Practical considerations for using low-cost sensors to assess wildfire smoke exposure in school and childcare settings.

BACKGROUND: More frequent and intense wildfires will increase concentrations of smoke in schools and childcare settings. Low-cost sensors can assess fine particulate matter (PM2.5) concentrations with high spatial and temporal resolution. OBJECTIVE: We sought to optimize the use of sensors for decision-making in schools and childcare settings during wildfire smoke to reduce children's exposure to PM2.5. METHODS: We measured PM2.5 concentrations indoors and outdoors at four schools in Washington State during wildfire smoke in 2020-2021 using low-cost sensors and gravimetric samplers. We randomly sampled 5-min segments of low-cost sensor data to create simulations of brief portable handheld measurements. RESULTS: During wildfire smoke episodes (lasting 4-19 days), median hourly PM2.5 concentrations at different locations inside a single facility varied by up to 49.6 µg/m3 (maximum difference) during school hours. Median hourly indoor/outdoor ratios across schools ranged from 0.22 to 0.91. Within-school differences in concentrations indicated that it is important to collect measurements throughout a facility. Simulation results suggested that making handheld measurements more often and over multiple days better approximates indoor/outdoor ratios for wildfire smoke. During a period of unstable air quality, PM2.5 over the next hour indoors was more highly correlated with the last 10-min of data (mean R2 = 0.94) compared with the last 3-h (mean R2 = 0.60), indicating that higher temporal resolution data is most informative for decisions about near-term activities indoors. IMPACT STATEMENT: As wildfires continue to increase in frequency and severity, staff at schools and childcare facilities are increasingly faced with decisions around youth activities, building use, and air filtration needs during wildfire smoke episodes. Staff are increasingly using low-cost sensors for localized outdoor and indoor PM2.5 measurements, but guidance in using and interpreting low-cost sensor data is lacking. This paper provides relevant information applicable for guidance in using low-cost sensors for wildfire smoke response.

Leveraging low-cost sensors to predict nitrogen dioxide for epidemiologic exposure assessment.

BACKGROUND: Statistical models of air pollution enable intra-urban characterization of pollutant concentrations, benefiting exposure assessment for environmental epidemiology. The new generation of low-cost sensors facilitate the deployment of dense monitoring networks and can potentially be used to improve intra-urban models of air pollution. OBJECTIVE: Develop and evaluate a spatiotemporal model for nitrogen dioxide (NO2) in the Puget Sound region of WA, USA for the Adult Changes in Thought Air Pollution (ACT-AP) study and assess the contribution of low-cost sensor data to the model's performance through cross-validation. METHODS: We developed a spatiotemporal NO2 model for the study region incorporating data from 11 agency locations, 364 supplementary monitoring locations, and 117 low-cost sensor (LCS) locations for the 1996-2020 time period. Model features included long-term time trends and dimension-reduced land use regression. We evaluated the contribution of LCS network data by comparing models fit with and without sensor data using cross-validated (CV) summary performance statistics. RESULTS: The best performing model had one time trend and geographic covariates summarized into three partial least squares components. The model, fit with LCS data, performed as well as other recent studies (agency cross-validation: CV- root mean square error (RMSE) = 2.5 ppb NO2; CV- coefficient of determination ( R 2 ) = 0.85). Predictions of NO2 concentrations developed with LCS were higher at residential locations compared to a model without LCS, especially in recent years. While LCS did not provide a strong performance gain at agency sites (CV-RMSE = 2.8 ppb NO2; CV- R 2 = 0.82 without LCS), at residential locations, the improvement was substantial, with RMSE = 3.8 ppb NO2 and R 2 = 0.08 (without LCS), compared to CV-RMSE = 2.8 ppb NO2 and CV- R 2 = 0.51 (with LCS). IMPACT: We developed a spatiotemporal model for nitrogen dioxide (NO2) pollution in Washington's Puget Sound region for epidemiologic exposure assessment for the Adult Changes in Thought Air Pollution study. We examined the impact of including low-cost sensor data in the NO2 model and found the additional spatial information the sensors provided predicted NO2 concentrations that were higher than without low-cost sensors, particularly in recent years. We did not observe a clear, substantial improvement in cross-validation performance over a similar model fit without low-cost sensor data; however, the prediction improvement with low-cost sensors at residential locations was substantial. The performance gains from low-cost sensors may have been attenuated due to spatial information provided by other supplementary monitoring data.

