Thesis Paper

Modeling of PM2.5 and Health Effects in Seven California Counties

COH 692: Master of Public Health Capstone Project

National University

Dr. Gina M. Piane

August 27, 2011

1

Table of Contents

Abstract

Acknowledgement

Introduction

Literature Review

PM2.5 Studies in the East Coast and other 50 States

PM2.5 Studies in California

Methodology

Sampling Sites:

Table 1. Description of air pollution monitoring stations.

Sampling Method and Laboratory Chemical Analysis

Table 2. Samples collected by season for each county.

Quality Assurance/Quality Control (QA/QC):

PM2.5 Modeling:

Table 3. Pooled estimates of percent changes in daily mortality categories

Results

Characterization of PM2.5, Yearly-Seasonal Trends and Potential Contributing Sources

Fresno

Sacramento

Sacramento-site-1:Del Paso Manor

Sacramento-site-2-T Street

1

Table of Contents

San Diego County

El Cajon

Escondido

Los Angeles

Riverside

Kern County (Bakersfield)

Santa Clara County (San Jose)

Figure 1. Daily PM2.5 Mass Concentration .

Figure 2. Seasonal Averaged PM2.5 Mass Concentration

Figure 3. Chemical Composition of Daily PM2.5 Concentrations Health Effects

Figure 4a. The Number of Population in the three Major Mortality Categories

Figure 4b. The Percent Increase of the three Major Mortality Categories

Figure 4c. The Percent Increase of other factors under Mortality

Table 4a. The Percent Increase of all Mortality Categories in Bakersfield

Table 4b. The Percent Increase of all Mortality Categories in El Cajon

Table 4c. The Percent Increase of all Mortality Categories in Escondido

Table 4d. The Percent Increase of all Mortality Categories in Fresno

Table 4e. The Percent Increase of all Mortality Categories in Los Angeles

Table 4f. The Percent Increase of all Mortality Categories in Riverside

Table of Contents

Table 4g. The Percent Increase of all Mortality Categories in Sacramento Del Paso

Manor(Kern County).

Table 4h. The Percent Increase of all Mortality Categories in Sacramento T Street(Kern

County).

Table 4i. The Percent Increase of all Mortality Categories in San Jose (Santa Clara County).

Discussion

Limitations

Conclusion

References

Abstract

Mortality due to respiratory disease, cardiovascular disease, and diabetes has been shown

to have an association with an elevated fine particle concentration on a daily basis. This study

describes the characterization of the particulate matter (PM 2.5) and identifies the potential

contributing sources of PM2.5 in nine air sampling sites in seven California counties. The study

also uses an established mathematical model to analyze and compare the percent change and

increase in negative health effects under several mortality subcategories, which included

respiratory disease, cardiovascular disease, diabetes, age > 65 years, females, deaths out of the

hospital, and non-high school graduates. Particulate matter (PM2.5) composition data for this

study, collected from years 2000 to 2005 were retrieved from the California Air Resources

Board. The data sources provided raw data of the concentrations PM2.5 and other elements that

could potentially be or have made up of PM2.5 mass concentrations. The completed and

calculated results show the ranking of counties by daily average PM2.5 mass concentrations and

the variance in the three major mortality categories. Potential contributing sources of PM2.5 and

their yearly and seasonal trends in each county are also identified. This study was not intended

to study cause and effect between particulate matter pollutants and adverse health effects, but the

findings provide links to useful sources, data, information and a model that could be used to

further study air pollutants.

Introduction

PM2.5 is a mixture of multiple constituents, including both directly emitted particles

(“primary particles”) and particles that form in the atmosphere (“secondary particles”) through

chemical reactions and physical transformations, (California Air Resources Board/California

Environment Protection Agency(CAARB/CAEPA), 2010). Particles less than 2.5 micrometers

in diameter (PM2.5) are referred to as "fine" particles and are believed to pose the largest health

risks, and these fine particles, because of their small size (less than one-seventh the average

width of a human hair), can lodge deeply into the lungs, (EPA, 2011). Epidemiological analyses

throughout the world have shown that high 24-hour average levels of ambient particulate air

pollution are associated with an increase in all-cause, respiratory, and cardiovascular disease

mortality, (Peters, A., Dockery, D. W., Muller, J. E., & Mittleman, A. M., et. al., and (2001). .

According to CAARB/CAEPA report of 2010, the 2009 science assessment discusses scientific

studies linking PM2.5 to a variety of health effects, including cardiovascular and respiratory

effects. These effects are evaluated for both short-term and long-term exposures. Various

populations are more susceptible to the effects of particulate matter exposures, including the

young, elderly, and individuals with pre-existing disease, (CAARB/CAEPA, 2010).

Researchers have consistently indicated that there is association between high PM2.5

concentrations and adverse health effects. Research conducted by Watkinson and others (1998)

indicated that the recent epidemiological studies of that time had reported a positive association

between exposure to ambient concentrations of particulate matter (PM) and the incidence of

cardiopulmonary-related morbidity and mortality, (Watkinson, Campen, & Costa, 1998). The

results of their study were claimed to support previous epidemiological studies that suggested a

link between preexisting cardiopulmonary disease and potentiation of adverse health effects

following exposure to anthropogenic particulates such as fugitive residual oil fly ash (ROFA)

particulate matter (PM).

The compositions and contributing sources further complicate the associations. The

Compositions and sources of PM2.5 could vary depending on location. For example, PM2.5 in

California could have different compositions and sources than Silver Springs, Maryland. The

main contributing sources of PM2.5 are combustion processes. Sources variable by county also

contribute to PM2.5 mass concentrations. In California, both primary and secondary particles

significantly contribute to non-compliant PM2.5 standards, and local air districts regulate PM2.5

pollution from industrial sources while U.S EPA and ARB regulate PM2.5 emissions from

mobile sources, including both gasoline and diesel engines, (CAARB/CAEPA, 2010).

Many other studies have investigated PM2.5 mass concentrations and health effects using

several approaches, aspects, and conclusions, but none has been found to specifically elaborate

the correlation between PM2.5 mass concentrations and hospital admissions, number of

respiratory and ischemic heart disease cases, and the elaborate details of potential sources and

that are contributing to the elevated PM2.5 in these seven California counties from years 2000 to

2005.

This study entails the analysis of raw data sets of PM2.5 and their components from

California Air Resources Board and the analysis of the raw data of the adverse health effects and

hospital admissions from California Fine Particles data sources CALFINE. The seven counties

were selected because these seven CA counties are some of the leading, original air sampling

sites that were established by CAARB and NAAQS boards to take accountability of sources of

air pollutions.

This study will determine if the elevated annual average concentrations of PM2.5 and

hospital admissions from seven California counties are correlated. The main objectives of this

study include: 1). Characterizing fine particulate matter (PM2.5) in seven California counties;

2). Identifying potential contributing sources of seven California counties; 3). Modeling PM2.5

and health effects in those seven counties. The elevated annual PM2.5 mass concentrations will

not be used to investigate a cause-effect study of the adverse health effects. The extent that

elevated concentrations of PM2.5 contributed to the health effects and hospital admissions will

remain unknown. However, the analysis and results of this study will identify, and connect,

and/or provide the link between the potential PM2.5 contributing sources to specific locations

and health effects. Finally, this study tests the utility of a model to be for other further studies in

the air pollution and health effects in the environmental public health field.

Literature Review

Health Effects and Particulate Matter (PM 2.5)

Mortality categories such as respiratory disease, cardiovascular disease, and diabetes,

from other studies have shown the associations between elevated PM2.5 and adverse health

effects on a daily basis. The current National Ambient Air Quality Standard (NAAQS) requires

PM2.5’s reduction. This implies that all PM2.5 mass concentrations are equally bad or equally

causing adverse health effects. However, The United States Environmental Protection Agency

(EPA) has candidly admitted that it has based its conclusions almost entirely on the

epidemiological literature, which reveals more or less consistent statistical “associations”

between significant increases in ambient PM2.5 concentrations and the foregoing adverse health

effects. The EPA acknowledges that “the relevant toxicological and controlled human studies

published to date have not identified any accepted mechanism(s) that would explain how [the]

relatively low concentrations of ambient PM [regulated by the PM2.5 NAAQS] might cause the

health effects reported in the epidemiological literature, (Karmel & FitzGibbon, 2002).

What the EPA and many other research studies have not been able to fully investigate,

explain, and confirm were the strength of the correlation between the two factors, health effects

and elevated PM2.5 mass concentrations. Karmel & FitzGibbon, (2002) also indicated that the

health effects of ambient PM2.5 have been analyzed and debated in a vast body of

epidemiological and toxicological literature, and substantial research in this area is underway.

The EPA’s bottom line conclusion after a review of these studies, in a judgment that has now

been upheld by the federal courts, is that exposure to PM2.5 at the ambient concentrations that

presently exist in some areas of the country, including those in compliance with the PM10

NAAQS, can result in serious health consequences, including premature mortality, exacerbation

of respiratory and cardiovascular disease, decreased lung function, increased respiratory

symptoms from pre-existing pulmonary disease, and aggravation of symptoms associated with

asthma, (Karmel, P.E., & FitzGibbon, T.N., 2002). At the same time, the EPA has candidly

admitted that it has based its conclusions almost entirely on the epidemiological literature, which

reveals more or less consistent statistical “associations” between significant increases in ambient

PM2.5 concentrations and the foregoing adverse health effects, (Karmel, P.E., & FitzGibbon,

T.N., 2002).

Particulate Matter (PM2.5) and Ultrafine Particulates (UFPs)

It is imperative to include a discussion of Ultrafine Particulates (UFPs) when studying the PM2.5

and health effects. PM2.5’s composition consists of any particulate that is 2.5 micrometer or

smaller. UFPs are therefore included. The smaller particulates are easier to travel and pass

through human respiratory organs, i.e., nostrils and lungs. Delfino, R. J., Sioutas, C., and Malik,

S. (2005) indicate that the ultrafine particles < 0.1 μm (UFPs) dominate particle number

concentrations and surface area and are therefore capable of carrying large concentrations of

adsorbed or condensed toxic air pollutants. Delfino et. al. (2005) also suggests that it is likely

that redox-active components in UFPs from fossil fuel combustion reach cardiovascular target

sites. High UFP exposures may lead to systemic inflammation through oxidative stress

responses to reactive oxygen species and thereby promote the progression of atherosclerosis and

precipitate acute cardiovascular responses ranging from increased blood pressure to myocardial

infarction, (Delfino et. al., 2005).

Delfino et. al. (2005) concluded in their research that numerous studies have implicated

particulate air pollution as an important contributor to morbidity and mortality from

cardiovascular causes, and most of these data have been epidemiologic and have used available

air pollution data from governmental monitoring stations. These data may have met the

governmental regulatory standards, but they may not necessarily meet the needs of researchers

attempting to investigate the causal pollutant components that could possibly contribute to

certain adverse health effects. Ultra-Fine Particulates (UFPs) and related toxic constituents and

precursors are examples of air pollutants that have not been fully investigated, in part due to lack

of available data, (Delfino et. el., 2005). Delfino and his colleagues also concluded that, to date,

data from epidemiologic studies indirectly implicate traffic- and other combustion-related

pollutants, which include UFPs. Exposure assessment issues for UFPs are complex and need to

be considered before undertaking epidemiologic investigations of UFP health effects (Sioutas et

al. 2005). A large body of evidence shows that inflammation and oxidative stress are related to

both acute changes in cardiovascular health and chronic processes, including atherosclerosis. It

is likely that redox-active components in UFPs from fossil fuel combustion reach target sites in

the lungs, vasculature, and heart to induce inflammation and oxidative stress, adding to the

burden of known lifestyle risk factors for cardiovascular disease such as diet, tobacco smoke, and

stress, (Delfino et. al., 2005).

