ANALYSIS OF AIR TOXICS COLLECTED AS PART OF THE JOINT AIR TOXICS ASSESSMENT PROJECT FINAL REPORT STI-905039.03-3016-FR By: Hilary R. Hafner Theresa E. O’Brien Sonoma Technology, Inc. Petaluma, CA Prepared for: Arizona Department of Environmental Quality Phoenix, AZ December 2006 This PDF document contains blank pages to accommodate two?sided printing. ANALYSIS OF AIR TOXICS COLLECTED AS PART OF THE JOINT AIR TOXICS ASSESSMENT PROJECT FINAL REPORT STI-905039.03-3016-FR By: Hilary R. Hafner Theresa E. O’Brien Sonoma Technology, Inc. 1360 Redwood Way, Suite C Petaluma, CA 94954-1169 Prepared for: Peter Hyde Arizona Department of Environmental Quality 1110 W. Washington Street Phoenix, AZ 85007 December 29, 2006 TABLE OF CONTENTS Section Page LIST OF FIGURES.........................................................................................................................v LIST OF TABLES.........................................................................................................................vii 1. OVERVIEW....................................................................................................................... 1-1 2. DATA VALIDATION ....................................................................................................... 2-1 2.1 Data Availability....................................................................................................... 2-1 2.2 Data Validation Approach ........................................................................................ 2-3 2.3 Validation Results..................................................................................................... 2-4 2.3.1 MDL Summary............................................................................................. 2-4 2.3.2 Other Findings.............................................................................................. 2-7 2.4 Sampling and Analytical Precision: Analysis of Collocated Measurements, Interlaboratory Comparisons, and Analytical Methods Comparisons ..................... 2-7 2.4.1 Collocated Analyses ..................................................................................... 2-8 2.4.2 Interlaboratory Comparison ....................................................................... 2-10 2.4.3 Analytical Methods Comparison................................................................ 2-11 2.5 Summary of Data Validation .................................................................................. 2-13 3. DATA ANALYSIS RESULTS.......................................................................................... 3-1 3.1 Can the Ambient Data Be Used in Risk Assessment?.............................................. 3-1 3.2 Putting Air Toxics Concentrations in Perspective.................................................... 3-3 3.2.1 What Are Local Concentrations of Air Toxics?........................................... 3-3 3.2.2 How Do Air Toxics Concentrations Compare with Typical National Levels?.......................................................................................................... 3-6 3.2.3 What Are Background Concentrations of Air Toxics? ................................ 3-6 3.2.4 Which Air Toxics Present Possible Health Risks?....................................... 3-6 3.2.5 How Do Ambient Concentrations Compare with NATA Results?.............. 3-7 3.3 Changes in Air Toxics Concentrations Over Time................................................. 3-10 3.3.1 Seasonal Variation...................................................................................... 3-10 3.3.2 Weekday-Weekend Variation .................................................................... 3-10 3.3.3 Interannual Trends...................................................................................... 3-10 4. SUMMARY OF RESULTS............................................................................................... 4-1 5. RECOMMENDATIONS ................................................................................................... 5-1 6. REFERENCES................................................................................................................... 6-1 iii LIST OF FIGURES Figure Page 1-1. Monitoring sites contributing data to JATAP 2005 and discussed in this report. ............... 1-2 2-1. Scatter plot comparing average 2005 MDLs to 2004 MDLs by species............................. 2-5 2-2. Scatter plots of correlations between benzene and toluene concentrations at South Phoenix, West Phoenix, and Senior Center sites ................................................................ 2-7 2-3. Scatter plot of collocated measurements of 1,3,5-trimethylbenzene collected at Phoenix sites. ...................................................................................................................... 2-8 2-4. Scatter plot of collocated measurements of all species collected on March 17, 2005, at Phoenix sites. .................................................................................................................. 2-9 2-5. Scatter plot of benzene concentrations analyzed at EAS and EPA Region 9 laboratories........................................................................................................................ 2-11 2-6. Percentage below MDL for species measured using both TO-14 and TO-15 SIM methods. ............................................................................................................................ 2-12 2-7. Scatter plot of styrene concentrations made with EPA methods TO-14 and TO-15 SIM. .................................................................................................................................. 2-12 3-1. Annual average concentrations at JATAP sites compared with national concentration ranges. ................................................................................................................................. 3-4 3-2. Annual average concentrations at JATAP sites compared with national concentration ranges, background concentrations, and cancer benchmarks. ............................................ 3-5 3-3. Annual average concentrations estimated by the 1999 NATA model compared with 2005 annual average concentrations at JATAP sites. ...........Error! Bookmark not defined. 3-4. Illustration of box whisker plots and notched box whisker plots as defined by SYSTAT statistical software............................................................................................. 3-11 3-5. Twenty-four hour average concentrations by year at the Supersite and South Phoenix site from 2001 through 2005............................................................................................. 3-12 4-1. Annual average concentrations at JATAP sites compared with national concentration ranges, background concentrations, and cancer benchmarks. ............................................ 4-2 v LIST OF TABLES Table Page 1-1. Monitoring details and sampling objectives for JATAP 2005. ........................................... 1-3 2-1. Analysis methods used for data collected at JATAP sites in 2005. .................................... 2-2 2-2. List of available data collected at JATAP sites in 2005. ..................................................... 2-3 2-3. The percentage of cartridge samples collected in 2005 at the Greenwood, South Phoenix, and JLG Supersite sites for each species with 24-hr or averaged 24-hr concentrations above reported MDLs................................................................................. 2-5 2-4. The percentage of canister samples collected in 2005 for each species with 24-hr or averaged 24-hr concentrations above reported MDLs........................................................ 2-6 2-5. Sample outliers identified from the interlaboratory comparison of EAS and EPA Region 9. ........................................................................................................................... 2-10 3-1. Summary of the suitability of the 2005 JATAP air toxics measurements for assessing risk....................................................................................................................................... 3-2 3-2. The top five air toxics of concern based on risk-weighted annual mean concentrations at JATAP sites in 2005. ...................................................................................................... 3-7 3-3. Annual mean concentrations at JATAP sites in 2005 compared with cancer benchmarks, RfCs, and AAAQG........................................................................................ 3-9 vii 1. OVERVIEW The Joint Air Toxics Assessment Project (JATAP) is a consortium of federal, state, local, and tribal air pollution control officials designed to address the risk from air toxics in the greater Phoenix Metropolitan area. Participants include the U.S. Environmental Protection Agency (EPA) Region 9, the EPA Office of Air Quality, Planning, and Standards (OAQPS), Arizona Department of Environmental Quality (ADEQ), the Maricopa County Environmental Services Division, the Pinal County Air Quality Control District (PCAQCD), the Intertribal Council of Arizona, the Gila River Indian Community (GRIC), the Salt River–Pima Maricopa Indian Community (SRPMIC), and the Fort McDowell Yavapai Nation. The purpose of the study is to determine which air toxics are of most concern to South Phoenix and tribal communities. The measurements made in 2005 are the second phase of the study. Phase I consisted of air toxics measurements at the South Phoenix and West 43rd Avenue sites from August 2001 through March 2004 (McCarthy et al., 2004a). The ultimate goal of this consortium is to obtain a metropolitan-wide assessment of the risk associated with airborne toxics in greater Phoenix. Currently, 188 hazardous air pollutants (HAPs), or air toxics, regulated by the federal Clean Air Act have been associated with a wide variety of adverse health effects, including cancer, neurological effects, reproductive effects, and developmental effects (U.S. Environmental Protection Agency, 2000). Air toxics are emitted by a range of anthropogenic sources such as automobiles, commercial and retail entities, and industrial sources. Air toxics monitoring data are needed to characterize ambient concentrations in representative areas (i.e., at regional levels) to support and evaluate dispersion modeling efforts, and to quantify trends and the effectiveness of air toxics reduction strategies (e.g., the reduction of source-specific pollutant concentrations). For JATAP 2005, air toxics samples (24-hr average or two 12-hr samples) were collected every sixth day at five sites operated by ADEQ: South Phoenix, West Phoenix, Greenwood, Phoenix Supersite (typical urban site, National Air Toxics Trends Site [NATTS]), and Queen Valley (background/downwind site). The SRPMIC collected air toxics samples at its Senior Center site, and the GRIC collected air toxics samples at its St. Johns site. Air toxics samples were also collected at Fort McDowell, but were assessed separately. ADEQ contracted with Sonoma Technology, Inc. (STI) to validate and assess gaseous air toxics data collected from January 2005 through January 2006. The sites are shown in Figure 1-1. Monitoring details and objectives are provided in Table 1-1 (Sundblom et al., 2006). 1-1 Queen Valley qr Le??nd if ADED sites I St Johns site 40 1? Salt River site Ki'omete" Urban Areas Figure 1-1. Monitoring sites contributing data to JATAP 2005 and discussed in thisreport 1-2 Table 1-1. Monitoring details and sampling objectives for JATAP 2005. Location Phoenix - JLG Supersite Measurements Sampling Frequency Objective Population Exposure VOCs, PAHs, carbonyls, 1 – 24-hr Sample continuous BC, continuous Every 6 Days ECOC, continuous NMHC, PM metals Quality Assurance Collocated VOCs, 1 – 24-hr Sample, carbonyls, PM metals Schedule Varies by Sample Type West Phoenix VOCs, PM metals 1 – 24-hr Sample Population Exposure Every 6 Days South Phoenix VOCs, carbonyls, 1 – 24-hr Sample Population Exposure PM metals Every 6 Days Transport/ Gradient Sampling Every Gila River Indian VOCs, PM metals 6 Days, Alternate Community, St. 2 - 12-hr Samples and Johns 1 – 24-hr Sample Transport/Gradient Sampling Every Salt River Pima- VOCs, PM metals 6 Days, Alternate Maricopa Indian 2 - 12-hr Samples and Community, 1 – 24-hr Sample Senior Center Fort McDowell VOCs 1 – 24-hr Sample Transport/Gradient Yavapai Nation Every 12 Days Greenwood, SW VOCs, carbonyls, 1 – 24-hr Sample Maximum Corner of I-10/I-17 PM metals Every 6 Days Concentration Queen Valley VOCs, EC, PM metals 1 – 24-hr Sample Background Every 6 Days Volatile organic compounds (VOCs), organic and elemental carbon (OCEC), particulate matter (PM), nonmethane hydrocarbon (NMHC), black carbon (BC) The three primary tasks for STI in this data validation and analysis project were to (1) validate the 2005 gaseous air toxics data (Section 2), (2) analyze gaseous air toxics data collected as part of JATAP in 2005 (Section 3), and (3) interpret and communicate the results (this report). STI previously validated and analyzed the gaseous air toxics data collected in 2003-2004 at ADEQ sites (Hafner et al., 2004; McCarthy and Hafner, 2004) and GRIC sites (McCarthy et al., 2004b) as part of JATAP. 