Three Years of High Time-resolution Air Pollution Monitoring in the Complex Multi-source Harbor of New York and New Jersey.

In densely developed port areas with numerous emissions sources, relating measured air quality changes to emissions is challenging given the geographic density of sources without unique pollutant composition signatures. To better understand air quality during increasing emission controls at the Port of New York and New Jersey ("Port"), an air monitoring station was sited to minimize collinearity of sources along ordinal directions. The study area includes an international airport, interstate highway, port terminals and shipping lanes, and industrial sources, as well as typical urban emissions of a megacity. Because air flow travel time from sources to the monitor were usually much less than one hour, minute-by-minute, high-precision data were collected for three years (2013-2015) for sulfur dioxide (SO2), carbon monoxide (CO), oxides of nitrogen (NO, NO2), black carbon (BC), fine particulate matter (PM2.5), and meteorology (wind speed, wind direction, temperature, humidity). From summer 2014 to spring 2015, hourly metals data were also collected. A high degree of temporal variability was observed for pollutants associated with direct emissions, with highest hourly average coefficient of variation observed for NO (2.65), SO2 (1.45) and BC (1.21). Nonparametric trajectory analysis (NTA) was utilized to separate the source areas influencing the monitoring data and observe how they changed over time, with over 1.6 million trajectories computed in total. Comparing the last 5 quarters of the study to the first 5 quarters, concentrations at the monitoring site associated with three port-related geographic areas decreased by 34-41%, 11-17%, and 28-41% for SO2, NOx, and BC, respectively. Over the same period, indicators of shipping and cargo activity at the port remained relatively constant; therefore, a shift in emission factors is likely the cause of the change. This study demonstrates the value of high-time resolution, accurate monitoring data along with careful siting to understand source area influences.

Source apportionment of polycyclic aromatic hydrocarbons (PAHs) in small craft harbor (SCH) surficial sediments in Nova Scotia, Canada.

Multiple source apportionment approaches were employed to investigate PAH sources which contribute to small craft harbor (SCH) sediments in Nova Scotia (NS), Canada. A total of 580 sediment samples were analyzed using PAH diagnostic ratios, Unmix Optimum receptor modeling, and by assessment of the composition of the PAH profile. PAH diagnostic ratios suggest PAHs are primarily of pyrogenic (thermal) origin, while UnmixO modeling identifies four individual sources which best describe surficial sediments and suggests contributions from both pyrogenic and petrogenic origins. These include coal combustion, automobile exhaust, and biomass incineration. PAH profile assessment determined an overwhelming contribution of high molecular weight PAHs, which exhibited a strong correlation with total PAH concentrations.

Unmix Optimum analysis of PAH sediment sources.

Unmix Optimum (UnmixO) was developed to analyze data, such as sediment PAH data, that were resistant to existing methods of multivariate analysis. Using a geometrical approach, UnmixO uses multiple advanced nonlinear optimization algorithms to find potential sources that obey non-negativity constraints while optimally fitting the data. UnmixO does not require specific knowledge of the uncertainties in the data and will work better for smaller data sets than other multivariate models. UnmixO was able to identify polycyclic aromatic hydrocarbon (PAH) contaminant sources contributing to sediment samples based on sample composition data with good diagnostic values. Results were compared to published EPA Chemical Mass Balance (CMB) sediment results from Lady Bird Lake (LBL) Austin, TX and 40 lakes (40LKS) across the U.S. A Chi-sum approach determined which UnmixO source profile best matched profiles used in CMB sediment studies; two coal tar (CT) sealcoat sources and a mixed combustion source contributed to the sediment PAHs. These results were consistent with CMB results for the LBL and 40LKS studies that estimated CT sealcoats contribute over 80% of PAHs to urban lakes. UnmixO results also showed that CT sealant's contribution to sediments decreased after the City of Austin ban in 2006.

Separating the air quality impact of a major highway and nearby sources by nonparametric trajectory analysis.

Nonparametric Trajectory Analysis (NTA), a receptor-oriented model, was used to assess the impact of local sources of air pollution at monitoring sites located adjacent to highway I-15 in Las Vegas, NV. Measurements of black carbon, carbon monoxide, nitrogen oxides, and sulfur dioxide concentrations were collected from December 2008 to December 2009. The purpose of the study was to determine the impact of the highway at three downwind monitoring stations using an upwind station to measure background concentrations. NTA was used to precisely determine the contribution of the highway to the average concentrations measured at the monitoring stations accounting for the spatially heterogeneous contributions of other local urban sources. NTA uses short time average concentrations, 5 min in this case, and constructed local back-trajectories from similarly short time average wind speed and direction to locate and quantify contributions from local source regions. Averaged over an entire year, the decrease of concentrations with distance from the highway was found to be consistent with previous studies. For this study, the NTA model is shown to be a reliable approach to quantify the impact of the highway on local air quality in an urban area with other local sources.