PM2.5 Studies in the East Coast and other 50 States

A study by Peters, A., Dockery, D. W., Muller, J. E., & Mittleman, A. M. (2001) suggests

that elevated concentrations of ambient particulate air pollution have been associated with

increased hospital admissions for cardiovascular disease, but whether high concentrations of

ambient particles can trigger the onset of acute myocardial infarction (MI), however, remains

unknown, (Peters et. al., 2001). In their study, they interviewed 772 patients with MI in the

greater Boston area between January 1995 and May 1996 as part of the Determinants of

Myocardial Infarction Onset Study, and the hourly concentrations of particle mass ,2.5 mm

(PM2.5), carbon black, and gaseous air pollutants were measured, (Peters et. al., 2005). The

results showed the risk of MI onset increased in association with elevated concentrations of fine

particles in the previous 2-hour period, and delayed response associated with 24-hour average

exposure 1 day before the onset of symptoms was also observed. Multivariate analyses

considering both time windows jointly revealed an estimated odds ratio of 1.48 associated with

an increase of 25 mg/m3 PM2.5 during a 2-hour period before the onset and an odds ratio of 1.69

for an increase of 20 mg/m3 PM2.5 in the 24-hour period 1 day before the onset (95% CIs 1.09,

2.02 and 1.13, 2.34, respectively), (Peters et. al., 2005).

Peters and colleagues (2005) concluded in their study that elevated concentrations of fine

particles in the air may transiently elevate the risk of MIs within a few hours and 1 day after

exposure, and further studies in other locations are needed to clarify the importance of this

potentially preventable trigger of MI, (Peters et. al., 2001). Bell, M.L., Dominici, F., Ebisu, K.,

Zeger, S.L., and Samet, J.M. (2007) conducted their research on the characterization of spatial

and temporal variability of PM2.5 components in the United States. Their objective was to

identify components for assessment in epidemiologic studies. The results of the study by Bell

and her colleagues found strong seasonal and geographic variations in PM2.5 chemical

composition. Only seven of the 52 components contributed ≥ 1% to total mass for yearly or

seasonal averages [ammonium (NH++), elemental carbon (EC), organic carbon matter (OCM),

nitrate (NO3), silicon, sodium (Na+), and sulfate (SO42–)], (Bell et. al., 2007). Each of these

components would have shown a number of recorded concentrations in a particular time of the

year or a period of the season. The strongest correlations with PM2.5 total mass were with

ammonium (NH4+) (yearly), organic carbon matter (OCM) (especially winter), nitrate (NO3)

(winter), and sulfate (SO42-) (yearly, spring, autumn, and summer), with particularly strong

correlations for ammonium (NH4+) and sulfate (SO42–) in summer. Components that co-varied

with PM2.5 total mass, based on daily detrended data, were ammonium (NH4+), sulfate (SO42–)

organic carbon matter (OCM), nitrate (NO3-), bromine, and Elemental Carbon (Bell et. al., 2007).

Pope III, C. A., Burnett, R. T., Thun, M.J., Calle, E.E., Ito, K., & Thurston, G.D. (2002)

analyzed lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air

pollution in 50 states. The study results showed that fine particulate and sulfur oxide–related

pollution were associated with all cause, lung cancer, and cardiopulmonary mortality. Using

adjusted mortality Relative Risk (RR) associated with a 10-μg/m3 change in PM2.5, they found

that each 10-μg/m3 elevation in fine particulate air pollution was associated with approximately

a 4%, 6%, and 8% increased risk of all-cause, cardiopulmonary, and lung cancer mortality,

respectively.

PM2.5 Studies in California

Ostro et al. (2006) derived a similar relationship for estimating mortality from exposure

to PM2.5 air pollution within California. Ostro et. al., (2006) used Poisson multiple regression

models incorporating natural or penalized splines to control for covariates that affect daily counts

of mortality, including time, seasonality, temperature, humidity, and day of the week. Spline

smoothing is the model used for curve fitting when each data point does not have to exactly

match each other in linear regression. Spline functions are piecewise polynomials, with the

polynomial pieces joining in the so called knots and fulfilling continuity conditions for the

function itself and some of its derivatives. They also used meta-analyses using random-effects

models to pool the observations in nine of the largest California counties in their study. The

modeled estimates used for this study were derived from Ostro and colleagues (2006) since their

study was more applicable to California than Pope and colleague’s study. As Ostro et al. (2006),

described, relatively few studies have been conducted in California, where particle sources,

chemistry, size distribution, and temporal patterns of exposure are quite different. Specifically,

existing evidence suggests that, in California, a) nitrates comprise a larger fraction of PM2.5than

they do in other regions, and b) mobile sources represent the predominant source of PM2.5,

whereas a mix of mobile and stationary sources predominate elsewhere (Blanchard, 2003).

Moreover, in the Los Angeles air basin, peak PM2.5 exposures occur in both winter and non-

winter months (Ostro et al., 2006). The modeled estimates of percent changes in daily mortality

categories per 10-μg/m3 increment in PM2.5 using penalized splines from Ostro et al. (2006)

included all-cause, cardiovascular, respiratory, age > 65 years, ischemic heart disease, diabetes,

males, females, whites, blacks, Hispanics, in hospital, out of hospital, high school graduates, and

non-high school graduates.

Other researchers concluded that the subset of identified PM2.5 components should be

investigated further to determine whether their daily variation is associated with daily variation

of health indicators, and whether their seasonal and regional patterns can explain the seasonal

and regional heterogeneity in PM10 (PM with aerodynamic diameter < 10 μm) and PM2.5 health

risks, (Bell et. al., 2007). One of the important findings in the study by Bell and colleagues was

that the occurrence of certain PM2.5 components in certain times a year showed a seasonal and

yearly trend. This study could relate to the manner of the yearly and seasonal trends in the

research of this study that includes seven counties of California. The seasonal and yearly trends

could indicate or assist in identifying the activities or processes that occurred in area of the

studied and sampling sites. The agricultural burning in San Joaquin valley would, for instance,

correspond to elevated SO4 2--in the dry season of December and January when many high

agricultural burning permits were issued. The finding is significant for accurately identifying

and confirming the link of each particular component of PM2.5 to a specific health risk.

Ostro, B., Broadwin, R., Green, S., Feng, W., & Lipsett, M. in 2006. Ostro et. al. (2006)

examined associations between PM2.5 and daily mortality in nine heavily populated California

counties using data from 1999 through 2002. They considered daily counts of all-cause

mortality and several cause-specific subcategories (respiratory, cardiovascular, ischemic heart

disease, and diabetes), their study also examined these associations among several

subpopulations, including the elderly (> 65 years of age), males, females, non-high school

graduates, whites, and Hispanics, (Ostro et.al., 2006). The raw data sets were from the

California Fine Particulates from California Air Resources Board data sources. Ostro and his

colleagues used Poisson multiple regression models incorporating natural or penalized splines to

control for covariates that could affect daily counts of mortality, including time, seasonality,

temperature, humidity, and day of the week. We used meta-analyses using random-effects

models to pool the observations in all nine counties, (Ostro et. al., 2006). The analysis revealed

associations of PM2.5 levels with several mortality categories. Specifically, a 10-μg/m3 change

in 2-day average PM2.5 concentration corresponded to a 0.6% (95% confidence interval, 0.2–

1.0%) increase in all-cause mortality, with similar or greater effect estimates for several other

subpopulations and mortality subcategories, including respiratory disease, cardiovascular

disease, diabetes, age > 65 years, females, deaths out of the hospital, and non-high school

graduates. The results were generally insensitive to model specification and the type of spline

model used.

While the analysis of the study by Ostro and his colleagues adds to the growing body of

evidence linking PM2.5 with daily mortality, contributing sources of PM2.5 and their locations

were neither included nor discussed in their study. Contributing sources to PM2.5 and their

locations are important and should add more value to the evidence link and investigation of

PM2.5 and adverse health effects. Therefore, the analysis of PM2.5 concentrations as well as

the direct and indirect contributing sources in each of seven counties would be worth further

research and exploration.

Particulate matter can be directly emitted into the air (primary PM) or it can be formed in

the atmosphere from the reaction of gaseous precursors such as nitric oxides (Nox), sulfuric

oxides (Sox), reactive organic gases (ROG), and ammonia (secondary PM). The PM2.5 emission

inventory includes only directly emitted particulate emissions. On an annual average basis,

directly emitted PM2.5 emissions contribute approximately 40 percent of the ambient PM2.5 in

the San Francisco Bay Area Air Basin, (ARB Almanac, 2003).

As many of other studies indicated, the topic of PM2.5 and adverse health effects have

been extensively studied but the data that have been used were mostly, if not all, collected to

meet governmental standards and regulations, and do not necessarily meet the same standards for

researchers to study the causal effects. The elevated annual PM2.5 mass concentrations will not

be used to investigate a cause-effect study of the adverse health effects. The extent that elevated

concentrations of PM2.5 and each of its components contributing to the health effects and

hospital admissions will remain unknown. However, the analysis and results of this study,

especially the analysis of the PM2.5 mass concentrations and contributing sources and locations,

will identify, and connect, and/or provide the link between the potential PM2.5 contributing

sources to specific locations and health effects.

Methodology

In order to estimate health effects from exposure to particulate matter air pollution,

PM2.5 mass concentration and chemical composition data from 2000 to 2005 was obtained from

the United States Environmental Protection Agency (USEPA)’s Speciation Trends Network

(STN). This dataset represented nine sampling locations within seven of the largest counties in

California. Ostro and others (2006) obtained and linked daily readings of PM2.5 with mortality

in these seven heavily populated counties in California. Ostro et. al., (2006) believed that the

ability to explore hypotheses of association with adverse health in multiple cities has several

distinct advantages since it enhances the power of the statistical analysis and reduces the

likelihood of spurious results or publication bias that might result from the analysis of a single

city (Anderson et al. 2005). Because of this, we focus our study in these seven large California

counties.

Sampling Sites:

Detailed descriptions of the nine sampling locations can be obtained from Table 1. Seven

of these nine STN sampling locations fall under the National Air Monitoring Stations (NAMS)

and the remaining two under the State and Local Air Monitoring Stations (SLAMS). Data from

both of these stations provide information needed for developing effective air quality attainment

plans. The focus of the SLAMS PM2.5 speciation network is to enhance the spatial coverage of

the NAMs sites in areas with a diversity of PM problems. Bakersfield, El Cajon, Fresno, two

locations in Sacramento, San Jose, and Riverside are part of the NAMS and Los Angeles and

Escondido are part of the SLAMS. The selections of these 9 sites are primarily based on

population density, meteorological and geographical characteristics, air pollution emissions, and

support for ongoing health studies.

