Over the last 20 years, the Kirk R. Smith research group at the University of California Berkeley—in collaboration with Electronically Monitored Ecosystems, Berkeley Air Monitoring Group, and other academic institutions—has developed a suite of relatively inexpensive, rugged, battery-operated, microchip-based devices to quantify parameters related to household air pollution. These devices include two generations of particle monitors; data-logging temperature sensors to assess time of use of household energy devices; a time-activity monitoring system using ultrasound; and a CO₂-based tracer-decay system to assess ventilation rates. Development of each system involved numerous iterations of custom hardware, software, and data processing and visualization routines along with both lab and field validation. The devices have been used in hundreds of studies globally and have greatly enhanced our understanding of heterogeneous household air pollution (HAP) concentrations and exposures and factors influencing them.

Winter air pollution in Ulaanbaatar, Mongolia is among the worst in the world. The health impacts of policy decisions affecting air pollution exposures in Ulaanbaatar were modeled and evaluated under business as usual and two more-strict alternative emissions pathways through 2024. Previous studies have relied on either outdoor or indoor concentrations to assesses the health risks of air pollution, but the burden is really a function of total exposure. This study combined projections of indoor and outdoor concentrations of PM_2.5 with population time-activity estimates to develop trajectories of total age-specific PM_2.5 exposure for the Ulaanbaatar population. Indoor PM_2.5 contributions from secondhand tobacco smoke (SHS) were estimated in order to fill out total exposures, and changes in population and background disease were modeled. The health impacts were derived using integrated exposure-response curves from the Global Burden of Disease Study. Annual average population-weighted PM_2.5 exposures at baseline (2014) were estimated at 59 μg/m³. These were dominated by exposures occurring indoors, influenced considerably by infiltrated outdoor pollution. Under current control policies, exposures increased slightly to 60 μg/m³ by 2024; under moderate emissions reductions and under a switch to clean technologies, exposures were reduced from baseline levels by 45% and 80%, respectively. The moderate improvement pathway decreased per capita annual disability-adjusted life year (DALY) and death burdens by approximately 40%. A switch to clean fuels decreased per capita annual DALY and death burdens by about 85% by 2024 with the relative SHS contribution increasing substantially. This study demonstrates a way to combine estimated changes in total exposure, background disease and population levels, and exposure-response functions to project the health impacts of alternative policy pathways. The resulting burden analysis highlights the need for aggressive action, including the elimination of residential coal burning and the reduction of current smoking rates.

Black carbon (BC) or soot contains ultrafine combustion particles that are associated with a wide range of health impacts, leading to respiratory and cardiovascular diseases. Both long-term and short-term health impacts of BC have been documented, with even low-level exposures to BC resulting in negative health outcomes for vulnerable groups. Two aethalometers—AethLabs MA350 and Aerosol Magee Scientific AE33—were co-located at a Utah Division of Air Quality site in Bountiful, Utah for just under a year. The aethalometer comparison showed a close relationship between instruments for IR BC, Blue BC, and fossil fuel source-specific BC estimates. The biomass source-specific BC estimates were markedly different between instruments at the minute and hour scale but became more similar and perhaps less-affected by high-leverage outliers at the daily time scale. The greater inter-device difference for biomass BC may have been confounded by very low biomass-specific BC concentrations during the study period. These findings at a mountainous, high-elevation, Greater Salt Lake City Area site support previous study results and broaden the body of evidence validating the performance of the MA350.

Ambient fine particulate matter pollution (PM2.5) is the leading global environmental risk factor for human health. This study leverages three years (April 2022-March 2025) of continuous, in situ ambient PM2.5 and black carbon (BC) measurements from the Multi-Angle Imager for Aerosols (MAIA) investigation’s surface monitoring network to characterize particulate air pollution in Addis Ababa, Ethiopia. Continuous PM2.5 data were collected from 10 PurpleAir sensors deployed across Addis Ababa and calibrated using collocated Beta Attenuation Monitors at two contrasting U.S. Embassy sites (Central and Jacros). Continuous BC data were collected with AethLabs MA350 microAethalometers operating at these two sites. During the study period, the average PM2.5 concentration across all 10 sites was 30 μg/m3, and the average BC concentrations were 3.7 μg/m3 and 7.7 μg/m3 at Central and Jacros, respectively. Both PM2.5 and BC concentrations exhibited strong diurnal and seasonal variation, with peaks during morning traffic and nighttime inversion, especially during the wet season. PM2.5 levels were highest at sites west of the city center and exceeded the WHO guideline for daily PM2.5 of 15 μg/m3 on 99% of days. The Aethalometer Model for source apportionment estimated that, on average, 94% of total BC in Addis Ababa was attributable to fossil fuel combustion, while 6% was associated with biomass burning. Directional and trajectory analyses indicate distinct local and regional sources of biomass and fossil fuel BC, including potential long-range transport of shipping emissions from the Gulf of Aden. This study presents, to our knowledge, the first long-term, multisite characterization of continuous PM2.5 and source-apportioned BC in Ethiopia. Findings from this study establish a pollution baseline for this MAIA target area ahead of the satellite’s launch, improving understanding of particulate pollution levels and sources and supporting future assessment of air quality trends in this region.

The portable microAeth® MA200 (MA200) is widely applied for measuring black carbon (BC) in human exposure characterization and mobile air quality monitoring. However, the field lacks information about this instrument's performance under various settings. This study evaluated the real-time performance of the MA200 in an urban area, Augsburg, Germany. Noise reduction and negative value mitigation were explored using different data processing methods: local polynomial regression (LPR), optimized noise reduction averaging (ONA), and centered moving average (CMA) under different interval time (5 s, 10 s, and 30 s). After noise reduction, the data were evaluated and compared by (1) the relative number of negative values; (2) more detailed microenvironmental change information retained after noise reduction; (3) the reduction of the peak values and number of peak samples; (4) more detailed microenvironmental change retained after the background correction. Our results showed that CMA showed a good prospect to analyze the raw BC concentration data in terms of the interval time due to its proportions of negative values and the detail microenvironmental change. Moreover, the CMA method has the highest reduction peak values and the number of peak samples compared to ONA and LPR. Furthermore, after background correction, the CMA treatment results remained more detailed microenvironmental changes in pollutants than others. Therefore, based on a comprehensive comparison, CMA offered a good approach to post-process the raw BC concentration data. These findings provide new insight for the noise reduction approach that applied in mobile monitoring campaign using BC instruments.