In recent years, the devastating impacts of wildfires have surged dramatically across wildfire-prone regions, particularly in the western United States. The catastrophic fires that engulfed Southern California earlier this year serve as a stark reminder of this growing crisis, claiming 30 lives, razing over 18,000 homes, and scorching more than 57,000 acres of land. Beyond the immediate destruction of property and loss of life, these wildfires unleash dense smoke clouds that drift far beyond the flames, exposing millions to hazardous air pollutants. The pervasive presence of particulate matter in wildfire smoke, especially fine particles known as PM2.5, poses serious and long-term health risks even to populations located miles downwind from the origin of the fire.
A groundbreaking development from a Harvard atmospheric research team aims to directly address this critical but often overlooked dimension of wildfire impacts: smoke exposure. Led by Loretta Mickley, a senior research fellow at Harvard’s John A. Paulson School of Engineering and Applied Sciences and head of the Atmospheric Chemistry Modeling Group, the team has developed an innovative online platform intended to aid fire managers and policymakers in precisely targeting land management strategies to curtail smoke exposure. This tool, known as SMRT-Flames, integrates advanced atmospheric modeling with population data, offering a novel approach that shifts focus from mere fire risk prediction to the quantification of smoke exposure risk affecting human communities.
At the heart of SMRT-Flames lies a sophisticated computational framework that assimilates meteorological, chemical, and geographical information to simulate fire behavior and forecast smoke dispersion patterns on a high spatial resolution. This enables identification of grid cells where wildfires would produce the greatest downwind population-weighted smoke exposure. Unlike traditional fire risk models that only estimate where fires are likely to ignite and spread, SMRT-Flames models the human health dimension by estimating who and how many individuals are exposed to hazardous smoke concentrations, allowing for a more nuanced understanding of wildfire impacts.
This research focused initially on Northern California, one of the United States’ most fire-prone regions, notorious for both the frequency and severity of its wildfires. Using retrospective data from the 2020 fire season, the team applied their methodology to estimate the potential benefits of targeted land management interventions. Their results were compelling: by conducting controlled burns or similar strategies in just 3.5% of the region’s highest smoke-risk areas, overall smoke exposure could have been reduced by as much as 18% during that year. Such targeted fuel management strategies effectively reduce the accumulation of combustible vegetation, thus lowering the likelihood of catastrophic fires generating widespread smoke pollution.
PM2.5 particles—fine airborne particulates with diameters less than 2.5 microns—constitute the main hazardous agent in wildfire smoke, carrying significant health risks. Because of their minute size, PM2.5 can penetrate deep into the lungs and enter the bloodstream, exacerbating respiratory illnesses like asthma and cardiovascular conditions, and increasing the mortality risk among vulnerable populations including the elderly. The team estimated that in 2020 alone, complications linked to smoke exposure contributed to approximately 36,400 premature deaths across the western United States, underscoring the urgency of incorporating smoke risk into wildfire management policies.
The SMRT-Flames application empowers stakeholders by allowing them to explore hypothetical fire scenarios and simulate the effects of prescribed burns on reducing smoke exposure across broader geographic regions. This regional-scale perspective is pivotal because it takes into account how smoke from prescribed fires disperses downwind, affecting areas well beyond the immediate vicinity of controlled burns. Fire managers gain the ability to strategically plan burns in manners that minimize public health impacts while still achieving ecological and wildfire prevention objectives.
Underpinning the SMRT-Flames platform is the GEOS-Chem model, a community-developed atmospheric chemistry transport model widely recognized for its capability to synthesize meteorological, chemical, and physical data for air quality forecasting. By simulating fire emissions and atmospheric transport processes, GEOS-Chem provides a dynamic, multidimensional representation of smoke behavior. This integration offers unprecedented precision in mapping how smoke from specific wildfire events or prescribed burns would translate into exposure risks for individual populations downwind.
Prescribed burns emerge from this research as a potent land management strategy. These controlled, low-intensity fires intentionally reduce understory fuel loads accumulated due to a century of fire suppression policies, which ironically have created conditions for more devastating wildfires. By safely removing excess vegetation, prescribed burns can mitigate the intensity and spread of larger wildfires, thereby diminishing resultant smoke emissions. Though common in some parts of the U.S., prescribed burning remains underutilized in much of the West, primarily due to public perception, regulatory complexities, and logistical challenges.
The research team points to the unique challenges posed by the wildland-urban interface—zones where built environments meet undeveloped wildlands. Populations residing in these transitional areas were found to be disproportionately vulnerable to smoke exposure. This finding heightens the need for targeted fuel treatments and prescribed burning protocols in proximity to residential zones, an approach that has sparked renewed debate balancing public health, safety, and ecological stewardship.
Methodologically, integrating diverse disciplinary insights—from atmospheric science to land cover analysis—was imperative to overcome the complex confounding factors that obscure accurate smoke risk estimation. Variations in meteorological conditions, topography, vegetation types, and fire behavior introduce significant modeling challenges. The team’s multidisciplinary approach allowed for a comprehensive conceptualization and quantitative representation of smoke risk that explicitly accounts for these factors, enabling more reliable predictions relevant to practical fire management decisions.
Co-led by alumnae Tianjia (Tina) Liu and Makoto Kelp, who now respectively hold academic positions at the University of British Columbia and Stanford University, the project showcases the power of collaborative climate research networks. Their work builds on earlier studies suggesting that prescribed burns across key wildfire zones in the West—from Northern California to Eastern Washington and Western Oregon—could dramatically reduce smoke pollution region-wide, potentially saving thousands of lives annually.
Looking forward, the researchers envision expanding the smoke risk modeling platform beyond Northern California to inform wildfire management strategies on a national and global scale. With wildfires projected to increase in severity and frequency due to ongoing climate change, tools like SMRT-Flames that explicitly incorporate smoke exposure metrics hold promise for enhancing wildfire resilience and public health protection in vulnerable communities worldwide.
This pioneering research received key support from the NOAA Climate Program Office’s Modeling, Analysis, Predictions, and Projections Program, as well as postdoctoral fellowships awarded to Liu and Kelp through the NOAA Climate and Global Change initiative. Together with co-authors Karn Vohra, Dana Skelly, Matthew Carroll, and Joel Schwartz, the team’s findings mark a significant step toward bridging atmospheric science and practical fire management, illuminating pathways to reduce the collateral damage inflicted by wildfires in an increasingly fire-prone era.
Subject of Research: Not applicable
Article Title: Managing Smoke Risk from Wildland Fires: Northern California as a Case Study
News Publication Date: 30-Jun-2025
Web References:
– SMRT-Flames application: https://smoke-policy-tool.projects.earthengine.app/view/smrt-flames
– GEOS-Chem model: https://geoschem.github.io/index.html
– Article DOI: https://doi.org/10.1021/acs.est.5c01914
References:
– Mickley Lab / Harvard SEAS study published in Environmental Science & Technology, 2025
Image Credits: Mickley Lab / Harvard SEAS
Keywords: Forest fires, Atmospheric science, Climatology, Earth systems science, Natural disasters, Wildfires, Environmental chemistry, Atmospheric chemistry, Greenhouse gases, Geography