TABLE 1. Description of air pollution monitoring stations. Address AIRS ID*1 CARB ID*2 GPS coordinates SLAMS

Y/N NAMS Y/N

Description of site (based on air polluting sources)

3415 N. First Street, FRESNO, CA 93726

060190008 1000246 Lat. 36046’55” N Long.119046’23”W elevation 98 m

N Y

STN

Located approximately 1 km north of the downtown commercial district. First Street is a four-lane artery with moderate traffic levels. Commercial establishments, office buildings, churches, and schools are located north and south of the monitor. Medium- density single-family homes and some apartments are located in the blocks to the east and west of First Street.

5558 California Ave., Bakersfield, CA 93309

060290014 15255 Lat. 35o 21’ 24” N, Long. 119o 3’ 46” W, elevation117 ml

N Y

STN

Distance to road of 300 meters. Ground cover is paved concrete. Avg. traffic count of 10,000 per day. Sampling method R&P2025, analysis method Gravimetric, Unrestricted airflow in 360 degrees.

1630 North Main St Los Angeles, CA 90012

060371103 70087 Lat. 34° 03' 59"N Long. 118° 13' 36"W, elevation 89 m

Y N STN

Located at General Warehouse Building. 51-71 meters of distance to road. Traffic counts 15276 vehicles per day. Groundcover: Asphalt.

Rubidoux 5888 Mission Blvd Riverside, CA 92509

060658001 20021141 Lat. 33°59' 58"N Long. 117° 24' 57"W (Elevation:248)

N Y STN

Located in a vacant lot. 119 meters to roadway. Traffic count: 20,000 vehicles per day. Ground cover: gravel. Airflow is 360 degrees.

Del Paso Manor, 2701 AVALON DR, Sacramento, CA

060670006 3400295 Lat. N 38o 36’ 50” Long, W 121o 22’ 5”

Y Y

STN

Located about 7 miles east-northeast of downtown Sacramento. Ground cover is vegetated.30 meters distance from roadway. Average daily traffic about 1000 vehicles per day. PM2.5 Main FRM Sampler measure elevated wintertime PM2.5 from motor vehicles and residential wood combustion. .

T Street 060670010 3400305 Lat. N 38o 34’ 6” Long. W 121o 29’ 35”

Y Y Residential area located in downtown Sacramento. 30meters from roadway. Avg. traffic about 5000vehicles per day ;ground cover-rooftop site(residential area is paved);Sampling method-low vol./very sharp cut cyclone, analysis method-Beta attenuation, Make/Model-Met-One BAM

4th Street 060850004 4300382 N Y STN

Distance to road from gaseous probe 4th St: 34.7 meters. San Jose was chosen for an air monitoring site because it is the largest city in Santa Clara; ground paved. County and the largest city in the Bay Area, with an estimated 2009 population of 1,023,083.

158E. Jackson St, San Jose CA 95112

060850005 Lat.37.3485° N, Long.121.8949° W

N Y STN

Top floor of two-story commercial building. Distance to road from gaseous probe Jackson St: 15.1 meters. The air monitoring site is located in the center of northern Santa Clara Valley, in a commercial and residential part of downtown San Jose. This area is completely encircled by major freeways, and has a large airport just to the northwest.

1155Redwood Ave., El Cajon, CA

060730003 80131 Lat. 32o 47’ 28” N, Long. 116o 56’ 32” W, elevation 144 m

N Y STN

This site represents a major population center located in an inland valley downwind of the heavily populated coastal zone in San Diego. It is impacted from the transportation corridor of Interstate 8 and its major arteries. Local sources are residential and school traffic with a traffic count of 2000 vehicles/day.

600 E. Valley Parkway 060731002 80115 Lat. 33o 07’ 40” N, Long 117o 04’ 31” W, elevation 200 m

Y N STN

The Escondido site represents a major population center for the Inland North County/I- 15 Corridor. It is a receptor of emissions from the burgeoning Highway 78 communities, their associated base of small business and industry, and vehicular traffic in the Highway 78 and I-15 corridors. Its inland location is comparable to the El Cajon site; it provides valuable data concerning the fate of coastal zone/I-5 emissions.

18

*1 AIRS ID: United States Environmental Protection Agency designated Identification number

for an air sampling/monitoring station in National Air Monitoring Station (NAMS).

*2 CARB ID: California Air Resources Board designated Identification number for an air

sampling/monitoring site in the State Local Air Monitoring Station (SLAMS).

Sampling Method and Laboratory Chemical Analysis:

Fine particle air pollution speciation data (from 2000-2005) from the above seven

California counties were collected on filters using ambient samplers (model RAAS2.5-300;

Andersen Instruments, Inc., Smyrna, GA) by the California Air Resources Board (CARB) as

outlined by the STN (CARB, 2002). These data measurements are collected using three

channels: Teflon filter analyzed for elements by X-ray Fluoroscence (XRF) and mass by

gravimetric analysis, nylon filter analyzed for ions by Ion Chromatography (IC), and quartz filter

analyzed for organic carbon (OC) and elemental carbon (EC) by Thermal Optical Transmittance

(TOT), (CARB, 2003). The air pollution monitoring samplers were operated on an every third

day sampling interval. Regular trip and field blanks are included among the samples shipped to

the field sites. Field blanks were run at a frequency of 10% and trip blanks were run at a

frequency of 3%. (CARB, 2002).

Table 2 shows the summary of the speciation data availability for each monitoring site.

Seasons were defined by the standard procedure from calculating the beginning and ending of

the equinoxes and solstices. As seen from Table 2, Sacramento, Riverside, Kern, and San Diego

(El Cajon and Escondido) counties had two monitors where speciation data was available, and

the remaining counties had only one monitoring site with speciation data. The number of

observations usually was every 3 days. For most counties, the number of observations per

season was usually above 25 except for Los Angeles where the observations were half as much.

There were some seasons with no speciation work conducted. Because the two sites in Riverside

and Kern counties were close to each other, the particulate matter concentrations were averaged

for overlapping days; however, El Cajon and Escondido were not averaged to provide a single

measurement for San Diego County because of the differences in air pollution emissions and

transport.

TABLE 2. Samples collected by season for each county.

Site

Season

Fresno Sacramento San Diego

Los Angeles

Riverside Kern San Jose

Site 1 Site 2 El Cajon Escondido Average *1 Average*2

WINTER 2001

24 23 3 3 4 4 24

SPRING 2001

26 29 7 13 6 5 25

SUMME R2001

20 24 14 15 26 30 22

FALL 2001

29 24 25 9 30 30 25

WINTER 2002

24 24 14 23 17 1 30 26 23

SPRING 2002

25 15 7 21 16 4 30 31 12

SUMME R2002

5 30 15 26 15 15 31 31

FALL 2002

32 29 15 21 15 15 30 28 17

WINTER 2003

28 29 14 21 17 14 29 22 22

SPRING 2003

27 30 17 24 16 15 30 27 25

SUMME R2003

27 30 15 25 15 15 30 26 25

FALL 2003

21 28 15 22 15 15 28 28 23

WINTER 2004

24 28 14 25 15 15 28 55 18

SPRING 2004

21 31 16 27 16 15 27 28 21

SUMME R2004

23 31 16 27 15 14 27 10 25

FALL 2004

21 30 14 25 17 30 15 24

WINTER 2005

20 30 15 23 15 14 29 16 24

SPRING 2005

30 31 14 26 15 15 30 22 25

SUMME R2005

31 31 16 26 16 16 27 27 26

FALL 2005

29 29 15 25 15 15 29 24 24

Quality Assurance/ Quality Control (QA/QC):

Description of actual procedures used for QA/QC for the ambient dataset is beyond the

scope of this paper and is not addressed in detail; a full description can be found on the

Environmental Protection Agency (EPA) website (US EPA, 2005). The sampling network and

the sampling locations were designed by using Experimental Design and the samples were

collected and handled using Research Triangle International’s (RTI) sample handling system.

RTI conducted meticulous procedures for sample handling and chain of custody system. For

Gravimetric analysis of samples, QC required that at least one laboratory blank was weighed

during each weighing session with not more than 15 µg difference from initial weight; field

blanks were analyzed periodically in order to affirm post-sampling weights did not exceed the

initial weight by more than 30µg.

The accuracy of temperature and relative humidity recorders were verified annually,

working mass standards and primary mass standards references were rectified each year. QC

criteria for Ion Chromatography included daily multipoint calibration, daily analysis of reagent

blanks, and precise sample preparation with laboratory reagents at acceptable concentrations.

Method Detection Limit (MDL) was checked annually or after major instrument change. The

XRF instruments were subjected to routine testing and maintenance and energy calibration was

performed using a copper calibration standard and calibration verification was performed weekly

by analyzing the following NIST thin-film standards provided by the EPA. For OC/EC analyses

the three thermal-optical transmittance (TOT) analyzers and the five thermal-optical reflectance

(TOR) and five dual reflectance/transmittance (TOR/TOT) analyzers were routinely tested and

maintained with daily calibration check, weekly three-point calibration check and daily weighing

of instrument blank with values less than 0.3 µg/cm3 as accepted . It is advisable that for more

detailed information on QA/QC procedures readers will refer to the EPA publication (US EPA,

2005).

Missing data was adjusted before air pollution trend analysis and health modeling were

conducted. Every missing Minimum Detection Limit (MDL) was replaced with the average

Maximum Detection Limit (MDL) for that monitor and species. For the two monitors where all

MDLs were missing, MDLs were replaced with the average MDL for the other monitor within

that county. All concentrations that were less than MDL (which includes zeroes) were replaced

with ½ the MDL. For missing uncertainties, if concentration was less than MDL, then

uncertainty was considered 5/6 MDL. On the other side, if concentration was more than MDL

then uncertainty was calculated using following formula: SQRT [(Percentage * concentration) ^2

+ MDL^2)]. Uncertainties were not replaced unless it was zero in which case it was replaced

with 5/6 of MDL. Using SAS 9.1, simple descriptive analysis was performed after data editing.

Average uncertainties were calculated by propagating the uncertainties of the measurement. For

two sites (Escondido & Sacramento T- street) with missing uncertainties, standard errors of the

mean were used for source apportionment.

PM2.5 Modeling:

In this study, modeling of PM2.5 health effects was conducted by applying the

parameters from Ostro et. al. (2006) in a linear relationship as shown in equation (1):

∆ HFi=CFi× C 10……………………………………………………………………………

………….. (1)

where,

ΔHFi = Changes in Health Effects from ‘i” category associated with an increase of

PM2.5 concentration

CFi = Conversion Factor for “i” category from Ostro et al. (2006)

C = Average PM2.5 Concentration in µg/m3

“i” = Mortality categories (All-Cause, Cardiovascular, Respiratory, include ALL)

The above linear relationship has been established to estimate mortality from PM2.5

concentration by Pope III, C. A., Burnett, R. T., Thun, M.J., Calle, E.E., Ito, K., & Thurston,

G.D. (2002).

