ANALYSIS OF TEMPORAL AND SPATIAL DICHOTOMOUS PARTICULATE MATTER (PM) AIR SAMPLES IN THE EL PASO-CD.

JUAREZ AIR QUALITY BASIN PROJECT NUMBER: D-2
WEN-WHAI LI, UNIVERSITY OF TEXAS AT EL PASO RUBEN ORQUIZ, UNIVERSITY OF TEXAS AT EL PASO NICHOLAS E. PINGITORE, UNIVERSITY OF TEXAS AT EL PASO JORGE GARDEA-TORRESDEY, UNIVERSITY OF TEXAS AT EL PASO JOSE H. GARCIA, UNIVERSITY OF TEXAS AT EL PASO TANIA T. ESPINO, UNIVERSITY OF TEXAS AT EL PASO JUDITH CHOW, UNIVERSITY OF TEXAS AT EL PASO JOHN G. WATSON, DESERT RESEARCH INSTITUTE
NARRARATIVE SUMMARY This paper presents and discusses the results obtained from the gravimetric and chemical analyses of the 24-hr-average dichotomous samples collected from five sites in the El Paso-Cd. Juárez air quality basin between August 1999 and March 2000. Gravimetric weighing was performed to determine the temporal and spatial variations of PMfine (particulate matter less than 2.5 µm in diameter, or “PM2.5”) and PMcoarse (particulate matter less than 10 µm but greater than 2.5 µm in diameter, or abbreviated “PM10”) mass concentrations. The results indicate that nearly 80% of the PM10 (i.e., PMfine + PMcoarse) concentration is composed of coarse particulate. Concurrent measurements of hourly PM concentrations, wind speed, and temperature showed strong diurnal patterns of the regional PM pollution. Results of XRF elemental analyses were compared to similar but limited studies performed by the Texas Natural Resource Conservation Commission (TNRCC) in 1990 and 1997. Major elements - Al, Si, Ca, Na, K, Fe, and Ti accounted for 35% of the total concentrations in the PMcoarse fraction, indicating that geologic sources in the area are the prevailing PM sources. Levels of heavy metals, mainly considered as products of anthropogenic activities, have decreased significantly from those observed in 1990 and 1997.

ANALYSIS OF TEMPORAL AND SPATIAL DICHOTOMOUS PARTICULATE MATTER (PM) AIR SAMPLES IN THE EL PASO-CD. JUAREZ AIR QUALITY BASIN PROJECT NUMBER: D-2
WEN-WHAI LI, UNIVERSITY OF TEXAS AT EL PASO RUBEN ORQUIZ, UNIVERSITY OF TEXAS AT EL PASO NICHOLAS E. PINGITORE, UNIVERSITY OF TEXAS AT EL PASO JORGE GARDEA-TORRESDEY, UNIVERSITY OF TEXAS AT EL PASO JOSE H. GARCIA, UNIVERSITY OF TEXAS AT EL PASO TANIA T. ESPINO, UNIVERSITY OF TEXAS AT EL PASO JUDITH CHOW, UNIVERSITY OF TEXAS AT EL PASO JOHN G. WATSON, DESERT RESEARCH INSTITUTE
INTRODUCTION The Paso del Norte (PdN) air quality basin contains three cities: El Paso, Texas; Sunland Park, New Mexico; and Ciudad Juárez, Chihuahua, Mexico. Ambient PM10 concentrations in the region frequently exceed the national ambient air quality standards (NAAQS) of both countries. PM in the region derives from geologic sources, vehicle exhaust, residential cooking and heating, and other unidentified sources. The arid weather, occasional high winds, frequent stagnations, shallow nighttime and morning mixing depths, and complex topography preclude simple explanations for excessive PM10 levels. Air quality in El Paso has improved gradually since 1990, as shown in Figure 1. Although El Paso is still classified as nonattainment for ozone, CO, and PM10, reductions in all three criteria pollutants have been reported by the Texas Natural Resource Conservation Commission (TNRCC)1-3. Ozone and CO concentrations have decreased to levels close to or below their respective NAAQS. The annual average ambient PM10 level decreased from the peak of 67 µ g/m3 in 1989 to 37 µg/m3 in 1993; but increased to 55 µg/m3 from 1993 to 19972. A brief modeling feasibility study in December 1990 collected 12-hr (day and night) dichotomous samples at five sites in El Paso for gravimetric, elemental, and carbon analyses 4. During that study, PM10 exceeded the 150 µg/m3 24-hr standard on 15 out of 31 sampled days. Substantial spatial variation of PM10 during air pollution episodes was observed and nighttime concentrations were reported to be greater than daytime concentrations. Geologic material accounted for most of the mass in PMcoarse. Trace metal (Cr, Cu, As, Pb, Cd) concentrations were higher in PM2.5 than in PMcoarse. Surprisingly, the amount of chlorine present in El Paso air during 1990 was also higher than chlorine found in Texas coastal cities. A recent study5 in downtown El Paso showed that geologic material ([AlO+AlO2] average + SiO2 + [FeO+FeO2] average) accounted for 22% of the mass in PM2.5, while trace elements (Sum of XRF species – Al + Si + Ca + Fe + S + 2

