Study of the Tsunami Aftermath and Recovery
Overview Research Data STAR Investigators FAQ Contact


Disasters, Imagery, and Artificial Intelligence

Our goal is to use AI to develop efficient scalable methods that process satellite imagery and generate measures of landscape changes that result both from extreme events and from the rehabilitation and reconstruction of the natural and built environment in their aftermath. We link resulting measures of change to longitudinal information on population health, wealth, and well-being to examine variation in outcomes as a function of both disaster impact and rebuilding.









Using High Resolution Imagery and Neural Networks to Measure Destruction and Reconstruction after a Disaster. Elizabeth Frankenberg, Keenan Karrigan, Peter Katz, Eric Peshkin, Cecep Sumantri, and Duncan Thomas.

The quantification of markers of economic development from time-series satellite imagery using deep learning. Eric Peshkin, 2018.

Nighttime lights time series of tsunami damage and recovery in Sumatra, Indonesia. Thomas Gillespie, Elizabeth Frankenberg, Kai Fung Chum and Duncan Thomas. Remote Sensing Letters, 5.3:286-294, 2014.

Assessment of Natural Hazard Damage and Reconstruction: A Case Study from Banda Aceh, Sumatra. Thomas Gillespie, Elizabeth Frankenberg, Matt Braughton, Abigail Cook, Tiffany Armenta and Duncan Thomas. California Center for Population Research, December, 2009.

Assessment and prediction of natural hazards from satellite imagery. Thomas Gillespie, Jasmine Chu, Elizabeth Frankenberg, and Duncan Thomas. Progress in Physical Geography, 31.5:459-70, 2007.