Cohort profile: The Assessing Imperial Valley Respiratory Health and the Environment (AIRE) study.

BACKGROUND: The Children's Assessing Imperial Valley Respiratory Health and the Environment (AIRE) study is a prospective cohort study of environmental influences on respiratory health in a rural, southeastern region of California (CA), which aims to longitudinally examine the contribution of a drying saline lake to adverse health impacts in children. OBJECTIVES: This cohort was established through a community-academic partnership with the goal of assessing the health effects of childhood exposures to wind-blown particulate matter (PM) and inform public health action. We hypothesize that local PM sources are related to poorer children's respiratory health. POPULATION: Elementary school children in Imperial Valley, CA. DESIGN: Prospective cohort study. METHODS: Between 2017 and 2019, we collected baseline information on 731 children, then follow-up assessments yearly or twice-yearly since 2019. Data have been collected on children's respiratory health, demographics, household characteristics, physical activity and lifestyle, via questionnaires completed by parents or primary caregivers. In-person measurements, conducted since 2019, repeatedly assessed lung function, height, weight and blood pressure. Exposure to air pollutants has been assessed by multiple methods and individually assigned to participants using residential and school addresses. Health data will be linked to ambient and local sources of PM, during and preceding the study period to understand how spatiotemporal trends in these environmental exposures may relate to respiratory health. PRELIMINARY RESULTS: Analyses of respiratory symptoms indicate a high prevalence of allergies, bronchitic symptoms and wheezing. Asthma diagnosis was reported in 24% of children at enrolment, which exceeds both CA state and US national prevalence estimates for children. CONCLUSIONS: The Children's AIRE cohort, while focused on the health impacts of the drying Salton Sea and air quality in Imperial Valley, is poised to elucidate the growing threat of drying saline lakes and wind-blown dust sources to respiratory health worldwide, as sources of wind-blown dust emerge in our changing climate.

Evaluation of Orthotrichum lyellii moss as a biomonitor of diesel exhaust.

Exhaust from diesel combustion engines is an important contributor to urban air pollution and poses significant risk to human health. Diesel exhaust contains a chemical class known as nitrated polycyclic aromatic hydrocarbons (nitro-PAHs) and is enriched in 1-nitropyrene (1-NP), which has the potential to serve as a marker of diesel exhaust. The isomeric nitro-PAHs 2-nitropyrene (2-NP) and 2-nitrofluoranthene (2-NFL) are secondary pollutants arising from photochemical oxidation of pyrene and fluoranthene, respectively. Like other important air toxics, there is not extensive monitoring of nitro-PAHs, leading to gaps in knowledge about relative exposures and urban hotspots. Epiphytic moss absorbs water, nutrients, and pollutants from the atmosphere and may hold potential as an effective biomonitor for nitro-PAHs. In this study we investigate the suitability of Orthotrichum lyellii as a biomonitor of diesel exhaust by analyzing samples of the moss for 1-NP, 2-NP, and 2-NFL in the Seattle, WA metropolitan area. Samples were collected from rural parks, urban parks, residential, and commercial/industrial areas (N = 22 locations) and exhibited increasing concentrations across these land types. Sampling and laboratory method performance varied by nitro-PAH, but was generally good. We observed moderate to moderately strong correlation between 1-NP and select geographic variables, including summer normalized difference vegetation index (NDVI) within 250 m (r = -0.88, R2 = 0.77), percent impervious surface within 50 m (r = 0.83, R2 = 0.70), percent high development land use within 500 m (r = 0.77, R2 = 0.60), and distance to nearest secondary and connecting road (r = -0.75, R2 = 0.56). The relationships between 2-NP and 2-NFL and the geographic variables were generally weaker. Our results suggest O. lyellii is a promising biomonitor of diesel exhaust, specifically for 1-NP. To our knowledge this pilot study is the first to evaluate using moss concentrations of nitro-PAHs as biomonitors of diesel exhaust.

