Indoor Exposure to Ambient Particles and Its Estimation Using Fixed Site Monitors

Enhancing indoor PM2.5 predictions based on land use and indoor environmental factors by applying machine learning and spatial modeling approaches

The presence of fine particulate matter (PM2.5) indoors constitutes a significant component of overall PM2.5 exposure, as individuals spend 90% of their time indoors; however, personal monitoring for large cohorts is often impractical. In light of this, this study seeks to employ a novel geospatial artificial intelligence (Geo-AI) coupled with machine learning (ML) approaches to develop indoor PM2.5 models. Multiple predictor variables were collected from 102 residential households, including meteorological data; elevation; land use; indoor environmental factors including human activities, building characteristics, infiltration factors, and real-time measurements; and various other factors. Geo-AI, which integrates land use regression, inverse distance weighting, and ML algorithms, was utilized to construct outdoor PM2.5 and PM10 estimates for residential households. The most influential variables were identified via correlation analysis and stepwise regression. Three ML methods, namely support vector machine, multiple linear regression, and multilayer perceptron (MLP) were used to estimate indoor PM2.5 concentration. Then, MLP was employed to blend three ML algorithms. The resulting model demonstrated commendable performance, achieving a 10-fold cross-validation R2 of 0.92 and a root mean square error of 2.3 μg/m3 for indoor PM2.5 estimations. Notably, the combination of Geo-AI and ensembled ML models in this study outperformed all other individual models. In addition, the present study pointed out the most influential factors for indoor PM2.5 model were outdoor PM2.5, PM2.5/PM10 ratio, sampling month, infiltration factor, located near factory, cleaning frequency, number of door entrance linked with outdoor, and wall material. Further exploration of diverse ensemble model formats to integrate estimates from different models could enhance overall performance. Consequently, the potential applications of this model extend to estimating real individual exposure to PM2.5 for further epidemiological research. Moreover, the model offers valuable insights for efficient indoor air quality management and control strategies.

High spatial-resolved source-specific exposure and risk in the city scale: Influence of spatial interrelationship between PM2.5 sources and population on exposure

Research on High Spatial-Resolved Source-Specific Exposure and Risk (HSRSSER) was conducted based on multiple-year, multiple-site synchronous measurement of PM2.5-bound (particulate matter with aerodynamic diameter<2.5 μm) toxic components in a Chinese megacity. The developed HSRSSER model combined the Positive Matrix Factorization (PMF) and Land Use Regression (LUR) to predict high spatial-resolved source contributions, and estimated the source-specific exposure and risk by personal activity time- and population-weighting. A total of 287 PM2.5 samples were collected at ten sites in 2018-2020, and toxic species including heavy metals (HMs), polycyclic aromatic hydrocarbons (PAHs) and organophosphate esters (OPEs) were analyzed. The percentage non-cancer risk were in the order of traffic emission (48 %) > industrial emission (22 %) > coal combustion (12 %) > waste incineration (11 %) > resuspend dust (7 %) > OPE-related products (0 %) ≈ secondary particles (0 %). Similar orders were observed in cancer risk. For traffic emission, due to its higher source contributions and large population in central area, non-cancer and cancer risk fraction increased from 23 % to 48 % and 20 % to 46 % after exposure estimation; while for industrial emission, higher source contributions but small population in suburb area decreased the percentage non-cancer and cancer risk from 38 % to 22 % and 39 % to 24 %, respectively.

Construction and evaluation of hourly average indoor PM2.5 concentration prediction models based on multiple types of places

Background

People usually spend most of their time indoors, so indoor fine particulate matter (PM2.5) concentrations are crucial for refining individual PM2.5 exposure evaluation. The development of indoor PM2.5 concentration prediction models is essential for the health risk assessment of PM2.5 in epidemiological studies involving large populations.

Methods

In this study, based on the monitoring data of multiple types of places, the classical multiple linear regression (MLR) method and random forest regression (RFR) algorithm of machine learning were used to develop hourly average indoor PM2.5 concentration prediction models. Indoor PM2.5 concentration data, which included 11,712 records from five types of places, were obtained by on-site monitoring. Moreover, the potential predictor variable data were derived from outdoor monitoring stations and meteorological databases. A ten-fold cross-validation was conducted to examine the performance of all proposed models.