Cl) accounted for 2% of the total mass in PM2.5. In addition, unexplained chlorine concentrations continued to be higher in El Paso than in any city in Texas during the study period. The recurring increase of PM10 concentration in recent years and the temporal and spatial characteristics of PM pollution in the air basin are not understood. To improve understanding, a PM air-monitoring program, the UTEP (University of Texas at El Paso) Study, was conducted in 1999/2000 and PM2.5 and PMcoarse samples were collected at several locations in the air basin. The objectives were to determine: 1) the temporal and spatial variations of PM2.5 and PMcoarse; 2) relationships between PM2.5 and PMcoarse; and 3) the elemental constituents of PM2.5 and PMcoarse. THE AIR MONITORING PROGRAM Site Description The PM monitoring program began August 1, 1999 and ended July 31, 2000. During this period, 24-hr dichotomous samples were collected on alternate days at two El Paso sites: Chamizal National Park (Chamizal) and Sun Metro Bus Terminal (Sun Metro), as shown in Figure 2. Only air samples collected before March 7, 2000 are discussed in this paper. Three additional sites in Cd. Juarez (Club 20-30, Advanced Transformer, and Mission) were added to the sampling program during the winter months (January 3 – March 7, 2000). Figure 2 shows the Paso del Norte air basin and the locations of the five monitoring sites. In the U.S., the Chamizal site is located in south El Paso, west of the Cordova International Bridge, in a mixed residential, semi-industrial area. It is part of the USEPA’s State and Local Air Monitoring Stations (SLAMS) and Photochemical Air Monitoring Stations (PAMS) networks where daily, 24-hr Federal Reference Method (FRM) PM2.5 samples are collected, and hourly PM10 and meteorological data are continuously recorded. The Sun Metro site, an urban commercial/industrial site about 4 km from Chamizal, is located in southwest El Paso just north of the Rio Grande and south of Interstate Highway I-10. The site is collocated with a TNRCC Continuous Air Monitoring Site (CAMS) where hourly PM10 and meteorological data are continuously recorded. In Mexico, the Club 20-30 site is located in a residential area east of downtown Cd. Juárez. The Advanced Transformer site is located near maquiladoras that make electronic components and automotive parts; brick kilns operate to the south. These two sites are co-managed by the Environmental Department of Cd. Juárez and the El Paso City and County Health and Environmental District. The Mission site is located in the Juárez Mountains foothills southwest of the air basin. The site is surrounded by unpaved roads and is adjacent to a cement factory; it represents a typical residential location in the outskirts of Cd. Juárez. These locations represent different activities in the air basin and supplement the study measurements with FRM PM2.5, hourly PM10, and meteorological data. Sample Collection, Handling and Processing Two dichotomous air samplers6-7 were placed at each of the two U.S. sites, where one sampler was operated every other day to collect 24-hr air samples and the other was operated selectively 3

for collocated samples. Only one dichotomous sampler was operated at each of the three Mexico sites. Twenty-four-hour samples were collected from 0001 to 2359 MST on 37-mm-diameter ringed Teflon filters (Gelman Science Inc., ID No. R2PJ037) at an actual (not adjusted to STP) flow rate of 1 m3/hr. The filter has high particle collection efficiency, 99%, measured using the DOP test with a 0.3-µm particle at the sampler’s operating face velocity.8 Quality control was managed by following EPA guidelines and procedures for particulate matter monitoring9 and gravimetric weighing10. A mini-Buck bubble calibrator (Model M-30), a primary standard calibration device traceable to NIST, was used to calibrate the rotameters on the dichotomous samplers11. Collocated samples were collected at the El Paso sites for every ten samples. Analysis of Mass Concentrations Filters were conditioned at 25 °C ± 5°C and 30% ± 5% RH for 24 hours, pre-weighed, and stored in petri dishes for less than 30 days prior to sampling. Loaded filters were removed from the field and transported to the laboratory at the University of Texas at El Paso (UTEP) for gravimetric analysis with a CAHN model C-33 microbalance (± 1 µg sensitivity)12 after conditioning. Mass concentrations were reported as micrograms of PM per cubic meter of air (µ g/m3), adjusted for the 10% of fine mass in the coarse mass, due to the use of virtual impactor in the dichotomous samplers, and to EPA’s standard conditions of 298 K and 760 mm Hg. The additional approximately 13% adjustment for EPA standard conditions is required for determining compliance with the federal PM10 standard, but not for compliance with the federal PM2.5 standard. PM2.5 and PMcoarse use STP sample volumes for this study. X-ray Fluorescence Elemental Analysis PM2.5 and PMcoarse Teflon filters were analyzed by x-ray fluorescence (XRF) analysis for 38 elements (Al, Si, P, S, Cl, K, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, As, Se, Br, Rb, Sr, Y, Zr, Mo, Pd, Ag, Cd, In, Sn, Sb, Ba, La, Au, Hg, Tl, Pb, and U). Calibration standards, sensitivity factors for each excitation condition, quality control standards and procedures, and detailed laboratory methods and operation procedures are kept the same as that used at the Desert Research Institute.13-14

RESULTS AND DISCUSSION
Mass Concentrations Collocated Samples. Figure 3 shows superior collocated precision of ±1% for both PM2.5 and PMcoarse. Regression statistics for samples collected at the Chamizal and Sun Metro sites show high correlations (r2 = 0.99), near-unity slope (1.03), and low intercepts (<3 µg/m3). Comparison to TNRCC’s Beta Attenuation Monitor. Figures 4 shows substantially poorer comparisons between the dichotomous sampler PM10 and that derived from beta attenuation monitors. On average, beta attenuation monitors reported lower PM10 than the dichotomous samplers by 20 to 40%. The difference became more pronounced at high concentrations.