Evaluating low-cost monitoring designs for PM2.5 exposure assessment with a spatiotemporal modeling approach.

Determining the most feasible and cost-effective approaches to improving PM2.5 exposure assessment with low-cost monitors (LCMs) can considerably enhance the quality of its epidemiological inferences. We investigated features of fixed-site LCM designs that most impact PM2.5 exposure estimates to be used in long-term epidemiological inference for the Adult Changes in Thought Air Pollution (ACT-AP) study. We used ACT-AP collected and calibrated LCM PM2.5 measurements at the two-week level from April 2017 to September 2020 (N of monitors [measurements] = 82 [502]). We also acquired reference-grade PM2.5 measurements from January 2010 to September 2020 (N = 78 [6186]). We used a spatiotemporal modeling approach to predict PM2.5 exposures with either all LCM measurements or varying subsets with reduced temporal or spatial coverage. We evaluated the models based on a combination of cross-validation and external validation at locations of LCMs included in the models (N = 82), and also based on an independent external validation with a set of LCMs not used for the modeling (N = 30). We found that the model's performance declined substantially when LCM measurements were entirely excluded (spatiotemporal validation R2 [RMSE] = 0.69 [1.2 µg/m3]) compared to the model with all LCM measurements (0.84 [0.9 µg/m3]). Temporally, using the farthest apart measurements (i.e., the first and last) from each LCM resulted in the closest model's performance (0.79 [1.0 µg/m3]) to the model with all LCM data. The models with only the first or last measurement had decreased performance (0.77 [1.1 µg/m3]). Spatially, the model's performance decreased linearly to 0.74 (1.1 µg/m3) when only 10% of LCMs were included. Our analysis also showed that LCMs located in densely populated, road-proximate areas improved the model more than those placed in moderately populated, road-distant areas.

Within-City Variation in Ambient Carbon Monoxide Concentrations: Leveraging Low-Cost Monitors in a Spatiotemporal Modeling Framework.

BACKGROUND: Based on human and animal experimental studies, exposure to ambient carbon monoxide (CO) may be associated with cardiovascular disease outcomes, but epidemiological evidence of this link is limited. The number and distribution of ground-level regulatory agency monitors are insufficient to characterize fine-scale variations in CO concentrations. OBJECTIVES: To develop a daily, high-resolution ambient CO exposure prediction model at the city scale. METHODS: We developed a CO prediction model in Baltimore, Maryland, based on a spatiotemporal statistical algorithm with regulatory agency monitoring data and measurements from calibrated low-cost gas monitors. We also evaluated the contribution of three novel parameters to model performance: high-resolution meteorological data, satellite remote sensing data, and copollutant (PM2.5, NO2, and NOx) concentrations. RESULTS: The CO model had spatial cross-validation (CV) R2 and root-mean-square error (RMSE) of 0.70 and 0.02 parts per million (ppm), respectively; the model had temporal CV R2 and RMSE of 0.61 and 0.04 ppm, respectively. The predictions revealed spatially resolved CO hot spots associated with population, traffic, and other nonroad emission sources (e.g., railroads and airport), as well as sharp concentration decreases within short distances from primary roads. DISCUSSION: The three novel parameters did not substantially improve model performance, suggesting that, on its own, our spatiotemporal modeling framework based on geographic features was reliable and robust. As low-cost air monitors become increasingly available, this approach to CO concentration modeling can be generalized to resource-restricted environments to facilitate comprehensive epidemiological research. https://doi.org/10.1289/EHP10889.

Potential Impacts of Different Occupational Outdoor Heat Exposure Thresholds among Washington State Crop and Construction Workers and Implications for Other Jurisdictions.