Selecting an appropriate multivariate source apportionment model result.

Two easily available multivariate source apportionment models, Unmix and positive matrix factorization (PMF), often produce nearly the same source apportionment. However, this paper gives two examples in which this is not the case: a simulated air pollution data set of 8 species and 200 samples and a water quality data set of 32 PCB congeners and 106 sediment core samples from Sheboygan River Inner Harbor, WI. In the first case, a basic form of PMF fails primarily because the source compositions do not have any species with zero or near zero concentrations. Unmix produces source compositions and contributions that are much closer to the true values. A version of PMF with an adjustable parameter also gives good results. In the second case, each model found 5 sources for the Sheboygan PCB sediment data. PMF determined sources compositions were consistent with the original 50/50% Aroclor 1248/1254 mixture, a previously determined prominent dechlorination profile (processes H' + M), and three other partially dechlorinated profiles. The Unmix determined source compositions were not as successful as the Unmix results depended heavily on just three data points. Source apportionment results favor Unmix when edges in the data are well-defined and PMF when several zeros are present in the loading and score matrices. Since both models are seen to have potential weaknesses, both should be applied in all cases. If the two methods do not produce similar results the methods given in the paper can be used to select the model result most likely to be closest to the truth.

Harmonic analysis of environmental time series with missing data or irregular sample spacing.

The Lomb periodogram and discrete Fourier transform are described and applied to harmonic analysis of two typical data sets, one air quality time series and one water quality time series. The air quality data is a 13 year series of 24 hour average particulate elemental carbon data from the IMPROVE station in Washington, D.C. The water quality data are from the stormwater monitoring network in Milwaukee, WI and cover almost 2 years of precipitation events. These data have irregular sampling periods and missing data that preclude the straightforward application of the fast Fourier transform (FFT). In both cases, an anthropogenic periodicity is identified; a 7-day weekday/ weekend effect in the Washington elemental carbon series and a 1 month cycle in several constituents of stormwater. Practical aspects of application of the Lomb periodogram are discussed, particularly quantifying the effects of random noise. The proper application of the FFT to data that are irregularly spaced with missing values is demonstrated on the air quality data. Recommendations are given when to use the Lomb periodogram and when to use the FFT.

PM source apportionment and health effects. 3. Investigation of inter-method variations in associations between estimated source contributions of PM2.5 and daily mortality in Phoenix, AZ.

As part of an EPA-sponsored workshop to investigate the use of source apportionment in health effects analyses, the associations between the participant's estimated source contributions of PM(2.5) for Phoenix, AZ for the period from 1995-1997 and cardiovascular and total nonaccidental mortality were analyzed using Poisson generalized linear models (GLM). The base model controlled for extreme temperatures, relative humidity, day of week, and time trends using natural spline smoothers. The same mortality model was applied to all of the apportionment results to provide a consistent comparison across source components and investigators/methods. Of the apportioned anthropogenic PM(2.5) source categories, secondary sulfate, traffic, and copper smelter-derived particles were most consistently associated with cardiovascular mortality. The sources with the largest cardiovascular mortality effect size were secondary sulfate (median estimate=16.0% per 5th-to-95th percentile increment at lag 0 day among eight investigators/methods) and traffic (median estimate=13.2% per 5th-to-95th percentile increment at lag 1 day among nine investigators/methods). For total mortality, the associations were weaker. Sea salt was also found to be associated with both total and cardiovascular mortality, but at 5 days lag. Fine particle soil and biomass burning factors were not associated with increased risks. Variations in the maximum effect lag varied by source category suggesting that past analyses considering only single lags of PM(2.5) may have underestimated health impact contributions at different lags. Further research is needed on the possibility that different PM(2.5) source components may have different effect lag structure. There was considerable consistency in the health effects results across source apportionments in their effect estimates and their lag structures. Variations in results across investigators/methods were small compared to the variations across source categories. These results indicate reproducibility of source apportionment results across investigative groups and support applicability of these methods to effects studies. However, future research will also need to investigate a number of other important issues including accuracy of results.

PM source apportionment and health effects: 1. Intercomparison of source apportionment results.