4

Temporal Variation of PM Concentrations. Figure 5 presents the time series plot of PM10 and PM2.5 concentration for samples acquired on an every other day schedule at Chamizal and Sun Metro. Regardless of the PM10 concentrations, PM2.5 consists of only a small, but steady, fraction of PM10, indicating that PM2.5 in this area is rather independent of the temporal variations of regional sandstorms or mobile emissions. The average PM2.5 and PM10 concentrations of 22.5 µg/m3 and 109 µg/m3 at Sun Metro are considerably higher than Chamizal’s 11.0 µg/m 3 and 56.9 µg/m3, respectively. Table 1 summarizes the monthly average PM concentrations and temperatures obtained for the two El Paso sites. Both PM2.5 and PM10 increase during winter months, as temperature inversions increase in frequency and duration, and wood burning intensifies. Because of the nearby highway and unpaved residential area in Cd. Juárez, the PM concentrations at Sun Metro were expected to be higher than at Chamizal. PM 10 concentrations are dominated mainly by the PMcoarse fraction and the PM2.5 fraction is weakly associated with the PM10 concentrations at the two sites (Table 2). The average wintertime PM 2.5/PM10 ratio at Sun Metro, 0.28, is at a level similar to that observed at Chamizal of 0.21. These PM 2.5/PM10 ratios are significantly lower than 0.5, a value reported for a typical arid city15 where the ratio varies from 0.20 to 0.37 during dust storms and 0.33 to 0.75 during non-dust storm days16. Based on the observed PM ratios and PM10 concentrations, many of the high PM10 days would have been attributed to fugitive dust generated by high winds. However, concurrent wind measurements at the sites do not support such argument. For instance, two of the highest 24-hr PM10 concentrations measured at Chamizal were 275 µg/m3 on October 15th and 258 µg/m3 on November 18th. The respective PM2.5 concentrations for these two days at Chamizal were 11.2 µ g/m3 and 14.7 µg/m3, which resulted in PM2.5/PM10 ratios of 0.04 and 0.06. The average wind speeds for these two days, however, were not considered high, 4.3 m/s (with occasional gusts up to 8.9 m/s) and 2.8 m/s, respectively. Wind gusts reaching 8.9 m/s apparently had a significant impact on the elevated PM10 concentration for that day. The low-wind-high-PM and high-windhigh-PM phenomena have been observed at several locations and throughout the year in the air basin. In fact, Figure 6 shows that days with low or extremely high wind speeds (exceeding the wind erosion threshold wind speed of approximately 7 m/sec) tend to yield higher hourly PM 10 concentrations than days with moderate wind speeds17. Additionally, on January 19th at Sun Metro, the average wind speed for the day was 3.2 m/s (with maximum wind gusts of up to 4.6 m/s) and the PM10 and PM2.5 concentrations were 203 µg/m3 and 43.1 µg/m3. On February 12th, the PM10 and PM2.5 concentrations were 188 µg/m3 and 24.8 µg/m3 with average wind speed of 5.1 m/s and occasional wind gusts up to 10.3 m/s. When wind gusts reach levels of 13.1 m/s, as on February 24th, the PM10 and PM2.5 concentrations reached 337 µg/m3 and 37.2 µg/m3. The PM2.5/PM10 ratio was 0.21 for Jan. 19th, 0.13 for Feb. 12th, and reduced to 0.11 on Feb. 24th which is consistent with the changes in wind speed and maximum wind gusts. PM pollution in the area appears to be dominated by fugitive dust. A small proportion (Figure 6) results from wind erosion, with most cases caused by suspension of road or construction dust when wind speeds exceeded 7 m/s. Contributions to PM pollution by mobile emissions (primarily as PM 2.5) may be quite localized and do not significantly affect the overall PM10 concentrations in the air basin. Perhaps, PM2.5 in the area is dominated by re-suspension of urban dust and other meteorological conditions that are likely to trap PM in the air basin. Using hourly Beta Gauge PM10 and TEOM PM2.5 data18, 19, diurnal averages for the study period were plotted for both sites. At both locations, PM peaked during two distinct time intervals. The 5

first PM (both fine and coarse fractions) peak was during the morning hours when ground-based inversions occurred and traffic began (Figures 7 and 8). The second PM peak occurred in the evening when radiation inversions started to form and wood burning and home cooking prevailed in the air basin. Similar diurnal variations in PM 10 were observed in southern California. 20 Pronounced morning and evening peaks in PM10 were observed at urban and rural sites. However, the most pronounced peak was observed at a suburban location where shifts in winds and atmospheric pressure were considered the major causes of the peak. Characterization of the daily peaks in our air basin would help determine the localized PM emission sources and, consequently, reduce the PM pollution. Spatial Variation of PM Concentrations. Figure 9a contains the temporal variation of 24-hr PM10 and PM2.5 at Club 20-30. The PM2.5 concentrations visibly followed the PM10 trend. The average PM2.5/PM10 ratio was 0.36, with a S.D. ( ) of 0.11, obviously affected by the increased mobile emissions in downtown Cd. Juárez. At the Mission site, it is expected that the PM10 concentration be high and mostly made up of PMcoarse. Indeed, Table 2 shows that the average PM coarse concentration at this site was 142 µg/m3, a value much higher than that monitored in El Paso or downtown Cd. Juárez. The high PM concentrations could be attributed to a cement factory located in the vicinity of the site and the large number of wood stoves, unpaved roads, and heaters in the area. Figure 9b shows the temporal variation at Mission during the winter months. As expected, the PM2.5/PM10 ratio, 0.16 (with a value of 0.06), is significantly lower than at other sites and is a good indication of the dominance of PM pollution by geologic sources. Figure 9c contains the winter study temporal variation at Advanced Transformer. The average PM2.5 concentration of 50.9 µg/m3 and PM10 concentration of 197 µg/m3 are the highest of all sites (Table 2), reflecting the unique emission sources (brick kilns, automobiles, unpaved roads, and industrial sources) in the immediate vicinity of the site. For the Juárez sites, PM2.5 correlates moderately to PM10, as seen in Table 2, indicating that anthropogenic emissions and fugitive dust are more pronounced in Cd. Juárez than in El Paso. Elemental Analysis A total of 149 filters collected at the five sites were analyzed using XRF for rapid evaluation of the elemental compositions of PM. XRF analysis was performed for samples collected at Chamizal and Sun Metro for the month of September 1999 and from December 2, 1999, to March 5, 2000. At the Cd. Juárez sites, filters were analyzed for samples collected from January 3, 2000, to March 7, 2000. Figures 10 and 11 show the ambient trace metal concentrations (Cu, Cr, As, Cd, and Pb) at the two El Paso sites. These elements were selected for their association with operations of local industrial sources. Trace metal concentrations were higher in PM coarse than in PM2.5 for the El Paso sites, but lower in PM2.5 than in PMcoarse for the Cd. Juárez sites (Figure 12). In general, arsenic (As), chromium (Cr), and lead (Pb) in either PM 2.5 or PMcoarse were lower in 1997 and 2000 than a decade ago at the two El Paso sites. Arsenic and chromium levels were consistently 6

