caused by natural disasters are a source of growing debate and uncertainty as the frequency and intensity of disasters increase due to climate change. Local economic and political conditions can influence the quality of self-reported damage estimates and skew estimates of a disaster’s impact. To help correct for this data limitation, I constructed a novel data set of building-level damages using satellite images of communities impacted by hurricanes, and I developed supervised machine learning methods. This new source of damage data will allow researchers and policymakers to better identify impacted households in order to evaluate the effectiveness of disaster relief and determine how disasters impact local economic conditions. I gained a broader understanding of the importance of computer science and economics in understanding the urban impacts of climate change through my research, and I hope to continue utilizing machine learning and data analysis for beneficial global change.

Internship Year


Project Category

Climate and Environmental Science


Center for Policy Research on Energy and the Environment (C-PREE), School of Public and International Affairs, Princeton University


Michael Oppenheimer, Albert G. Milbank Professor of Geosciences and International Affairs and the High Meadows Environmental Institute; Rachel Young, Ph.D. candidate, Princeton School of Public and International Affairs