Air Quality Sensor Experts Convene: Current Quality Assurance Considerations for Credible Data

Air sensors can provide valuable non-regulatory and supplemental data as they can be affordably deployed in large numbers and stationed in remote areas far away from regulatory air monitoring stations. Air sensors have inherent limitations that are critical to understand before collecting and interpreting the data. Many of these limitations are mechanistic in nature, which will require technological advances. However, there are documented quality assurance (QA) methods to promote data quality. These include laboratory and field evaluation to quantitatively assess performance, the application of corrections to improve precision and accuracy, and active management of the condition or state of health of deployed air quality sensors. This paper summarizes perspectives presented at the U.S. Environmental Protection Agency's 2023 Air Sensors Quality Assurance Workshop (https://www.epa.gov/air-sensor-toolbox/quality-assurance-air-sensors#QAworkshop) by stakeholders (e.g., manufacturers, researchers, air agencies) and identifies the most pressing needs. These include QA protocols, streamlined data processing, improved total volatile organic compound (TVOC) data interpretation, development of speciated VOC sensors, and increased documentation of hardware and data handling. Community members using air sensors need training and resources, timely data, accessible QA approaches, and shared responsibility with other stakeholders. In addition to identifying the vital next steps, this work provides a set of common QA and QC actions aimed at improving and homogenizing air sensor QA that will allow stakeholders with varying fields and levels of expertise to effectively leverage air sensor data to protect human health.

Evaluation of Two Low-Cost Optical Particle Counters for the Measurement of Ambient Aerosol Scattering Coefficient and Ångström Exponent

The aerosol scattering coefficient and Ångström exponent (AE) are important parameters in the understanding of aerosol optical properties and aerosol direct effect. These parameters are usually measured by a nephelometer network which is under-represented geographically; however, a rapid growth of air-pollution monitoring, using low-cost particle sensors, may extend observation networks. This paper presents the results of co-located measurements of aerosol optical properties, such as the aerosol scattering coefficient and the scattering AE, using low-cost sensors and using a scientific-grade polar Aurora 4000 nephelometer. A high Pearson correlation coefficient (0.94-0.96) between the low-cost particulate matter (PM) mass concentration and the aerosol scattering coefficient was found. For the PM10 mass concentration, the aerosol scattering coefficient relation is linear for the Dfrobot SEN0177 sensor and non-linear for the Alphasense OPC-N2 device. After regression analyses, both low-cost instruments provided the aerosol scattering coefficient with a similar mean square error difference (RMSE) of about 20 Mm-1, which corresponds to about 27% of the mean aerosol scattering coefficient. The relative uncertainty is independent of the pollution level. In addition, the ratio of aerosol number concentration between different bins showed a significant statistical (95% of confidence level) correlation with the scattering AE. For the SEN0177, the ratio of the particle number in bin 1 (radius of 0.15-0.25 µm) to bin 4 (radius of 1.25-2.5 µm) was a linear function of the scattering AE, with a Pearson correlation coefficient of 0.74. In the case of OPC-N2, the best correlation (r = 0.66) was found for the ratio between bin 1 (radius of 0.19-0.27 µm) and bin 2 (radius of 0.27-0.39 µm). Comparisons of an estimated scattering AE from a low-cost sensor with Aurora 4000 are given with the RMSE of 0.23-0.24, which corresponds to 16-19%. In addition, a three-year (2016-2019) observation by SEN0177 indicates that this sensor can be used to determine an annual cycle as well as a short-term variability.

Investigation of distribution, transportation, and impact factors of atmospheric black carbon in the Arctic region based on a regional climate-chemistry model