low in the basin throughout the study period, which may reflect the closure of a local copper smelting operation (Table 3). Levels of lead and copper, although reduced, are still high in the air basin. Geologic elements (Al, Si, Ca, Fe) appear to dominate the coarse fraction of PM 10 (Figures 13 and 14). These elements are expected to be of higher concentrations as low PM2.5/PM10 ratios and high correlations between PMcoarse and PM10 were observed at all sites. Based on the results of the XRF analysis, geologic elements account for 35 % of PMcoarse and 12 % of PM2.5 concentrations. PM concentrations at the rural Cd. Juárez site (Figure 15) are particularly affected by the geologic elements, signifying the impacts of unpaved roads and surrounding desert on local air quality. Figure 15 also shows that sulfur concentrations are low, but similar in the fine and coarse fractions of PM. Potential emission sources in the region for sulfur are fuel combustion and re-entrainment of fallout from past smelting of sulfide-containing ores. Occasional high chlorine concentrations were detected in PM2.5, indicating the existence of possible anthropogenic sources of chlorine-containing substances or long range transport of sea salt from the Gulf of Mexico, as shown in Figure 15. Chlorine levels at Sun Metro and Advance Transformer appeared to be higher than in other areas in the air basin. Nevertheless, the level has decreased significantly from what was observed in 1990, but remained at approximately the same level as reported in TNRCC’s 1997 statewide PM2.5 study (Table 3). Table 3 compares elemental compositions of PM2.5 observed in this study to those observed in 1990 and 1997. One notes that the sampling periods and sampling locations for the 1990 and 1997 studies are different from those for the current study. Although the comparison may include uncertainties, it provides the best available information of historical PM data for the air basin. Almost all elemental compositions decreased from 1990 to 2000 at both El Paso sites. PM 2.5 lead concentrations are significantly lower today than in the previous studies. Reasons for the lower lead concentration lie in the elimination of lead from gasoline (eliminated recently in Cd. Juárez) and the shutdown of a major smelting operation in the city. Arsenic and other smelter emissions have experienced the same decrease in concentration as lead. PM2.5 chlorine and sulfur concentrations have also experienced decreases in concentration. The dominance of geologic elements in PM2.5 decreased over the past decade. For instance, aluminum concentrations in El Paso decreased from an average of 0.47 µg/m3 in 1990, to 0.24 µ g/m3 in 1997 and to 0.27 µg/m3 in 2000 (average of the two El Paso sites). PM2.5 calcium increased from 1.27 µg/m3 in 1990, to 1.41 µg/m3 in 1997, but decreased to 1.04 µg/m3 in 2000 while PM2.5 silicon increased from 0.76 µg/m3 in 1990, to 0.93 µg/m3 in 1997 and in 2000. In the meantime, the mass concentration of PM2.5 decreased from 32.8 and 55.6 µg/m3 in 1990 at Chamizal and Sun Metro to 11.0 µg/m3 and 22.5 µg/m3 in 2000, respectively. The fraction of geologic material in PM2.5 (based on the sum of Al, Si, Ca, and Fe), however, actually increased from 6 % to 12 %. One notices from Table 3 that chlorine concentration decreased from 1990 to 2000 in PM2.5, but somehow, increased in PMcoarse. The seasonal and spatial variations in chlorine concentrations require further investigation. 7

Table 4 compares the compositions of PMcoarse in the present study to those measured in 1990. As expected, elements associated with geologic sources, such as Al, Si, Ca, and Fe, are high in PMcoarse. Based on the sum of the mass of these four geologic elements and the mass of the PMcoarse derived from Table 1, geologic materials account for about 35 % of the mass for the coarse fraction of PM10. This confirms that geologic materials contribute more to PMcoarse than to PM2.5 and that carbon compounds, sulfates, and nitrates roughly account for the other portion of PMcoarse. As stated previously, coarse chlorine concentration was high at Chamizal but not at Sun Metro while fine chlorine concentration was low at Chamizal but high at Sun Metro. Elements associated with industrial emissions in the air basin, such as Cr, Cu, As, Pb and Cd, remained at low levels in PMcoarse, indicating that industrial emissions contribute more to PM2.5 than PMcoarse. Levels of trace metals in the air decreased significantly from 1990 to 2000. For instance, arsenic concentration in PM2.5 decreased by 57-fold (from 0.073 to 0.0013 µg/m3) and lead by 18-fold (from 0.24 to 0.014 µg/m3) in El Paso. The ratio of trace metal to PM2.5 concentration also improved. The ratio of copper to PM2.5 actually increased by about 40% between 1990 and 1997, but decreased to less than 40% in 2000. Table 5 shows that ratios for the dominant trace metals in the region (except those for chromium and copper) decreased less impressively from 1990 to 1997, but rather significantly (except that for chromium) to 2000 after a major smelter halted its operations in early 2000. This is indicative of improvement in both geologic and industrial emissions in the past decade and further reduction of trace metals from industrial emissions in the past year.

SUMMARY AND CONCLUSIONS
Our seven-month study of PM concentrations in El Paso shows that the average PM 2.5 and PM10 concentrations are 11 and 57 µg/m3 for Chamizal and 22 and 109 µg/m3 for Sun Metro, respectively. The PM10 concentration increases towards the suburban area of Cd. Juárez while PM2.5 peaks in areas of that city surrounded by brick kiln emissions and unpaved roads. The PM2.5/PM10 ratio for the Paso del Norte air basin varied from 0.16 to 0.36, compared to 0.38 in 1990 (two sites for two weeks) and 0.45 in 1997 (one site for eleven months determined with two different types of instrument). Downtown locations of the air basin yield higher ratios (0.36 in Cd. Juárez during 2000 and 0.45 in El Paso during 1997). The basin-wide average ratio is 0.24, a low value indicating that fugitive dust from geologic and anthropogenic sources (within and outside the air basin) dominates PM10 pollution in the region. This study showed that PM2.5 pollution is rather ubiquitous in the air basin regardless of the seasonal/spatial variations of PM10. PMcoarse dominates the PM10 mass concentration, and geologic sources are the major contributors to PMcoarse. The diurnal variation of PM concentrations at the Sun Metro site shows that both PM2.5 and PM10 concentrations peak in the morning and at night. Characterization of hourly PM chemical composition, both organic and elemental, will be extremely helpful in determining the sources of the high morning and nighttime pollution. Trace metal concentrations are lower today than historical values, and we are continuing to evaluate the causes of this reduction. Elements of geologic origin dominate the coarse fraction of PM10 and are persistent due to the abundance of unpaved roads and complex terrain. Further 8

investigation using source fingerprints and chemical, both organic and elemental, compositions of air samples could provide mitigation alternatives for controlling PM pollution in the El PasoCd. Juárez air-quality basin.