Results

The final predictor variables incorporated in the MLR model were outdoor PM2.5 concentration, type of place, season, wind direction, surface wind speed, hour, precipitation, air pressure, and relative humidity. The ten-fold cross-validation results indicated that both models constructed had good predictive performance, with the determination coefficients (R2) of RFR and MLR were 72.20 and 60.35%, respectively. Generally, the RFR model had better predictive performance than the MLR model (RFR model developed using the same predictor variables as the MLR model, R2 = 71.86%). In terms of predictors, the importance results of predictor variables for both types of models suggested that outdoor PM2.5 concentration, type of place, season, hour, wind direction, and surface wind speed were the most important predictor variables.

Conclusion

In this research, hourly average indoor PM2.5 concentration prediction models based on multiple types of places were developed for the first time. Both the MLR and RFR models based on easily accessible indicators displayed promising predictive performance, in which the machine learning domain RFR model outperformed the classical MLR model, and this result suggests the potential application of RFR algorithms for indoor air pollutant concentration prediction.

Impacts of pollution heterogeneity on population exposure in dense urban areas using ultra-fine resolution air quality data

Traditional air quality data have a spatial resolution of 1 km or above, making it challenging to resolve detailed air pollution exposure in complex urban areas. Combining urban morphology, dynamic traffic emission, regional and local meteorology, physicochemical transformations in air quality models using big data fusion technology, an ultra-fine resolution modeling system was developed to provide air quality data down to street level. Based on one-year ultra-fine resolution data, this study investigated the effects of pollution heterogeneity on the individual and population exposure to particulate matter (PM_2.5 and PM₁₀), nitrogen dioxide (NO₂), and ozone (O₃) in Hong Kong, one of the most densely populated and urbanized cities. Sharp fine-scale variabilities in air pollution were revealed within individual city blocks. Using traditional 1 km average to represent individual exposure resulted in a positively skewed deviation of up to 200% for high-end exposure individuals. Citizens were disproportionally affected by air pollution, with annual pollutant concentrations varied by factors of 2 to 5 among 452 District Council Constituency Areas (DCCAs) in Hong Kong, indicating great environmental inequities among the population. Unfavorable city planning resulted in a positive spatial coincidence between pollution and population, which increased public exposure to air pollutants by as large as 46% among districts in Hong Kong. Our results highlight the importance of ultra-fine pollutant data in quantifying the heterogeneity in pollution exposure in the dense urban area and the critical role of smart urban planning in reducing exposure inequities.

Development and Performance Evaluation of a Low-Cost Portable PM2.5 Monitor for Mobile Deployment

The concentration of fine particulate matter (PM2.5) is known to vary spatially across a city landscape. Current networks of regulatory air quality monitoring are too sparse to capture these intra-city variations. In this study, we developed a low-cost (60 USD) portable PM2.5 monitor called Smart-P, for use on bicycles, with the goal of mapping street-level variations in PM2.5 concentration. The Smart-P is compact in size (85 × 85 × 42 mm) and light in weight (147 g). Data communication and geolocation are achieved with the cyclist’s smartphone with the help of a user-friendly app. Good agreement was observed between the Smart-P monitors and a regulatory-grade monitor (mean bias error: −3.0 to 1.5 μg m−3 for the four monitors tested) in ambient conditions with relative humidity ranging from 38 to 100%. Monitor performance decreased in humidity > 70% condition. The measurement precision, represented as coefficient of variation, was 6 to 9% in stationary mode and 6% in biking mode across the four tested monitors. Street tests in a city with low background PM2.5 concentrations (8 to 9 μg m−3) and in two cities with high background concentrations (41 to 74 μg m−3) showed that the Smart-P was capable of observing local emission hotspots and that its measurement was not sensitive to bicycle speed. The low-cost and user-friendly nature are two features that make the Smart-P a good choice for empowering citizen scientists to participate in local air quality monitoring.