Table 3. Pooled estimates of percent changes in daily mortality categories and 95% CIs per 10-μg/m3 increment in PM2.5 using penalized splines*. __________________________________________ Mortality category% Change (95% CI) __________________________________________ All-cause 0.6 (0.2 to 1.0) Cardiovascular 0.6 (0.0 to 1.1) Respiratory 2.2 (0.6 to 3.9) Age > 65 years 0.7 (0.2 to 1.1) Ischemic heart disease 0.3 (–0.5 to 1.0) Diabetes 2.4 (0.6 to 4.2) Males 0.5 (–0.2 to 1.2) Females 0.8 (0.3 to 1.3) Whites 0.8 (0.2 to 1.3) Blacks 0.1 (–0.9 to 1.2) Hispanics 0.8 (–0.1 to 1.6) In hospital 0.6 (–0.1 to 1.3) Out of hospital 0.6 (0.1 to 1.1) High school graduates 0.4 (0.0 to 0.8) Non-high school graduates 0.9 (–0.1 to 1.9) *Model includes average of 0- and 1-day lags of PM2.5, day of week, spline smoothers of temperature and humidity, and two spline smoothers of time. Pooled results based on meta-analysis using a random-effects model.

Table 3 provides with the Conversion Factors (CF) to be used in equation (1) above for

the various “i” mortality categories. In the present study, average seasonal PM2.5 concentration

for each sampling site as described in the “Sampling Sites” section was used to derive the

changes in health effects from equation (1) on a seasonal basis. It is worth examining and

exploring the seasonal health effects to determine any existing relationship between the elevation

of average daily PM2.5 mass concentration and adverse health effects in a particular county in a

given season. The health effects could and do occur in a daily basis, but if the average daily

health effects are found reflecting in a seasonal trend, then the link or correlation of the two

factors could be further investigated. The main objectives of this study include: 1).

Characterizing fine particulate matter (PM2.5) in seven California counties; 2). Identifying

potential contributing sources of seven California counties; 3). Modeling PM2.5 and health

effects in those seven counties.

Results, Discussion and Conclusion

Results

Fine particulate matter (PM2.5) in seven California counties can be characterized as

shown Figure 3. Figure 3 shows the chemical composition of daily PM2.5 Concentrations

during Winter 2003-2004. There are some differences in the level of averaged PM in some

counties in the winter. Seasonal averaged PM2.5 mass concentration from are shown on Figure

2; the seasonal trends of elevation of the particulate matter (PM2.5) are indicative of the potential

contributing sources of seven California counties. Figure 1 shows the daily PM2.5 mass

concentration from seven California counties (nine sites) between 2000-2005. The percent

increase of the mortality categories that were affected by PM2.5 are shown Tables 4-12 and

Figures 4a to 4c.

Characterization of PM2.5, Yearly-Seasonal Trends and Potential Contributing Sources

Fresno

The concentration of PM2.5 mass in Fresno County is at its highest concentration during

December and January of every year from year 2000 to 2005. The seasonal trend appears to be

that the highest average 24-hour PM 2.5 mass concentration reaches the highest in the fall and

winter and reaches the lowest during summer and spring of every year from 2000 to 2005.

Years 2000 and 2002 have PM2.5 average concentration of 23 µg/m3, and year 2001 shows the

highest average concentration in the six year period, with a total average concentration of 25

µg/m. Then the declining trend starts in 2003 to 2005, with average annual concentrations of 19

µg/m3, 18 µg/m3, and 18 µg/m3 respectively. These annual averages of PM2.5 mass

concentrations exceeded the NAAQS in effect (15 µg/m3) during that time. There were several

occurrences of highest concentrations in 2000, 2001, 2002, and 2005 that were exceeding

NAAQS standard; the concentrations recorded 117 µg/m3, 188 µg/m3, 93 µg/m3, and 96 µg/m3

respectively. These entire concentrations exceeded current California ARB PM2.5 annual

average standard of 12 µg/m3. PM2.5 begins to increase in late October, reaches its highest

concentration in the middle of December, and declines to the lowest in May and June of every

year. Generally, the highest concentrations recorded in fall and winter, and the lowest recorded

in summer and spring. (See Figure 1-Fresno)

About 41% of PM 2.5 for this analysis is attributed to OC and only 5% to EC, and both

combined to be 46% of PM 2.5. The remaining portions of PM 2.5 are potentially contributed

by Al, Si, Fe, Ti, and Ca from fugitive dust. Fugitive dust is an important contributor to PM2.5

during the summer and fall in California’s San Joaquin Valley (SJV) (Chow et al., 1992, 1993a).

SJV fugitive dust is believed to originate from paved and unpaved roads and parking lots,

agricultural field preparation, cultivation, harvesting, wind erosion of fallow land, and

construction of buildings and roadways, (Chow et. al, 2003). (See Figure 3 for Chemical

composition).

Possible reasons for these seasonal trends include agricultural burning and inversion

layer. According to Final Report of 2000 California Agricultural Burning Database, most

agriculture burning take place in October, November, December, and January of every year.

According to our data analysis, Potassium shows highest overall (aggregated) peaks in December

and January of every year from 2000 to 2005. Similar to Potassium, Levoglucosan is also used as

a biomarker for biomass combustion [cite]. Rinehart et al. (2006) and Schauer and Cass (2000)

reported higher concentration of biomass burning during winter episodes in Fresno county by

using Levoglucosan as their biomarker for biomass combustion. San Joaquin Valley Air District

recorded 11,476 acres and 877 tons of agricultural burning for Fresno County in 2000, (1).

According to the district permit data source, approximately 4600 tons in 2002, 4500 tons in

2003, 3500 tons in 2004, and 2750 tons in 2005 of agricultural burning were allowed in the San

Joaquin Valley, (CAEPA-ARB, 2010).

Agricultural burning is a significant contributor to PM2.5. According to CAARB Staff

Report on the San Joaquin Valley Smoke Management Program and Consideration of

Modifications to Agricultural Burning Requirements, the requirements of SB 705 are being

implemented in conjunction with California’s longstanding smoke management programs

adopted by air districts consistent with ARB regulations. ARB’s statewide regulations for smoke

management were comprehensively updated in 2000, and air districts were required to strengthen

their smoke management programs. The combined effect of both sets of requirements has been

an almost 70% reduction in total acreage of agricultural materials burned since 2002 in the San

Joaquin Valley, (CAARB, 2010). This could possibly be a contributing factor to the decreasing

trend of annual average of PM2.5 from 2003 to 2005.

Another possible reason for the seasonal and annual trends of PM2.5 in Fresno County

would be the inversion layer. Generally, inversion layers are more of a problem for PM2.5 in

the winter. An inversion layer is a zone of air near the ground that is colder than the air above it.

Because of the temperature and density difference, air in inversion layers do not interact and mix

with the rest of the atmosphere above it, (Wagner et. al, 2007). In other words, the subtropical

layer above has cooler air, and the next layer down that the pollutants are prevented from rising

through inversion and trapped below and created high concentrations. Fresno county has a high

mountain range to the east and ocean to the west, and it has colder air and slower wind in the

winter. Wintertime air pollution events in a lot of urban areas and their surrounding rural areas

throughout the world are often characterized by elevated particulate matter (PM) concentrations

and secondary PM such ammonium nitrate and/or ammonium sulfate is a major component of

the fine airborne particulate matter (PM) during these air pollution episodes, (Ying, 2011).

Nitrates and sulfate account for certain percentage in PM2.5. In Fresno, nitrate and ammonium

concentrations increase in the fall and winter. Sulfate increases in spring and summer.

Nitrogen deposition has increased in the western United States than the eastern side of the

country because of rapid increases in urbanization, population, distance driven, and large

concentrated animal feeding operations, (Fenn et. al., 2003). Nitrogen (N) deposition in the

western United States ranges from 1 to 4 kilograms (kg) per hectare (ha) per year over much of

the region to as high as 30 to 90 kg per ha per year downwind of major urban and agricultural

areas. Primary N emissions sources are transportation, agriculture, and industry, (Fenn et al.,

2003).

The central valley of California is a low-elevation geographic basin surrounded by

mountainous terrain (Fig. 1). The lower portion of the valley, is in nonattainment of PM2.5 and

PM10 standards (EPA, 2005). During the winter, PM concentrations in Fresno, CA often exceed

the 24-h PM2.5 National Ambient Air Quality Standard (NAAQS) of 65 µg/m3 with gravimetric

mass concentrations2–3 times higher than the standard (Chow et. al.,2002). Major particulate

emission sources include agricultural activities, biomass burning, cooking, on-road and off-road

engine exhaust, (Rinehart et. al, 2006).

In the study by Chen et al. (2007), UNMIX and Positive Factorization (PMF) solutions to

the Chemical Mass Balance (CMB) equations were applied to chemically speciated PM2.5

measurements from 23 sites in California San Joaquin Valley to estimate source contributions.

According to these applications for PM2.5 measurements, six and seven factors were determined

by UNMIX for the low PM 2.5 period of February to October and high PM 2.5 period of

November to January, respectively. The eight factors for each period that corresponded with the

UNMIX factors in chemical profiles and time series that were resolved by PMF. These

contributing factors include marine sea salt, fugitive dust, agriculture—dairy, cooking, secondary

aerosol, motor vehicle, and residential wood combustion (RWC) emissions, with secondary

aerosol and RWC accounting for over 70% of PM 2.5 mass during the high PM2.5 period. This

study further indicated that zinc was resolved by only PMF, and also found that the contribution

from motor vehicle was between 10 and 15% with higher percentages occurring in summer,

(Chen et. al, 2007). (See Figure 2)

According to our CAARB data analysis, PM2.5’s trend resembles and reflects that of OC

for this six year period, and this is almost true with EC, except EC trend has the lowest of the

highest average concentration in 2005. The OC/EC ration in regards to PM2.5 appears to

consistently have a significant difference for 2000 to 2005 period, with an average percentage of

41% for OC and 5% for EC of PM2.5. What this means is other major sources of particulate

matter include other sources such as agricultural activities, biomass burning, and cooking that

created high OC% and less of other man-made fuel combustion that created EC such as on-road

and off-road engine exhaust and oil refinement. (Figure 2)

Sacramento

Site-1: Del Paso Manor

PM 2.5 mass for Sacramento Del Paso Mano has the highest concentration during

December of every year. Late fall and early winter of every year are the times that show the high

peaks of PM2.5 mass. The peaks decline to the lowest point during late May, July and

throughout early October, and the peaks start increasing in late November and reach the highest

reading during December and January. The highest 24-hour concentration is 120 µg/m3 on

January 1, 2001. Year 2000 has its highest daily concentration of 90 µg/m3, and 2002 has its

highest of 84 µg/m3. Year 2005 has its highest daily concentration is 72 µg/m3 . With the

exception of 2003 and 2004, the other years exceeded the 2005 NAAQS standard of 65 µg/m3 .