ACKNOWLEDGEMENTS
This study was supported by a grant provided by the Southwest Center for Environmental Research and Policy (SCERP). R. Orquiz was sponsored by a scholarship provided by the GE Faculty of the Future Program. Assistance received from Mr. R. Currey and Dr. N.J. Parks of CERM/UTEP; Mr. D. R. van Schoik of SCERP; Drs. H. Meuzelaar and A. Sarofim of U. of Utah; Mr. S. Kohl of DRI; Mr. J. Saenz, Mr. M. Munoz, Mr. A. Clouse and Mr. V. Valenzuela of TNRCC Region 6; and Mr. G. Tarin of the Environmental Department of Cd. Juárez is appreciated. This study would not have been completed without the help of the following people and program support: I. Becerril (SCERP), N. J. Castro and A. Martinez (UTEP MIE Undergraduate Research Program), K.C. Gallardo (EPA Undergraduate Research Program), L. Ledesma (EPA Undergraduate Research Program), J. Chianelli (SCERP), and J. Sanchez (SCERP).

REFERENCES
1. Air Pollution Trends in Texas, Nonattainment Areas. http://www.tnrcc.state.tx.us/cgibin/monops/elpaso/O.html (accessed December 1999). 2. Air Pollution Trends in Texas. http://www.tnrcc.state.tx.us/air/monops/e/10.html (accessed December 1999). 3. Air Pollution Trends in Texas. http://www.tnrcc.state.tx.us/air/monops/e/oz.html (accessed December 1999). 4. Dattner, S. El Paso/Juárez 1990 PM10 Receptor Modeling Feasibility Study, TNRCC December 1994. 5. Price, J.H.; Dattner, S.L.; Lambeth, B.; Kamrath, J.; Aguirre, M., Jr.; McMullen, G.; Loos, K.; Crow, W.; Tropp, R.J.; and Chow, J.C. (1998). Preliminary results of early PM2.5 monitoring in Texas: Separating the impacts of transport and local contributions. In Proceedings, PM2.5: A Fine Particle Standard, Chow, J.C. and Koutrakis, P., Eds. Air & Waste Management Association, Pittsburgh, PA, pp. 191-202. 6. Evans, J.S.; Ryan, P.B. (1983). “Statistical uncertainties in aerosol mass concentrations measured by virtual impactors,” Aerosol Sci. Technol., 1983, 2:531-536. 7. Quality Assurance Document, Method Compendium: Field Standard Operating Procedure for the PM2.5 Performance Evaluation Program, U.S. Environmental Protection Agency: Research Triangle Park, NC., October 1998. 8. Lodge, J. P., Jr., (1989). Methods of Sampling Analysis. Lewis Publishing, Inc. 9. Reference Method for the Determination of Particulate Matter as PM 10 in the Atmosphere (Dichotomous Sampler Method); U.S. Environmental Protection Agency, Quality Assurance Handbook for Air Pollution Measurements Systems, Vol. II, Section 2.10 (1994). 10. Quality Assurance Guidance Document, Method Compendium: PM2.5 Mass Weighing Laboratory Standard Operating Procedures for the Evaluation Performance Program, U.S. Environmental Protection Agency: Research Triangle Park, NC, November 1998. 11. A.P. Buck, Inc.(1987), Mini-Buck Calibrator Instruction Manual. 9

12. Orion Research, Cahn Model C-33 Microbalance Instruction Manual, 1997. 13. Watson, J.G.; Chow, J.C.; Frazier, C.A. (1998), Elemental Analysis of Airborne Particles, Landsberger, S. and Creatchman, M., Ed.; Gordon and Breach Publishers, U.S.A., pp 67-96. 14. Chow, J. C., J. (1995) Air and Waste Manage. Ass., 48:320. 15. Haller, L.; Clairborn, C.; Larson, T.; Norris, G; Edgar, R.(1999) “Airborne Particulate Matter Size Distributions in an Arid Urban Area,” J. Air & Waste Manage. Assoc., Vol.49: 161-168. 16. Claiborn, C.S.; Finn, D.; Larson, T.; Koenig J.(2000), “Windblown Dust Contributes to High PM2.5 Concentrations,” J. Air & Waste Manage. Assoc., Vol. 50: 1440-1445. 17. Becerril, I, Li,; W.-W., Gallardo, K.G., (1999), “Implication of potential PM sources based on hourly meteorological and PM concentrations in El Paso,” Presented in the U.S.-Mexico Border: An Exchange of Ideas, SCERP Technical Conference, Las Cruces, New Mexico. 18. CAMS 41 Data by Month by Monitoring Site by Parameter. http://www.tnrcc.state.tx.us/cgibin/monops/monthly_summary?41 (accessed December 1999). 19. CAMS 40 Data by Month by Monitoring Site by Parameter. http://www.tnrcc.state.tx.us/cgibin/monops/monthly_summary?40 (accessed December 1999). 20. Dolislager, L.; Motallebi, N., (1999), “Characterization of Particulate Matter in California,” J. Air & Waste Manage. Assoc., Vol. 49: 45-56.