1-3 2. DATA VALIDATION This section summarizes the available data from 2005 and the validation findings. Additional details were provided in a technical memorandum dated May 10, 2006. 2.1 DATA AVAILABILITY Due to physical and chemical differences in air toxics, multiple sampling and analytical methods are necessary to characterize their concentrations in the atmosphere. Gaseous air toxics such as benzene and formaldehyde were the focus of the JATAP project, rather than particulate matter (PM) metals such as arsenic or chromium VI, or semivolatiles such as chrysene or fluoranthene. Two methods were used to collect gaseous air toxics in the Phoenix area for JATAP: stainless steel canisters and dinitro-phenyl hydrazine (DNPH) cartridges. Canister samples were used to collect a variety of hydrocarbons and chlorocarbon compounds; these canisters were analyzed by Environmental Analytical Services (EAS) or the San Diego Air Pollution Control District (SDAPCD), depending on the site (see Table 2-1). A subset of canister samples were also analyzed by the EPA Region 9 laboratory for quality assurance (QA) purposes. Cartridges were used to collect a set of oxygenated hydrocarbons known as carbonyl compounds (e.g., formaldehyde and acetaldehyde) at South Phoenix, JLG Supersite, and Greenwood. Table 2-2 lists the sites where analysis was complete as of November 2006. Samples were analyzed using a number of methods (as summarized in Table 2-2). Data were supplied to STI electronically by EAS, SDAPCD, and the EPA Region 9 laboratory and were then assessed and validated as part of this task. SDAPCD delivered the canister data collected at JLG Supersite, Queen Valley, and Greenwood; however, the data reported with dates after October 2005 have not been completely quality assured by the laboratory. Extensive data processing was required to convert data into a standardized format. Ambient concentration, log sheet, and laboratory sheet data were combined in a database. Canister numbers listed on sampling and analytical log sheets were checked to ensure the correct sample was associated with the correct date and time. 2-1 Table 2-1. Analysis methods used for data collected at JATAP sites in 2005. Site Sampling Method Analysis Method Laboratory Greenwood Cartridges Canisters TO-11 TO-15 EAS SDAPCD JLG Supersite Cartridges Canisters TO-11 TO-15 SDAPCD SDAPCD Queen Valley Canisters TO-15 SDAPCD St Johns Canisters TO-15 EAS Senior Center Canisters TO-15 TO-15 SIM EAS Cartridges TO-11 SDAPCD Canisters TO-14 TO-15 TO-15 SIM EAS Canisters TO-14 TO-15 TO-15 SIM EAS South Phoenix West Phoenix TO-15 = gas-chromatograph mass-spectrometer (GC-MS) SIM = Selective Ion Monitoring TO-14 = GC-flame ionization detection (FID) TO-11 = high performance liquid chromatography (HPLC) with ultraviolet absorption 2-2 Table 2-2. List of available data collected at JATAP sites in 2005. Samples Available Valid Samples Percent Valid 24-hr 61 60 60 98 24-hr 24-hr 61 61 61 61 59 49 97 80 61 30 (24-hr) 62 (12-hr) 30 (24-hr) 62 (12-hr) 61 61 37 (24-hr) 44 (12-hr) 37 (24-hr) 46 (12-hr) 60 55 90 79 95b 83 98b Cartridgesa 24-hr 24-hr and 12-hr 24-hr and 12-hr 24-hr 52 85 Canisters 24-hr 61 60 59 97 Canisters Canisters 24-hr 24-hr 32 61 32 60 30 94 59 97 Greenwood Cartridgesa Canisters JLG Supersite Cartridgesa Canisters St. Johns Canisters Senior Center Canisters Queen Valley West Phoenix a Samples Expected Sampling South Phoenix b Sampling Duration Site Carbonyls only. Alternating schedule was not consistent; 24-hr samples were collected in place of some 12-hr samples. This percentage is based on total number of sample days. 2.2 DATA VALIDATION APPROACH Many air toxics are present in concentrations too low to be detected using standard analysis techniques. Species with concentrations below the reported minimum detection limits (MDLs) were tabulated for the South Phoenix and West Phoenix sites. Further validation of species with more than 75% of measurements below the MDL is difficult because we cannot easily check species relationships or time series for patterns. Data displays—scatter, fingerprint, and time series plots—were inspected to identify problems and inconsistencies. Scatter plots enable investigation of the relationships among species at one site or at a number of sites. Fingerprint plots show the concentration of each species in a sample and help identify unique characteristics of the samples. Time series plots show the concentrations of species in every sample over a specified time period and are useful in showing the diurnal behavior of a species. Specific investigations were performed during validation of the air toxics data using VOCDat (Hafner and Prouty, 2004)and database tools: • Quantified the percentage of measurements with concentrations above the MDL at each site. • Inspected time series plots of every species, looking for seasonal variations, high and low values, and relationships to other species. 2-3 • Compared concentrations of species to well-characterized remote background concentrations (McCarthy et al., 2006c). • Prepared scatter plots of VOCs that are likely to be emitted by the same source and that typically correlate well, including benzene and acetylene, the xylene isomers (o-xylene should be less than the sum of m-&p-xylenes), benzene and toluene (toluene concentrations are typically higher and toluene outliers are expected), and 1,3-butadiene and benzene. • Inspected fingerprint plots of every sample with an emphasis on the flagged samples (based on the investigations above) and on the surrounding samples (looking for gaps in data, odd fingerprints, abrupt changes, and missing data). • Entered comments about why quality control (QC) codes were changed when we identified samples or individual species that needed to be flagged. • Examined duplicate analyses of samples to determine the precision of measurements, including sample collection and laboratory analysis. 2.3 VALIDATION RESULTS This section summarizes findings, provides additional information about suspect data, and includes example plots from VOCDat that illustrate validation findings. This section should be useful to monitoring personnel interested in analytical issues and to data analysts interested in observations of unusual features in the data. The tables in Appendix A list all samples we identified as suspect or invalid and provide the reason for the QC code change. 2.3.1 MDL Summary Species with reported concentrations below the MDL or concentrations reported as nondetects were counted by species and site. The number of samples above MDL was divided by the number of valid samples to find the percentage of each species above the reported MDL, as displayed in Table 2-3 for cartridge species and Table 2-4 for canister samples. Species highlighted in green have 75% or more measurements above the reported MDL. Species with 25 to 75% of measurements above the reported 2005 MDL are highlighted in yellow. Species with less than 25% of measurements above the 2005 MDL will have relatively uncertain annual average values and may be biased by a substitution; these species are highlighted in red. When such a high percentage of data are below the MDL, the MDL can be used as an upper limit on the annual average concentration (i.e., of red species). Subsequent risk assessments to understand which air toxics contribute to risk in local populations may be limited if annual averages are uncertain due to high MDLs or a low percentage of species above MDL. The species with the largest differences in percent of data above MDL between sites are 1,3-butadiene and styrene. MDL values are reported for every sample and provided in the database. One of the goals of the 2005 sampling campaign was to obtain improved MDLs compared with 2004 and this goal was met (Figure 2-1). 2-4 Table 2-3. The percentage of cartridge samples collected in 2005 at the Greenwood, South Phoenix, and JLG Supersite sites for each species with 24-hr or averaged 24-hr concentrations above reported MDLs. Greenwood 2005 % Above MDL 100 97 Species Acetaldehyde Formaldehyde South Phoenix 2005 % Above MDL JLG Supersite 2005 % Above MDL 98 98 100 100 Most species had a similar percentage of samples above or below MDL at the South Phoenix, West Phoenix, and Senior Center sites, indicating that concentrations at these three sites are relatively consistent. There were two exceptions to this trend. Only 17% of 1,3-butadiene measurements were above MDL at the Senior Center compared with 78% at South Phoenix and 81% at West Phoenix. This finding is consistent with the Senior Center’s proximity to roadway emissions—the traffic density near the Senior Center is much lower than that near the other two sites, and 1,3-butadiene reacts away quickly in the atmosphere. The amount of styrene data above the MDL was much lower at South Phoenix (57%) compared with West Phoenix (78%) and the Senior Center (81%). The spatial variability in styrene is difficult to explain and we found no obvious explanation. 0.09 Dichloromethane 0.08 1:1 Higher at ERG 1:1 1,1-Dichl 1,2,4-Trim 1,2-Dichl 1,3,5-Trim 0.07 1,3-Butad 1,2,4-Trimethylbenzene 1,3-Butadiene 1,2-Dichloroethane Benzene 2004 MDL ppb 0.06 Benzene Bromomethane 0.05 1,1-Dichloroethene 0.04 Ethylbenzene Vinyl Chloride Chloroform Bromome Carbon Tetrachloride Tetrachloroethylene Trichloroethylene Toluene m,p-Xylene Carbon T Chlorofor Dichlorom Ethylben Styrene m,p-Xyle 1,3,5-Trimethylbenzene o-Xylene o-Xylene 0.03 Styrene Tetrachlo Toluene Trichloro 0.02 Higher at EAS Vinyl Chl 0.01 0.00 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 Average 2005 MDL ppb Figure 2-1. Scatter plot comparing average 2005 MDLs to 2004 MDLs by species. The MDLs of species above the 1:1 line were higher in 2004 while MDLS of those below the line were higher in 2005. MDLs of most species analyzed by EAS were lower in 2005. 2-5 0.10 Table 2-4. The percentage of canister samples collected in 2005 for each species with 24-hr or averaged 24-hr concentrations above reported MDLs. Percent Above MDL Species 2-6 1,3-Butadiene 1,1-Dichloroethene 1,2,4-Trimethylbenzene 1,2-Dichloroethane 1,2-Dichloropropane 1,3,5-Trimethylbenzene Benzene Bromomethane Carbon tetrachloride Chloroform Dichloromethane Ethylbenzene Hexachlorobutadiene m,p-Xylene o-Xylene Styrene Tetrachloroethene Toluene Trichloroethene Vinyl chloride St. Johns 10 24 73 2 1 60 100 40 89 43 76 71 0 89 73 40 21 99 6 12 Senior Center 17 17 86 5 1 71 99 36 89 90 94 92 0 96 92 89 30 98 18 1 South Phoenix West Phoenix 78 9 86 5 0 68 100 37 89 77 97 92 0 94 94 57 54 100 17 3 81 13 92 6 2 76 100 49 83 83 98 94 0 97 94 78 63 98 17 3 Greenwood 100 2 98 42 0 98 100 24 100 98 100 100 2 100 100 98 100 100 90 2 JLG Supersite 100 0 100 25 0 100 100 33 100 100 100 100 4 100 100 100 100 100 84 0 Queen Valley 17 0 83 20 3 23 100 23 100 53 97 93 0 97 87 37 83 100 37 0 2.3.2 Other Findings Many species that were present in detectable concentrations (such as benzene, toluene, 1,3-butadiene, and xylenes) are likely emitted from mobile sources. Mobile source air toxics generally correlated well with one another as shown in Figure 2-2. These findings are consistent with other urban areas. South Phoenix West Phoenix Senior Center Figure 2-2. Scatter plots of correlations between benzene (y-axis) and toluene (x-axis) concentrations (ppb) at South Phoenix, West Phoenix, and Senior Center sites. These plots indicate that the species were probably emitted by the same source (i.e., motor vehicle exhaust); however, several outliers (circled in red) at West Phoenix may indicate a source that only emits toluene. We noted that concentrations of 1,3-butadiene were higher after October 7, 2005, at the South Phoenix, West Phoenix, and Senior Center sites than in the months prior. In addition, hexachlorobutadiene detection limits were higher at the South Phoenix, West Phoenix, and Senior Center sites after October 7, 2005. This may be related to the higher 1,3-butadiene concentrations. The changes appeared to be analytical rather than outliers from a change in source emissions. Details regarding data validation for each site are included in Appendix B. Appendix C contains a list of some of the more common gaseous air toxics and their sources in the atmosphere. 2.4 SAMPLING AND ANALYTICAL PRECISION: ANALYSIS OF COLLOCATED MEASUREMENTS, INTERLABORATORY COMPARISONS, AND ANALYTICAL METHODS COMPARISONS Collocated (i.e., duplicate) samples were taken at the South Phoenix and West Phoenix sites at relatively regular intervals. The collocated samples were sent to EAS for analysis. Using 2-7 the collocated measurements from these two sites, we can estimate the combined analytical and sampling precision for species collected using canisters. 