The trend of PM 2.5 almost mirrors the trend of EC and OC, and EC has much lower

concentration than OC, with about 1 to 6 ratio of EC/OC. EC is about 5% of PM2.5, OC is about

46% of PM2.5. Generally, PM 2.5 mass peaks are high in late fall and winter and lower in

spring and summer of every year from 2000 to 2005. (Figure 1)

Site-2: Sacramento-T Street

PM 2.5 mass observations for Sacramento T Street site are from 2002 to 2005. In 2002,

the highest daily concentration occurs in November 28, 2002, which reaches above 74 µg/m3,

which exceeded the 2005 NAAQS standard of 65 µg/m3. The annual average PM 2.5 mass

concentrations are almost always higher in fall and winter every year, with the exception of

winter 2002 that has 11.29 µg/m3 and summer 2002 with 12.0 µg/m3, and annual average for

every year met NAAQS annual standard, (Figure 1). There is a significant difference between

OC and EC. OC has an average of 52% of PM 2.5 and EC only has 5% for the four year period.

The two sites in Sacramento show similar PM 2.5 mass concentrations in the seasonal

and yearly measurement trends. Both sites indicate high concentrations in fall and winter and

show the lowest during summer and spring of every year. Both locations have PM 2.5 average

mass concentrations exceeding yearly NAAQS standards of 15 µg/m3 and daily standard of 65

µg/m3 in 2002, (Figure 2).

According to google earth map, these two sites are located approximately less than 9

miles apart with T street being at the a slightly northeast of Del Paso Manor,. T Street is

located nearby major highways (Interstate 80, highways 99 and 50). Del Paso Manor is located

nearby a dense residential area. Del Paso Manor site show a slightly higher average annual and

daily concentrations than T street site.

Possible explanations for this difference could be the PM 2.5 emission source key

contributors in the area. A preliminary Chemical Mass Balance (CMB) analysis from year 2005

shows that at the Sacramento Del Paso Manor site, wood smoke and cooking are significant

contributors to Sacramento’s concentrations, wood smoke and cooking are significant

contributors to Sacramento’s concentrations, (SMAQD, 2006). The report also states that the

winter Sacramento PM2.5 inventory is dominated by area wide sources by about 79%, most of

which is residential wood burning at 44%, followed by road dust at 17%, and construction and

demolition at 8%. Other large area-wide sources include farming and open burn, at 7% each.

Stationary and mobile sources contribute to about 20% of the PM2.5 winter inventory,

(SMAQD, 2006).

The seasonal trend found in this analysis appears to be the same as what Sacramento

Metropolitan Air Quality Management District’s ( SMAQD) Report to community of 2005, as

the report indicated that particulate matter is primarily a problem during the fall and winter

months, (SMAQD, 2005). The report also stated that the county PM 2.5 was designated as non-

attainment for State standard and had 5 exceedances for federal 24-hour standard during

December and zero exceedance from January to November 2005, (SMAQD, 2005).

San Diego County:

El Cajon

Annual average of PM 2.5 mass concentration for El Cajon met the 2005 NAAQS daily

standard of 65 µg/m3 every year from 2001 to 2005. The highest 24-hour average PM 2.5 mass

concentration was recorded on 01/01/2004 as 43.3 µg/m3. There is a declining annual trend of

PM 2.5 mass concentrations every year from 2001 to 2005, with the highest annual average of 17

µg/m3 in 2001 to the lowest of 12.58 µg/m3 in 2005, (Figure 1). The seasonal trend of high

peaks of PM2.5 in fall and winter and low peaks in spring and summer are still prevalent

throughout the five year period of collected data, (Figure 2).

OC and EC annual and seasonal trends also reflect PM2.5 trends. OC annual average

concentration was recorded as 6.42 µg/m3 in 2001 and continued to decline every year to 4.69

µg/m3 in 2005, (Figure 1). EC annual average concentration started at 1.03 µg/m3 in 2001 and

continued declining annually to 0.76 µg/m3 in 2005, (Figure 3). Seasonal trends of OC and EC

and OC/EC ratios reflect PM2.5 as well; higher concentrations in fall and winter and lower

concentrations in spring and summer of every year. Ammonium, sulfate, and nitrate yield 12%,

20%, 25%, respectively, to PM 2.5 mass for the total average in the five year period, (Figure 3).

Escondido

The data sets for Escondido were available or collected for only from 2002 to 2005. The

annual average PM 2.5 mass concentration is 15.43 µg/m3 for 2002 and continued annually to

13.70 µg/m3 in 2005. The highest concentration was recorded in December 25, 2002 as 42.0

µg/m3, and the highest average annual concentration started to decline every year to 33.0 µg/m3

in 2005. All of the highest average annual concentrations were recorded in fall or winter for this

four year period. All annual average PM2.5 mass concentrations met NAAQS standard of 2005

of 65 µg/m3 and even the current NAAQS daily standard of 35 µg/m3, (Figure 1). The OC and

EC percentage to PM2.5 also reflect on the seasonal and annual trends of PM 2.5. That means

the average concentrations are high in 2002 and continue to decline annually to the lowest in

2005 as well as having high concentrations in fall and winter and low in spring and summer. OC

consists of average of 42% in PM2.5, and EC has 4% in PM2.5 for the four year period.

Ammonium, sulfate, and nitrate contribute good percentage to PM 2.5 in Escondido site, 11%,

17%, and 22% respectively, (Figure 3).

Seasonal, yearly trends, and all other measurements of the two sites in San Diego County

almost mirror each other. Both sites did not meet an annual average PM 2.5 mass concentration

for NAAQS standard in 2002. However, El Cajon site exceeded NAAQS standard in 2001 and

2003. Escondido site did not data for 2001. As San Diego Air Pollution Control District report

stated, San Diego County is an Attainment Area for federal PM standards and a Nonattainment

Area for state PM standards, which are more stringent, (SDAPCD, 2005).

PM 2.5, OC, and EC seasonal and yearly trends are consistent with the San Diego Air

Pollution Control District report in 2005. Detailed analysis of PM2.5 monitoring data from

Escondido and El Cajon indicates a large majority (92%) of ambient PM2.5 results from

combustion sources, and those key combustion sources in San Diego County are off-road

construction equipment, trucks, passenger cars, aircraft, ships, residential wood combustion, and

power plants, (SDAPCD, 2005).

Annual average PM 2.5 composition in Escondido 2002-2003 consists of 52%

combustion organic carbon, 19% of combustion Nox (Ammonium Nitrate), 18% of combustion

Sox (Ammonium Sulfate), 4% of road and other dust, 3% of combustion elemental carbon, and

4% of other unknowns, and annual average PM 2.5 composition in El Cajon 2002-2003 consists

of 41% combustion organic carbon, 26% of combustion Nox (Ammonium Nitrate), 20% of

combustion Sox (Ammonium Sulfate), 4% of road and other dust, 5% of combustion elemental

carbon, and 4% of other unknowns, (SDACP, 2005).The PM2.5 emission inventory includes

only directly emitted particulate emissions; on an annual average basis, directly emitted PM2.5

emissions contribute approximately 50 percent of the ambient PM2.5 in the San Diego Air Basin,

(ARB Almanac, 2003).

Los Angeles

Air pollution is a serious issue in the Los Angeles Basin. According to the American

Lung Association (2007), the Los Angeles-Orange-Riverside counties combined together rank as

the most polluted region in multiple categories, including most polluted year-round and short

term by PM2.5 , (Wagner et. al, 2007).

Data sets were recorded from 2002 to 2005 for Los Angeles County. The highest annual

average concentration was recorded as 23.30 µg/m3 in 2002 and decreased in 2003 and 2004 as

21.65 µg/m3 and 19.34 µg/m3 respectively, and it increased to 20.85 µg/m3 in 2005. The

highest daily concentration was recorded for 39.90 µg/m3 on December 4, 2002, 69.90 µg/m3

on July 5, 2003, 57.30 µg/m3 on March 16, 2004, and 68.50 µg/m3 on March 11, 2005. Years

2003 and 2005’s highest daily concentrations did not comply with 2004/2005 NAAQS standard

of 65 µg/m3, and all exceeded the current NAAQS daily standard of 35 µg/m3. The annual

average PM2.5 mass concentrations are about the same among all years. Figure 1 shows annual

and daily average PM mass concentration.

One unique observation of Los Angeles is that there is no obvious seasonal or yearly

trend of annual or daily PM 2.5 mass concentrations in this four year period. However, the

highest average annual concentrations were recorded in fall and winter. Another unique

observation from this analysis is the percentages of OC and EC in PM2.5 are closer compared to

other counties. OC consists of an average of 31% and EC of 7% of PM 2.5 for the four year

period; OC is only 4 times more than EC. Percentages of ammonium, sulfate, and nitrate to PM

2.5 mass concentrations are high, with ammonium of 13%, sulfate 18%, and nitrate of 28%.

In the study by A.H. Miguel et. al., (2004) during nearly a 1-year period, a comprehensive set of

measurements was carried out in Claremont, California, as part of the Southern California

Particle Center and Supersite (SCPCS) activities. This receptor area location is characterized by

high concentrations of secondary air pollutant species including ozone, nitric acid, sulfate (SO4

2), nitrate NO3, and particulate organic carbon (OC) (Blumenthal et al., 1987; Kim et al., 2000,

2002), (Miguel et. al., 2004).

Miguel et. al, (2004) explained in their study that PAHs(Polycyclic Aromatic

Hydrocardons) emitted by vehicular emissions are associated with fine EC. Emissions from

diesel engines including both on-highway and off-highway usages are the main contributors to

EC concentrations in urban area air. EC and OC (total particulate carbon) result from the

accumulation of small increments from a variety of emission source types such as gasoline and

diesel powered highway vehicles, stationary source fuel oil and gas combustion, industrial

processes, paved road dust, fireplaces, cigarettes and food cooking, (Miguel et al., 2004).

Additionally recent experiments carried out in the Caldecott tunnel (east of Berkeley, CA.)

showed that, on a mass/volume of fuel basis, the contribution of EC by heavy-duty diesel

vehicles is ca. 50-fold of that estimated for light-duty gasoline vehicles (Miguel et al., 2004).

Wagner et. al., (2007) concludes in their study in Claremont (an eastern town 30 miles

east of downtown Los Angeles) that the presence or absence of an inversion layer is the single

most important factor in determining PM2.5 levels. The topography of the Los Angeles Basin,

with the ocean to the west and mountains ringing the basin to the north, east, and south, is ideal

for the formation of an inversion layer because pollutants are trapped by the mountains.

Inversion layers play a serious role determining air pollution levels by trapping pollutants

emitted from ground-based sources and preventing them from escaping to other parts of the

atmosphere, (Wagner et. al., 2007).

Riverside

According to the American Lung Association State of the Air Annual nationwide report

2010, Riverside county is ranked number four on the most polluted county by Short-term

Particle Pollution (24-hour PM2.5) and ranked third on the most polluted by Year-Round Particle

Pollution (Annual PM2.5), ( ALA, 2010). This is an improvement compared to the American

Lung Association State of the Air Annual nationwide report in 2005 when the county was ranked

first of the most polluted county by short-term particle pollution (24-Hour PM2.5) and also the

first of the most polluted by long-term Particle Pollution (Annual PM2.5), (ALA, 2005).

