Stanford’s SmokeCast project is testing truck-mounted sensors to forecast toxic metals in wildfire smoke, addressing a gap in firefighter exposure monitoring.
In early July 2026, researchers from Stanford University mounted a briefcase-sized prototype on a Colorado fire truck to collect real-time data for forecasting toxic metals in wildfire smoke.
Stanford's SmokeCast project is developing a forecasting system intended to show where wildfire smoke carries toxic metals released from soil. The project is led by Stanford soil scientist and biogeochemist Scott Fendorf. Postdoctoral researcher Alex Honeyman and graduate student Mark Leone are contributing to the hardware, field sampling, and model development. This team is working to address a critical gap in conventional smoke monitoring.
Current air-quality ratings generally focus on fine particulate matter such as PM2.5. These standard measurements indicate particle concentration, but they do not identify the chemistry of those particles or the specific metals they contain. Fendorf noted there is currently no way to monitor the metal content of wildfire smoke in real time. There is also no current way to forecast where that metal content will travel as a plume moves.
The underlying science rests on the finding that naturally occurring metals locked in soil can be transformed into toxic airborne particles by a fire. The researchers are studying how local geology and fire conditions influence that specific release. Honeyman described the lack of chemical information in conventional air-quality data as a massive gap for both scientists and frontline personnel. His concern is operational as well as scientific.
The project team asked what firefighters could do differently on a daily basis if they received more detailed information about the smoke around them. SmokeCast's model combines local geology, fire temperature, smoke distribution, wind, and weather conditions. It also integrates historical EPA smoke-particle measurements, truck-mounted filter samples, and soil measurements. The model is intended to estimate whether metals in local soil are likely to become airborne and where those particles may travel.
The reality of the fire line is that smoke contains a complex mixture of hazards. A 2026 real-world study of respirator cartridges used by wildland firefighters in British Columbia clearly illustrated this chemical density. Researchers analyzed samples from 19 firefighters and four types of replaceable filter cartridges using thermal-desorption gas chromatography-mass spectrometry. The study detected 496 distinct wildfire-smoke compounds in the respirators.
Of those captured compounds, 24 appeared in public databases as carcinogens or endocrine disruptors. The researchers identified Group 1 carcinogens including benzene. They also found Group 2B compounds that are possibly carcinogenic to humans. Understanding this chemical density changes how crews should approach their protective equipment.
Because respirators capture both particles and gases, a toxic metal forecast alone cannot dictate the full protection requirement. The British Columbia study found that approximately 55 percent of the compounds identified in paper or microfiber filter layers were oxygenated aromatics. Meanwhile, 36 percent of compounds captured in activated-carbon layers were aromatic hydrocarbons. Polycyclic aromatic hydrocarbons were detected in both the particle-filter and activated-carbon layers.
This data validates the need for comprehensive exposure tracking across a long career. Integrating these findings into daily operations means treating smoke as chemically variable rather than just visibly dense. By understanding the ongoing research around respiratory health after fire, smoke and emergency-service exposure, departments can better prepare their personnel. The goal is to build a clearer picture of occupational hazards.
The complex composition of wildfire smoke means that a toxic metal forecast will eventually serve as an additional layer of information, not a replacement for existing safety protocols. Firefighters must continue tracking weather patterns, standard PM2.5 data, and incident-command guidance. A narrow focus on just one element of smoke chemistry risks ignoring other hazardous constituents. Broadening the scope of hazard awareness directly supports long-term physical capability.
A persistent assumption on the fire line is that low visibility equals high exposure. Fendorf and the SmokeCast team actively challenge this specific operational heuristic. Current smoke and air-quality forecasts may successfully identify the arrival of the main visible plume. However, Fendorf pointed out that toxic metals could potentially be present earlier or travel differently from the most visible portion of the smoke.
This distinction is critically important for working firefighters. A low-visibility or changing smoke environment does not necessarily indicate low chemical exposure. Relying on visual smoke cues to dictate voluntary respirator use leaves crews vulnerable to invisible chemical loads.