10

Table 1. Summary of 24-hr-average PM concentrations (µg/m3) and temperatures (°F) at the two El Paso sites.
Chamizal Month Aug-99 Sep-99 Oct-99 Nov-99 Dec-99 Jan-00 Feb-00 Average PM2.5 ±STD 8.6 ± 2.4 7.2 ± 3.6 11.2 ± 4.3 11.5 ± 4.6 11.6 ± 8.0 12.7 ± 7.5 12.5 ± 6.7 10.95 PM10 ± STD 43.5 ± 24.6 31.5 ± 18.4 73.8 ± 61.8 69.2 ± 58.2 37.3 ± 15.8 58.2 ± 26 75.1 ± 78.5 56.9 Temp. 84.5 78.5 68 60.9 47.3 53.4 57.5 PM2.5 ± STD N/C N/C 10.5 ± 2.7 22.6 ± 9.9 34.6 ± 28.8 18.2 ± 11 28.4 ± 17.5 21.5 ± 9.4 22.5 Sun Metro PM10 ± STD N/C N/C 47.8 ± 18.8 138.9 ± 76.9 155.8 ± 88.5 85.2 ± 73.3 106.4 ± 50.5 133 ± 106 109 Temp. N/C 77.1 66.6 59.3 45.9 51.6 55.8

N/C: No Samples Collected

Table 2. Summary of 24-hr-average winter month PM concentrations (µg/m3) at all sites.
UTEP Study Average Values (Jan. 3 through March 7, 2000) PM2.5 µg/m3 Avg ± S.D. Chamizal a Sun Metro
a

Site Name

PMcoarse µg/m3 Avg ± S.D. 57.7 ± 54.7 90.0 ± 71 36.0 ± 17.0 146 ± 68.5 142 ± 50.7

PM10 µg/m3 Avg ± S.D. 70.3 ± 57.9 113 ± 79.2 56.3 ± 26 197 ± 107 169 ± 58.2

PM2.5/PM10 Avg ± S.D. 0.22 ± 0.118 0.23 ± 0.099 0.36 ± 0.105 0.23 ± 0.112 0.16 ± 0.055

r2 for PM2.5 to PM10 0.21 0.28 0.73 0.56 0.49

r2 for PMcoarse to PM10 0.97 0.97 0.87 0.67 0.97

12.6 ± 6.5 23.1 ± 14.2 20.0 ± 11.5 50.9 ± 59.3 26.8 ± 11.6

Club 20-30 b Adv. Transformer b Mission b
a

El Paso, Texas, U.S.A. b Juárez, Chihuahua, Mexico

11

Table 3. Comparison of UTEP winter study PM2.5 elemental composition to those of the 1990 and the 1997 TNRCC studies.
Chamizal
ELEMENT

Sun Metro UTEP Study ng/m3 54.00 69.00 357.00 1260.00 2.41 341.00 625.00 217.00 1540.00 19.00 0.51 1.95 8.87 508.00 0.11 0.15 22.20 38.10 0.04 1.48 0.05 0.63 0.63 5.54 0.38 1.25 0.26 0.45 0.27 1.41 0.72 2.95 14.30 22.30 12.30 0.00 0.00 0.07 20.40 0.04 1990 TNRCC Study ng/m3 497.00 1070.00 1350.00 1740.00 365.00 1990.00 20.00 9.00 5.00 17.00 411.00 119.00 242.00 94.00 21.00 38.00 8.00 8.00 7.00 18.00 12.00 301.00 -

AdvT UTEP Study ng/m3 83.10 74.70 761.00 1440.00 0.57 492.00 742.00 381.00 3410.00 27.70 1.37 1.17 14.60 397.00 0.26 1.36 35.00 159.00 0.00 1.39 0.04 51.70 0.95 9.69 0.34 1.32 0.15 0.78 0.18 1.62 0.83 2.99 23.10 21.50 9.08 0.00 0.00 0.15 36.00 0.09

Club 20-30 UTEP Study ng/m3 53.80 42.60 218.00 726.00 1.84 514.00 349.00 183.00 1260.00 10.90 0.83 0.48 5.59 243.00 0.03 0.34 24.80 97.50 0.03 0.74 0.01 14.30 0.46 4.94 0.29 0.78 0.12 0.72 0.71 0.75 0.86 2.79 10.50 22.10 14.70 0.00 0.00 0.01 13.80 0.12

Mission UTEP Study ng/m3 66.30 73.20 285.00 1010.00 1.63 374.00 405.00 242.00 3060.00 17.70 1.35 0.58 10.60 291.00 0.14 0.43 37.60 72.50 0.00 1.21 0.06 15.60 0.70 7.40 0.24 1.01 0.16 0.98 0.00 0.76 0.00 3.30 15.40 18.40 11.90 0.00 0.00 0.06 39.10 0.08

PMfine AVG. OF 17 3/97 TO 3/98 ng/m3 18.00 53.00 243.00 928.00 0.10 594.00 362.00 191.00 1410.00 20.00 3.00 1.00 9.00 324.00 1.00 1.00 61.00 64.00 0.10 19.00 8.00 9.00 1.00 5.00 0.00 1.00 1.00 1.00 1.00 2.00 1.00 3.00 8.00 9.00 8.00 0.00 0.00 0.00 37.00 0.00

Na Mg Al Si P S Cl K Ca Ti V Cr Mn Fe Co Ni Cu Zn Ga As Se Br Rb Sr Y Zr Mo Pd Ag Cd In Sn Sb Ba La Au Hg Tl Pb U

UTEP Study ng/m3 63.90 40.90 190.00 593.00 1.52 409.00 89.20 128.00 542.00 7.91 0.54 0.44 4.49 231.00 0.10 0.15 14.70 24.10 0.00 1.08 0.04 7.16 0.32 2.59 0.16 0.65 0.13 0.25 0.08 0.26 0.10 0.76 2.27 17.10 9.65 0.00 0.00 0.02 6.52 0.02

1990 TNRCC Study ng/m3 440.00 440.00 1170.00 565.00 186.00 559.00 10.00 7.00 4.00 13.00 179.00 49.00 90.00 51.00 6.00 33.00 3.00 5.00 3.00 8.00 7.00 188.00 -

Table 4. Comparison of elemental PMcoarse between the 1990 study and the UTEP winter study.