2.4.1 Collocated Analyses Collocated measurements were analyzed to assess the combined sampling and analytical error in air toxics collected. Scatter plots of collocated measurements were created to examine differences in the collocated samples. Deming regressions were performed on collocated sample species to check for systematic bias and outliers. In general, the collocated samples showed good correlations at low concentrations of air toxics. However, air toxics with high concentrations were often significantly different from the expected 1:1 relationship. For example, Figure 2-3 shows the scatter plot of concentrations for 1,3,5-trimethylbenzene from all collocated canisters. The high outlier concentration from a sample collected March 17, 2005, is significantly above the 1:1 line. Removing this sampling date, as shown in the inset of Figure 2-3, reveals additional outliers, but the slope is less biased (slope = 0.82 ± 0.30). The March 17, 2005, samples showed that one of the two canisters was likely compromised (Figure 2-4). TO15-SIM 1,3,5-Trimethylbenzene 3 3/17/05 y = -0.0064x + 0.2363 N = 24 Without outlier Standard Error Intercept: 0.011 Slope: 0.057 0.6 Duplicate 2 (ppbv) Duplicate 2 (ppbv) 2.5 2 1.5 0.4 0.2 0 1 0 0.2 0.4 0.6 Duplicate 1 (ppbv) 0.5 0 0 0.5 1 1.5 2 2.5 3 Duplicate 1 (ppbv) Figure 2-3. Scatter plot of collocated measurements of 1,3,5-trimethylbenzene collected at Phoenix sites. Samples circled in red are outliers. 2-8 Figure 2-4. Scatter plot of collocated measurements of all species collected on March 17, 2005, at Phoenix sites. Samples circled in red are outliers. Both the main figure and the inset show that concentrations for sample 2 were a factor of 10 to 20 higher than those for sample 1. There are multiple possible sources of imprecision in the duplicate measurements. One type of error could occur in sampling. If flow rates or pressures are not equal, we would expect a systematic bias that could alter the slope to be greater than or less than one. This is typically not the case for these samples because 75% of the slopes are within the 1.00 ± 0.15 range (when the March 17, 2005, outlier is removed). A second issue is that of analytical and sampling precision. If the slope is typically close to 1 but the standard error of the slope is high (>0.2), then the samples are likely accurate but are being imprecisely measured. The average standard error of the slope was 0.2, which is at our threshold for high imprecision. It is also possible that some of the canisters could be contaminated prior to analysis, resulting in large discrepancies in concentrations affecting only a few compounds. However, only a few sample dates appeared to be systematically biased, including the February 15, March 17, and August 27 sample dates at West Phoenix. The primary sample at West Phoenix on February 15 was flagged as suspect because of low initial canister pressure, and the March 17 primary sample was also flagged as suspect because the final canister pressure was low. In addition, although the August 27 primary sample was not originally flagged, it was noted as a sample with concentrations not typical for the West Phoenix site. 2-9 2.4.2 Interlaboratory Comparison An interlaboratory comparison of EAS and EPA Region 9 laboratories was performed to assess possible systematic differences in concentration measurements. Eleven pairs of collocated canisters were collected at the South Phoenix and West Phoenix sites. One of each sampling pair was sent to EAS; the other was sent to the EPA Region 9 laboratory. Six toxics species were measured by each laboratory and are listed in Table 2-5. Scatter plots of laboratory measurements were created to examine differences. Any biases or problem species were noted. Outliers were identified visually from the scatter plots and verified statistically (i.e., found to leverage the linear regression results). Table 2-5 shows the outlier dates at South Phoenix and West Phoenix sites for samples and species measured at both EAS and EPA Region 9 laboratories. Table 2-5. Sample outliers identified from the interlaboratory comparison of EAS and EPA Region 9. Method Parameter South Phoenix West Phoenix Total Outliers – – 0 TO-15 SIM Benzene TO-15 SIM Carbon tetrachloride TO-15 SIM Chloroform TO-15 SIM Dichloromethane 2/27/2005 TO-15 SIM Tetrachloroethene – 12/6/2005 1 TO-15 SIM Trichloroethene 2/9/2005, 9/19/2005, 12/6/2005 – 3 7/3/2005 – 1 – 0 1 Outliers did not systematically occur on certain dates or sites or for certain species. In general, most species did not show outliers that skewed regression results. However, the slope calculated from regression of concentrations from the EAS and EPA Region 9 laboratories was outside the expected range of 1.0 ± 0.15. Analysis showed that EAS generally measured lower concentrations than EPA Region 9 as shown in Figure 2-5. The precision of the measurements between laboratories appears reasonable, but the systematic differences in concentrations suggest that one of the laboratories is inaccurate. Examination of the percent recovery of internal laboratory calibration spikes revealed that the EAS laboratory recovery rate was 88 ± 4% compared with the EPA Region 9 laboratory recovery rate of 97 ± 2%. Moreover, for the species that were measured by both laboratories, the difference in spike recovery rate was more than 8% higher on average. This difference suggests that measurements made by EAS are systematically biased low. 2-10 Benzene 1:1 EAS Result (ppbv) 2.2 y = 0.7162x + 0.0316 N = 11 Standard Error Intercept: 0.13 Slope: 0.10 1.7 1.2 0.7 0.2 0.2 0.7 1.2 1.7 2.2 EPA Region 9 Result (ppbv) Figure 2-5. Scatter plot of benzene concentrations (ppb) analyzed at EAS and EPA region 9 laboratories. The regression slope, shown in blue, is 0.72 ± 0.03. The red line shows the expected 1:1 slope. 2.4.3 Analytical Methods Comparison EAS used various analytical methods for its laboratory analysis: TO-15 SIM, TO-15, and TO-14. TO-15 SIM and TO-15 both rely on mass spectrometry, but TO-15 SIM focuses on selected species for lower MDLs while TO-15 performs a full scan of species resulting in higher MDLs. TO-14 uses flame ionization detection. MDL summaries for each method show that the percentage of data below detection was highest for TO-14 and lowest for TO-15 SIM, as shown in Figure 2-6. Scatter plots were created to examine differences between concentrations of species measured with more than one analytical method. TO-14 is generally a less accurate method due to the higher MDL ranges. Figure 2-7 shows an example of the TO-14 method with a majority of concentrations reported at MDL. Because TO-14 method data are less accurate, they were only used if TO-15 or TO-15 SIM method data were not available. 2-11 1,2,4-Trimethylbenzene 1,3,5-Trimethylbenzene 1,3-Butadiene Benzene TO-14 TO-15 SIM Ethylbenzene o-xylene Styrene Toluene 0 20 40 60 80 100 Percent Below MDL Figure 2-6. Percentage below MDL for species measured using both TO-14 and TO-15 SIM methods. 3.5 y = 0.3921x - 0.2838 N = 120 Standard Error Intercept: 0.29 Slope: 0.044 TO-15 SIM (ppbv) 3 2.5 2 Repeat TO-14 avg MDL concentrations 1.5 1 0.5 0 0 0.5 1 1.5 2 2.5 3 3.5 TO-14 (ppbv) Figure 2-7. Scatter plot of styrene concentrations (ppb) made with EPA methods TO-14 and TO-15 SIM. 2-12 2.5 SUMMARY OF DATA VALIDATION In summary, the following conclusions were drawn from the 2005 measurements: • Data quality appeared reasonable. • Concentrations of two-thirds of the species at all sites were above detection limits most of the time. • Numerous carbonyl compound samples were invalidated because sample flow rates were not within acceptable limits. Low flow rates resulted from a clogged needle valve flow controller; this problem has been corrected for future measurements. • The interlaboratory analyses showed that generally EAS measures VOC concentrations lower than EPA Region 9 by about 30%. EAS has an average blank recovery rate of 88 ± 4% versus 97 ± 2% at EPA Region 9 for the laboratory QC analysis. The blank recovery rate is consistent with the hypothesis that EAS concentrations are systematically biased low. • Collocated measurements showed that the precision of canister species as measured by the coefficient of variation (CV = standard deviation/mean) was, on average, 23%. This CV is higher than the goal set for the NATTS program of 15% CV for collocated data. In total, 16 of 20 pollutants showed CVs of 15% or more. • At the West Phoenix site, collocated samples collected on three days showed very poor correlations. This may indicate a problem with the replicate sampler at West Phoenix or, possibly, sample contamination. These collocated samples were flagged as suspect. • EAS used TO-15 and TO-15 SIM analytical methods in 2005. A comparison of analytical methods showed that TO-15 and TO-15 SIM methods have lower MDLs and higher percentages of data above the MDL than does the TO-14 method. In addition, JATAP paid for certification of every canister, meaning an MDL was determined by sample rather than on average. Both the analytical methodology and the sample certification led to an increase in the number of species for which there are 75% or more data above MDL, compared to the results for 2003-2004 data. Interlaboratory comparisons and collocated measurements indicate that JATAP data collected at multiple sites are likely biased low and are relatively imprecise. The systematically low bias of canister data analyzed by EAS reduces our ability to accurately characterize risk. It may be necessary to adjust data upwards to accurately characterize risk from air toxics in the JATAP area. Comparisons with data from other data sets (i.e., future years, other areas, past years) should be made with caution. Imprecision in the data will make spatial and temporal comparisons more difficult because of the larger uncertainties in the concentrations at any given site or time. 2-13 3. DATA ANALYSIS RESULTS The goal of this task was to analyze and interpret the gaseous air toxics data collected during JATAP 2005. Numerical, statistical, and graphical analyses of the data were performed to characterize the spatial and temporal variation of the air toxics concentrations. The focus of the analyses was on air toxics with cancer benchmarks. 3.1 CAN THE AMBIENT DATA BE USED IN RISK ASSESSMENT? One of the ultimate goals of air toxics monitoring is to assess the adverse health risk in ambient air from individual air toxics. To assess risk with confidence, average concentrations at a site must be present at levels that can either be measured accurately or determined to be below adverse health risk levels. The risk decision matrix in Table 3-1 shows one method of assessing the suitability of ambient measurements for assessing risk. One way of determining if the risk associated with a given species can be determined with confidence is to assess how the MDL compares to both the median concentration and the cancer benchmark. If the ambient concentration and the cancer benchmark are both below the MDL, we have low confidence in their estimated risk (i.e., the pink column). We have some confidence in our ability to estimate risk for species when either the ambient average concentration and/or cancer benchmark is slightly above MDL (i.e., the yellow column). We have high confidence in the risk associated with species whose quantifiable concentrations or MDLs are well below the cancer benchmark (i.e., the green column). Pollutants for which we have modest to good confidence in estimating risk at most sites using 2005 JATAP data included carbon tetrachloride, acetaldehyde, benzene, dichloromethane, and formaldehyde. Only hexachlorobutadiene was routinely placed in the low confidence column. Chloroform and 1,3-butadiene were on the border between the low and medium confidence columns. Better MDLs may be necessary to estimate concentrations for these pollutants. 3-1 Table 3-1. Summary of the suitability of the 2005 JATAP air toxics measurements for assessing risk. Site Greenwood Uncharacterizable None Risk can be estimated Hexachlorobutadiene Vinyl Chloride Chloroform Carbon Tetrachloride Tetrachloroethene JLG Supersite None West Phoenix Hexachlorobutadiene South Phoenix Hexachlorobutadiene Senior Center 1,3-Butadiene Hexachlorobutadiene St. Johns 1,3-Butadiene Hexachlorobutadiene Chloroform Queen Valley Chloroform Hexachlorobutadiene Vinyl Chloride Chloroform Carbon Tetrachloride Tetrachloroethene 1,3-butadiene 1,3-Butadiene Chloroform Vinyl chloride Carbon Tetrachloride Tetrachloroethene Trichloroethene 1,3-Butadiene Chloroform Vinyl chloride Carbon Tetrachloride Tetrachloroethene Chloroform Vinyl chloride Tetrachloroethene Trichloroethene Benzene Carbon Tetrachloride Tetrachloroethene Trichloroethene Vinyl chloride Benzene Carbon tetrachloride 1,3-butadiene Hexachlorobutadiene Vinyl chloride Carbon tetrachloride Benzene Tetrachloroethene 3-2 Risk can be quantified Trichloroethene 1,3-butadiene Benzene Acetaldehyde Formaldehyde Dichloromethane Trichloroethene Benzene Acetaldehyde Formaldehyde Dichloromethane Benzene Dichloromethane Benzene Trichloroethene Dichloromethane Dichloromethane Dichloromethane Trichloroethene Dichloromethane 3.2 PUTTING AIR TOXICS CONCENTRATIONS IN PERSPECTIVE The following sections build on previous and ongoing analyses for the EPA regarding the national air toxics picture (Hafner and McCarthy, 2004; Kenski et al., 2005; McCarthy et al., 2006a, b; McCarthy et al., 2006c). 