Black carbon (BC) as the main component of pollutants in the Arctic plays an important role on regional climate change. In this study, we applied the regional climate-chemistry model, WRF-Chem, to investigate the spatial distribution, transportation, and impact factors of BC in the Arctic. Compared with reanalysis data and observations, the WRF-Chem performed well in terms of the seasonal variations of meteorological parameters and BC concentrations, indicating the applicability of this model on Arctic BC simulation works. Our results showed that the BC concentrations in the Arctic had an obviously seasonalvariation pattern. Surface BC concentrations peaked during winter and spring seasons, while the minimum occurred during summer and autumn seasons. For the vertical distribution, BC aerosols mainly concentrated in the Arctic lower troposphere, and most of BC distributed near the surface during winter and spring seasons and in the higher altitude during other seasons. The seasonality of BC was associated with the seasonal change of meteorological field. During winter, the significant northward airflow prevailing in northern Eurasia caused the transport of accumulated pollutants from this region into the Arctic. The similar but weakened northward airflow pattern and the anticyclone activity during spring can allow pollutants to be transported to the Arctic lower troposphere. Moreover, the more stable atmosphere during winter and spring seasons made BC accumulated mainly near the surface. During summer and autumn seasons, the less stable boundary layer and the cyclone activity in the Arctic facilitated the diffusion of pollutants into the higher altitude. Meanwhile, the higher relative humidity can promote the wet removal process and lead to the relatively lower BC concentrations near the surface. Compared with the seasonal change of emission, our analysis showed that the seasonal variation of meteorological field was the main contributor for the seasonality of BC in the Arctic.

The impact of air pollutant deposition on solar energy system efficiency: An approach to estimate PV soiling effects with the Community Multiscale Air Quality (CMAQ) model

Deposition and accumulation of aerosol particles on photovoltaics (PV) panels, which is commonly referred to as "soiling of PV panels," impacts the performance of the PV energy system. It is desirable to estimate the soiling effect at different locations and times for modeling the PV system performance and devising cost-effective mitigation. This study presents an approach to estimate the soiling effect by utilizing particulate matter (PM) dry deposition estimates from air quality model simulations. The Community Multiscale Air Quality (CMAQ) modeling system used in this study was developed by the U.S. Environmental Protection Agency (U.S. EPA) for air quality assessments, rule-making, and research. Three deposition estimates based on different surface roughness length parameters assumed in CMAQ were used to illustrate the soling effect in different land-use types. The results were analyzed for three locations in the U.S. for year 2011. One urban and one suburban location in Colorado were selected because there have been field measurements of particle deposition on solar panels and analysis on the consequent soiling effect performed at these locations. The third location is a coastal city in Texas, the City of Brownsville. These three locations have distinct ambient environments. CMAQ underestimates particle deposition by 40% to 80% when compared to the field measurements at the two sites in Colorado due to the underestimations in both the ambient PM₁₀ concentration and deposition velocity. The estimated panel transmittance sensitivity due to the deposited particles is higher than the sensitivity obtained from the measurements in Colorado. The final soiling effect, which is transmittance loss, is estimated as 3.17 ± 4.20% for the Texas site, 0.45 ± 0.33%, and 0.31 ± 0.25% for the Colorado sites. Although the numbers are lower compared to the measurements in Colorado, the results are comparable with the soiling effects observed in U.S.

Changes in visibility with PM2.5 composition and relative humidity at a background site in the Pearl River Delta region

In fall-winter, 2007-2013, visibility and light scattering coefficients (bsp) were measured along with PM2.5 mass concentrations and chemical compositions at a background site in the Pearl River Delta (PRD) region. The daily average visibility increased significantly (p<0.01) at a rate of 1.1 km/year, yet its median stabilized at ~13 km. No haze days occurred when the 24-hr mean PM2.5 mass concentration was below 75 μg/m(3). By multiple linear regression on the chemical budget of particle scattering coefficient (bsp), we obtained site-specific mass scattering efficiency (MSE) values of 6.5 ± 0.2, 2.6 ± 0.3, 2.4 ± 0.7 and 7.3 ± 1.2m(2)/g, respectively, for organic matter (OM), ammonium sulfate (AS), ammonium nitrate (AN) and sea salt (SS). The reconstructed light extinction coefficient (bext) based on the Interagency Monitoring of Protected Visual Environments (IMPROVE) algorithm with our site-specific MSE revealed that OM, AS, AN, SS and light-absorbing carbon (LAC) on average contributed 45.9% ± 1.6%, 25.6% ± 1.2%, 12.0% ± 0.7%, 11.2% ± 0.9% and 5.4% ± 0.3% to light extinction, respectively. Averaged bext displayed a significant reduction rate of 14.1/Mm·year (p<0.05); this rate would be 82% higher if it were not counteracted by increasing relative humidity (RH) and hygroscopic growth factor (f(RH)) at rates of 2.5% and 0.16/year(-1) (p<0.01), respectively, during the fall-winter, 2007-2013. This growth of RH and f(RH) partly offsets the positive effects of lowered AS in improving visibility, and aggravated the negative effects of increasing AN to impair visibility.

