On January 9, roughly a full day after the Eaton blaze ravaged the foothill region of Altadena, situated north of the Caltech campus in Pasadena, California, various specialized scientific devices in a shipping container located in Pico Rivera, around 11 miles to the south, registered a surge in atmospheric levels, especially of chlorine and lead—both recognized as harmful even at minimal concentrations. At their highest that day, chlorine concentrations surged to nearly 40 times the regular levels, while lead peaked at over 100 times the standard amount.
It was not mere chance that the air-monitoring devices were functioning during and directly after the fire. In reality, they operate almost continuously, 24 hours a day, 365 days a year.
Since July 2023, Caltech has managed the Pico Rivera site of what is referred to as the Atmospheric Science and Chemistry mEasurement NeTwork (ASCENT), a network comprising 12 such locations. Funded by the National Science Foundation and directed by Nga Lee (Sally) Ng (PhD ’07) from the Georgia Institute of Technology, ASCENT aims to continuously track particulate matter in the atmosphere to gain a more comprehensive understanding of the evolving chemical makeup and physical attributes of these aerosols. Real-time data from this site is now accessible online.
“This network was established to provide high-quality real-time, high-resolution characterization of particulate matter,” clarifies Haroula Baliaka (MS ’23), a graduate scholar at Caltech who, along with alumnus Ryan Ward (PhD ’25), aided in the installation of the ASCENT instruments in the shipping container at an air quality monitoring station operated by the South Coast Air Quality Management District (SCAQMD). “When the Canadian wildfires occurred last year, the New York ASCENT site was tremendously significant as it received the aged plume days later. And currently, we’re gathering data regarding the Eaton fire as well,” Baliaka states.
During and right after the Los Angeles fires, the winds primarily blew southward, directing the plumes from the Eaton fire straight towards the Pico Rivera ASCENT station. As a result, the researchers believe that the observations collected there are chiefly related to the Eaton fire.
Although the peaks observed by the instruments in the days following the fires are significantly higher than normal, Baliaka emphasizes the necessity of contextualizing the data. Lead and chlorine are considered trace elements in the atmosphere, as they are typically found within the nanograms-per-cubic-meter range. Following the fire, these elements reached levels of micrograms per cubic meter. However, Baliaka points out, “As recently as 2010, lead concentrations regularly recorded were greater than what we noted after the fires due to factors like leaded gasoline. The further back you look, the more frequently these higher concentrations were deemed ‘normal.’ Fortunately, we have now adapted to cleaner air.”
Information from the Environmental Protection Agency (EPA) reveals that as a result of regulations, including the ban on lead in gasoline, atmospheric lead levels dropped by 98 percent from 1980 to 2014.
The Caltech researchers observe that the elemental measurements recorded at the Pico Rivera site have now reverted to standard levels.
A noteworthy aspect of the data is that it lacks the typical characteristics of a wildfire regarding the elements detected, states Richard Flagan, the Irma and Ross McCollum-William H. Corcoran Professor of Chemical Engineering and Environmental Science and Engineering at Caltech and co-principal investigator of the Pico Rivera ASCENT site alongside Caltech’s John Seinfeld, the Louis E. Nohl Professor of Chemical Engineering, Emeritus.
“Decades of research have provided a very clear signature of what a wildfire resembles,” Flagan states. For instance, potassium, vital for plant life, is generally employed as a tracer for biomass combustion. However, after the Eaton fire, potassium concentrations were eclipsed by unexpected elements. “Our findings deliver a clear indication that this was not a wildfire scenario. This was a wind-driven urban firestorm, which significantly differs in terms of air pollution,” he asserts.
A pivotal aspect of the ASCENT network is that all 12 sites employ the same suite of four instruments conforming to identical operating protocols, allowing for data comparison nationwide.
One instrument, known as a scanning mobility particle sizer (SMPS), was developed by Flagan. It provides a comprehensive distribution of particle sizes detected over time. Initially, when the fires ignited, the SMPS data indicated that concentrations of submicron particles declined as high winds cleared the atmosphere. However, when the wind calmed and smoke from the fires reached Pico Rivera, the SMPS recorded substantial increases in both particle concentration and size.
“We observe that a few days post-fires, when the smoke arrives at Pico Rivera, we experience a significant spike in the total mass of aerosols,” remarks Vine Blankenship, a graduate student at Caltech who has concentrated on the SMPS data.
“There are evident statistical relationships between aerosols—particulate matter smaller than 2.5 microns in diameter, referred to as PM2.5—and health,” notes Flagan. He emphasizes that it took two decades of data collection to evolve from merely defining PM2.5 in the late 1970s to establishing the statistical health associations necessary for the EPA to integrate PM2.5 into air-quality standards. “Typically, air-quality measurements lack the granularity required for establishing those connections. Most data have been collected by networks established by research teams aiming to comprehend the effects of air pollution. That underlines the ongoing significance of work like this.”
The ASCENT data have piqued the interest of numerous air-quality researchers and organizations. Additional institutions have already visited the Pico Rivera site to gather data with supplementary instruments. The SCAQMD is also in the process of setting up two more aerosol measurement stations to collect further data points.