3.2.1 What Are Local Concentrations of Air Toxics? Several questions about the air toxics data collected during JATAP can be answered using Figures 3-1 and 3-2. Of interest to the stakeholders is not only how the concentrations vary from site to site, but also how the concentrations compare with other urban areas and to cancer benchmarks. Typical urban concentrations were calculated as part of the national air toxics analysis (Hafner and McCarthy, 2004). Seasonal average concentrations from over 800 monitoring sites collected from 1990-2005 were compiled into interquartile ranges (i.e., 25th-75th percentile range). Figure 3-1 displays these interquartile ranges with the 2005 mean concentrations for the JATAP sites. Data below detection were substituted with MDL/2. Species with an asterisk (*) indicate more than 50% of data were below the MDL at some sites (see Tables 2-3 and 2-4 for specifics). Figure 3-2 is the same as Figure 3-1 with cancer benchmarks from EPA OAQPS and remote background concentrations added to the display of mean concentrations at each site. Understanding which species are present at concentrations known to exceed levels set by the EPA OAQPS, the Integrated Risk Information System (IRIS), or the California EPA can aid policy makers in regulating sources of potentially dangerous pollutants. While inspecting Figures 3-1 and 3-2, note that less than 50% of data in the following species-site combinations was above MDL, and the annual average concentrations should be viewed with caution because they are biased by MDL substitution: • 1,3-butadiene at St. Johns, Senior Center, and Queen Valley • Chloroform at St. Johns and Queen Valley • Tetrachloroethene at St. Johns and Senior Center • Trichloroethene at St. Johns, Senior Center, South Phoenix, West Phoenix, and Queen Valley • Vinyl chloride at all sites 3-3 Figure 3-1. Annual average concentrations at JATAP sites (2005) compared with national concentration ranges. Species with an asterisk (*) indicate ≥50% of data are below the MDL at some sites. 3-4 Figure 3-2. Annual average concentrations at JATAP sites compared with national concentration ranges, background concentrations, and cancer benchmarks. Species with an asterisk (*) indicate ≥50% of data are below the MDL at some sites. Locally, concentrations varied across the JATAP sites and concentration differences were generally consistent with proximity to emissions. As expected, sites located closer to the urban core of Phoenix (i.e., JLG Supersite, South Phoenix, Greenwood, and West Phoenix) had the highest annual average concentrations, while those located on the periphery of the urban area had lower concentrations (i.e., Salt River>St. Johns>Queen Valley). The following are specific findings: • Greenwood carbonyl concentrations were higher than other sites; of note, concentrations were above the national urban 95th percentile. This finding may indicate additional fresh emissions at a local scale, likely from mobile sources given the location of the monitor. Carbonyl compound concentrations at South Phoenix and JLG Supersite were also high relative to other urban areas. • Phoenix urban 1,3-butadiene concentrations were higher than concentrations at Queen Valley, St. Johns, and Senior Center sites. These three sites are located close enough to Phoenix to have been impacted by transport of air toxics from the urban core, but far 3-5 enough removed to show less influence from fresh emissions. Highly reactive air toxics, such as 1,3-butadiene, may be depleted before arriving at these sites as well. • Benzene concentrations show a similar spatial pattern to that of 1,3-butadiene, but concentrations at the Senior Center are nearly as high as the urban sites. Benzene is removed much more slowly from the atmosphere than 1,3-butadiene and the Senior Center site is closer to a major roadway than either the St. Johns or Queen Valley sites. • Chloroform concentrations were highest at the JLG Supersite. Concentrations at the other urban Phoenix sites were significantly higher than those at the downwind sites. 3.2.2 How Do Air Toxics Concentrations Compare with Typical National Levels? Phoenix area urban concentrations of 1,3-butadiene, acetaldehyde, formaldehyde, chloroform, benzene, and tetrachloroethene were typically on the high end of the national urban scale (i.e., above the 75th percentile). Other air toxics concentrations were typically within the interquartile range of national concentrations (25th to 75th percentile). Note that vinyl chloride concentrations were typically below the MDL at all sites and are plotted at MDL/2. 3.2.3 What Are Background Concentrations of Air Toxics? Background concentrations can be defined as remote or regional phenomena. Remote background is the mean concentration of a compound measured at a site not impacted by local or regional emission sources. Examples of remote sites include islands in the middle of the ocean or polar regions. These data are useful to track global scale changes in long-lived air toxics, for example. Regional background is the mean concentration of a compound measured at a site not impacted by local sources. A regional background site might be one located in a coastal area with predominantly onshore flow, or a remote continental site uninfluenced by local sources. These data are useful to help understand what is transported into an area. Remote and rural background air toxics concentrations were compiled by McCarthy et al. (2004d). Concentrations at JATAP sites in 2005 were typically above remote background concentrations as expected. Air toxics concentrations at Queen Valley were lowest among the sites, consistent with its location far from the urban area. Air toxics concentrations at St. Johns, and sometimes at the Senior Center, were also typically on the lower end of the concentration ranges. Typical remote background concentrations contributed less than 10% of the total concentrations measured at the urban JATAP sites (Greenwood, Supersite, South Phoenix, and West Phoenix). Only carbon tetrachloride background concentrations were as high as those seen in Phoenix. Since carbon tetrachloride is a globally distributed long-lived chlorofluorocarbon (CFC), it was not surprising to see that background concentrations were responsible for 100% of the observed values at all sites. 3.2.4 Which Air Toxics Present Possible Health Risks? Annual mean concentrations of air toxics concentrations at JATAP sites were often higher than the one-in-a-million cancer benchmark. 1,3-butadiene, acetaldehyde, benzene, carbon tetrachloride, chloroform, and tetrachloroethene concentrations were well above the one3-6 in-a-million risk level. Table 3-2 lists the top five air toxics of concern based on risk-weighted annual averages at the JATAP sites in 2005. Species in the list were colored in red, yellow, or green to indicate the confidence in the assessment of each species as shown in Table 3-1. Table 3-2. The top five air toxics of concern based on risk-weighted annual mean concentrations at JATAP sites in 2005 – numbers represent rank. Species Benzene 1,3-Butadiene Acetaldehyde Carbon tetrachloride Chloroform Tetrachloroethene Hexachlorobutadiene Greenwood JLG Supersite Queen Valley St. Johns Senior Center South Phoenix West Phoenix 1 2 3 4 5 1 3 2 3 4 2 3 2 3 2 4 2 5 1 4 5 3 2 5 4 3 5 1 4 5 5 4 1 1 1 Colors indicate the confidence in the risk estimate (green = high, yellow = medium, red = low confidence). Blank cells indicate the species was not among the five most abundant risk-weighted species for that year. Table 3-3 lists the annual mean concentrations (µg/m3) at each site compared with health effect levels (cancer benchmarks, noncancer reference concentrations [RfCs], and Arizona Ambient Air Quality Guidelines [AAAQG]). Typically, the AAAQG levels are within a factor of two of the cancer benchmarks. If a concentration was above the cancer benchmark, it was also usually above the AAAQG. All species were well below their RfC values at the site. Appendix D lists the top five concentrations of the pollutants in Table 3-3 by site. Appendix E contains the summary statistics for all the air toxics by site. Hazardous air pollutants can also cause noncancer health effects, such as respiratory irritation. Reference concentrations provide an indication of levels in which chronic exposures may be of concern. For the air toxics in Phoenix, only formaldehyde at the Greenwood site exhibited concentrations that exceeded the reference concentration (see Table 3-3). All other species were below their reference concentration. 3.2.5 How Do Ambient Concentrations Compare with NATA Results? The EPA performed a national-scale air toxics assessment (NATA) to identify and prioritize air toxics, emissions source types, and locations that are of greatest potential concern in terms of contributing to population risk. The results of the 1999 NATA are available from EPA (http://www.epa.gov/ttn/atw/nata1999/). The 10 most important air toxics to inhalation cancer risk in the JATAP area identified in the 1999 NATA included benzene, ethylene dibromide, 1,3-butadiene, 1,1,2,2-tetrachloroethane, carbon tetrachloride, acetaldehyde, bis(2-ethylhexyl)phthalate, naphthalene, tetrachloroethylene, and ethylene dichloride. Of these air toxics, measurements of bis(2-ethylhexyl)phthalate, naphthalene, and ethylene dichloride 3-7 MODELMONITOR were not available from the JATAP project; these air toxics are not commonly monitored across the United States. The 1999 NATA county-level model results were compared with the 2005 annual average concentrations measured at JATAP sites (Figure 3-3). In comparing model to monitor concentrations, the NATA web site noted that results within a factor of two were considered to be in reasonable agreement. For the JATAP sites, acetaldehyde, benzene, dichloromethane, and trichloroethene typically agreed within a factor of two of the model consistent with national-level comparisons of model and monitor data. Carbon tetrachloride was not modeled well because the background concentrations assumed in the model were too low; chloroform modeling was similar. Ethylbenzene-, formaldehyde-, and tetrachloroethylene-monitored concentrations at the JATAP sites were more than a factor of two higher than model estimates. 1.00 0.10 th th ne de oe de ne tra rm hy nze Te rofo ome nze ehy hlor loroe e r ld Be bon hlo lo lbe ald rac ich r eta C ich thy orm Tet Tr c Ca D E F A Figure 3-3. Annual average concentrations estimated by the 1999 NATA model compared with 2005 annual average concentrations at JATAP sites. County-level model results were used. Only species with more than 15% of data above detection are reported for the monitored data. Boxes show the 25th, 50th (median), and 75th percentiles. Horizontal lines at 0.5 and 2 represent agreement between the model and monitor values within a factor of two. 3-8 Table 3-3. Annual mean concentrations (µg/m3) at JATAP sites in 2005 compared with cancer benchmarks, RfCs, and AAAQG). Species in boldface type indicate >50% of data are above the MDL. Concentrations for species not boldfaced should be used with caution. Species 3-9 1,3-Butadiene Acetaldehyde Benzene Carbon tetrachloride Chloroform Dichloromethane Ethylbenzene Formaldehyde Hexachlorobutadiene m,p-Xylene o-Xylene Styrene Tetrachloroethene Toluene Trichloroethene Vinyl chloride Greenwood Annual Mean (µg/m3) 0.62 5.07 2.79 0.63 0.33 1.15 2.06 9.81 0.11 5.43 2.08 1.71 0.89 8.82 0.27 0.03 JLG Supersite Annual Mean (µg/m3) 0.47 3.13 2.50 0.62 0.59 0.83 1.61 5.61 0.12 4.32 0.78 0.76 1.43 7.18 0.18 0.03 Queen Valley Annual Mean (µg/m3) 0.03 St. Johns Annual Mean (µg/m3) 0.13 Salt River Annual Mean (µg/m3) 0.15 0.38 0.61 0.05 0.12 0.82 0.61 0.56 0.11 0.26 0.37 1.65 0.57 0.35 0.46 0.71 0.11 0.82 0.38 0.11 0.18 0.59 0.09 0.03 2.26 0.88 0.36 0.35 0.35 2.51 0.18 0.03 1.91 1.83 0.79 1.96 0.76 7.23 0.18 0.02 South Phoenix Annual Mean (µg/m3) 0.64 3.15 2.33 0.60 0.32 0.64 1.12 4.20 2.49 3.46 1.20 0.40 1.32 6.86 0.22 0.03 West Phoenix Annual Mean (µg/m3) 0.71 2.43 0.54 0.34 1.04 2.38 2.91 4.84 1.67 0.82 0.94 12.87 0.42 0.02 Cancer Benchmark (µg/m3) 0.033 0.45 0.13 0.07 2.13 181.8 0.05 0.17 0.5 0.11 RfC (µg/m3) AAAQG (µg/m3) 2 9 30 40 98 1000 1000 9.8 90 100 100 1000 270 400 600 100 0.067 0.5 0.14 0.036 0.043 5.6 0.08 0.067 2.1 0.76 0.012 3.3 CHANGES IN AIR TOXICS CONCENTRATIONS OVER TIME 3.3.1 Seasonal Variation It is important to understand how air toxics concentrations vary by season to gain insight into source types (e.g., winter residential wood smoke versus year-round mobile source emissions) and the effects of meteorology on concentrations (e.g., differences in photochemistry and mixing heights). At all sites, concentrations of 1,3-butadiene, dichloromethane, tetrachloroethene, trichloroethene, styrene, benzene, and other mobile source-related species (e.g., toluene) were higher in the cooler months than during the warmer months. Lower mixing heights and less photochemistry likely contribute to these high concentrations more than changes in emissions for these pollutants. These findings are consistent with the pilot study ((McCarthy et al., 2004c) and with urban national-level analyses. 