Occupational heat exposure is associated with substantial morbidity and mortality among outdoor workers. We sought to descriptively evaluate spatiotemporal variability in heat threshold exceedances and describe potential impacts of these exposures for crop and construction workers. We also present general considerations for approaching heat policy-relevant analyses. We analyzed county-level 2011-2020 monthly employment (Bureau of Labor Statistics Quarterly Census of Employment and Wages) and environmental exposure (Parameter-elevation Relationships on Independent Slopes Model (PRISM)) data for Washington State (WA), USA, crop (North American Industry Classification System (NAICS) 111 and 1151) and construction (NAICS 23) sectors. Days exceeding maximum daily temperature thresholds, averaged per county, were linked with employment estimates to generate employment days of exceedances. We found spatiotemporal variability in WA temperature threshold exceedances and crop and construction employment. Maximum temperature exceedances peaked in July and August and were most numerous in Central WA counties. Counties with high employment and/or high numbers of threshold exceedance days, led by Yakima and King Counties, experienced the greatest total employment days of exceedances. Crop employment contributed to the largest proportion of total state-wide employment days of exceedances with Central WA counties experiencing the greatest potential workforce burden of exposure. Considerations from this analysis can help inform decision-making regarding thresholds, timing of provisions for heat rules, and tailoring of best practices in different industries and areas.

Spatial predictors of heavy metal concentrations in epiphytic moss samples in Seattle, WA.

The use of bio-indicators is an emerging, cost-effective alternative approach to identifying air pollution and assessing the need for additional air monitoring. This community science project explores the use of moss samples as bio-indicators of the distribution of metal air particulates in two residential neighborhoods of the industrial Duwamish Valley located in Seattle, WA (USA). We applied geographically weighted regression to data from 61 youth-collected samples to assess the location-specific area-level spatial predictors of the concentrations of 25 elements with focus on five heavy metals of concern due to health and environmental considerations. Spatial predictors included traffic volume, industrial land uses, major roadways, the airport, dirt roads, the Duwamish River, impervious surfaces, tree canopy cover, and sociodemographics. Traffic volume surrounding sample locations was the most consistent positive predictor of increasing heavy metal concentration. Greater distance from the heavy-industry corridor surrounding the Duwamish River predicted lower concentrations of all metals, with statistically significant associations for chromium and lead in some areas. As the distance from dirt roads increased, the concentration of arsenic and chromium decreased significantly. Percent tree canopy within 200 m of sample locations was a significant protective factor for cadmium concentrations. In addition, percent people of color was significantly positively associated with increasing lead, chromium and nickel concentrations. Our findings underscore the potential influence of heavy industry and mobile sources on heavy metal concentrations, the buffering potential of trees in local environments, and persistent opportunity to improve environmental justice.

Probabilistic Machine Learning with Low-Cost Sensor Networks for Occupational Exposure Assessment and Industrial Hygiene Decision Making.

Occupational exposure assessments are dominated by small sample sizes and low spatial and temporal resolution with a focus on conducting Occupational Safety and Health Administration regulatory compliance sampling. However, this style of exposure assessment is likely to underestimate true exposures and their variability in sampled areas, and entirely fail to characterize exposures in unsampled areas. The American Industrial Hygiene Association (AIHA) has developed a more realistic system of exposure ratings based on estimating the 95th percentiles of the exposures that can be used to better represent exposure uncertainty and exposure variability for decision-making; however, the ratings can still fail to capture realistic exposure with small sample sizes. Therefore, low-cost sensor networks consisting of numerous lower-quality sensors have been used to measure occupational exposures at a high spatiotemporal scale. However, the sensors must be calibrated in the laboratory or field to a reference standard. Using data from carbon monoxide (CO) sensors deployed in a heavy equipment manufacturing facility for eight months from August 2017 to March 2018, we demonstrate that machine learning with probabilistic gradient boosted decision trees (GBDT) can model raw sensor readings to reference data highly accurately, entirely removing the need for laboratory calibration. Further, we indicate how the machine learning models can produce probabilistic hazard maps of the manufacturing floor, creating a visual tool for assessing facility-wide exposures. Additionally, the ability to have a fully modeled prediction distribution for each measurement enables the use of the AIHA exposure ratings, which provide an enhanced industrial decision-making framework as opposed to simply determining if a small number of measurements were above or below a pertinent occupational exposure limit. Lastly, we show how a probabilistic modeling exposure assessment with high spatiotemporal resolution data can prevent exposure misclassifications associated with traditional models that rely exclusively on mean or point predictions.

Potential impacts of Washington State's wildfire worker protection rule on construction workers.