As shown in figure 1, the annual average PM 2.5 mass concentration started to steadily

decrease every year from 32.21 µg/m3 in 2001 to 21.44 µg/m3 in 2005. However, the highest

24-hour concentrations of every year exceeded the 2005 NAAQS standard of 65 µg/m3, with the

highest daily concentration of 112.6 µg/m3 recorded on October 22, 2005. The most noticeable

seasonal trend for Riverside is that the concentrations are high in spring and summer and low in

fall and winter. The percentage of OC in PM2.5 is lower than EC. OC consists only 5%

whereas EC consists of 26% in PM 2.5, (Figure 2). EC is the major contributor in OC/EC ratio

and of PM 2.5 mass in this case. Annual average concentration of OC/EC ratio shows 1 to 7 for

the five year period. The annual trend of highest daily concentration of OC and EC also reflect

that of PM2.5. The percentages of NO3 (38%), SO4 (11%), and NH4(14%) that made up portion

of PM 2.5 are noticeably high, especially NO3.

In the study by Miguel et. al., (2004) indicates that coarse mode nitrate may result from

the reaction of HNO3 with sea salt and crustal material (Seinfeld and Pandis, 1998). Ammonia

concentrations from a nearby area of large livestock and agricultural activities, contribute with

increased concentrations of nitrate measured by Hughes et al. (2000) downwind in Riverside,

and constitute a major source of precursor of nitrate. The largest effect of nitric-acid produced

coarse nitrate is expected to occur during the summer, due to increased photochemical activity,

(Miguel et.al., 2004).

Salmon et. al., (2004), stated in their Southern California Children’s Health Study of

1999-2001 that highest annual average EC concentrations were found inland at Mira Loma (a

college town in west Riverside County) since 1996. Upland and Riverside were also among the

communities with high EC levels. EC is produced in combustion processes and strongly

influenced by diesel engine exhaust which is prevalent in all of these communities. Mira Loma,

Upland, and Riverside are home to numerous warehouse distribution centers serviced by

thousands of diesel trucks each day, and the study’s conclusion also indicated the EC

concentrations have shown a gradual increase in most sites with the 2001 EC concentrations

being equal to or greater than the level of 1994, (Salmon et. al., 2004).

Kern County (Bakersfield)

Bakersfield’s PM 2.5 annual concentrations do not show a particular annual trend. The

annual average was recorded as 22.29 µg/m3 in 2001, and the highest annual average

concentration is 23.72 µg/m3 in 2002 and the lowest is 18.33 µg/m3 in 2003. It increased to

20.0 µg/m3 in 2004 and decreased to 18.69 µg/m3 in 2005, (Figure 1). The seasonal trend has

been consistent for the entire five year period; high annual PM 2.5 mass concentrations were

recorded in fall and winter and the low ones were recorded in spring and summer of every year,

(Figure 2). All of the highest 24-hour concentrations from every year exceeded the 2005

NAAQS daily standard of 65 µg/m3, with the highest concentration of 87.30 µg/m3 recorded on

February 4, 2002. An average OC to EC ratio is 7 to 1 for the entire five year period, and PM

2.5 consists of 35% of OC and 5% of EC. The increasing and decreasing annual and seasonal

trends of the highest annual concentrations, OC and EC concentrations also reflect the annual

and seasonal trends of PM 2.5. Ammonium, Sulfate, and Nitrate also make up higher percentage

of PM 2.5, with Ammonium and Sulfate have 11% each and 27% for Nitrate. Chen et. al.,

(2007) explained in their study that PM2.5 concentrations varied with elevation. While the

valley floor experienced annual PM 2.5 up to 30 µg/m3, concentrations generally decreased to <

5 µg/m3 at the surrounding coastal, mountain, and desert monitors. For non-urban sites in the

SJV, elevated PM2.5 in late fall and winter was mostly driven by ammonium nitrate (NH4 NO3),

while carbonaceous material exacerbated PM2.5 pollution in urban areas such as Fresno and

Bakersfield, (Chen et. al., 2007).

The seasonality is most pronounced in the San Joaquin Valley Air Basin, where the

November-December-January concentrations were on the order of four to five times greater than

those for March through August, (CEPA-ARB, 2003), (Figure 3). Chen et. al., (2007) also

stated that the contributing factors to PM2.5 in the valley include marine sea salt, fugitive dust,

agriculture—dairy, cooking, secondary aerosol, motor vehicle, and residential wood combustion

(RWC) emissions, with secondary aerosol and RWC accounting for over 70% of PM2.5 mass

during the high PM2.5 period in San Joaquin Valley, (Chen et. al., 2007).

Santa Clara County (San Jose)

San Jose’s annual average of PM 2.5 mass concentration for each and every single year

from 2000 to 2005 is below 18.0 µg/m3. The highest daily concentration is 62.0 µg/m3 recorded

on January 4, 2001, and this met the 2005 NAAQS daily standard during that period, (Figure 1).

The highest annual average concentration is 17.82 µg/m3 for 2002 and started a decreasing trend

every year to 11.95 µg/m3 in 2005. Annual average PM 2.5 mass concentrations are high in fall

and winter and low in spring and summer of every year, (Figure 2). The annual trend maximum

concentrations, OC, and EC concentrations also reflect the PM2.5 annual trend. The average

OC/EC ratio for the six year period is 6 to 1, with 39% of OC and 7% of EC in PM 2.5, (Figure

3). The nitrate also shows 19% in PM2.5, and sulfate has 11%. Ammonium is 6% and almost as

high as EC percentage. The changes in the seasonal and yearly trend could be explained by many

factors. San Jose’s unique local climate also plays an important role. As Fairley and Burch

(2010) described in their report that Bay Area PM2.5 concentrations are highly seasonal because

meteorology varies by time of year (wind speeds and direction, and atmospheric stability in

particular, (Fairley et Burch, 2010).

The seasonal trend that has high PM 2.5 in San Jose is also consistent with the Air

Quality Existing Conditions Report of San Jose by Reyff. Reyff (2009) indicates in his report

that the highest PM2.5 levels in the region are measured in San José. Episodes of high

particulate levels occur in late fall and winter when the Pacific High can combine with high

pressure over the interior regions of the western United States (known as the Great Basin High)

to produce extended periods of light winds and low-level temperature inversions, (Reyff, 2009).

San José is a large urban area with abundant emission sources, it does lie downwind of

other major urban portions of the San Francisco Bay Area, and as a result, emissions from human

activities (primarily traffic) within San José and upwind locations (the Peninsula and central Bay

Area) contribute to air quality problems experienced in San José, (Reyff, 2009). Sources of air

pollution in and around San Jose are primarily traffic or on road-vehicles, and area wide sources

include construction activities, residential wood smoke, off-road travel, and agriculture account

for over 50 percent of PM2.5, (Reyff 2009).

Even though declining, diesel vehicles are one of the on-road sources that directly

emitted PM 2.5 in the Bay area, (ARB Almanac 2003). Like all fuel-burning equipment, diesel

engines produce nitrogen oxides, a common air pollutant in California, and diesel exhaust is still

one of the most widespread and toxic substances in California's air, (OEHHA, 2011). Diesel

PM2.5 is both part of overall PM2.5 and also the Bay Area's major known ambient carcinogen

(OEHHA, 1998; Fairley & Burch, 2010).

Particulate matter can be directly emitted into the air (primary PM) or it can be formed in

the atmosphere from the reaction of gaseous precursors such as NOx, SOx,ROG, and ammonia

(secondary PM). The PM2.5 emission inventory includes only directly emitted particulate

emissions. On an annual average basis, directly emitted PM2.5 emissions contribute

approximately 40 percent of the ambient PM2.5 in the San Francisco Bay Area Air Basin, (ARB

Almanac, 2003).

Figure 1. Daily PM2.5 Mass Concentration from seven California counties (nine sites) between 2000-2005.

Figure 2. Seasonal Averaged PM2.5 Mass Concentration from seven California counties (nine sites) between 2000-2005.

Figure 3. Chemical Composition of Daily PM2.5 Concentrations During Winter 2003-2004 in seven California counties (nine sites).

Health Effects

The percent increase of the mortality categories that were affected by PM2.5

Number of Population in the three Major Mortality Categories (Figure 4a)

The current study results show distinct differences of the three major mortality categories in each

of nine sampling sites of the seven counties. The three major categories which included cardiovascular disease,

respiratory disease, and diabetes showed highest occurrences in Riverside , San Jose, Los Angeles, Fresno,

Sacramento Del Paso Manor, Sacramento T Street, Bakersfield, Escondido, and El Cajon, respectively (Figure

4a). In figure 4a, number of population affected by mortality subcategory, diabetes, is highest, and the

respiratory disease shows the second highest while cardiovascular disease affected population number is ranked

the third. These observations appear to be true for all counties. The sites that have the highest population

absorb the highest impact in terms of a number of people affected. Los Angeles County, for instance, had over

nine million people in year 2000, and with even a 1.250% increase of the cardiovascular disease there would be

476,754 related mortality cases, (Figure 4a and Table 4).

Percent Increase of the three Major Mortality Categories (Figure 4b)

The percent increase of each major mortality category shows differences in each county, but the

contributing percentage of each category is the same with and proportional to other counties. Diabetes

contributes 46% while respiratory disease takes 42%, and cardiovascular disease makes up the rest of 12% of

the total mortality during the study period, (Figure 4b). The results of this study show the same proportion or

percentage throughout the nine sites.

The Percent Increase of other factors under Mortality Categories (Figure 4c & Tables 4a to 4i)

As shown in Figure 4c and Tables 4a to 4i, other specific factors of mortality categories such as

male, females, whites, blacks, Hispanics, in-hospital, out-of-hospital, high school graduates, and non-high

school graduates show different percent increases mortality. There are differences among counties and among

each factor.

. Figure 4c further illustrates the results of the percent increase in each factor of all nine sampling sites of seven

counties. The mortality subcategory that has the highest percent increase in every county is non-high school

graduate. Hispanics, females, whites have the second highest percent increase, and the third highest increase

subcategories are in-hospital, and out-of-hospital; the fourth highest increase is males, and the lowest percent

increase is black subcategory.

Figure 4a. The Number of Population in the three Major Mortality Categories in seven California counties between 2000-2005

Figure 4b. The Percent Increase of the three Major Mortality Categories in seven California counties between 2000-2005

Figure 4c. The Percent Increase of other factors under Mortality Categories in seven California counties between 2000-2005.

Table 4a. The Percent Increase of all Mortality Categories in Bakersfield County between 2000- 2005.