3.3.2 Weekday-Weekend Variation It is important to look at day-of-week variation in air toxics concentrations to gain insight into anthropogenic source types (e.g., mobile source emissions, point source emissions). With only one year of data collected on a 1-in-6-day schedule, the confidence intervals on the median concentrations are too large to discern statistically significant differences in concentration between days. In 2005, none of the species exhibited a statistically significant weekdayweekend difference in concentration; this is consistent with previous findings. Even when Supersite data across several years (2001-2005) were inspected, no significant day-of-week trend was evident. Our expectation was that mobile source-dominated air toxics, such as benzene, 1,3-butadiene, and formaldehyde, would display lower concentrations on weekends. However, this pattern was not observed. In locations with 1-hr or 3-hr duration samples, morning mobile source air toxic concentrations have shown declines on the weekends. It may be that over a 24-hr averaging period, there is less reduction in mobile source emissions on weekends in the Phoenix area than we anticipated. 3.3.3 Interannual Trends A useful way to compare sites and obtain an overall understanding of the data is to inspect various stratifications of selected air toxics. In box whisker plots (an example is shown in Figure 3-3), the box shows the 25th, 50th (median), and 75th percentiles. The whiskers always end on a data point, so when the plots show no data points beyond the end of a whisker, the whisker shows the value of the highest or lowest data point. The whiskers have a maximum length equal to 1.5 times the length of the box (the interquartile range). If there are data outside this range, the points are shown on the plot and the whisker ends on the highest or lowest data point within the range of the whisker. The “outliers” are further identified with asterisks representing the points that fall within three times the interquartile range from the end of the box and with circles representing points beyond this. 3-10 Because we were interested in how similar or dissimilar the data are among time periods, we used an option called a notched box whisker plot to analyze data in this study (also Figure 3-3). These plots include notches that mark confidence intervals. The boxes are notched (narrowed) at the median and return to full width at the 95% lower and upper confidence interval values. If the 95% confidence interval is beyond the 25th or 75th percentile, then the notches extend beyond the box (hence a “folded” appearance). Confidence intervals are a function of sample size; small sample size will increase these intervals. Benzene concentrations from 2001 to 2005 have remained relatively flat at the Supersite and South Phoenix site as shown in Figure 3-4. Other air toxics showed similar interannual trends—higher emissions in some years and lower in others—but no statistically significant change over this period. 300 20 15 200 Measure Ou tliers Outliers Data with in 3*IR 10 Whisker 5 Data within 1.5*IR Whisker 100 75t h percen tile Bo x (Inte rq uartile ran ge) Notch around median Median 25t h percen tile 95 % C.I. Median 95 % C.I. 0 0 Figure 3-4. Illustration of box whisker plots and notched box whisker plots as defined by SYSTAT statistical software. Annual average concentrations of air toxics from 2005 were also compared with concentrations from a study in 1994/1995 by (Zielinska et al., 1998). Measurements were made at the Supersite for a year (roughly spring 1994 through spring 1995). When comparing two data sets measured a decade apart, it is important to consider that the measurements were from different sampling and analytical methods with different detection limits, different sampling intervals, and different months; therefore, these comparisons should be considered qualitative. The most dramatic difference from the 1994/1995 time period compared with 2005 is that benzene concentrations at the Supersite have decreased by more than a factor of three (from roughly 7.7 µg/m3 to near 2 µg/m3). This drop is consistent with efforts over the past decade to reduce benzene in gasoline, paint, and other consumer products. Carbon tetrachloride concentrations have been shown to decline at roughly 10 ppt per year, and the 2005 measurements reflect this reduction compared to the 1994/1995 concentrations. Formaldehyde concentrations are similar between the two time periods while acetaldehyde concentrations in 2005 are significantly lower than a decade ago. 3-11 PSAZ 8 7 BENZENE 6 5 4 3 2 1 0 2000 2001 2002 2003 2004 2005 2006 2007 YEAR SPAZ 8 7 BENZENE 6 5 4 3 2 1 0 2000 2001 2002 2003 2004 2005 2006 2007 YEAR Figure 3-5. Twenty-four hour average concentrations by year at the Supersite (PSAZ) and South Phoenix (SPAZ) site from 2001 through 2005. 3-12 4. SUMMARY OF RESULTS The JATAP is a consortium of federal, state, local, and tribal air pollution control officials designed to address the risk from air toxics in the greater Phoenix Metropolitan area. Participants include U.S. Environmental Protection Agency (EPA) Region 9, the EPA OAQPS, ADEQ, the Maricopa County Environmental Services Division, the Pinal County Air Quality Control District, the Intertribal Council of Arizona, the Gila River Indian Community, the Salt River–Pima Maricopa Indian Community, and the Fort McDowell Yavapai Nation. The purpose of the study is to determine which air toxics are of most concern to South Phoenix and tribal communities. The measurements made in 2005 were the second phase of the study. Phase I consisted of air toxics measurements at the South Phoenix and West 43rd Avenue sites from August 2001 through March 2004 ((McCarthy et al., 2004a)). The ultimate goal of this consortium is to obtain a metropolitan-wide assessment of the risk associated with airborne toxics in greater Phoenix. Validation and analysis of the gaseous air toxics data collected during the 2005 JATAP field project focused on four questions: 1. Can air toxics data be used to assess risk in the Phoenix area? 2. Which species were identified as the most important contributors to risk? 3. What are the likely sources of air toxics in the Phoenix area? 4. How do concentrations of air toxics in the Phoenix area compare with other urban areas and to important health benchmarks? In addressing these questions, the following conclusions and recommendations were made based on the air toxics data collected at the JATAP sites. Can air toxics data be used to assess risk in the Phoenix area? • Data quality appeared reasonable and concentrations of two-thirds of the species at all sites were above detection limits most of the time. The amount of data above detection in 2005 was improved over 2004 because of generally lower detection limits. We noted that interlaboratory comparisons and collocated measurements indicate that JATAP concentrations collected at multiple sites may be biased low. • Pollutants for which we have modest to good confidence in estimating risk at most sites using 2005 JATAP data included carbon tetrachloride, acetaldehyde, benzene, dichloromethane, and formaldehyde. How do concentrations of air toxics in the Phoenix area compare with other urban areas and to important health benchmarks? • Phoenix area urban concentrations of 1,3-butadiene, acetaldehyde, formaldehyde, chloroform, benzene, and tetrachloroethene were typically on the high end of the national urban scale (i.e., above the 75th percentile). Other air toxics concentrations were typically 4-1 within the interquartile range of national concentrations (25th to 75th percentile) (Figure 4-1). Figure 4-1. Annual average concentrations at JATAP sites compared with national concentration ranges, background concentrations, and cancer benchmarks. Species with an asterisk (*) indicate ≥50% of data are below the MDL at some sites. • Greenwood carbonyl compound concentrations were higher than those at other sites and were above the national 95th percentile. This finding may indicate additional fresh emissions at a local scale, likely from the nearby mobile sources extremely close to the monitor. Carbonyl compound concentrations at South Phoenix and JLG Supersite were also high relative to other urban areas (i.e., above the 75th percentile). • Concentrations at JATAP sites in 2005 were typically above remote background concentrations, as expected. Air toxics concentrations at Queen Valley were lowest among the sites, consistent with its location far from the urban area. Air toxics concentrations at St. Johns, and sometimes at the Senior Center, were also typically on the lower end of the concentration ranges. Typical remote background concentrations contributed less than 10% of the total concentrations measured at the urban JATAP sites (Greenwood, Supersite, South Phoenix, and West Phoenix). 4-2 • Benzene concentrations at the Supersite decreased between 1995 and 2005 by more than a factor of three (from 7.7 µg/m3 to near 2 µg/m3). This drop is consistent with efforts over the past decade to reduce benzene in gasoline, paint, and other consumer products. Carbon tetrachloride concentrations have been shown to decline at roughly 10 ppt per year (McCarthy et al., 2006), and the 2005 measurements reflect this reduction compared with the 1994/1995 concentrations. Formaldehyde concentrations are similar between the two time periods while acetaldehyde concentrations in 2005 are significantly lower than a decade ago. Which species were identified as the most important contributors to risk? What are the likely sources of air toxics in the Phoenix area? • Annual mean concentrations of air toxics concentrations at JATAP sites were often higher than the one-in-a-million cancer benchmark. 1,3-butadiene, acetaldehyde, benzene, carbon tetrachloride, chloroform, and tetrachloroethene were well above the one-in-a-million risk level. These pollutants are emitted by mobile sources (1,3-butadiene, benzene), are formed in the atmosphere through reactions of other volatile organic compounds (acetaldehyde), are due to global background (carbon tetrachloride), and come from solvent use and other sources (chloroform, tetrachloroethene). However, concentrations of 1,3-butadiene, acetaldehyde, benzene, carbon tetrachloride and chloroform in urban areas in the United States are typically above the cancer benchmark. • Some species that are not monitored are considered important contributors to risk or health hazard (e.g., acrolein, arsenic, chromium VI, particulate organic matter [POM]). Of the 10 air toxics identified by the 1999 NATA as most important to inhalation cancer risk, ethylene dibromide and 1,1,2,2-tetrachloroethane were monitored for but were below detection while bis(2-ethylhexyl)phthalate, naphthalene, and ethylene dichloride are not currently monitored. • For the JATAP sites, acetaldehyde, benzene, dichloromethane, and trichloroethene typically agreed within a factor of two of the 1999 NATA model results consistent with national level comparisons of model and monitor data. Ethylbenzene-, formaldehyde-, and tetrachloroethylene-monitored concentrations at the JATAP sites were more than a factor of two higher than model estimates. 