Driven by climate change, wildfires are increasing in frequency, duration, and intensity across the Western United States. Outdoor workers are being exposed to increasing wildfire-related particulate matter and smoke. Recognizing this emerging risk, Washington adopted an emergency rule and is presently engaged in creating a permanent rule to protect outdoor workers from wildfire smoke exposure. While there are growing bodies of literature on the exposure to and health effects of wildfire smoke in the general public and wildland firefighters, there is a gap in knowledge about wildfire smoke exposure among outdoor workers generally and construction workers specifically-a large category of outdoor workers in Washington totaling 200,000 people. Several data sources were linked in this study-including state-collected employment data and national ambient air quality data-to gain insight into the risk of PM2.5 exposure among construction workers and evaluate the impacts of different air quality thresholds that would have triggered a new Washington emergency wildfire smoke rule aimed at protecting workers from high PM2.5 exposure. Results indicate the number of poor air quality days has increased in August and September in recent years. Over the last decade, these months with the greatest potential for particulate matter exposure coincided with an annual peak in construction employment that was typically 9.4-42.7% larger across Washington counties (one county was 75.8%). Lastly, the 'encouraged' threshold of the Washington emergency rule (20.5 µg m-3) would have resulted in 5.5 times more days subject to the wildfire rule on average across all Washington counties compared to its 'required' threshold (55.5 µg m-3), and in 2020, the rule could have created demand for 1.35 million N-95 filtering facepiece respirators among construction workers. These results have important implications for both employers and policy makers as rules are developed. The potential policy implications of wildfire smoke exposure, exposure control strategies, and data gaps that would improve understanding of construction worker exposure to wildfire smoke are also discussed.

Deployment, Calibration, and Cross-Validation of Low-Cost Electrochemical Sensors for Carbon Monoxide, Nitrogen Oxides, and Ozone for an Epidemiological Study.

We designed and built a network of monitors for ambient air pollution equipped with low-cost gas sensors to be used to supplement regulatory agency monitoring for exposure assessment within a large epidemiological study. This paper describes the development of a series of hourly and daily field calibration models for Alphasense sensors for carbon monoxide (CO; CO-B4), nitric oxide (NO; NO-B4), nitrogen dioxide (NO2; NO2-B43F), and oxidizing gases (OX-B431)-which refers to ozone (O3) and NO2. The monitor network was deployed in the Puget Sound region of Washington, USA, from May 2017 to March 2019. Monitors were rotated throughout the region, including at two Puget Sound Clean Air Agency monitoring sites for calibration purposes, and over 100 residences, including the homes of epidemiological study participants, with the goal of improving long-term pollutant exposure predictions at participant locations. Calibration models improved when accounting for individual sensor performance, ambient temperature and humidity, and concentrations of co-pollutants as measured by other low-cost sensors in the monitors. Predictions from the final daily models for CO and NO performed the best considering agreement with regulatory monitors in cross-validated root-mean-square error (RMSE) and R2 measures (CO: RMSE = 18 ppb, R2 = 0.97; NO: RMSE = 2 ppb, R2 = 0.97). Performance measures for NO2 and O3 were somewhat lower (NO2: RMSE = 3 ppb, R2 = 0.79; O3: RMSE = 4 ppb, R2 = 0.81). These high levels of calibration performance add confidence that low-cost sensor measurements collected at the homes of epidemiological study participants can be integrated into spatiotemporal models of pollutant concentrations, improving exposure assessment for epidemiological inference.

Field measurements of PM2.5 infiltration factor and portable air cleaner effectiveness during wildfire episodes in US residences.