12

CHAMIZAL ELEMENT UTEP STUDY ng/m3 126.00 179.00 2230.00 7490.00 14.40 249.00 396.00 665.00 5500.00 89.70 0.88 3.20 24.30 1190.00 0.85 0.62 17.90 29.50 0.03 0.73 0.02 1.18 3.01 20.10 1.01 4.80 0.81 0.54 0.82 0.91 0.36 1.82 1.50 45.60 7.03 0.00 0.00 0.11 9.29 0.12 1990 TNRCC Study ng/m3 934.00 4240.00 283.00 138.00 614.00 6670.00 100.00 6.00 12.00 36.00 1250.00 57.00 49.00 7.00 1.00 6.00 33.00 27.00 4.00 3.00 6.00 1.00 40.00 55.00 -

Na Mg Al Si P S Cl K Ca Ti V Cr Mn Fe Co Ni Cu Zn Ga As Se Br Rb Sr Y Zr Mo Pd Ag Cd In Sn Sb Ba La Au Hg Tl Pb U

SUN METRO UTEP Study ng/m3 91.50 254.00 3730.00 12700.00 37.90 335.00 215.00 1070.00 12600.00 150.00 1.05 10.30 49.00 2710.00 1.35 1.99 36.40 51.20 0.00 2.53 0.01 5.25 5.25 38.30 1.89 8.28 1.85 0.00 2.10 3.04 0.00 1.99 1.20 64.40 12.60 0.00 0.17 0.57 22.60 0.39

Advanced Trans. UTEP Study ng/m3 162.00 324.00 3990.00 14500.00 64.30 622.00 284.00 1400.00 30400.00 200.00 3.25 3.61 50.20 2280.00 2.54 3.13 13.00 70.10 0.00 0.98 0.13 4.08 6.98 78.20 2.06 10.70 2.33 2.41 0.00 0.11 3.22 0.34 3.00 87.30 3.25 0.00 0.21 0.57 11.50 0.30

Club 20-30 UTEP Study ng/m3 73.30 149.00 1920.00 6010.00 28.10 265.00 110.00 487.00 6920.00 71.50 1.32 1.80 17.40 852.00 0.22 0.73 9.72 68.40 0.00 0.44 0.00 1.79 2.12 21.40 0.82 3.42 0.42 0.00 0.24 0.00 0.32 0.74 1.67 31.60 11.60 0.00 0.00 0.00 5.61 0.14

Mission UTEP Study ng/m3 153.00 355.00 3500.00 12300.00 64.00 569.00 183.00 1080.00 31800.00 159.00 4.15 3.14 38.70 1930.00 0.82 2.62 16.30 49.10 0.04 0.60 0.01 3.08 5.16 74.20 1.97 8.20 2.39 7.09 0.00 0.00 1.90 1.61 4.46 57.50 3.29 0.00 0.00 0.55 16.20 0.60

13

Table 5. Fraction of trace metals in PM2.5.

1990 TNRCC Element Study

1997 Study

UTEP Winter Study (2000) El Paso Chamizal 0.005 0.110 0.009 0.003 0.046 Ciudad Juárez, Mx. Sun Advanced Club Mission Metro Transformer 20-30 0.011 0.098 0.009 0.006 0.078 0.001 0.068 0.003 0.003 0.070 0.001 0.124 0.004 0.003 0.122 0.002 0.104 0.002 0.003 0.055

Chami/Sun Central Metro El Paso Cr Cu As Cd Pb 0.001 0.19 0.164 0.015 0.552 0.004 0.27 0.086 0.008 0.163

14

Figure 1. Annual average PM10 and maximum 1-hr ozone and CO concentrations for El Paso.
Ozone 180 165 150 PM10 (µ g/m3) & O3 (ppb) 135 120 105 90 75 60 45 30 88 89 90 91 92 Year 93 94 95 96 97 PM 10 CO 15 14 13 12 CO (ppm) 11 10 9 8 7 6 5

Figure 2. Locations of air monitoring sites.

Cd. Juarez Mountains

15

Figure 3. Comparison of collocated PM2.5 and PMcoarse samples at Chamizal and Sun Metro for the period of August 1999 to February 2000.
PM(2.5) 160 PM(coarse) Linear (PM(2.5)) Linear (PM(coarse))

140

120

Sampler 2 (µg/m 3)

100

80 PM2.5 (Sampler 2)= 1.03PM2.5 (Sampler 1) - 0.54 60 r2 = 0.99 PMcoarse (Sampler 2) = 1.03PMcoarse (Sampler 1) - 2.41 r = 0.99
2

40

20

0 0 20 40 60 80 Sampler 1 (µg/m3) 100 120 140 160

Figure 4. Comparison of PM10 concentrations acquired by dichotomous and Beta Gauge samplers at Chamizal and Sun Metro for the period of Dec. 2, 1999 to March 7, 2000.

16

Chamizal 350

Sun Metro

Linear (Chamizal)

Linear (Sun Metro)

300 24-hour Beta Gauge PM 10 (µ g/m3)

250

Chamizal PM10 (beta) = 0.68 PM10 (dichot) + 13.02 r2 = 0.81

200

150

100

Sun Metro PM10 (beta) = 0.49PM10 (dichot) + 10.58 r2 = 0.71

50

0 0 50 100 150 200
3

250

300

350

Dichotomous Sampler PM10 (µg/m )

17

9/

PM µg/m3

07 /2 9 /9
100 150 200 250 300 350 50 0

PM µg/m3
9

18
100 150 200 250 300 350

1/

50

0

99

08 /1 2/ 99 99 9 9 6/ /9 /9 99 /2 /0 9 /2 3 /0 7/ /9 9 9 /2 1 /0 4 /9 /9 9 9 99 12 01 01 02 02 03 /3 0/ /1 3 /2 7 /1 /2 99 /0 0 /0 0/ 4/ /0 9 0 00 00 /0 0 /1 8 /0 2 /9 /1 6/ 08 09 09 10 10 11 11

9/

15

/9 9

9/ 2

9/

99

10

/1

3/

99

10 /2

7/

99

11 /1

0/

99

PM2.5

11 /2

4/

99

PM2.5

Figure 5b. Temporal PM variations at Sun Metro.

PM10

12 /

Date
12 12

Figure 5a. Temporal PM10 and PM2.5 variation at Chamizal.