4-3 5. RECOMMENDATIONS As this second phase of JATAP ends, several tasks are yet to be performed according to the original JATAP blueprint, including emission inventory development, urban airshed modeling of air toxics, risk assessment, and communication of risk to the general population. The work performed in this project—analysis of air toxics collected as part of the JATAP 2005—leads to additional recommendations for further research: • Part of this work included exploring the use of source apportionment with the air toxics data. Exploratory analysis using positive matrix factorization (PMF) showed that there was an insufficiently large matrix of air toxics samples and other species to obtain meaningful results. However, STI’s recent work with EPA exploring multiple pollutant source apportionment indicates that it may be useful to perform source apportionment on a large data set of combined air toxics and speciated PM2.5, such as that available from the Supersite from 2000-2005. • This work focused on gaseous air toxics. Validation and analysis of speciated PM2.5 data collected at the Supersite should be performed. This site has an especially rich record of data including data from the national networks, STN and IMPROVE. Particulate toxics can be important in terms of risk in urban areas, and the PM2.5 data could be placed in a broader national context: how do toxics metals fit into the national picture? Are concentrations above benchmarks? • Another area of concern to the air toxics community is diesel particulate matter (DPM), which was not examined in this work. Aethalometer™ black carbon measurement occurring at the Supersite can help us better understand DPM: . Validation and analysis of Aethalometer data from the Supersite should be performed with the goal of better understanding the importance of DPM. • A formal analysis of spatial variability could be performed to statistically compare concentrations and the coefficient of variation by pollutant among the sites. However, in qualitative review of the data, the Supersite appears to be fairly representative of most of the other urban sites, indicating that some of the other urban sites may be redundant. The Greenwood site was very interesting in that concentrations of many pollutants were higher at this site, likely due to its proximity to the freeways. The concentrations at the West and South Phoenix sites were similar to each other and to those at the Supersite, indicating that these sites may not be as important to maintain in the future. For the more rural sites, Queen Valley continues to be useful as a remote site, but concentrations of air toxics are very low and often below detection. While the measurements provide information, ADEQ may want to work with EPA and others to consider ways to lower the detection limits to improve data quality at the low concentration sites (e.g., longer duration samples). The tribal land sites are different from each other and show some features different from Queen Valley, indicating that retaining these sites may be important. • The air toxics data collected during JATAP 2005 are sufficient to perform a risk assessment screening. 5-1 • Analyses focused on the air toxics with cancer benchmarks; however, further exploration of species with noncancer benchmarks, ozone precursors, and chlorofluorocarbons could be made, including trends over time and spatial variability. • For the Supersite, trends in air toxics should be compared with trends in criteria pollutants and known emission control programs to assess whether control programs targeting criteria and other pollutants may have a beneficial multipollutant effect, including on air toxics. 5-2 6. REFERENCES Hafner H.R., Brown S.G., McCarthy M.C., Arkinson H.L., and Sullivan D.C. (2004) Validation and preliminary analysis of JATAP air toxics data. Presentation to the Arizona Department of Environmental Quality, Phoenix, AZ, June 11 by Sonoma Technology, Inc., Petaluma, CA, STI-903774-2564. Hafner H.R. and McCarthy M.C. (2004) Phase III air toxics data analysis workbook. Workbook prepared for the Lake Michigan Air Directors Consortium, Des Plaines, IL, by Sonoma Technology, Inc., Petaluma, CA, STI-903553-2592-WB, August. Hafner H.R. and Prouty J.D. (2004) VOCDat user’s guide. Prepared for U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Research Triangle Park, NC by Sonoma Technology, Inc., Petaluma, CA, STI-902620-2015-UG, September. Kenski D., Koerber M., Hafner H.R., McCarthy M.C., and Wheeler N. (2005) Lessons learned from air toxics data: a national perspective. Environ. Man. J., 19-22. McCarthy M.C., Brown S.G., Frankel A., and Hafner H.R. (2004a) Data analysis for air toxics collected at JATAP sites from 2001 to 2004: Task 2 results. Final report prepared for Arizona Department of Environmental Quality, Phoenix, AZ, by Sonoma Technology, Inc., Petaluma, CA, STI-903772-2620-FR, November. McCarthy M.C., Brown S.G., Hafner H.R., and Frankel A. (2004b) Data analyses for Gila River air toxics collected from 2003 to 2004. Final report prepared for the Gila River Indian Community Department of Environmental Quality, Sacaton, AZ, by Sonoma Technology, Inc., Petaluma, CA, STI-904063-2608-FR, September. McCarthy M.C., Brown S.G., Hafner H.R., and Shields E. (2004c) Prescribed data analysis for Phoenix, Arizona, hydrocarbon data collected in 2002 and 2003. Final report prepared for Arizona Department of Environmental Quality, Phoenix, AZ, by Sonoma Technology, Inc., Petaluma, CA, STI-903701-2600-FR, September. McCarthy M.C. and Hafner H.R. (2004) Data validation of South Phoenix and West 43rd Street 2003 toxics data as part of JATAP Small Area Monitoring Project. Technical memorandum prepared for the Arizona Department of Environmental Quality, Phoenix, AZ, by Sonoma Technology, Inc., Petaluma, CA, Purchase order number EV04-0084, STI-903772-2593, August. McCarthy M.C., Hafner H.R., and Montzka S.A. (2004d) Background concentrations of 18 core air toxics in the northern hemisphere. Journal of Air and Waste Management Association (to be submitted) (STI-903550-2589). McCarthy M.C., Hafner H.R., and Chinkin L.R. (2006a) Temporal variability of selected air toxics: a national perspective. (Submitted) (STI-2894). McCarthy M.C., Hafner H.R., and Chinkin L.R. (2006b) Spatial variability of selected air toxics: a national perspective. (Submitted) (STI-2893). McCarthy M.C., Hafner H.R., and Montzka S.A. (2006c) Background concentrations of 18 air toxics for North America. J. Air & Waste Manage. Assoc. 56, 3-11 (STI-903550-2589). 6-1 Available on the Internet at http://www.awma.org/journal/ShowAbstract.asp?Year=&PaperID=1509. Sundblom M., Armijo C., and Hafner H. (2006) Joint Air Toxics Assessment Project (JATAP) for the Maricopa/Pinal urban area, Arizona. Presentation for the EPA National Air Monitoring Conference, Las Vegas, NV, November 8, by the Arizona Department of Environmental Quality, the Salt River Pima Maricopa Indian Community, and Sonoma Technology, Inc., Petaluma, CA. U.S. Environmental Protection Agency (2000) Air toxics monitoring concept paper. Revised peer review draft for the Science Advisory Committee, Air Toxics Monitoring Strategy Subcommittee FY-00 prepared by the U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, February 29. Available on the Internet at . Zielinska B., Fujita E., Sagebiel J., Harshfield G., Uberna E., Hayes T., and Keene F. (1998) Arizona hazardous air pollutants monitoring program. J. Air & Waste Manag. Assoc. 48, 1038-1050 (11). 6-2 APPENDIX A LIST OF FLAGGED SAMPLES Data validation, when done well, is labor intensive. To assess the validity of the data, we inspected a time series plot of every species, prepared numerous scatter plots, including a plot of every species and species group versus the total mass, and inspected the fingerprint plot of every sample. Whole samples are flagged as suspect only if a large percentage of above MDL species shows suspicious concentrations or temporal variability. Individual species are flagged as suspect when problems are noted. In some cases, the samples may have an analytical error that needs to be corrected, may have a sampling problem that was inadvertently missed, or may just appear odd. A review of the data by ADEQ or EAS may indicate that some suspect data are invalid, some data need to be reprocessed and re-reported, and other data remain suspect (cause unknown). These suspect samples remain flagged in our working database, and in this report, for the attention of data analysts. Table A-1 lists the samples and individual VOCs collected at the JATAP sites in 2005 that were flagged as suspect or noted as having potential problems, Table A-2 lists the missing or invalid samples from the JATAP sites in 2005, and Table A-3 lists average minimum detection limits for species collected at JATAP sites in 2005. A-1 Table A-1. Samples collected at JATAP sites in 2005 that were flagged by STI. Sample Date QC Codea Speciesb Comment Greenwood 8/2/2005 8/8/2005 7 7 Formaldehyde Formaldehyde below background concentration below background concentration Senior Center 3/11/2005 7 Sample final pressure < 5 psi 2/3/2005 8 channel never came on, no air to canister 1/4/2005 7 1/28/2005 2/15/2005 2/21/2005 3/5/2005 3/11/2005 4/16/2005 4/22/2005 5/10/2005 5/28/2005 6/9/2005 8/2/2005 10/1/2005 11/12/2005 11/30/2005 West Phoenix 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Sample Formaldehyde, Acetaldehyde Chloroform, Carbon tetrachloride Sample Tetrachloroethene Sample Sample Sample Sample Sample Sample Sample Sample Toluene Sample Sample 10/7/2005 8 Sample 2/15/2005 7 Sample 3/17/2005 4/22/2005 8/27/2005 10/13/2005 7 7 7 7 Sample Sample Sample Sample 10/25/2005 10/31/2005 12/12/2005 12/18/2005 12/30/2005 7 7 7 7 7 Sample Sample Sample Sample Sample South Phoenix below background concentration spike, probably real, not typical final pressure < 5 psi spike final pressure -1 psi final pressure < 5 psi dip, initial canister pressure of -22 in Hg final pressure 1.5 psi final pressure 2 psi final pressure -1 psi final pressure 2 psi dip all species, power outage toluene:benz outlier, low compared to benz dip all species, below background final pressure 0 psi final canister pressure < 5 psi, canister broken initial canister pressure of -20 in Hg; collocated analysis discrepancies line to canister was loose, final canister pressure < 5 psi; collocated analysis discrepancies final canister pressure 1 psi collocated analysis discrepancies final canister pressure 2 psi actual run time unknown, removal flow rate not in acceptable range. final pressure < 5 psi final pressure negative. final pressure negative. final pressure 0 psi a QC Code = Flag code. A flag of 7 indicates a suspect species record, 8 indicates an invalid record, 9 indicates a species record that has been designated as missing, and “note” is a species record that could have potential issues. b Species = A designation of the individual hydrocarbon or the entire sample record to which the flag applies. A-2 Table A-2. Missing or invalid samples from JATAP sites in 2005. Site Date of Expected Sample Sample Type Reason Sample is Missing or Invalid Greenwood December 30, 2005 Cartridge no sample taken Senior Center January 4, 2005 Canister no sample taken South Phoenix January 10, 2005 Cartridge sample flow rate out of limits South Phoenix January 19, 2005 Cartridge sample flow rate out of limits South Phoenix February 3, 2003 Canister no air to canister South Phoenix July 15, 2005 Canister voided by EAS lab South Phoenix July 21, 2005 Canister make up run on July 24 South Phoenix October 19, 2005 Cartridge sample flow rate out of limits South Phoenix November 24, 2005 Cartridge sample flow rate out of limits South Phoenix November 30, 2005 Cartridge sample flow rate out of limits South Phoenix December 6, 2005 Cartridge sample flow rate out of limits South Phoenix December 12, 2005 Cartridge lost South Phoenix December 18, 2005 Cartridge sample flow rate out of limits South Phoenix December 24, 2005 Cartridge sample flow rate out of limits Supersite January 10, 2005 Cartridge sample flow rate out of limits Supersite January 16, 2005 Cartridge sample flow rate out of limits Supersite January 28, 2005 Cartridge sample flow rate out of limits Supersite February 9, 2005 Cartridge sample flow rate out of limits Supersite October 25, 2005 Cartridge sample flow rate out of limits Supersite November 24, 2005 Cartridge sample flow rate out of limits Supersite November 30, 2005 Cartridge sample flow rate out of limits Supersite December 6, 2005 Cartridge sample flow rate out of limits Supersite December 12, 2005 Cartridge sample flow rate out of limits Supersite December 12, 2005 Cartridge sample flow rate out of limits Supersite December 18, 2005 Cartridge sample flow rate out of limits Supersite December 24, 2005 Cartridge sample flow rate out of limits West Phoenix January 4, 2005 Canister canister not analyzed West Phoenix October 7, 2005 Canister canister broken A-3 Table A-3. Average minimum detection limits for species collected at JATAP sites in 2005. Method AQS Code Parameter Average MDL (ppb) TO-15 43218 1,3-Butadiene 0.10 TO-15 43844 Hexachlorobutadiene 0.42 TO-15 SIM 43802 Dichloromethane 0.03 TO-15 SIM 43803 Chloroform 0.03 TO-15 SIM 43804 Carbon tetrachloride 0.04 TO-15 SIM 43815 1,2-Dichloroethane 0.03 TO-15 SIM 43817 Tetrachloroethene 0.06 TO-15 SIM 43819 Bromomethane 0.02 TO-15 SIM 43824 Trichloroethene 0.06 TO-15 SIM 43826 1,1-Dichloroethene 0.02 TO-15 SIM 43829 1,2-Dichloropropane 0.04 TO-15 SIM 43860 Vinyl chloride 0.02 TO-15 SIM 45109 m & p-Xylene 0.06 TO-15 SIM 45201 Benzene 0.03 TO-15 SIM 45202 Toluene 0.04 TO-15 SIM 45203 Ethylbenzene 0.04 TO-15 SIM 45204 o-Xylene 0.04 TO-15 SIM 45207 1,3,5-Trimethylbenzene 0.03 TO-15 SIM 45208 1,2,4-Trimethylbenzene 0.03 TO-15 SIM 45220 Styrene 0.06 A-4 APPENDIX B DATA VALIDATION DETAILS BY SITE FOR 2005 Data Validation: Greenwood The Greenwood site ambient carbonyl compound data for 2005 had no major issues. On August 2 and August 8, 2005, formaldehyde concentrations were flagged as suspect because the concentrations were below typical background concentrations as shown in Figure B-1. Figure B-1. Low concentrations (ppb) in formaldehyde (form) (here shown with acetaldehyde – aceta) at Greenwood on August 2 and August 8, 2005, were flagged as