Wildfires have frequently occurred in the western United States (US) during the summer and fall seasons in recent years. This study measures the PM2.5 infiltration factor in seven residences recruited from five dense communities in Seattle, Washington, during a 2020 wildfire episode and evaluates the impacts of HEPA-based portable air cleaner (PAC) use on reducing indoor PM2.5 levels. All residences with windows closed went through an 18-to-24-h no filtration session, with five of seven following that period with an 18-to-24-h filtration session. Auto-mode PACs, which automatically adjust the fan speed based on the surrounding PM2.5 levels, were used for the filtration session. 10-s resolved indoor PM2.5 levels were measured in each residence's living room, while hourly outdoor levels were collected from the nearest governmental air quality monitoring station to each residence. Additionally, a time-activity diary in minute resolution was collected from each household. With the impacts of indoor sources excluded, indoor PM2.5 mass balance models were developed to estimate the PM2.5 indoor/outdoor (I/O) ratios, PAC effectiveness, and decay-related parameters. Among the seven residences, the mean infiltration factor ranged from 0.33 (standard deviation [SD]: 0.06) to 0.76 (SD: 0.05). The use of auto-mode PAC led to a 48%-78% decrease of indoor PM2.5 levels after adjusting for outdoor PM2.5 levels and indoor sources. The mean (SD) air exchange rates ranged from 0.30 (0.13) h-1 to 1.41 (3.18) h-1 while the PM2.5 deposition rate ranged from 0.10 (0.54) h-1 to 0.49 (0.47) h-1. These findings suggest that staying indoors, a common protective measure during wildfire episodes, is insufficient to prevent people's excess exposure to wildfire smoke, and provides quantitative evidence to support the utilization of auto-mode PACs during wildfire events in the US.

Toward Environmental Justice in Civic Science: Youth Performance and Experience Measuring Air Pollution Using Moss as a Bio-Indicator in Industrial-Adjacent Neighborhoods.

This article reports on an interdisciplinary evaluation of the pilot phase of a community-driven civic science project. The project investigates the distribution of heavy metals in air pollution using moss growing on street trees as a bio-indicator in two industrial-adjacent neighborhoods in Seattle, Washington (USA). One goal of the ongoing project is to meaningfully engage local urban youths (eighth to twelfth grade) in the scientific process as civic scientists, and teach them about environmental health, environmental justice, and urban forestry concepts in a place-based, urban-oriented environmental research project. We describe the collaborative context in which our project developed, evaluate the quality of youth-collected data through analysis of replicate samples, and assess participants' learning, career interests, and overall appraisal of the pilot. Our results indicate that youth scientists collected usable samples (with acceptable precision among repeated samples), learned project content (with statistically significant increases in scores of test-style survey questions; p = 0.002), and appraised their engagement favorably (with 69% of participants reporting they liked the project). We observed few changes in career interests, however. We discuss our intention to use these preliminary insights to further our community-driven education, research, and action model to address environmental injustices.

Estimating personal exposures from a multi-hazard sensor network.

Occupational exposure assessment is almost exclusively accomplished with personal sampling. However, personal sampling can be burdensome and suffers from low sample sizes, resulting in inadequately characterized workplace exposures. Sensor networks offer the opportunity to measure occupational hazards with a high degree of spatiotemporal resolution. Here, we demonstrate an approach to estimate personal exposure to respirable particulate matter (PM), carbon monoxide (CO), ozone (O3), and noise using hazard data from a sensor network. We simulated stationary and mobile employees that work at the study site, a heavy-vehicle manufacturing facility. Network-derived exposure estimates compared favorably to measurements taken with a suite of personal direct-reading instruments (DRIs) deployed to mimic personal sampling but varied by hazard and type of employee. The root mean square error (RMSE) between network-derived exposure estimates and personal DRI measurements for mobile employees was 0.15 mg/m3, 1 ppm, 82 ppb, and 3 dBA for PM, CO, O3, and noise, respectively. Pearson correlation between network-derived exposure estimates and DRI measurements ranged from 0.39 (noise for mobile employees) to 0.75 (noise for stationary employees). Despite the error observed estimating personal exposure to occupational hazards it holds promise as an additional tool to be used with traditional personal sampling due to the ability to frequently and easily collect exposure information on many employees.

Integrating Public Health into Climate Change Policy and Planning: State of Practice Update.

Policy action in the coming decade will be crucial to achieving globally agreed upon goals to decarbonize the economy and build resilience to a warmer, more extreme climate. Public health has an essential role in climate planning and action: "Co-benefits" to health help underpin greenhouse gas reduction strategies, while safeguarding health-particularly of the most vulnerable-is a frontline local adaptation goal. Using the structure of the core functions and essential services (CFES), we reviewed the literature documenting the evolution of public health's role in climate change action since the 2009 launch of the US CDC Climate and Health Program. We found that the public health response to climate change has been promising in the area of assessment (monitoring climate hazards, diagnosing health status, assessing vulnerability); mixed in the area of policy development (mobilizing partnerships, mitigation and adaptation activities); and relatively weak in assurance (communication, workforce development and evaluation). We suggest that the CFES model remains important, but is not aligned with three concepts-governance, implementation and adjustment-that have taken on increasing importance. Adding these concepts to the model can help ensure that public health fulfills its potential as a proactive partner fully integrated into climate policy planning and action in the coming decade.