All-cause Cardiovas cular Respiratory Age>65 years

Ischemi c heart disease Diabetes Males Females Whites Blacks Hispanics

In hospital

Out of hospital

High school graduates

Non-high school graduates

1.30% 1.30% 4.77% 1.52% 0.65% 5.21% 1.09% 1.74% 1.74% 0.22% 1.74% 1.30% 1.30% 0.87% 1.95% 0.78% 0.78% 2.84% 0.91% 0.39% 3.10% 0.65% 1.03% 1.03% 0.13% 1.03% 0.78% 0.78% 0.52% 1.16% 1.11% 1.11% 4.08% 1.30% 0.56% 4.46% 0.93% 1.49% 1.49% 0.19% 1.49% 1.11% 1.11% 0.74% 1.67% 1.04% 1.04% 3.81% 1.21% 0.52% 4.15% 0.87% 1.38% 1.38% 0.17% 1.38% 1.04% 1.04% 0.69% 1.56% 0.82% 0.82% 2.99% 0.95% 0.41% 3.26% 0.68% 1.09% 1.09% 0.14% 1.09% 0.82% 0.82% 0.54% 1.22% 0.91% 0.91% 3.32% 1.06% 0.45% 3.62% 0.76% 1.21% 1.21% 0.15% 1.21% 0.91% 0.91% 0.60% 1.36% 1.04% 1.04% 3.81% 1.21% 0.52% 4.16% 0.87% 1.39% 1.39% 0.17% 1.39% 1.04% 1.04% 0.69% 1.56% 1.03% 1.03% 3.78% 1.20% 0.51% 4.12% 0.86% 1.37% 1.37% 0.17% 1.37% 1.03% 1.03% 0.69% 1.54% 0.65% 0.65% 2.40% 0.76% 0.33% 2.62% 0.55% 0.87% 0.87% 0.11% 0.87% 0.65% 0.65% 0.44% 0.98% 0.86% 0.86% 3.14% 1.00% 0.43% 3.43% 0.71% 1.14% 1.14% 0.14% 1.14% 0.86% 0.86% 0.57% 1.29% 1.15% 1.15% 4.22% 1.34% 0.57% 4.60% 0.96% 1.53% 1.53% 0.19% 1.53% 1.15% 1.15% 0.77% 1.72% 0.85% 0.85% 3.11% 0.99% 0.42% 3.40% 0.71% 1.13% 1.13% 0.14% 1.13% 0.85% 0.85% 0.57% 1.27% 0.83% 0.83% 3.04% 0.97% 0.41% 3.31% 0.69% 1.10% 1.10% 0.14% 1.10% 0.83% 0.83% 0.55% 1.24% 0.75% 0.75% 2.76% 0.88% 0.38% 3.01% 0.63% 1.00% 1.00% 0.13% 1.00% 0.75% 0.75% 0.50% 1.13% 0.91% 0.91% 3.34% 1.06% 0.46% 3.65% 0.76% 1.22% 1.22% 0.15% 1.22% 0.91% 0.91% 0.61% 1.37% 1.01% 1.01% 3.72% 1.18% 0.51% 4.06% 0.84% 1.35% 1.35% 0.17% 1.35% 1.01% 1.01% 0.68% 1.52% 0.60% 0.60% 2.21% 0.70% 0.30% 2.41% 0.50% 0.80% 0.80% 0.10% 0.80% 0.60% 0.60% 0.40% 0.91% 0.77% 0.77% 2.81% 0.89% 0.38% 3.06% 0.64% 1.02% 1.02% 0.13% 1.02% 0.77% 0.77% 0.51% 1.15% 0.89% 0.89% 3.26% 1.04% 0.45% 3.56% 0.74% 1.19% 1.19% 0.15% 1.19% 0.89% 0.89% 0.59% 1.34% 0.77% 0.77% 2.81% 0.90% 0.38% 3.07% 0.64% 1.02% 1.02% 0.13% 1.02% 0.77% 0.77% 0.51% 1.15%

Average=> 0.90% 3.31% 1.05% 0.45% 3.61% 0.75% 1.20% 1.20% 0.15% 1.20% 0.90% 0.90% 0.60% 1.35%

Number Affected => 25413 93180 29648 12706 101651 21177 33884 33884 4235 33884 25413 25413 16942 38119

The Percent Increase of the Mortality Categories (El Cajon)

Table 4b. The Percent Increase of all Mortality Categories in El Cajon (San Diego County) between 2000-2005.

All-cause Cardiovascular Respiratory Age>65 years

Ischemic heart disease Diabetes Males Females Whites Blacks Hispanics

In hospital

Out of hospital

High school graduates

Non-high school graduates

0.81% 0.81% 2.96% 0.94% 0.40% 3.23% 0.67% 1.08% 1.08% 0.13% 1.08% 0.81% 0.81% 0.54% 1.21% 0.90% 0.90% 3.31% 1.05% 0.45% 3.62% 0.75% 1.21% 1.21% 0.15% 1.21% 0.90% 0.90% 0.60% 1.36% 0.93% 0.93% 3.42% 1.09% 0.47% 3.73% 0.78% 1.24% 1.24% 0.16% 1.24% 0.93% 0.93% 0.62% 1.40% 1.52% 1.52% 5.57% 1.77% 0.76% 6.08% 1.27% 2.03% 2.03% 0.25% 2.03% 1.52% 1.52% 1.01% 2.28% 0.77% 0.77% 2.81% 0.89% 0.38% 3.06% 0.64% 1.02% 1.02% 0.13% 1.02% 0.77% 0.77% 0.51% 1.15% 0.86% 0.86% 3.17% 1.01% 0.43% 3.46% 0.72% 1.15% 1.15% 0.14% 1.15% 0.86% 0.86% 0.58% 1.30% 1.17% 1.17% 4.28% 1.36% 0.58% 4.67% 0.97% 1.56% 1.56% 0.19% 1.56% 1.17% 1.17% 0.78% 1.75% 0.88% 0.88% 3.21% 1.02% 0.44% 3.50% 0.73% 1.17% 1.17% 0.15% 1.17% 0.88% 0.88% 0.58% 1.31% 0.83% 0.83% 3.03% 0.96% 0.41% 3.30% 0.69% 1.10% 1.10% 0.14% 1.10% 0.83% 0.83% 0.55% 1.24% 0.70% 0.70% 2.58% 0.82% 0.35% 2.82% 0.59% 0.94% 0.94% 0.12% 0.94% 0.70% 0.70% 0.47% 1.06% 0.82% 0.82% 2.99% 0.95% 0.41% 3.26% 0.68% 1.09% 1.09% 0.14% 1.09% 0.82% 0.82% 0.54% 1.22% 1.03% 1.03% 3.77% 1.20% 0.51% 4.11% 0.86% 1.37% 1.37% 0.17% 1.37% 1.03% 1.03% 0.69% 1.54% 0.68% 0.68% 2.48% 0.79% 0.34% 2.70% 0.56% 0.90% 0.90% 0.11% 0.90% 0.68% 0.68% 0.45% 1.01% 0.76% 0.76% 2.79% 0.89% 0.38% 3.05% 0.63% 1.02% 1.02% 0.13% 1.02% 0.76% 0.76% 0.51% 1.14% 1.04% 1.04% 3.81% 1.21% 0.52% 4.16% 0.87% 1.39% 1.39% 0.17% 1.39% 1.04% 1.04% 0.69% 1.56% 0.81% 0.81% 2.99% 0.95% 0.41% 3.26% 0.68% 1.09% 1.09% 0.14% 1.09% 0.81% 0.81% 0.54% 1.22%

Average=> 0.91% 3.32% 1.06% 0.45% 3.63% 0.76% 1.21% 1.21% 0.15% 1.21% 0.91% 0.91% 0.60% 1.36%

NumberAffected=> 25504 93513 29754 12752 102015 21253 34005 34005 4251 34005 25504 25504 17002 38255

The Percent Increase of all Mortality Categories (Escondido)

Table 4c. The Percent Increase of all Mortality Categories in Escondido (San Diego County) between 2000-2005.

Table 4d. The Percent Increase of all Mortality Categories in Fresno County between 2000-2005.

Table 4e. The Percent Increase of all Mortality Categories in Los Angeles County between 2000-2005.

Table 4f. The Percent Increase of all Mortality Categories in Riverside County between 2000-2005.

Table 4g. The Percent Increase of all Mortality Categories in Sacramento Del Paso Manor(Kern County) between 2000-2005.

Table 4h. The Percent Increase of all Mortality Categories in Sacramento T Street(Kern

County) between 2000-2005.

Table 4i. The Percent Increase of all Mortality Categories in San Jose (Santa Clara County) between 2000-2005.

Discussion

The results show unique characteristics of PM2.5 and potential contributing sources of

pollutants in each selected county. Additionally, the percent increase in mortality categories

from each county also reflect on its PM2.5 mass concentrations and yearly and seasonal trends of

PM2.5 mass concentrations. However, the potential contributing sources were mentioned or

identified in each county, but further studies would be necessary to specifically link or connect

those sources to specific health effects. The unique characteristics of PM2.5 are referred to the

chemical compositions as shown on figure 3 (Chemical Composition of Daily PM2.5

Concentrations during Winter 2003-2004 in seven California counties (nine sites).). Elemental

carbon, organic carbon, ammonium, sulfate, nitrate, potassium ion, sodium ion, crustal oxides,

metal oxides, metal oxides, and some unknown elements made up of PM2.5 mass. As shown on

figure 3, each county has different level concentration of each of these chemicals depending on

the level of the quantity of pollutants being released to each county environment. These different

levels are distinctively apparent and unique in each county even though there are some similar

seasonal trends in occurrences.

Riverside County had the highest average PM2.5 mass concentrations (26.37 µg/m3) in

the five year period (2001 to 2005). Riverside county is ranked number four on the most

polluted county by Short-term Particle Pollution (24-hour PM2.5) and ranked third on the most

polluted by Year-Round Particle Pollution (Annual PM2.5), ( ALA, 2010). Ammonia, nitrate,

and elemental carbon contributed to elevated PM2.5 in Riverside county. Riverside is located

downwind and receptive to coarse mode nitrate from the reaction of ammonium nitrate with sea

salt and crustal materials. Ammonium concentrations from nearby area of large livestock and

agricultural activities constitute a major source of precursor nitrate. Therefore higher

concentration of nitrate occurred in the area. The most significant sources of air pollution in

Riverside County were the diesel combustion processes from commercial diesel trucks prevalent

in the area.

The model was useful in this study for its application for assessing, comparing, and

predicting the percent increase of each mortality category for each county. Once the results were

formulated and calculated using the model, the information is useful in indicating the PM2.5

mass concentrations in the affected areas. The model would also be useful and applicable in the

areas where similar PM2.5 compositions exist. As mentioned earlier in the paper, the model was

also used by Pope & Dockery (2002) to study PM2.5 in other states.

Limitations

Limitations of this study include the specific and rigidity of the raw data set availability,

and there are many more implicated complexities in retrieving and analyzing the data sets. The

data sets were retrieved and applied as they were, and there was no possible to improve the

retrospective sampling results. These came with difficulties such identifying and confirming the

methodology, standard operating procedures of sampling sources, missing data in certain period

of time. This study is restricted by the validity and reliability of the secondary data available

from the California Air Resources Board. The sampling locations and sources of particulate

matter were determined by official government approval.

Another limitation of the study is application of the mathematical model. Even though

the model used by other authors who published many ground breaking air pollution subjects the

model could only estimate, and it was used with an estimated assumptions that all other

background parameters were similar.

The other limitation of the study is the constant and the sample population in the

mathematical model may not be representative of the actual population in those nine sites of the

seven California counties. California Fine Particulate data sets that Ostro and colleagues used in

the math model as well as sources of mortality category data sets did not include the Asian

population in the seven California counties even though there were 12.39% of total California in

2000 and 13.49% in 2005 (United States Census, 2000).