suspect. Data Validation: Senior Center For Senior Center, the 12-hr morning and evening samples were inspected individually and then averaged to create 24-hr average data for further validation. We made the following observations of the ambient 2005 Senior Center data: • Higher than usual dichloromethane concentrations seen in Figure B-2 on April 22, May 4, and December 24, 2005, were noted but not flagged. • A spike in tetrachloroethene concentration shown in Figure B-3 on March 29, 2005, was noted but not flagged. B-1 Figure B-2. Time series of toxics concentrations (ppb) at the Senior Center site. High dichloromethane (DCM) concentrations on April 22, May 4, and December 24, 2005, are circled in red. Figure B-3. Time series of toxics concentrations (ppb) at the Senior Center site. A high tetrachloroethene (TCE) concentration on March 29, 2005, is circled in red. Data Validation: South Phoenix The 2005 ambient South Phoenix air toxics data can be characterized by several specific issues and observations: • A spike in tetrachloroethene concentration on February 21, 2005, shown in Figure B-4, was flagged as suspect. The concentration is a factor of 40 greater than the average tetrachloroethene concentration at the site. • Spikes that occurred in chloroform and carbon tetrachloride concentrations on January 28, 2005, were flagged as suspect. The concentrations are possibly real, but not typical. B-2 • Eight samples were flagged as suspect because the final pressure of the canisters was below the acceptable final pressure of 5 psi. Most of these samples cannot be differentiated from those with acceptable final pressures; however, these data should be used with caution. Figure B-4. Time series of tetrachloroethene (TCE), ethylbenzene (ebenz), and trichloroethene (TCE3) concentrations (ppb) at South Phoenix. The high concentration for TCE (circled in red) on February 21, 2005 was flagged as suspect. The South Phoenix 2005 carbonyl compound data were analyzed by the SDAPCD. The flow rates of 12 samples were lower than acceptable limits and were, therefore, flagged as invalid by laboratory analysts. The reduced flow was caused by a clogged needle valve used to maintain flow rate during sample collection; this sampling problem has been addressed by SDAPCD in its more recent sampling equipment supplied to ADEQ. Cartridge samples that do not meet accepted flow rates are invalidated because concentrations of these samples are calculated by dividing a mass of analyte by the volume of air flowing through the cartridge; flow rates beyond acceptable limits mean a concentration cannot be calculated with any certainty. Data Validation: JLG Supersite The JLG Supersite ambient carbonyl compound data for 2005 were analyzed by the SDAPCD. The only issue with samples collected at this site was that the flow rates of 12 samples were lower than accepted limits and were therefore flagged as invalid by the laboratory; the same problem occurred at the South Phoenix site for carbonyl compound samples. Data Validation: West Phoenix Specific issues with respect to the West Phoenix ambient 2005 data were noted: • Eight samples were flagged as suspect because the final pressure of the canisters was below the acceptable final pressure of 5 psi. Most of these samples cannot be B-3 differentiated from those with acceptable final pressures; however, these data should be used with caution. • Spikes in toluene and trichloroethene concentrations on February 21, 2005, shown in Figure B-5, were noted. These species were not flagged; however, these concentrations were not typical. We speculate that these spikes may be due to nearby solvent use. • A high concentration sample of dichloromethane on June 9, 2005, was noted but not flagged. 29 ppb toluene 2/21/05 trichloroethene 1.2 ppb 5/4/05 Figure B-5. Comparison of fingerprint plots of data collected at the West Phoenix site on two days. The fingerprint plot on February 21, 2005, (top) shows significantly higher concentrations of toluene and trichloroethene concentrations (ppb) compared to typical concentrations as illustrated on May 4, 2005 (bottom). B-4 APPENDIX LIST OF SELECTED SPECIES AND THEIR EMISSIONS SOURCES. C-l Page 1 of 2 Species Likely Emission Sources Comments acetaldehyde Key photochemical reaction product. Also produced from fuel combustion (diesel?). Intermediate in synthesis of other chemicals (e.g., perfumes, polyester resins, and basic dyes); fruit and fish preservative and flavoring agent; also a solvent in rubber, tanning, and paper industries. benzene Constituent in motor fuels. Also present in fuel combustion processes, refining, and solvents. Secondary source from degradation of other volatile organic compounds (VOCs); most likely to correlate with formaldehyde Tracer for vehicle exhaust; significantly reduced since 1995 with the introduction of reformulated gasoline. Likely to correlate with benzene and other mobile source tracer species. 1,3-butadiene carbon tetrachloride C-2 chloroform formaldehyde dichloromethane (aka methylene chloride) Petroleum refining, motor vehicle exhaust, production of rubber and plastics/ copolymers. Also from forest fires and cigarette smoke. Completely phased out of production in 2000 by Clean Air Act Title VI. Limited uses expected in the United States post-2000. Was used as an intermediate for production of CFCs/freons (refrigerants), propellants, solvents, landfill disposal, grain fumigant, dry cleaning agent. Only industrial uses remain; no fumigant or commercial uses. Solvent for adhesives, pesticides, fats, oils, and rubbers. Used in pharmaceutical manufacturing and to make hydrochlorofluorocarbons (HCFC-22). By-product of chlorination from drinking water plants (and swimming pools?) and wood pulp chlorination. Key photochemical reaction product. Also produced by power plants, manufacturing facilities, incinerators, and automobile exhaust emissions. Used as a chemical intermediate, analytical reagent, concrete and plaster additives, cosmetics, disinfectants, fumigants, photography, and wood preservation; ureaformaldehyde resins and particle board products. Common solvent, agent in urethane foam blowing propellant, electroplating, and the paint industry; aerospace manufacturing and de-icing. Very long lifetime in the atmosphere. Concentrations do not vary much site to site on multi-year basis. Past uses: extraction of solvents, dry cleaning spot remover, fire extinguishers, fumigant, anesthetic – no longer used for these products. Secondary source from degradation of other VOCs; most likely to correlate with acetaldehyde. Ureaformaldehyde insulating material phased out in 1982. May have been used in semiconductor industry. Page 2 of 2 Species Likely Emission Sources Comments 1,2-dichloropropane (aka propylene chloride) Chemical intermediate in production of chlorinated organic chemicals; industrial solvent; ion exchange; toluene diisocyanate production; photographic file manufacture; paper coating; petroleum catalyst regeneration; soil fumigant; and degreaser. perchloroethylene (aka tetrachloroethylene) Dry cleaning and textile processing; chemical intermediate; vapor degreasing in metal cleaning operations. Pesticide formulations containing 1,2-dichloropropane are no longer available. Concentrations are typically below MDL at most sites. Chemical residence time in the atmosphere of a few months. Concentrations tend to correlate with population density. Was used as a general anesthetic. Often (>50%) below MDL in national studies. trichloroethylene vinyl chloride C-3 Solvent degreasing, dry cleaning, chemical intermediate, and refrigerant. Consumer products (paint remover, adhesives, spot removers, rug-cleaning fluid). Chemical and plastics industry (polyvinyl chloride [PVC] precursor) used in furniture and car upholstery, wall coverings; housewares and automotive parts refrigerant. Landfills and waste water treatment may release vinyl chloride. Almost entirely (>90%) below MDL in national studies. APPENDIX TOP FIVE 24-HR CONCENTRATIONS IN 2005 BY POLLUTANT AND SITE D-l Table D-1. The five highest 24-hour concentrations (µg/m3) at each JATAP site in 2005 compared to Arizona Ambient Air Quality Guidelines (AAAQG). Values ≥ AAAQG are in bold. Page 1 of 4 Species 1,3-Butadiene Acetaldehyde D-2 Benzene Carbon tetrachloride Chloroform Greenwood 24-hr (µg/m3) 1.7 1.3 1.3 1.3 1.2 17.3 14.1 10.7 9.6 9.4 6.7 6.2 5.6 5.5 5.2 0.8 0.7 0.7 0.7 0.7 0.8 0.6 0.6 0.6 0.6 JLG Supersite 24-hr (µg/m3) 1.5 1.4 1.1 1.1 1.1 5.8 8.8 4.8 4.7 4.5 6.6 5.9 5.5 5.0 5.0 0.7 0.7 0.7 0.7 0.7 1.7 1.5 1.4 1.4 1.3 Queen Valley 24-hr (µg/m3) 0.09 0.09 0.09 0.08 0.08 St. Johns 24-hr (µg/m3) 0.9 0.5 0.3 0.3 0.3 Salt River 24-hr (µg/m3) 1.0 0.7 0.5 0.4 0.3 a a a 0.6 0.4 0.4 0.4 0.4 0.8 0.7 0.7 0.7 0.7 0.1 0.1 0.1 0.1 0.1 1.6 1.4 1.3 1.3 1.2 1.2 1.0 0.9 0.8 0.8 0.3 0.2 0.2 0.2 0.2 23.2 18.3 9.1 8.5 2.1 1.0 0.9 0.9 0.9 0.8 0.9 0.9 0.9 0.7 0.6 South Phoenix 24-hr (µg/m3) 3.8 3.3 2.7 2.1 1.9 15.3 5.8 5.4 4.8 4.7 17.0 5.1 4.9 4.3 4.2 5.6 1.1 0.9 0.9 0.9 4.9 1.0 0.8 0.6 0.5 West Phoenix 24-hr (µg/m3) 2.9 2.7 2.5 2.1 2.1 a 8.0 7.7 6.4 5.9 5.5 0.9 0.9 0.8 0.8 0.8 1.7 1.0 0.8 0.7 0.6 AAAQG 24-hr (µg/m3) 1.9 1400 51 13 16 Table D-1. The five highest 24-hour concentrations (µg/m3) at each JATAP site in 2005 compared to Arizona Ambient Air Quality Guidelines (AAAQG). Values ≥ AAAQG are in bold. Page 2 of 4 Species Dichloromethane Ethylbenzene D-3 Formaldehyde Hexachlorobutadiene m,p-Xylene Greenwood 24-hr (µg/m3) 4.0 4.0 3.1 2.7 2.4 4.7 4.7 4.6 4.6 4.2 29.0 20.6 18.5 17.3 16.4 0.4 0.3 13.8 11.8 11.4 10.7 10.5 JLG Supersite 24-hr (µg/m3) 2.4 1.9 1.8 1.7 1.7 4.3 4.0 4.0 3.6 3.6 9.6 9.0 8.2 7.6 7.5 0.6 0.4 11.2 9.5 9.4 9.1 8.3 Queen Valley 24-hr (µg/m3) 0.2 0.2 0.2 0.2 0.2 0.5 0.4 0.4 0.4 0.4 St. Johns 24-hr (µg/m3) 1.4 1.3 0.7 0.5 0.5 1.4 1.2 1.1 1.0 0.9 Salt River 24-hr (µg/m3) 2.8 2.4 2.0 1.0 0.8 6.0 5.4 1.7 1.4 1.4 South Phoenix 24-hr (µg/m3) 3.0 1.9 1.7 1.3 1.3 3.0 2.9 2.8 2.8 2.6 7.1 7.0 6.6 6.0 6.0 West Phoenix 24-hr (µg/m3) 6.0 3.2 2.7 2.6 2.3 6.1 5.5 4.6 3.5 3.3 a a a a 12 b b b b b 1.9 1.1 1.0 1.0 1.0 1.0 4.6 3.3 2.1 2.1 2.0 14.3 5.2 4.9 4.8 4.3 9.1 8.5 8.4 7.8 7.6 16.7 9.5 9.0 8.8 8.7 3500 AAAQG 24-hr (µg/m3) 2000 3500 Table D-1. The five highest 24-hour concentrations (µg/m3) at each JATAP site in 2005 compared to Arizona Ambient Air Quality Guidelines (AAAQG). Values ≥ AAAQG are in bold. Page 3 of 4 Species o-Xylene Styrene D-4 Tetrachloroethene Toluene Trichloroethene Greenwood 24-hr (µg/m3) 5.1 4.7 4.4 4.3 4.1 4.6 4.5 4.5 4.2 3.8 8.4 3.3 2.4 1.7 1.7 26.3 20.0 19.3 18.5 18.1 1.7 0.6 0.6 0.5 0.5 JLG Supersite 24-hr (µg/m3) 4.0 3.8 3.6 3.4 3.4 2.9 1.9 1.4 1.4 1.3 5.5 4.4 4.2 4.0 3.5 18.4 18.2 16.0 15.3 15.1 0.5 0.4 0.4 0.4 0.3 Queen Valley 24-hr (µg/m3) 0.6 0.5 0.5 0.5 0.5 0.5 0.5 0.4 0.4 0.4 0.5 0.4 0.3 0.3 0.2 1.5 1.1 1.0 1.0 0.9 0.2 0.2 0.2 0.2 0.2 St. Johns 24-hr (µg/m3) 1.6 1.5 0.8 0.8 0.8 2.2 2.1 1.4 1.3 1.3 2.3 2.1 1.7 1.5 0.7 9.7 9.4 8.3 5.4 4.5 1.0 0.9 0.7 0.2 0.2 Salt River 24-hr (µg/m3) 6.2 2.7 2.6 2.0 1.8 6.8 4.7 4.5 3.9 3.4 8.1 6.6 4.8 2.0 1.4 62.8 40.6 37.1 29.0 25.1 0.7 0.5 0.4 0.4 0.4 South Phoenix 24-hr (µg/m3) 3.3 3.2 3.0 2.9 2.6 2.5 2.3 0.9 0.8 0.8 6.3 3.9 3.8 3.4 3.3 28.4 18.6 15.8 15.7 14.3 3 0.7 0.5 0.4 0.3 West Phoenix 24-hr (µg/m3) 5.6 3.4 3.0 3.0 2.9 3.5 3.5 3.4 3.0 3.0 6.2 2.0 2.0 1.4 1.2 124 21.9 21.8 21.0 20.0 12 0.5 0.4 0.4 0.4 AAAQG 24-hr (µg/m3) 3500 1700 770 3000 280 Table D-1. The five highest 24-hour concentrations (µg/m3) at each JATAP site in 2005 compared to Arizona Ambient Air Quality Guidelines (AAAQG). Values ≥ AAAQG are in bold. Page 4 of 4 Species Greenwood 24-hr (µg/m3) JLG Supersite 24-hr (µg/m3) Queen Valley 24-hr (µg/m3) 0.13 b b Vinyl chloride a Measured at Greenwood, JLG Supersite, and South Phoenix only. Below detection limit. b St. Johns 24-hr (µg/m3) 0.27 0.19 0.09 0.09 0.07 Salt River 24-hr (µg/m3) South Phoenix 24-hr (µg/m3) West Phoenix 24-hr (µg/m3) AAAQG 24-hr (µg/m3) 0.04 0.01 0.04 0.02 0.06 0.02 4.4 D-5 APPENDIX SUMMARY STATISTICS FOR 2005 BY SITE AND POLLUTANT E-l Table E1. Summary statistics for Greenwood in 2005. Page 1 of 2 SITE Species 1,1,2,2-Tetrachloroethane 1,1,2-Trichloroethane 1,1-Dichloroethane 1,1-Dichloroethene 1,2,4-Trichlorobenzene 1,2,4-Trimethylbenzene 1,2-Dichlorobenzene 1,2-Dichloroethane 1,2-Dichloropropane 1,3,5-Trimethylbenzene 1,3-Butadiene 1,3-Dichlorobenzene 1,4-Dichlorobenzene Acetaldehyde Benzene Benzyl Chloride Bromoform Bromomethane Carbon Tetrachloride Chlorobenzene Chloroethane Chloroform Chloromethane Cis-1,2-Dichloroethene Cis-1,3-Dichloropropylene Dichlorodifluoromethane Dichloromethane Ethylbenzene Ethylene Dibromide Formaldehyde Freon 113 Freon 114 Hexachlorobutadiene Isoprene M,P-Xylene Methyl Chloroform Methyl Ethyl Ketone Methyl Tert-Butyl Ether M-Xylene O-Xylene GWAZ Number of cases 59 59 59 59 59 59 59 59 59 59 59 59 59 60 59 59 59 59 59 59 59 59 59 59 59 59 59 59 59 60 59 59 59 59 59 59 59 59 59 59 GWAZ Mean (ppbv) 0.011 0.010 0.010 0.011 0.046 0.496 0.014 0.017 0.010 0.187 0.273 0.015 0.131 2.728 0.844 0.012 0.012 0.014 0.097 0.013 0.020 0.066 