Sources of error and variability in particulate matter sensor network measurements.

The quality of mass concentration estimates from increasingly popular networks of low-cost particulate matter sensors depends on accurate conversion of sensor output (e.g., voltage) into gravimetric-equivalent mass concentration, typically using a calibration procedure. This study evaluates two important sources of variability that lead to error in estimating gravimetric-equivalent mass concentration: the temporal changes in sensor calibration and the spatial and temporal variability in gravimetric correction factors. A 40-node sensor network was deployed in a heavy vehicle manufacturing facility for 8 months. At a central location in the facility, particulate matter was continuously measured with three sensors of the network and a traditional, higher-cost photometer, determining the calibration slope and intercept needed to translate sensor output to photometric-equivalent mass concentration. Throughout the facility, during three intensive sampling campaigns, respirable mass concentrations were measured with gravimetric samplers and photometers to determine correction factors needed to adjust photometric-equivalent to gravimetric-equivalent mass concentration. Both field-determined sensor calibration slopes and intercepts were statistically different than those estimated in the laboratory (α = 0.05), emphasizing the importance of aerosol properties when converting voltage to photometric-equivalent mass concentration and the need for field calibration to determine slope. Evidence suggested the sensors' weekly field calibration slope decreased and intercept increased, indicating the sensors were deteriorating over time. The mean correction factor in the cutting and shot blasting area (2.9) was substantially and statistically lower than that in the machining and welding area (4.6; p = 0.01). Therefore, different correction factors should be determined near different occupational processes to accurately estimate particle mass concentrations.

Mapping Occupational Hazards with a Multi-sensor Network in a Heavy-Vehicle Manufacturing Facility.

Due to their small size, low-power demands, and customizability, low-cost sensors can be deployed in collections that are spatially distributed in the environment, known as sensor networks. The literature contains examples of such networks in the ambient environment; this article describes the development and deployment of a 40-node multi-hazard network, constructed with low-cost sensors for particulate matter (SHARP GP2Y1010AU0F), carbon monoxide (Alphasense CO-B4), oxidizing gases (Alphasense OX-B421), and noise (developed in-house) in a heavy-vehicle manufacturing facility. Network nodes communicated wirelessly with a central database in order to record hazard measurements at 5-min intervals. Here, we report on the temporal and spatial measurements from the network, precision of network measurements, and accuracy of network measurements with respect to field reference instruments through 8 months of continuous deployment. During typical production periods, 1-h mean hazard levels ± standard deviation across all monitors for particulate matter (PM), carbon monoxide (CO), oxidizing gases (OX), and noise were 0.62 ± 0.2 mg m-3, 7 ± 2 ppm, 155 ± 58 ppb, and 82 ± 1 dBA, respectively. We observed clear diurnal and weekly temporal patterns for all hazards and daily, hazard-specific spatial patterns attributable to general manufacturing processes in the facility. Processes associated with the highest hazard levels were machining and welding (PM and noise), staging (CO), and manual and robotic welding (OX). Network sensors exhibited varying degrees of precision with 95% of measurements among three collocated nodes within 0.21 mg m-3 for PM, 0.4 ppm for CO, 9 ppb for OX, and 1 dBA for noise of each other. The median percent bias with reference to direct-reading instruments was 27%, 11%, 45%, and 1%, for PM, CO, OX, and noise, respectively. This study demonstrates the successful long-term deployment of a multi-hazard sensor network in an industrial manufacturing setting and illustrates the high temporal and spatial resolution of hazard data that sensor and monitor networks are capable of. We show that network-derived hazard measurements offer rich datasets to comprehensively assess occupational hazards. Our network sets the stage for the characterization of occupational exposures on the individual level with wireless sensor networks.