Conclusion

The objectives of this study were intended to, 1).explore the characterization of the

particulate matter (PM 2.5) in seven California counties at nine sampling sites, 2). Identify the

potential contributing sources of PM2.5 in each county, and 3). Model the health effects based

the three major mortality categories. PM2.5 was characterized by analyzing their daily mass

concentrations by plotting data points yearly and seasonally. Other chemical compositions that

made up PM2.5 mass were also plotted and analyzed. The results of the analysis show yearly

and seasonal trends that were also indicative of the potential releasing or contributing sources.

The seasonal trends of elevated PM2.5 mass concentrations showed the reflection or coincide

with the contributing sources of pollutants such fuel combustion from wood and agricultural

burning, vehicle exhausts, industrial fuel combustion, and other contributing sources in each

county were researched and described. Using a mathematical model from other researchers

(Ostro and colleagues 2006), the raw data, specifically the PM2.5 average daily concentrations

were calculated, analyzed and compared with each county. The completed and calculated

results show the counties with the highest to lowest elevation of daily averaged PM2.5 mass

concentrations and the most increased to the least of the three major mortality categories in each

county. This study was not intended to study cause and effect between particulate matter

pollutants and adverse health effects, but the findings could provide the links to useful sources,

data, information and a useful model that could be used to further study air pollutants.

Air pollution is a significant contributor to leading causes of death in California. Further studies

should investigate means to monitor and reduce fine particulate matter in the air.

References

ALA, (2005). American Lung Association: State of the Air Annual Report 2005. Accessed on June 15, 2011 from http://www.lungusa.org/assets/documents/publications/state-of-the-air/state-of-the-air- report-2005.pdf

ALA, (2010). American Lung Association: State of the Air Annual Report 2010 Accessed on June 15, 2011 from http://www.lungusa.org/assets/documents/publications/state-of-the-air/state-of-the-air- report-2010.pdf

Anderson HR, Atkinson RW, Peacock JL, Sweeting MJ, Marston L. 2005. Ambient particulate matter and health effects. Publication bias in studies of short-term associations. Epidemiology 16:155–163.

ARB Almanac, (2003). http://www.arb.ca.gov/aqd/almanac/almanac03/pdf/chap403.pdf San Diego Air Pollution Control District-SDAPCD, (2005). Measures to Reduce Particulate Matter in San Diego County. June 19, 2011. Available at : http://www.sdapcd.org/planning/PM-Measures.pdf

Bell, M.L., Dominici, F., Ebisu, K., Zeger, S.L., & Samet, J. M. (2007). Spatial and Temporal Variation in PM2.5 Chemical Composition in the United States for Health Effects Studies. Environmental Health Perspectives, 115(7), 989-995.

Blanchard C. 2003. Spatial and temporal characterization of particulate matter. In: Particulate Matter Science for Policy Makers: A NARSTO Assessment (McMurry PH, Shepherd MF, Vickery JS, eds). Cambridge, UK: Cambridge University Press, 191–231.

California Air Resources Board ( CAARB), 2010. Staff Report on the San Joaquin Valley Smoke Management Program and Consideration of Modifications to Agricultural Burning Requirements

California Environmental Protection Agency- Air Resources Board (CAEPA-ARB), ( 2010). Staff Report on the San Joaquin Valley Smoke Management Program and Consideration of Modifications to Agricultural Burning Requirements. June 19, 2011. Available: http://www.arb.ca.gov/smp/district/arbstaffreport.pdf

California Air Resources Board, CAARB/CAEPA, (2010). Estimate of Premature Deaths Associated with Fine Particle Pollution (PM2.5) in California Using a U.S. Environmental Protection Agency Methodology. Accessed on July 15, 2011 from http://www.arb.ca.gov/research/health/pm-mort/pm-report_2010.pdf

Chen et. al., (2007). Quantifying PM2.5 Source Contributions for the San Joaquin Valley with Multivariate Receptor Models. Environmental Science Technology, 41, 2818-2826. Retrieved from National University online Request on 12 June 2011.

Chow et. al, (2003). Similarities and differences in PM10 chemical source profiles for geological dust from the San Joaquin Valley, California. Atmospheric Environment 37 (2003) 1317–1340. June 2011. Available at: http://www.arb.ca.gov/ei/speciate/r01t20/rf13doc/Ref%2013%20- %20Chowetal2003AESJVFugDustProfiles.pdf

Delfino, R. J., Sioutas, C., and Malik, S. (2005). Potential Role of Ultrafine Particles in Associations between Airborne Particle Mass and Cardiovascular Health. Environmental Health Perspectives, 113(8), 934-946.

Environmental Protection Agency –EPA, (2011). PM2.5 NAAQS Implementation. Accessed on July 12, 2011 at http://www.epa.gov/ttnnaaqs/pm/pm25_index.html

Fairley, D. et., Burch, D., (2010). Multi-Pollutant Evaluation Method Technical Document. Bay Area Air Quality Management District. April 2010

Fenn et. al.,(2011). Nitrogen Emissions, Deposition, and Monitoring in the Western United States Accessed on 06-12-2011 from http://public.wsu.edu/~forda/CE%20543/N %20Paper%201.pdf

Green, L.C., Crouch, E.A.C, Ames, M.R., and Lash, T.L. (2002). What’s Wrong with the National Ambient Air Quality Standard (NAAQS) for Fine Particulate Matter (PM2.5)? Regulatory Toxicology and Pharmacology 35, 327–337. doi:10.1006/rtph.2002.1548

Jerrett, M., Burnett, R. T., Ma, R., Pope III, C.A., Krewski, D., Newbold, K. B., …Thun, M. J. (2005). Spatial Analysis of Air Pollution and Mortality in Los Angeles. Epidemiology, 16(6), 727-736.

Karmel, P.E., & FitzGibbon, T.N. (2002). PM2.5: Federal and California Regulation of Fine Particulate Air Pollution. Bryan Cave LLP. http://www.bryancave.com/files/Publication/7e0edf14-dd55-4ac5-8975-452a79ffe3e0/ Presentation/PublicationAttachment/185ff3d4-77c8-4f6b-91c2-47bb435fd129/Env %20Article-KarmelEnvLawinCA8-02.pdf

Kwangsam et. al., (2004). Primary and secondary carbonaceous species in the atmosphere of Western Riverside County, California. Atmospheric Environment 38 (2004) 1345–1355. http://cichlid.cert.ucr.edu/research/pubs/na-22.pdf

Miguel et. al., (2004). Seasonal variation of the particle size distribution of polycyclic

aromatic hydrocarbons and of major aerosol species in Claremont, California.Atmospheric Environment 38 (2004) 3241–3251

Office of Environmental Health Hazard Assessment (OEHHA), 1998. Health Effects of Diesel Exhaust. June 19, 2011 available at: http://oehha.ca.gov/public_info/facts/dieselfacts.html

Ostro, B., Roth, L., Malig, B., and Marty, M. (2009), The Effects of Fine Particle Components on Respiratory Hospital Admissions in Children. Environmental Health Perspectives, 117(3), 475-480.

Ostro, B., Broadwin, R., Green, S., Feng, W., & Lipsett, M. (2006). Fine Particulate Air Pollution and Mortality in Nine California Counties: Results from CALFINE. Environmental Health Perspectives, 114(1), 29-33.

Peters, A., Dockery, D. W., Muller, J. E., & Mittleman, A. M.,, (2001). Increased Particulate Air Pollution and the Triggering of Myocardial Infarction. Journal of the American Heart Association. CCirculation 2001, 103:2810-2815

Pope III, C. A., Burnett, R. T., Thun, M.J., Calle, E.E., Ito, K., & Thurston, G.D. (2002). Lung Cancer, Cardiopulmonary Mortality, and Long-term Exposure to Fine Particulate Air Pollution. JAMA, 2002;287:1132-1141. Downloaded from jama.ama-assn.org on July 17, 2011.

Reyff, J.A., (2009). Air Quality Existing Conditions Report San Jose, CA. June 12, 2009. June 19, 2011. Available at: http://www.sanjoseca.gov/planning/gp_update/documents/Existing_AirQuality061209.pd f

Rhinehart et. al., (2006). Spatial distribution of PM2.5 associated organic compounds in central California. Atmospheric Environment, Volume 40, Issue 2, January 2006, Pages 290-303

Sacramento Metropolitan Air Quality Management District, SMAQD, (2005). Report to Community. (Page 9). Accessed on June 18, 2011 from http://www.airquality.org/communicationsoffice/2005/2005ReportToCommunity.pdf

Salmon et. al., (2004). Determination of Elemental Carbon and Organic Carbon Concentrations During the Southern California Children’s Health Study, 1999-2001 Final report prepared for the California Air Resources Board and the California Environmental Protection Agency March, 2004

Scarborough et. al., (2002). Creating a Statewide Spatially and Temporally Allocated

Agricultural Burning Emissions. ARB Contract Number: 99-71. CAMFER LabUniversity of California, Berkeley June 19, 2011. Available at: http://www.arb.ca.gov/research/apr/past/99-714.pdf

Sacramento Metropolitan Air Quality Management District, SMAQD, (2006). Staff Report: 2006 PM2.5 Standard: Evaluating the Nine Factors in Setting Nonattainment Area Boundaries for the Sacramento Region. (Pages 7, 13) November 19, 2007 http://www.airquality.org/plans/federal/pm/PM2.5/PMDesignationFinal- AQMDStaffReport.pdf

Schauer, J.J. & Cass, G.R., (2000). Source appointment of wintertime gas-phase and particle- phase air pollutants using organic compounds as tracers Environ Sci Technol 34(9):1821–1832.

Sioutas, C., Delfino, R.J., and Singh, M. (2005). Exposure Assessment for Atmospheric Ultrafine Particles (UFPs) and Implications in Epidemiologic Research. Environmental Health Perspectives, 113(8), 947-955. Accessed on July 20, 2011 from http://www.ncbi.nlm.nih.gov/pubmed/16079062

United States Environmental Protection Agency, USEPA, (2005), Volume I Quality assurance project plan chemical speciation of PM 2.5 filter samples (RTI/08858/12-01S) published on September 27, 2005 http://www.epa.gov/ttnamti1/files/ambient/pm25/spec/rtiqapp05.pdf

Wagner et. al., (2007). Daily Patterns in PM2.5 Ion Concentrations in Claremont, California, The Journal of Young Investigators, 17 (2) retrieved 12 June 2011 from http://www.jyi.org/research/re.php?id=1181

Watkinson, W. P., Campen, M. J., and Costa, D. L. (1998). Cardiac Arrhythmia Induction after Exposure to

Residual Oil Fly Ash Particles in a Rodent Model of Pulmonary Hypertension. ToxicoL ScL 41, 209-216.

Watson, J., Chow, J., & Chen, A., (2007). Assessment of Carbon Sampling Artifacts in the IMPROVE, STN/CSN, and SEARCH Networks. Available on July 31, 2011 at http://www.dri.edu/images/stories/editors/eafeditor/Watsonetal2007EPAReportQuartzArt ifactCArbon.pdf

Ying, Q., (2011). Physical and Chemical Processes of Wintertime Secondary Nitrate Aerosol Formation. Zachry Department of Civil Engineering, Texas A&M University, College Station, TX 77945, USA June 19, 2011. Accessed from: https://ceprofs.civil.tamu.edu/qying/paper_v3_FESE.pdf