Optical and chemical characterization of aerosols emitted from coal, heavy and light fuel oil, and small-scale wood combustion

Particle emissions affect radiative forcing in the atmosphere. Therefore, it is essential to know the physical and chemical characteristics of them. This work studied the chemical, physical, and optical characteristics of particle emissions from small-scale wood combustion, coal combustion of a heating and power plant, as well as heavy and light fuel oil combustion at a district heating station. Fine particle (PM1) emissions were the highest in wood combustion with a high fraction of absorbing material. The emissions were lowest from coal combustion mostly because of efficient cleaning techniques used at the power plant. The chemical composition of aerosols from coal and oil combustion included mostly ions and trace elements with a rather low fraction of absorbing material. The single scattering albedo and aerosol forcing efficiency showed that primary particles emitted from wood combustion and some cases of oil combustion would have a clear climate warming effect even over dark earth surfaces. Instead, coal combustion particle emissions had a cooling effect. Secondary processes in the atmosphere will further change the radiative properties of these emissions but are not considered in this study.

Temporal trends in atmospheric PM₂.₅, PM₁₀, elemental carbon, organic carbon, water-soluble organic carbon, and optical properties: impact of biomass burning emissions in the Indo-Gangetic Plain

The first simultaneous measurements and analytical data on atmospheric concentrations of PM(2.5), PM(10), inorganic constituents, carbonaceous species, and their optical properties (aerosol optical depth, AOD; absorption coefficient, b(abs); mass absorption efficiency, σ(abs); and single scattering albedo, SSA) from an urban site (Kanpur) in the Indo-Gangetic Plain are reported here. Significantly high aerosol mass concentration (>100 μg m(-3)) and AOD (> 0.3) are seen as a characteristic feature throughout the sampling period, from October 2008 to April 2009. The temporal variability in the mass fractions of carbonaceous species (EC, OC, and WSOC) is pronounced during October-January when emissions from biomass burning are dominant and OC is a major constituent (∼30%) of PM(2.5) mass. The WSOC/OC ratio varies from 0.21 to 0.65, suggesting significant contribution from secondary organic aerosols (SOAs). The mass fraction of SO(4)(2-) in PM(2.5) (Av: 12.5%) exceeds that of NO(3)(-) and NH(4)(+). Aerosol absorption coefficient (@ 678 nm) decreases from 90 Mm(-1) (in December) to 20 Mm(-1) (in April), and a linear regression analysis of the data for b(abs) and EC (n = 54) provides a measure of the mass absorption efficiency of EC (9.6 m(2) g(-1)). In contrast, scattering coefficient (@ 678 nm) increases from 98 Mm(-1) (in January) to 1056 Mm(-1) (in April) and an average mass scattering efficiency of 3.0 ± 0.9 m(2) g(-1) is obtained for PM(10) samples. The highest b(scat) was associated with the dust storm event (April 17, 2009) over northern Iraq, eastern Syria, and southern Turkey; thus, resulting in high SSA (0.93 ± 0.02) during March-April compared to 0.82 ± 0.04 in October-February. These results have implications to large temporal variability in the atmospheric radiative forcing due to aerosols over northern India.

Intake fraction of urban wood smoke

Intake fraction (iF), the proportion of emissions inhaled by an exposed population, is useful for prioritizing sources with the greatest impact on population exposure per unit emissions. This article reports iF estimates for urban winter wood smoke emissions. We used two approaches, incorporating spatiotemporal statistical models for (1) winter wood smoke fine particulate matter (PM2.5) emissions and concentration and (2) concentrations of levoglucosan (a wood smoke particulate marker). Empirical data used in our models were measured in Vancouver, Canada during 2004-2005. We used Monte Carlo simulations to quantify uncertainty. The estimated geometric mean iF (units: per million) is 13 (one geometric standard deviation range: 6.6-24) for wood smoke PM2.5 and 15 (4.5-50) for levoglucosan. These iF estimates are comparable to or slightly larger than iF values for urban vehicle emissions reported in the literature. On average, higher-income areas have lower wood smoke PM2.5 concentrations and intake. Our results emphasize the importance of urban wood smoke as a source of PM2.5 exposure and highlight the comparatively large population exposure and potential environmental justice benefits from reducing wood smoke emissions.