Efficacy of Paired Electrochemical Sensors for Measuring Ozone Concentrations.

Typical low-cost electrochemical sensors for ozone (O3) are also highly responsive to nitrogen dioxide (NO2). Consequently, a single sensor's response to O3 is indistinguishable from its response to NO2. Recently, a method for quantifying O3 concentrations became commercially available (Alphasense Ltd., Essex, UK): collocating a pair of sensors, a typical oxidative gas sensor that responds to both O3 and NO2 (model OX-B431) and a second similar sensor that filters O3 and responds only to NO2 (model NO2-B43F). By pairing the two sensors, O3 concentrations can be calculated. We calibrated samples of three NO2-B43F sensors and three OX-B431 sensors with NO2 and O3 exclusively and conducted mixture experiments over a range of 0-1.0 ppm NO2 and 0-125 ppb O3 to evaluate the ability of the paired sensors to quantify NO2 and O3 concentrations in mixture. Although the slopes of the response among our samples of three sensors of each type varied by as much as 37%, the individual response of the NO2-B43F sensors to NO2 and OX-B431 sensors to NO2 and O3 were highly linear over the concentrations studied (R2 ≥ 0.99). The NO2-B43F sensors responded minimally to O3 gas with statistically non-significant slopes of response. In mixtures of NO2 and O3, quantification of NO2 was generally accurate with overestimates up to 29%, compared to O3, which was generally underestimated by as much as 187%. We observed changes in sensor baseline over 4 days of experiments equivalent to 34 ppb O3, prompting an alternate method of calculating concentrations by baseline-correcting sensor signal. The baseline-correction method resulted in underestimates of NO2 up to 44% and decreases in the underestimation of O3 up to 107% for O3. Both methods for calculating gas concentrations progressively underestimated O3 concentrations as the ratio of NO2 signal to O3 signal increased. Our results suggest that paired NO2-B43F and OX-B431 sensors permit quantification of NO2 and O3 in mixture, but that O3 concentration estimates are less accurate and precise than those for NO2.

Sensor Selection to Improve Estimates of Particulate Matter Concentration from a Low-Cost Network.

Deployment of low-cost sensors in the field is increasingly popular. However, each sensor requires on-site calibration to increase the accuracy of the measurements. We established a laboratory method, the Average Slope Method, to select sensors with similar response so that a single, on-site calibration for one sensor can be used for all other sensors. The laboratory method was performed with aerosolized salt. Based on linear regression, we calculated slopes for 100 particulate matter (PM) sensors, and 50% of the PM sensors fell within ±14% of the average slope. We then compared our Average Slope Method with an Individual Slope Method and concluded that our first method balanced convenience and precision for our application. Laboratory selection was tested in the field, where we deployed 40 PM sensors inside a heavy-manufacturing site at spatially optimal locations and performed a field calibration to calculate a slope for three PM sensors with a reference instrument at one location. The average slope was applied to all PM sensors for mass concentration calculations. The calculated percent differences in the field were similar to the laboratory results. Therefore, we established a method that reduces the time and cost associated with calibration of low-cost sensors in the field.

Low-Cost, Distributed Environmental Monitors for Factory Worker Health.

An integrated network of environmental monitors was developed to continuously measure several airborne hazards in a manufacturing facility. The monitors integrated low-cost sensors to measure particulate matter, carbon monoxide, ozone and nitrogen dioxide, noise, temperature and humidity. The monitors were developed and tested in situ for three months in several overlapping deployments, before a full cohort of 40 was deployed in a heavy vehicle manufacturing facility for a year of data collection. The monitors collect data from each sensor and report them to a central database every 5 min. The work includes an experimental validation of the particle, gas and noise monitors. The R² for the particle sensor ranges between 0.98 and 0.99 for particle mass densities up to 300 μg/m³. The R² for the carbon monoxide sensor is 0.99 for concentrations up to 15 ppm. The R² for the oxidizing gas sensor is 0.98 over the sensitive range from 20 to 180 ppb. The noise monitor is precise within 1% between 65 and 95 dBA. This work demonstrates the capability of distributed monitoring as a means to examine exposure variability in both space and time, building an important preliminary step towards a new approach for workplace hazard monitoring.