Receptor modeling of ambient particulate matter data using positive matrix factorization: review of existing methods

Methods for apportioning sources of ambient particulate matter (PM) using the positive matrix factorization (PMF) algorithm are reviewed. Numerous procedural decisions must be made and algorithmic parameters selected when analyzing PM data with PMF. However, few publications document enough of these details for readers to evaluate, reproduce, or compare results between different studies. For example, few studies document why some species were used and others not used in the modeling, how the number of factors was selected, or how much uncertainty exists in the solutions. More thorough documentation will aid the development of standard protocols for analyzing PM data with PMF and will reveal more clearly where research is needed to help future analysts select from the various possible procedures and parameters available in PMF. For example, research likely is needed to determine optimal approaches for handling data below detection limits, ways to apportion PM mass among sources identified by PMF, and ways to estimate uncertainties in the solution. The review closes with recommendations for documenting the methodological details of future PMF analyses.     

Errors in coarse particulate matter mass concentrations and spatiotemporal characteristics when using subtraction estimation methods

In studies of coarse particulate matter (PM_10-2.5), mass concentrations are often estimated through the subtraction of PM_2.5 from collocated PM₁₀ tapered element oscillating microbalance (TEOM) measurements. Though all field instruments have yet to be updated, the Filter Dynamic Measurement System (FDMS) was introduced to account for the loss of semivolatile material from heated TEOM filters. To assess errors in PM_10-2.5 estimation when using the possible combinations of PM₁₀ and PM_2.5 TEOM units with and without FDMS, data from three monitoring sites of the Colorado Coarse Rural–Urban Sources and Health (CCRUSH) study were used to simulate four possible subtraction methods for estimating PM_10-2.5 mass concentrations. Assuming all mass is accounted for using collocated TEOMs with FDMS, the three other subtraction methods were assessed for biases in absolute mass concentration, temporal variability, spatial correlation, and homogeneity. Results show collocated units without FDMS closely estimate actual PM_10-2.5 mass and spatial characteristics due to the very low semivolatile PM_10-2.5 concentrations in Colorado. Estimation using either a PM_2.5 or PM₁₀ monitor without FDMS introduced absolute biases of 2.4 µg/m³ (25%) to –2.3 µg/m³ (–24%), respectively. Such errors are directly related to the unmeasured semivolatile mass and alter measures of spatiotemporal variability and homogeneity, all of which have implications for the regulatory and epidemiology communities concerned about PM_10-2.5. Two monitoring sites operated by the state of Colorado were considered for inclusion in the CCRUSH acute health effects study, but concentrations were biased due to sampling with an FDMS-equipped PM_2.5 TEOM and PM₁₀ TEOM not corrected for semivolatile mass loss. A regression-based model was developed for removing the error in these measurements by estimating the semivolatile concentration of PM_2.5 from total PM_2.5 concentrations. By estimating nonvolatile PM_2.5 concentrations from this relationship, PM_10-2.5 was calculated as the difference between nonvolatile PM₁₀ and PM_2.5 concentrations.

Implications: Errors in the estimation of PM_10-2.5 concentrations using subtraction methods were shown to be related to the unmeasured semivolatile mass when using certain combinations of TEOM instruments. For the northeastern Colorado region, the absolute bias associated with this error significantly affects mean and 95th percentile values, which would affect assessment of compliance if PM_10-2.5 is regulated in the future. Estimating PM_10-2.5 mass concentrations using nonvolatile mass concentrations from collocated PM₁₀ and PM_2.5 TEOM monitors closely estimates the total PM_10-2.5 mass concentrations. A corrective model that removes the described error was developed and applied to data from two sites in Denver.

Supplemental Materials: Supplemental materials are available for this paper. Go to the publisher's online edition of the Journal of the Air & Waste Management Association.     