Operating under the assumption that clear air means clean air is a dangerous habit developed over decades of fire service culture. The British Columbia study reinforced that the chemical load in the air can be severe even when the immediate smoke density seems manageable. The study reported that there are no mandatory requirements for wildland firefighters to wear respirators in the setting examined. This lack of strict mandate places the burden of safety decisions squarely on the individual responder.
The project also challenges assumptions about how to collect useful field data. The research team originally considered using drones to collect smoke samples inside plumes. The researchers later recognized that fire trucks could gather samples near active incidents without requiring a dedicated drone pilot. This practical design choice links sampling directly to vehicles already moving through wildfire operations.
Former fire chief Michael Schmitt hosted the initial prototype on his truck. Schmitt advised the project on the data firefighters would find useful and how to recruit crews for testing. He also provided guidance on how to make the smartphone app usable in the physically demanding field environment.
While forecasting systems are still in development, the underlying research demands immediate adjustments to shift preparation. The primary adjustment is prioritizing respiratory protection equipment that crews will actually wear. The British Columbia study noted that wildland firefighters may be reluctant to wear respirators during hot and unstable work. Comfort, breathability, and communication concerns heavily affect compliance during physically demanding operations.
The researchers noted that a respirator that is less effective but worn consistently provides more real-world protection than a theoretically superior mask that stays in the truck. Crews must maintain their ordinary decontamination and respiratory protection practices regardless of how the smoke looks. The current reporting on the SmokeCast system does not specify a formal respirator selection protocol, fit testing requirement, or filter change schedule. The British Columbia study specifically identified filter change timing as an area requiring further investigation.
Firefighters should not wait for a heavy plume to justify wearing their masks. Do not infer safety solely from the absence of visible smoke. Departments should factor these chemical exposures into their understanding of environmental exposures and hormonal health. Consistent gear cleaning and strict adherence to available protocols remain essential for long-term operational capability.
SmokeCast is still a research and development project rather than a validated operational exposure control system. The reported Colorado deployment involved a single prototype-equipped truck traveling approximately 300 miles, or 480 kilometers, to a wildfire. The equipment combined air filters, smoke-sampling devices, and satellite-linked sensors. These sensors measured temperature, humidity, and toxic gases such as carbon monoxide.
The truck-mounted system sent information to a smartphone app in real time. According to the project reports, the app helped the crew monitor environmental conditions and assess the probability of additional ignition. It also allowed them to receive toxic metal estimates and make decisions about protective gear.
The truck returned to its station after a 14-day field period. At that point, the crew recovered a cartridge containing smoke samples for analysis at Fendorf's Stanford laboratory. The team plans to expand the program from this initial prototype to dozens or hundreds of fire trucks over the next few years. Fendorf said the project's near-term goal is a more accurate forecast within a year.
He hopes that people will eventually be able to check air toxicity through a phone within a few years. The longer-term ambition is a nationwide forecast of toxic metals in wildfire smoke. This progression moves from regional forecasting toward exposure estimates at the level of individual firefighters.
This expansion introduces necessary challenges involving calibration, maintenance, data consistency, and interpretation across different wildfire environments. A toxic metal forecast would also be only one part of a broader exposure picture. Measuring or forecasting metals cannot by itself characterize every hazardous constituent in a fire. The proposed individual-level exposure estimates should not be confused with medical diagnosis.
A forecast may help identify higher-risk conditions on a specific shift, but it does not establish whether a firefighter has absorbed a harmful dose. It also does not predict whether an individual will develop an illness later in life. Building exposure records over time remains a prospective benefit rather than a demonstrated outcome of the current prototype. Tracking these ongoing pilot programs will help departments prepare for a future where personal chemical dosimetry becomes standard.
Improving exposure monitoring on the fire line is a necessary operational step, and translating those environmental realities into sustainable recovery habits is where our work begins. When conflicting advice that prioritises short-term performance while ignoring long-term health complicates your routine, Rescuris provides clear guidance built for the reality of your work. Explore Resources
Rescuris publishes evidence-aware guidance for firefighters, police officers, paramedics and other first responders on fitness, recovery, sleep, nutrition, stress resilience and long-term capability.
Read practical guidance on first responder fitness, recovery, sleep, nutrition and resilience for demanding work and a long, active career.
read the blog