Date

8/ 9

9

PM10

12

/2 2/

99

1/

5/ 0

0

1/ 1

9/

00

2/

2/

00

2/

16

/0 0

3/

1/

00

Figure 6. Relationship of hourly PM10, wind speed, and wind direction at Sun Metro for the period of September 98 through August 99.

Figure 7. PM10 diurnal variation at Sun Metro for the period of December 99 to February 00.
300 280 260 240 220 200 180 PM10 ug/m
3

160 140 120 100 80 60 40 20 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Hour Max Min 75th % 25th % Median

19

Figure 8. PM2.5 diurnal variation at Sun Metro for the period of December 99 to February 00.
140

120

100 PM2.5 ug/m
3

80

60

40 Max Min 75th % 25th % Median

20

0

0

1

2

3

4

5

6

7

8

9 10

11 Hour

12

13

14

15

16

17

18

19

20

21

22

23

Figure

9:

Temporal
PM10 ADV

variation

of

PM10

and

PM2.5
PM 2.5 CLUB

at

the

Cd.

Juárez

sites.

PM2.5 ADV

PM 2.5 M ISSION

PM10 MISSION

PM 10 CLUB

520 5205 520 520 20 800 410
1000 600 210

(c) Advanced Transformer

( g/m )

3

0 400

10

(b) Mission
200 220

and PM
2.5

PM

0

0

(a) Club 20-30
-200 100

-400 40

0
-600

1/7/00

1/14/00

1/21/00

1/28/00

2/4/00
D AT E

2/11/00

2/18/00

2/25/00

3/3/00

20

u Cb P r C s Ad C

21
PM2.5 (F) and PM coarse (C) in µ g/m3
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 PM2.5 (F) and PM coarse (C) in µg/m3

0.00
9/7/99 9/7/99 9/11/99 9/11/99 9/15/99 9/15/99 9/19/99 9/19/99 9/21/99 9/21/99 9/23/99 9/23/99 12/2/99 12/2/99 12/14/99 12/14/99 12/20/99 12/20/99 1/7/00 1/7/00 1/11/00 1/11/00 1/19/00 1/19/00 1/23/00 1/23/00 1/27/00 1/27/00 2/4/00 2/4/00 2/16/00 2/16/00 2/20/00 2/20/00
PbCu AC Cd s r

0.01 Cd As
Date

0.01 Cr

0.02

0.02

0.03

0.03

0.04

0.04

Cd As Cr Pb

Figure 11. Summary of trace metal concentrations at Chamizal.

Figure 10. Summary of temporal trace metal concentrations at Sun Metro.

C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F

C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F

Date Pb Cu

Cu

9/7/99 9/7/99 9/11/99 9/11/99 9/15/99 9/15/99 9/19/99 9/19/99 9/23/99 9/23/99 12/2/99 12/2/99 12/14/99 12/14/99 12/22/99 12/22/99 1/7/00 1/7/00 1/11/00 1/11/00 1/19/00 1/19/00 1/23/00 1/23/00 1/27/00 1/27/00 2/4/00 2/4/00 2/8/00 2/8/00 2/16/00 2/16/00 2/20/00 2/20/00 2/24/00 2/24/00

Figure 12. Average trace metal concentrations at all sites.
Chamizal Avg. Sun Metro Avg. Club 20-30 Avg. Mission Avg. Adv. T. Avg.

0.0480 0.0440 0.0400 0.0360 0.0320 µ g/m
3

0.0280 0.0240 0.0200 0.0160 0.0120 0.0080 0.0040 0.0000 PM2.5 PMcoarse PMcoarse PMcoarse PMcoarse PM2.5 PM2.5 PM2.5 PM2.5 PMcoarse Pb
. vg . . A vg vg. .T A A . dv n 0 vg A issio 0-3 o A g. 2 M b tr v lu e A C n M izal Su am h C

Cr

Cu

As

Cd

Figure 13. Summary of selected elemental concentrations at Sun Metro.
S Cl Fe Al Si Ca

20

PM2.5 (F) and PM coarse (C) in µg/m3

18 16 14 12 10 8 6 4 2 0 1/27/00 2/4/00 9/23/99 12/2/99 1/19/00 1/23/00 12/14/99 12/14/99 12/20/99 2/16/00 9/7/99 9/11/99 9/15/99 9/21/99 9/21/99 9/23/99 1/11/00 1/19/00 2/16/00 12/20/99 2/20/00 2/20/00 9/7/99 1/7/00 1/7/00 9/19/99 9/11/99 9/15/99 9/19/99 12/2/99 1/23/00 1/27/00 1/11/00 2/4/00
Si Ca F Al S Cl e

C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F

Date

22

PM2.5 (F) and PM coarse (C) in µ g/m3 10 12 0 2 4 6 8

. vg . . . A vg vg . T A A g. dv on 30 v A issi 20- ro A vg. M lub et l A C n M iza Su ham C

23
µ g/m
10 12 14 4 6 8
3

0

2

PM2.5

Figure 14. Summary of selected elemental concentrations at Chamizal.

C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F C F

Figure 15. Average element concentrations of geologic origin and Cl & S at all sites.

Al Chamizal Avg. Sun Metro Avg.

PMcoarse

PM2.5

Si

PMcoarse

PM2.5

S

S

Cl

PMcoarse

Fe

Date

PM2.5

Al

Club 20-30 Avg.

Cl
31.8

Si

PMcoarse

Ca

PM2.5

Mission Avg.

Ca Adv. T. Avg.
30.4

PMcoarse

PM2.5

Fe

PMcoarse

20.9

9/7/99 9/7/99 9/11/99 9/11/99 9/15/99 9/15/99 9/19/99 9/19/99 9/23/99 9/23/99 12/2/99 12/2/99 12/14/99 12/14/99 12/22/99 12/22/99 1/7/00 1/7/00 1/11/00 1/11/00 1/19/00 1/19/00 1/23/00 1/23/00 1/27/00 1/27/00 2/4/00 2/4/00 2/8/00 2/8/00 2/16/00 2/16/00 2/20/00 2/20/00 2/20/00 2/20/00 2/24/00 2/24/00
S Ca F Al i S Cl e