AWS Public Sector Blog

Understanding wildfire risk in a changing climate with open data and AWS

A long-haired firefighter looks out across a forest landscape covered in flames.

A changing climate is impacting the risks facing American properties, communities, and businesses as perils like flood, fire, heat, and drought become more common and more severe. The First Street Foundation (First Street), a nonprofit research and technology group, is committed to making climate risk information accessible, simple to understand, and actionable for individuals, governments, and industry. Building upon modern modeling techniques validated through peer-reviewed science, the foundation’s team of data scientists, modelers, and researchers has calculated the present and future wildfire risk of every home and property in the contiguous United States (CONUS). The public now has access to the data at no cost through Risk FactorTM, an online tool that makes it simple to learn about climate-driven risk from wildfire, flood, and heat.

 As part of the Amazon Sustainability Data Initiative (ASDI), Amazon Web Services (AWS) invited Dr. Ed Kearns, the Chief Data Officer of First Street Foundation, to share how AWS technologies and open data are supporting their mission to provide accurate and up-to-date information on climate related risks.

Linking wildfire risk and climate science

 Wildfires across the United States are becoming more frequent and more severe every year. However, most Americans are not aware of the level of wildfire risk they are personally exposed to today, much less how their risk will change over the next 30 years as the climate changes. To address this, First Street created the first-ever property-specific and climate-adjusted assessment of wildfire risk for CONUS, and in May 2022 began sharing that information for approximately 143 million properties through Risk FactorTM.

The wildfire risk estimates are generated by running wildfire behavior models created by the Pyregence Consortium of wildfire scientists—a group of leading researchers from 18 institutions across industry, academia, and government, as well as software developers and designers—who have partnered with First Street to create over 100,000,000 simulated wildfires. These simulations account for a wide range of possible weather conditions, informed by current and future modeled climate conditions under the Intergovernmental Panel on Climate Change’s emission scenarios.

The First Street Foundation Wildfire Model shows a specific location’s probabilistic risk of wildfire based on the vegetation, topography, and weather in the surrounding area. The process starts by asking the question: “What would burn if a wildfire were to occur?” The First Street Foundation Wildfire Model uses data from the United States Forest Service to identify the type, quantity, age, and condition of the vegetation across the Continental US that would provide “fuel” for a potential wildfire. One of the things that makes this model unique is that it also incorporates fuels in the Wildland Urban Interface (WUI), which are areas with residences built into or near the wild landscape in the zone of transition between unoccupied land and human development, to account for wildfire spreading to homes within those areas based on patterns observed in 500 similar historic wildfires. Learn more about the model works in this published paper on MDPI Open Access Journals.

The result of these efforts generates wildfire hazards at 30 meters horizontal resolution across all of CONUS, expressed as the likelihood of wildfire exposure, the likely mean and maximum flame lengths to be experienced, and whether exposure to flying embers can be expected. While this exposure information is expressed as a 30-year aggregate 1-10 score for each parcel, the model also assesses the vulnerability of buildings on a parcel given the characteristics known about that home or business. First Street has partnered with global engineering experts at Arup, a global collective of designers, engineering and sustainability consultants, advisors, and experts dedicated to sustainable development and more, to estimate the likelihood that any building will ignite and be lost if exposed to flame or embers. This information may be used to estimate the expected wildfire losses for individual homes, communities, counties, or states today and 30 years into the future.

Leveraging the AWS Cloud to generate, access, and distribute key datasets

Modeling wildfire requires bringing together datasets from many US federal agencies like the National Oceanic and Atmospheric Administration (NOAA), the National Aeronautics and Space Administration (NASA), the United States Forest Service (USFS), and the United States Geological Survey (USGS). Some of which are made available at no cost on AWS through the ASDI data catalog. Then these datasets are used to drive the wildfire behavioral model under many possible scenarios.

The power and scalability of the AWS Cloud enables small nonprofit organizations like First Street to conduct this type of large-scale, high resolution modeling and data analyses, and supports success and reliability. Modern processing schemes allow the fire behavior model and input data to be distributed, and use processors where available, either on Amazon Elastic Compute Cloud (Amazon EC2) or First Street’s more limited on-premises systems. First Street uses AWS infrastructure, including the Amazon Relational Database Service (Amazon RDS) for PostgreSQL to analyze large amounts of data for individual property and building assessments. The generated wildfire hazard data is stored as cloud-optimized GeoTIFFS and made accessible through Amazon Simple Storage Service (Amazon S3) for both internal First Street and external consumer use. The Risk FactorTM website uses Amazon Athena, Amazon CloudWatch, Amazon OpenSearch Service, and Amazon CloudFront, among other AWS services, to deliver the wildfire risk information directly to users and to public-facing collaborators, such as Realtor.com and Redfin.com, who help deliver climate risk information for use by their many customers. First Street’s US National aggregated risk data are also now available at no cost through the AWS Data Exchange for non-commercial uses.

Using open science to inform preparation for current and future risks

 As governments, individuals, and businesses prepare for the impacts of climate change, access to trusted and high quality information is critical. By sharing climate risk information publicly, First Street seeks to enable a wide and common understanding of the risks facing the nation, and hopes to remove the asymmetry of information that can lead to inequities in responses. First Street believe that by using open data sources, publishing all its methodologies in peer-reviewed journals, and employing open science practices, First Street can build trust among its data users and facilitate effective action to address climate change. The nonprofit has established data sharing agreements with over 30 US federal agencies; numerous private sector firms in the real estate, reinsurance, banking, investment, and financial industries; and welcomes many thousands of interested consumers to its Risk FactorTM web site every day. In the coming years, First Street will continue to update and refine its estimates of wildfire, flood, and heat risks, and will be adding new assessments of risk from other climate perils such as drought, air quality, and wind.

The wildfire risk work from First Street Foundation was partially supported through an ASDI cloud grant. Discover more about AWS Promotional Credit for sustainability research here. Learn more about First Street Foundation and the Amazon Sustainability Data Initiative (ASDI).

Read more about ASDI on the AWS Public Sector Blog:


Subscribe to the AWS Public Sector Blog newsletter to get the latest in AWS tools, solutions, and innovations from the public sector delivered to your inbox, or contact us.

Please take a few minutes to share insights regarding your experience with the AWS Public Sector Blog in this survey, and we’ll use feedback from the survey to create more content aligned with the preferences of our readers.

Dr. Ed Kearns

Dr. Ed Kearns

Dr. Ed Kearns is the chief data officer at the nonprofit First Street Foundation. After 15 years of working in the federal government and 10 years as an academic at the University of Miami, he joined First Street to explore new, more effective ways of sharing and using climate change information. From his time as a research professor in oceanography to his role as a senior executive in the National Oceanic and Atmospheric Administration (NOAA), Ed has pursued novel ways of using data to unlock new insights and knowledge about our environment.

Ana Pinheiro Privette

Ana Pinheiro Privette

Dr. Ana Pinheiro Privette is the Head for Sustainability for AWS Impact Computing, and the Global Lead for the Amazon Sustainability Data Initiative (ASDI). ASDI seeks to accelerate sustainability research and innovation by minimizing the cost and time required to acquire and analyze large sustainability datasets. Ana was trained as an environmental engineer and as an earth scientist at the New University of Lisbon (Portugal) and at MIT. She spent most of her career as a research scientist at NASA and NOAA. Later, Ana worked on the US National Climate Assessment (NCA) focusing on bringing more transparency and traceability of the data sources supporting this climate report, and led projects for the White House climate portfolio, including the Obama Climate Data Initiative (CDI) and the Partnership for Resilience and Preparedness (PREP).