Andre Coleman, PhD

PNNL featured experts are available to provide scientific and technical expertise to the news media. To arrange an interview with a PNNL expert, contact the PNNL News & Media team. To search the entire database, visit Featured Experts.

Andre Coleman, PhD

Data Scientist

Biography

Topographical maps of Utah’s Uinta Mountains spread across the family kitchen table and in front of young Andre Coleman. The grade schooler was looking forward to an outdoors adventure with his dad. He did not expect to be dazzled by something else: maps.

“That ultimately led to this,” said Coleman.

“This” would be life as a PNNL data scientist and expert at the intersection of geography and computing to help inform decision-makers about natural, social, and built systems including sustainable energy planning, water security, extreme event mitigation and resilience planning, geointelligence, social justice impacts, and disaster management.

One aspect of Coleman’s research has led to advances in monitoring active wildfires and determining future wildfire risk by assembling numerous sources of disparate data, implementing machine-learning models for data processing and image analytics, and developing intuitive mapping products and feeding wildfire behavior models.

Coleman is credited with creating a wildfire modeling system that uses satellites to monitor wildfires and assess risk. The innovation, called RADR-Fire, is a wildfire-focused extension to the Rapid Analytics for Disaster Response (RADR) software (2020 R&D 100 winner) and provides decision-makers with the data to prepare, respond, and mitigate current and future fires.

A Department of Homeland Security Science and Technology Office request prompted Coleman and colleagues to initially develop the RADR system.

“We already had a track record at PNNL in conducting remote sensing and image analytics from satellite images,” Coleman said. “When the Homeland Security S&T opportunity arose, we refined our approach, choosing a large mix of satellites, some of which are experimental, that we thought would provide the best data at the fastest rate.”