Inter- and Intra-Individual Variability of Personal Health Risk of Combined Particle and Gaseous Pollutants across Selected Urban Microenvironments

Exposure surrogates, such as air quality measured at a fixed-site monitor (FSM) or residence, are typically used for health estimates. However, people spend various amounts of time in different microenvironments, including the home, office, outdoors and in transit, where they are exposed to different magnitudes of particle and gaseous air pollutants. Health risks caused by air pollution exposure differ among individuals due to differences in activity, microenvironmental concentration, as well as the toxicity of pollutants. We evaluated individual and combined added health risks (AR) of exposure to PM_2.5, NO_2, and O₃ for 21 participants in their daily life based on real-world personal exposure measurements. Exposure errors from using surrogates were quantified. Inter- and intra-individual variability in health risks and key contributors in variations were investigated using linear mixed-effects models and correlation analysis, respectively. Substantial errors were found between personal exposure concentrations and ambient concentrations when using air quality measurements at either FSM or the residence location. The mean exposure errors based on the measurements taken at either the FSM or residence as exposure surrogates was higher for NO₂ than PM_2.5, because of the larger spatial variability in NO₂ concentrations in urban areas. The daily time-integrated AR for the combined PM_2.5, NO₂, and O₃ (TIAR_combine) ranged by a factor of 2.5 among participants and by a factor up to 2.5 for a given person across measured days. Inter- and intra-individual variability in TIAR_combine is almost equally important. Several factors were identified to be significantly correlated with daily TIAR_combine, with the top five factors, including PM_2.5, NO₂ and O₃ concentrations at ‘home indoor’, O₃ concentrations at ‘office indoor’ and ambient PM_2.5 concentrations. The results on the contributors of variability in the daily TIAR_combine could help in targeting interventions to reduce daily health damage related to air pollutants.

A practical framework for predicting residential indoor PM2.5 concentration using land-use regression and machine learning methods

People typically spend most of their time indoors. It is of importance to establish prediction models to estimate PM_2.5 concentration in indoor environments (e.g., residential households) to allow accurate assessments of exposure in epidemiological studies. This study aimed to develop models to predict PM_2.5 concentration in residential households. PM_2.5 concentration and related parameters (e.g., basic information about the households and ventilation settings) were collected in 116 households during the winter and summer seasons in Hong Kong. Outdoor PM_2.5 concentration at households was estimated using a land-use regression model. The random forest machine learning algorithm was then applied to develop indoor PM_2.5 prediction models. The results show that the random forest model achieved a promising predictive accuracy, with R² and cross-validation R² values of 0.93 and 0.65, respectively. Outdoor PM_2.5 concentration was the most important predictor variable, followed in descending order by the household marked number, outdoor temperature, outdoor relative humidity, average household area and air conditioning. The external validation result using an independent dataset confirmed the potential application of the random forest model, with an R² value of 0.47. Overall, this study shows the value of a combined land-use regression and machine learning approach in establishing indoor PM_2.5 prediction models that provide a relatively accurate assessment of exposure for use in epidemiological studies.

Factors affecting variability in gaseous and particle microenvironmental air pollutant concentrations in Hong Kong primary and secondary schools

School-age children are particularly susceptible to exposure to air pollutants. To quantify factors affecting children's exposure at school, indoor and outdoor microenvironmental air pollutant concentrations were measured at 32 selected primary and secondary schools in Hong Kong. Real-time PM₁₀ , PM_2.5 , NO_2, and O₃ concentrations were measured in 76 classrooms and 23 non-classrooms. Potential explanatory factors related to building characteristics, ventilation practice, and occupant activities were measured or recorded. Their relationship with indoor measured concentrations was examined using mixed linear regression models. Ten factors were significantly associated with indoor microenvironmental concentrations, together accounting for 74%, 61%, 46%, and 38% of variations observed for PM_2.5 , PM₁₀ , O_3, and NO₂ microenvironmental concentrations, respectively. Outdoor concentration is the single largest predictor for indoor concentrations. Infiltrated outdoor air pollution contributes to 90%, 70%, 75%, and 50% of PM_2.5 , PM₁₀ , O_3, and NO₂ microenvironmental concentrations, respectively, in classrooms during school hours. Interventions to reduce indoor microenvironmental concentrations can be prioritized in reducing ambient air pollution and infiltration of outdoor pollution. Infiltration factors derived from linear regression models provide useful information on outdoor infiltration and help address the gap in generalizable parameter values that can be used to predict school microenvironmental concentrations.