During the past three decades, receptor models have been used to identify and apportion ambient concentrations to sources. A number of groups are employing these methods to provide input into air quality management planning. A workshop has explored the use of resolved source contributions in health effects models. Multiple groups have analyzed particulate composition data sets from Washington, DC and Phoenix, AZ. Similar source profiles were extracted from these data sets by the investigators using different factor analysis methods. There was good agreement among the major resolved source types. Crustal (soil), sulfate, oil, and salt were the sources that were most unambiguously identified (generally highest correlation across the sites). Traffic and vegetative burning showed considerable variability among the results with variability in the ability of the methods to partition the motor vehicle contributions between gasoline and diesel vehicles. However, if the total motor vehicle contributions are estimated, good correspondence was obtained among the results. The source impacts were especially similar across various analyses for the larger mass contributors (e.g., in Washington, secondary sulfate SE=7% and 11% for traffic; in Phoenix, secondary sulfate SE=17% and 7% for traffic). Especially important for time-series health effects assessment, the source-specific impacts were found to be highly correlated across analysis methods/researchers for the major components (e.g., mean analysis to analysis correlation, r>0.9 for traffic and secondary sulfates in Phoenix and for traffic and secondary nitrates in Washington. The sulfate mean r value is >0.75 in Washington.). Overall, although these intercomparisons suggest areas where further research is needed (e.g., better division of traffic emissions between diesel and gasoline vehicles), they provide support the contention that PM(2.5) mass source apportionment results are consistent across users and methods, and that today's source apportionment methods are robust enough for application to PM(2.5) health effects assessments.

PM source apportionment and health effects: 2. An investigation of intermethod variability in associations between source-apportioned fine particle mass and daily mortality in Washington, DC.

Source apportionment may be useful in epidemiological investigation of PM health effects, but variations and options in these methods leave uncertainties. An EPA-sponsored workshop investigated source apportionment and health effects analyses by examining the associations between daily mortality and the investigators' estimated source-apportioned PM(2.5) for Washington, DC for 1988-1997. A Poisson Generalized Linear Model (GLM) was used to estimate source-specific relative risks at lags 0-4 days for total non-accidental, cardiovascular, and cardiorespiratory mortality adjusting for weather, seasonal/temporal trends, and day-of-week. Source-related effect estimates and their lagged association patterns were similar across investigators/methods. The varying lag structure of associations across source types, combined with the Wednesday/Saturday sampling frequency made it difficult to compare the source-specific effect sizes in a simple manner. The largest (and most significant) percent excess deaths per 5-95(th) percentile increment of apportioned PM(2.5) for total mortality was for secondary sulfate (variance-weighted mean percent excess mortality=6.7% (95% CI: 1.7, 11.7)), but with a peculiar lag structure (lag 3 day). Primary coal-related PM(2.5) (only three teams) was similarly significantly associated with total mortality with the same 3-day lag as sulfate. Risk estimates for traffic-related PM(2.5), while significant in some cases, were more variable. Soil-related PM showed smaller effect size estimates, but they were more consistently positive at multiple lags. The cardiovascular and cardiorespiratory mortality associations were generally similar to those for total mortality. Alternative weather models generally gave similar patterns, but sometimes affected the lag structure (e.g., for sulfate). Overall, the variations in relative risks across investigators/methods were found to be much smaller than those across estimated source types or across lag days for these data. This consistency suggests the robustness of the source apportionment in health effects analyses, but remaining issues, including accuracy of source apportionment and source-specific sensitivity to weather models, need to be investigated.

Workgroup report: workshop on source apportionment of particulate matter health effects--intercomparison of results and implications.

Although the association between exposure to ambient fine particulate matter with aerodynamic diameter < 2.5 microm (PM2.5) and human mortality is well established, the most responsible particle types/sources are not yet certain. In May 2003, the U.S. Environmental Protection Agency's Particulate Matter Centers Program sponsored the Workshop on the Source Apportionment of PM Health Effects. The goal was to evaluate the consistency of the various source apportionment methods in assessing source contributions to daily PM2.5 mass-mortality associations. Seven research institutions, using varying methods, participated in the estimation of source apportionments of PM2.5 mass samples collected in Washington, DC, and Phoenix, Arizona, USA. Apportionments were evaluated for their respective associations with mortality using Poisson regressions, allowing a comparative assessment of the extent to which variations in the apportionments contributed to variability in the source-specific mortality results. The various research groups generally identified the same major source types, each with similar elemental makeups. Intergroup correlation analyses indicated that soil-, sulfate-, residual oil-, and salt-associated mass were most unambiguously identified by various methods, whereas vegetative burning and traffic were less consistent. Aggregate source-specific mortality relative risk (RR) estimate confidence intervals overlapped each other, but the sulfate-related PM2.5 component was most consistently significant across analyses in these cities. Analyses indicated that source types were a significant predictor of RR, whereas apportionment group differences were not. Variations in the source apportionments added only some 15% to the mortality regression uncertainties. These results provide supportive evidence that existing PM2.5 source apportionment methods can be used to derive reliable insights into the source components that contribute to PM2.5 health effects.