Effects of Ceiling-Mounted HEPA-UV Air Filters on Airborne Bacteria Concentrations in an Indoor Therapy Pool Building

Abstract

The purpose of this study was to assess the effectiveness of a new generation of high-volume, ceiling-mounted high-efficiency particulate air (HEPA)-ultraviolet (UV) air filters (HUVAFs) for their ability to remove or inactivate bacterial aerosol. In an environmentally controlled full-scale laboratory chamber (87 m³), and an indoor therapy pool building, the mitigation ability of air filters was assessed by comparing concentrations of total bacteria, culturable bacteria, and airborne endotoxin with and without the air filters operating under otherwise similar conditions. Controlled chamber tests with pure cultures of aerosolized Mycobacterium parafortuitum cells showed that the HUVAF unit tested provided an equivalent air-exchange rate of 11 hr^?1. Using this equivalent air-exchange rate as a design basis, three HUVAFs were installed in an indoor therapy pool building for bioaerosol mitigation, and their effectiveness was studied over a 2-year period. The HUVAFs reduced concentrations of culturable bacteria by 69 and 80% during monitoring periods executed in respective years. The HUVAFs reduced concentrations of total bacteria by 12 and 76% during the same monitoring period, respectively. Airborne endotoxin concentrations were not affected by the HUVAF operation.     

Using continuous PM2.5 monitoring data to report an air quality index

As stated in 40 CFR 58, Appendix G (2000), statistical linear regression models can be applied to relate PM2.5 continuous monitoring (CM) measurements with federal reference method (FRM) measurements, collocated or otherwise, for the purpose of reporting the air quality index (AQI). The CM measurements can then be transformed via the model to remove any bias relative to FRM measurements. The resulting FRM-like modeled measurements may be used to provide more timely reporting of a metropolitan statistical area's (MSA's) AQI. Of considerable importance is the quality of the model used to relate the CM and FRM measurements. The use of a poor model could result in misleading AQI reporting in the form of incorrectly claiming either good or bad air quality. This paper describes a measure of adequacy for deciding whether a statistical linear regression model that relates FRM and continuous PM2.5 measurements is sufficient for use in AQI reporting. The approach is the U.S. Environmental Protection Agency's (EPA's) data quality objectives (DQO) process, a seven-step strategic planning approach to determine the most appropriate data type, quality, quantity, and synthesis for a given activity. The chosen measure of model adequacy is r2, the square of the correlation coefficient between FRM measurements and their modeled counterparts. The paper concludes by developing regression models that meet this desired level of adequacy for the MSAs of Greensboro/Winston-Salem/High Point, NC; and Davenport/Moline/Rock Island, IA/IL. In both cases, a log transformation of the data appeared most appropriate. For the data from the Greensboro/Winston-Salem/High Point MSA, a simple linear regression model of the FRM and CM measurements had an r2 of 0.96, based on 227 paired observations. For the data from the Davenport/Moline/Rock Island MSA, due to seasonal differences between CM and FRM measurements, the simple linear regression model had to be expanded to include a temperature dependency, resulting in an r2 of 0.86, based on 214 paired observations.     

Prescribed burns and wildfires in Colorado: impacts of mitigation measures on indoor air particulate matter

Wildfires and prescribed burns are receiving increasing attention as sources of fine particulate matter (PM2.5). The goal of this research project was to understand the impact of mitigation strategies for residences impacted by scheduled prescribed burns and wildfires. Pairs of residences were solicited to have PM2.5 concentrations monitored inside and outside of their houses during four fires. The effect of using air cleaners on indoor PM2.5 was investigated, as well as the effect of keeping windows closed. Appropriately sized air cleaners were provided to one of each pair of residences; occupants of all of the residences were asked to keep windows shut and minimize opening of exterior doors. Additionally, residents were asked to record all of the activities that may be a source of particulate matter, such as cooking and cleaning. Measurements were made during one prescribed burn and three wildfires during the 2002 fire season. Outdoor 24-hr average PM2.5 concentrations ranging from 6 to 38 microg/m3 were measured during the fires, compared with levels of 2-5 microg/m3 during background measurements when no fires were burning. During the fires, PM2.5 was < 3 microg/m3 inside all of the houses with air cleaners installed. This corresponds with a decrease of 63-88% in homes with the air cleaners operating when compared with homes without air cleaners. In the homes without the air cleaners, measured indoor concentrations were 58-100% of the concentrations measured outdoors.     

Recycling as a planned behavior: the moderating role of perceived behavioral control

Environment, Development and Sustainability - This study examines the effectiveness of a public service announcement (PSA) video designed based on the theory of planned behavior (TPB) in motivating...