0.435 0.014 0.016 0.447 0.321 0.457 0.011 7.479 0.067 0.014 0.010 0.461 1.211 0.024 0.728 0.010 0.831 0.464 GWAZ Median (ppbv) 0.010 0.010 0.010 0.010 0.010 0.406 0.010 0.010 0.010 0.169 0.234 0.010 0.120 2.410 0.731 0.010 0.010 0.010 0.097 0.010 0.010 0.056 0.449 0.010 0.010 0.454 0.228 0.355 0.010 6.893 0.067 0.014 0.010 0.389 1.115 0.024 0.717 0.010 0.775 0.377 E-2 GWAZ GWAZ Minimum Maximum 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.035 0.010 0.010 0.740 0.203 0.010 0.010 0.005 0.068 0.010 0.010 0.010 0.250 0.010 0.010 0.093 0.056 0.115 0.010 0.071 0.042 0.010 0.010 0.119 0.114 0.009 0.050 0.010 0.104 0.083 0.020 0.010 0.010 0.048 0.192 1.069 0.096 0.091 0.010 0.362 0.763 0.107 0.244 9.290 2.023 0.157 0.032 0.060 0.115 0.035 0.066 0.149 0.566 0.266 0.123 0.842 1.107 1.049 0.040 22.871 0.087 0.020 0.035 1.130 3.066 0.069 1.597 0.010 2.062 1.130 GWAZ Standard Dev 0.002 0.000 0.000 0.005 0.057 0.275 0.015 0.013 0.000 0.068 0.149 0.018 0.044 1.598 0.435 0.019 0.005 0.011 0.010 0.006 0.014 0.033 0.073 0.033 0.020 0.108 0.245 0.265 0.007 3.895 0.011 0.002 0.003 0.289 0.678 0.010 0.328 0.000 0.467 0.265 GWAZ Average MDL 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.20 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.15 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 Table E1. Summary statistics for Greenwood in 2005. Page 2 of 2 SITE Species P-Ethyltoluene P-Xylene Styrene Tetrachloroethene Toluene Trans-1,2Dichloroethylene Trans-1,3Dichloropropylene Trichloroethene Trichlorofluoromethane Vinyl Chloride GWAZ Number of cases 59 59 59 59 59 GWAZ Mean (ppbv) 0.213 0.381 0.390 0.128 2.266 GWAZ Median (ppbv) 0.188 0.332 0.286 0.090 1.858 GWAZ GWAZ 0.456 1.004 1.035 1.196 6.766 GWAZ Standard Dev 0.100 0.214 0.249 0.165 1.458 GWAZ Average MDL 0.02 0.02 0.02 0.02 0.02 Minimum Maximum 0.010 0.010 0.010 0.024 0.234 59 0.013 0.010 0.010 0.177 0.022 0.02 59 0.014 0.010 0.010 0.106 0.016 0.02 59 59 59 0.049 0.311 0.011 0.042 0.283 0.010 0.010 0.220 0.010 0.309 1.405 0.050 0.041 0.152 0.005 0.02 0.02 0.02 E-3 Table E2. Summary statistics for JLG Supersite in 2005. Page 1 of 2 SITE Species 1,1,2,2-Tetrachloroethane 1,1,2-Trichloroethane 1,1-Dichloroethane 1,1-Dichloroethene 1,2,4-Trichlorobenzene 1,2,4-Trimethylbenzene 1,2-Dichlorobenzene 1,2-Dichloroethane 1,2-Dichloropropane 1,3,5-Trimethylbenzene 1,3-Butadiene 1,3-Dichlorobenzene 1,4-Dichlorobenzene Acetaldehyde Benzene Benzyl Chloride Bromoform Bromomethane Carbon Tetrachloride Chlorobenzene Chloroethane Chloroform Chloromethane Cis-1,2-Dichloroethene Cis-1,3-Dichloropropylene Dichlorodifluoromethane Dichloromethane Ethylbenzene Ethylene Dibromide Formaldehyde Freon 113 Freon 114 Hexachlorobutadiene Isoprene M,P-Xylene Methyl Chloroform Methyl Ethyl Ketone Methyl Tert-Butyl Ether M-Xylene O-Xylene PSAZ Number of cases 55 55 55 55 55 55 55 55 55 55 55 55 55 49 55 55 55 55 55 55 55 55 55 55 55 55 55 55 55 49 55 55 55 55 55 55 55 55 55 55 PSAZ Mean (ppbv) 0.011 0.010 0.010 0.010 0.018 0.393 0.027 0.013 0.010 0.161 0.207 0.027 0.185 1.688 0.758 0.010 0.012 0.013 0.095 0.012 0.017 0.117 0.448 0.010 0.015 0.460 0.231 0.359 0.012 4.424 0.067 0.015 0.011 0.433 0.964 0.028 1.372 0.011 0.658 0.360 PSAZ Median (ppbv) 0.010 0.010 0.010 0.010 0.010 0.304 0.010 0.010 0.010 0.153 0.149 0.010 0.174 1.700 0.576 0.010 0.010 0.010 0.100 0.010 0.010 0.105 0.454 0.010 0.010 0.467 0.171 0.246 0.010 4.300 0.067 0.014 0.010 0.326 0.662 0.028 1.276 0.010 0.446 0.240 E-4 PSAZ PSAZ Minimum Maximum 0.010 0.010 0.010 0.010 0.009 0.105 0.010 0.009 0.010 0.085 0.048 0.010 0.076 0.700 0.176 0.010 0.001 0.010 0.038 0.010 0.010 0.019 0.268 0.010 0.010 0.204 0.038 0.094 0.010 2.200 0.048 0.003 0.010 0.073 0.230 0.010 0.050 0.010 0.142 0.104 0.023 0.026 0.010 0.010 0.120 0.948 0.076 0.032 0.010 0.262 0.647 0.090 0.426 3.100 2.011 0.010 0.033 0.032 0.114 0.026 0.047 0.329 0.636 0.010 0.103 0.623 0.669 0.960 0.040 7.600 0.086 0.020 0.054 1.069 2.486 0.046 3.274 0.067 1.779 0.881 PSAZ Standard Dev 0.003 0.002 0.000 0.000 0.010 0.221 0.022 0.006 0.000 0.074 0.089 0.024 0.152 0.581 0.459 0.000 0.007 0.005 0.015 0.004 0.018 0.071 0.081 0.000 0.003 0.424 0.150 0.233 0.007 1.170 0.010 0.028 0.007 0.297 0.602 0.007 1.234 0.008 0.084 0.181 PSAZ Average MDL 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.10 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.10 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 Table E2. Summary statistics for JLG Supersite in 2005. Page 2 of 2 SITE Species P-Ethyltoluene P-Xylene Styrene Tetrachloroethene Toluene Trans-1,2Dichloroethylene Trans-1,3Dichloropropylene Trichloroethene Trichlorofluoromethane Vinyl Chloride PSAZ Number of cases 55 55 55 55 55 PSAZ Mean (ppbv) 0.179 0.306 0.174 0.205 1.845 PSAZ Median (ppbv) 0.169 0.219 0.143 0.155 1.174 PSAZ PSAZ 0.385 0.721 0.520 0.782 4.716 PSAZ Standard Dev 0.173 0.033 0.047 0.082 0.525 PSAZ Average MDL 0.02 0.02 0.02 0.02 0.02 Minimum Maximum 0.010 0.088 0.038 0.025 0.270 55 0.010 0.010 0.010 0.010 0.000 0.02 55 0.016 0.010 0.010 0.088 0.017 0.02 55 55 55 0.033 0.295 0.010 0.028 0.290 0.010 0.010 0.228 0.010 0.093 0.383 0.010 0.019 0.220 0.000 0.02 0.02 0.02 E-5 Table E3. Summary statistics for Queen Valley in 2005. Page 1 of 2 SITE Species 1,1,2,2-Tetrachloroethane 1,1,2-Trichloroethane 1,1-Dichloroethane 1,1-Dichloroethene 1,2,4-Trichlorobenzene 1,2,4-Trimethylbenzene 1,2-Dichlorobenzene 1,2-Dichloroethane 1,2-Dichloropropane 1,3,5-Trimethylbenzene 1,3-Butadiene 1,3-Dichlorobenzene 1,4-Dichlorobenzene Benzene Benzyl Chloride Bromoform Bromomethane Carbon Tetrachloride Chlorobenzene Chloroethane Chloroform Chloromethane Cis-1,2-Dichloroethene Cis-1,3-Dichloropropylene Dichlorodifluoromethane Dichloromethane Ethylbenzene Ethylene Dibromide Freon 113 Freon 114 Hexachlorobutadiene Isoprene M,P-Xylene Methyl Chloroform Methyl Ethyl Ketone Methyl Tert-Butyl Ether M-Xylene O-Xylene P-Ethyltoluene P-Xylene Styrene QVAZ Number of cases 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 30 QVAZ Mean (ppbv) 0.011 0.011 0.010 0.010 0.027 0.078 0.019 0.012 0.011 0.032 0.014 0.015 0.037 0.115 0.010 0.012 0.012 0.094 0.012 0.036 0.010 0.412 0.010 0.010 0.431 0.034 0.077 0.012 0.068 0.015 0.010 0.116 0.182 0.019 1.058 0.012 0.101 0.084 0.014 0.081 0.024 QVAZ Median (ppbv) 0.010 0.010 0.010 0.010 0.010 0.084 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.112 0.010 0.010 0.010 0.098 0.010 0.038 0.010 0.425 0.010 0.010 0.437 0.035 0.076 0.010 0.068 0.015 0.010 0.106 0.181 0.023 1.032 0.010 0.105 0.086 0.010 0.076 0.010 E-6 QVAZ QVAZ Minimum Maximum 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.010 0.084 0.010 0.010 0.010 0.054 0.010 0.010 0.001 0.273 0.010 0.010 0.049 0.010 0.010 0.010 0.047 0.010 0.010 0.010 0.020 0.004 0.050 0.010 0.010 0.010 0.010 0.010 0.002 0.020 0.022 0.010 0.010 0.140 0.151 0.057 0.030 0.027 0.131 0.040 0.059 0.092 0.174 0.010 0.030 0.020 0.121 0.022 0.080 0.023 0.550 0.010 0.010 0.545 0.067 0.100 0.038 0.086 0.021 0.010 0.414 0.241 0.028 4.070 0.056 0.158 0.125 0.073 0.112 0.118 QVAZ Standard Dev 0.002 0.003 0.000 0.000 0.037 0.037 0.016 0.005 0.003 0.042 0.010 0.013 0.033 0.018 0.000 0.005 0.003 0.015 0.003 0.019 0.006 0.070 0.000 0.000 0.101 0.015 0.020 0.007 0.011 0.003 0.000 0.110 0.044 0.007 0.814 0.008 0.022 0.034 0.015 0.027 0.034 QVAZ Average MDL 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 Table E3. Summary statistics for Queen Valley in 2005. Page 2 of 2 SITE Species Tetrachloroethene Toluene Trans-1,2Dichloroethylene Trans-1,3Dichloropropylene Trichloroethene Trichlorofluoromethane Vinyl Chloride QVAZ Number of cases 30 30 QVAZ Mean (ppbv) 0.026 0.166 QVAZ Median (ppbv) 0.024 0.151 QVAZ QVAZ 0.076 0.391 QVAZ Standard Dev 0.014 0.065 QVAZ Average MDL 0.02 0.02 Minimum Maximum 0.010 0.096 30 0.010 0.010 0.010 0.010 0.000 0.02 30 0.010 0.010 0.010 0.010 0.000 0.02 30 30 30 0.017 0.258 0.010 0.010 0.265 0.010 0.010 0.212 0.010 0.041 0.324 0.010 0.010 0.024 0.000 0.02 0.02 0.02 E-7 Table E4. Summary statistics for St. Johns in 2005. SITE Species 1,1-Dichloroethene 1,2,4Trimethylbenzene 1,2-Dichloroethane 1,2-Dichloropropane 1,3,5Trimethylbenzene 1,3-Butadiene Benzene Bromoethane Carbon tetrachloride Chloroform Dichloromethane Ethylbenzene Hexachlorobutadiene M,P-Xylene O-Xylene Styrene Tetrachloroethene Toluene Trichloroethene Vinyl chloride SJAZ Number of cases 58 SJAZ Mean (ppbv) 0.016 SJAZ Median (ppbv) 0.011 SJAZ SJAZ 0.113 SJAZ Standard Dev 0.017 SJAZ Average MDL 0.02 Minimum Maximum 0.005 58 0.088 0.064 0.006 0.618 0.105 0.02 58 58 0.019 0.016 0.020 0.016 0.010 0.008 0.076 0.103 0.010 0.013 0.03 0.04 58 0.054 0.032 0.008 0.419 0.071 0.03 58 58 58 58 58 58 58 58 58 58 58 58 58 58 58 0.058 0.185 0.034 0.087 0.022 0.072 0.082 0.205 0.196 0.081 0.079 0.050 0.646 0.032 0.012 0.039 0.156 0.013 0.086 0.018 0.056 0.063 0.148 0.157 0.068 0.041 0.028 0.525 0.026 0.008 0.035 0.062 0.005 0.012 0.006 0.018 0.012 0.130 0.018 0.013 0.009 0.013 0.088 0.013 0.004 0.380 0.479 0.389 0.180 0.072 0.396 0.305 0.660 1.030 0.362 0.489 0.322 2.484 0.191 0.108 0.057 0.092 0.060 0.029 0.012 0.069 0.071 0.114 0.178 0.069 0.102 0.065 0.488 0.033 0.016 0.09 0.02 0.02 0.03 0.02 0.03 0.03 0.41 0.05 0.03 0.06 0.05 0.03 0.05 0.02 E-8 Table E5. Summary statistics for Salt River (Senior Center) in 2005. SITE Species 1,1-Dichloroethene 1,2,4Trimethylbenzene 1,2-Dichloroethane 1,2-Dichloropropane 1,3,5Trimethylbenzene 1,3-Butadiene Benzene Bromoethane Carbon tetrachloride Chloroform Dichloromethane Ethylbenzene Hexachlorobutadiene M,P-Xylene O-Xylene Styrene Tetrachloroethene Toluene Trichloroethene Vinyl chloride SRAZ Number of cases 60 SRAZ Mean (ppbv) 0.013 SRAZ Median (ppbv) 0.010 SRAZ SRAZ 0.051 SRAZ Standard Dev 0.008 SRAZ Average MDL 0.02 Minimum Maximum 0.005 60 0.139 0.101 0.006 0.977 0.170 0.02 60 60 0.016 0.021 0.016 0.020 0.008 0.010 0.040 0.082 0.007 0.012 0.03 0.04 60 0.067 0.046 0.008 0.350 0.074 0.03 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 0.064 0.500 0.021 0.088 0.070 0.128 0.158 0.173 0.409 0.177 0.446 0.109 1.858 0.033 0.008 0.035 0.212 0.013 0.091 0.063 0.095 0.099 0.130 0.278 0.121 0.411 0.033 0.918 0.026 0.008 0.035 0.087 0.005 0.023 0.015 0.018 0.018 0.130 0.051 0.020 0.009 0.013 0.144 0.013 0.004 0.450 7.027 0.124 0.159 0.186 0.778 1.322 0.680 3.199 1.382 1.552 2.317 16.156 0.121 0.020 0.069 1.175 0.021 0.030 0.039 0.136 0.220 0.088 0.450 0.198 0.293 0.325 2.790 0.020 0.004 0.08 0.02 0.02 0.04 0.02 0.03 0.03 0.35 0.05 0.04 0.06 0.05 0.04 0.05 0.02 E-9 Table E6. Summary statistics for South Phoenix in 2005. SITE Species 1,1-Dichloroethene 1,2,4Trimethylbenzene 1,2-Dichloroethane 1,2-Dichloropropane 1,3,5Trimethylbenzene 1,3-Butadiene Acetaldehyde Benzene Bromoethane Carbon tetrachloride Chloroform Dichloromethane Ethylbenzene Formaldehyde Hexachlorobutadiene M,P-Xylene O-Xylene Styrene Tetrachloroethene Toluene Trichloroethene Vinyl chloride SPAZ Number of cases 59 SPAZ Mean (ppbv) 0.010 SPAZ Median (ppbv) 0.010 SPAZ SPAZ 0.027 SPAZ Standard Dev 0.004 SPAZ Average MDL 0.02 Minimum Maximum 0.005 59 0.238 0.131 0.006 1.644 0.303 0.02 59 59 0.015 0.019 0.016 0.020 0.008 0.010 0.035 0.044 0.006 0.007 0.03 0.04 59 0.094 0.072 0.008 0.639 0.101 0.03 59 52 59 59 59 59 59 59 52 59 59 59 59 59 59 59 59 0.280 1.695 0.706 0.025 0.093 0.063 0.179 0.248 3.310 0.226 0.771 0.268 0.090 0.189 1.763 0.039 0.008 0.160 1.500 0.619 0.013 0.081 0.041 0.137 0.184 3.350 0.145 0.618 0.201 0.070 0.057 1.351 0.026 0.008 0.035 0.020 0.021 0.005 0.009 0.006 0.024 0.010 0.015 0.130 0.018 0.010 0.009 0.013 0.021 0.013 0.004 1.640 8.200 5.158 0.279 0.856 0.979 0.833 0.663 5.600 0.660 2.030 0.736 0.575 4.913 7.295 0.582 0.018 0.332 1.093 0.690 0.039 0.106 0.127 0.145 0.177 1.106 0.170 0.542 0.194 0.102 0.647 1.428 0.074 0.003 0.10 0.10 0.02 0.02 0.03 0.02 0.03 0.03 0.10 0.44 0.05 0.03 0.05 0.05 0.03 0.05 0.02 E-10 Table E7. Summary statistics for West Phoenix in 2005. SITE Species 1,1-Dichloroethene 1,2,4Trimethylbenzene 1,2-Dichloroethane 1,2-Dichloropropane 1,3,5Trimethylbenzene 1,3-Butadiene Benzene Bromoethane Carbon tetrachloride Chloroform Dichloromethane Ethylbenzene Hexachlorobutadiene M,P-Xylene O-Xylene Styrene Tetrachloroethene Toluene Trichloroethene Vinyl chloride WPAZ Number of cases 59 WPAZ Mean (ppbv) 0.015 WPAZ Median (ppbv) 0.010 WPAZ WPAZ 0.176 WPAZ Standard Dev 0.023 WPAZ Average MDL 0.03 Minimum Maximum 0.005 59 0.292 0.215 0.006 3.069 0.434 0.04 59 59 0.023 0.026 0.016 0.020 0.008 0.010 0.281 0.351 0.040 0.044 0.05 0.06 59 0.131 0.092 0.008 1.357 0.187 0.05 59 58 59 59 59 59 59 59 59 59 59 59 59 59 59 0.311 0.736 0.026 0.083 0.067 0.290 0.528 0.265 1.080 0.374 0.186 0.135 3.306 0.075 0.011 0.185 0.600 0.016 0.076 0.050 0.223 0.205 0.130 0.634 0.237 0.084 0.066 1.860 0.028 0.008 0.035 0.440 0.005 0.013 0.006 0.018 0.014 0.130 0.018 0.014 0.009 0.013 0.023 0.013 0.004 1.270 2.419 0.176 0.316 0.346 1.676 13.319 1.675 7.986 1.813 1.005 2.012 31.936 2.127 0.140 0.302 0.537 0.030 0.043 0.056 0.265 1.717 0.287 1.239 0.354 0.239 0.281 5.777 0.278 0.018 0.11 0.03 0.03 0.05 0.03 0.04 0.05 0.49 0.07 0.05 0.08 0.08 0.05 0.08 0.03 E-11