Estimating the probability of the public perceiving a decrease in atmospheric haze.

Regional haze regulations require progress toward reducing atmospheric haze as measured by particle scattering coefficient of visible light. From a practical perspective, this raises the following question: Given a decrease in extinction, what is the probability that people will notice an improvement in visibility? This paper proposes a quantitative definition of the probability of a perceptible increase in visibility given a decrease in light extinction and a general method to estimate this probability from perception measurements made in the field under realistic conditions. Using data from a recent study of visibility perception by 8 observers, it is estimated that a 2-4 deciview change gives a 67% maximum probability of detecting the improvement. Stated another way, the odds of seeing a difference are at most 2:1 for a change of 2-4 deciviews. A 90% probability requires a change of at least 3.5-7.0 deciviews. The limitations and possible bias in the results of this study are discussed. These results may have a major effect on the cost-benefit analysis of regulatory actions to improve visibility.

Source apportionment of Phoenix PM2.5 aerosol with the Unmix receptor model.

The multivariate receptor model Unmix has been used to analyze a 3-yr PM2.5 ambient aerosol data set collected in Phoenix, AZ, beginning in 1995. The analysis generated source profiles and overall average percentage source contribution estimates (SCEs) for five source categories:gasoline engines (33 +/- 4%), diesel engines (16 +/- 2%), secondary SO4(2-) (19 +/- 2%), crustal/soil (22 +/- 2%), and vegetative burning (10 +/- 2%). The Unmix analysis was supplemented with scanning electron microscopy (SEM) of a limited number of filter samples for information on possible additional low-strength sources. Except for the diesel engine source category, the Unmix SCEs were generally consistent with an earlier multivariate receptor analysis of essentially the same data using the Positive Matrix Factorization (PMF) model. This article provides the first demonstration for an urban area of the capability of the Unmix receptor model.

Just-noticeable differences in atmospheric haze.

This article examines the only available experimental data taken in the natural environment on the ability of an observer to perceive small, incremental changes in the colorfulness of objects seen through atmospheric haze and estimates an appropriate just-noticeable difference (JND) from these data. This experimentally determined threshold of perception is compared to changes in the deciview scale. Based on these experimental results, the deciview scale is found to not be uniform over a wide range of visibility conditions, as has been previously claimed. In addition, a 1-deciview change never produces a perceptible change in haze, as defined by a 95% probability of producing a measurable change in the colorfulness of an object seen through the haze.

Diagnostics for Determining Influential Species in the Chemical Mass Balance Receptor Model.

The chemical mass balance (CMB) model can be applied to estimate the amount of airborne particulate matter (PM) coming from various sources given the ambient chemical composition of the particles measured at the receptor and the chemical composition of the source emissions. Of considerable practical importance is the identification of those chemical species that have a large effect on either the source contributions or errors estimated by the CMB model. This paper details a study of a number of influential diagnostics for application of the CMB software. Some of the diagnostics studied are standard regression diagnostics based on single-row deletion diagnostics. A number of new diagnostics were developed specifically for the CMB application, based on the pseudo-inverse of the source composition matrix and called nondeletion diagnostics to distinguish them from the standard deletion diagnostics. Simulated data sets were generated to compare the diagnostics and their response to controlled amounts of random error. A particular diagnostic called a modified pseudoinverse matrix (MPIN), developed for this study, was found to be the best choice for CMB model application. The MPIN diagnostic contains virtually all the information present in both deletion and nondeletion diagnostics. Since the MPIN diagnostic requires only the source profiles, it can be used to identify influential species in advance without sampling the ambient data and to improve CMB results through possible remedial actions for the influential species. Specific recommendations are given for interpretation and use of the MPIN diagnostic with the CMB model software. Elements with normalized MPIN absolute values of 1 to 0.5 are associated with influential elements. Noninfluential elements have normalized MPIN absolute values of 0.3 or less. Elements with absolute values between 0.3 and 0.5 are ambiguous but should generally be considered noninfluential.