&EPA
           United States
           Environmental Protection
           Agency
           Office of Air Quality
           Planning and Standards
           Research Triangle Park NC 27711
EPA-450/2-80-078
August 1980
           Air
Evaluation of Contribution
of Wind Blown Dust From
the  Desert to Levels of
Particulate Matter in
Desert Communities

-------
                                        EPA-450/2-80-078
 Evaluation of Contribution  of Wind  Blown
Dust From  the Desert to Levels of Participate
         Matter in Desert Communities
                            by

                    Frank A. Record and Lisa A. Baci

                        GCA Corporation
                      Bedford, Massachusetts
                      Contract No. 68-02-2607

                         Task No. 41
                         Prepared for

                U.S. ENVIRONMENTAL PROTECTION AGENCY
                    Office of Air, Noise, and Radiation
                 Office of Air Quality Planning and Standards
                Research Triangle Park, North Carolina 27711

                         August 1980

-------
This report is issued by the Environmental Protection Agency to report technical data of
interest to a limited number of readers.  Copies are available -  in limited quantities - from
the Library Services Office (MD-35), U.S.  Environmental Protection Agency, Research
Triangle Park, North Carolina 27711; or, for a fee, from the National Technical Infor-
mation Service, 5285 Port Royal Road, Springfield, Virginia 22161.
                       Publication No. EPA-450/2-80-078
                                       11

-------
                                  ABSTRACT
     This report uses existing data and studies to assess the impact of wind-
blown desert dust on the attainment of TSP standards in major cities situated
in desert environments in the Southwestern U.S.  Primary emphasis is placed
on four cities:  Phoenix and Tucson, Arizona; and Las Vegas and Reno, Nevada.
It is concluded that: (1) the contribution of wind-blown dust from the
undisturbed desert floor to particulate levels in desert communities is very
small and should be considered as part of the background; (2) if human activi-
ties repeatedly break up the desert crust, local violations of the 24-hour
standards are likely; (3) there is substantial agreement on the principal
source categories of fugitive dust contributing to the nonattainment problem
and the characteristics of urban areas most affected by each category.

     It is recommended that additional field programs be carried out to define
the level and sources of the inhalable fraction of suspended particulates.
                                     iii

-------
                                  CONTENTS
Abstract
Figures  	  v
Tables	viii

     1.   Introduction and Summary 	  1
               Major Findings	1
               Conclusions and recommendations 	  7
     2.   Analytical Techniques  	  8
               Introduction  	  8
               Analysis of temporal or spatial patterns of suspended
                 particulates  	  8
               Analysis of particulate emissions 	  9
               Chemical, elemental, and morphological analysis 	  9
               Meteorology and suspended particulates  	 10
               Overview	11
     3.   Background Suspended Particulates  	 13
               Background concentrations 	 13
               Characteristics and sources 	 20
     4.   Urban Particulates Levels  	 32
               Particulate levels in and around Las Vegas  	 32
               Particulate levels in and around Reno	38
               Particulate levels in and around Tucson 	 43
               Particulate levels in and around Phoenix  	 52
     5.   Urban Particulates:  Characteristics and Sources 	 60
               Particle size distributions 	 60
               Particulate sources 	 63
               Other studies	80

References	92

Appendix

     A.   Background Station Information 	 95
                                     iv

-------
                                   FIGURES

Number                                                                   Page

   1   Approximate bounds of study area 	     2

   2   1976 annual geometric mean TSP concentrations from National
         Aerometric Data Bank	    15

   3   1976 geometric mean TSP concentrations observed by Four
         Corners Ambient Air Monitoring Network operated by
         Ute Research Laboratories  	    16

   4   Frequency distribution of annual geometric mean TSP
         concentrations at 45 rural and remote sites in
         Southwestern U.S.  	    18

   5   1976 second-highest 24-hour TSP concentrations from
         National Aerometric Data Bank	    21

   6   1976 second-highest 24-hour TSP concentrations observed
         at Four Corners Ambient Air Monitor Network operated
         by Ute Research Laboratories	    22

   7   Frequency distributions of 24-hour TSP concentrations at
         45 rural and remote sites in Southwestern U.S	    23

   8   Volume distributions observed in the Mojave Desert, California,
         and Fort Collins, Colorado	    25

   9   Particle settling/suspension regimes 	    26

  10   Volume distribution as a function of height  	    28

  11   Percentage distribution of measured chemical species on
         respirable particles in the atmosphere at the
         Research Ranch 	    31

  12   Crustal enrichment factors of species measured on
         respirable particles in the atmosphere at the
         Research Ranch 	    31

  13   The  Clark County APCD sampling network for TSP	    33

  14   Annual mean TSP concentrations versus distance from
         urban core   	    35

                                      v

-------
                            FIGURES (continued)
Number                                                                   Page

 15    Plots of 24-hour TSP concentrations at 12 sites ordered
         by distance from city center (Las Vegas) ...........   37

 16    The Washoe County DEP sampling network for TSP .........   39

 17    Spatial distribution of TSP concentrations and land
         use in and around Reno ....................   41
 18    TSP monitoring network in Tucson and vicinity  .........   44

 19    Annual geometric mean TSP concentrations in
         and around Tucson in 1977  ..................   47

 20    Average TSP concentration versus distance from the
         urban core  ..........................   48

 21    Hi-vol monitoring sites within the Phoenix area  ........   53

 22    Expected annual geometric means in yg/m3 ............   56

 23    Expected maximal 24-hour concentrations in ug/m3  ........   57

 24    Effect of wind speed on ambient suspended particulate
       levels  ............................   59

 25    Average daily dust emissions from unpaved roads,  1975  .....   68

 26    Average daily street dust emissions entrained by motor
         vehicles on paved streets, 1975  ...............   69

 27    Average daily dust emissions from construction
         activities, 1975 .......................   70

 28    Emissions of fugitive dust arising from wind erosion,
         average daily emissions, 1975  ................   71

 29    Fugitive dust emissions arising from wind erosion of dis-
         turbed soil surfaces, first quarter, daily average, 1975   .  .   73

 30    Particulate fugitive emissions arising from wind erosion of
         undisturbed desert in Phoenix area, first quarter, daily
         average, 1975  ........................   74

 31    Dendogram of feature clustering for desert urban
         particulate matter ......................   77

 32    Dendogram of feature clustering for desert background
         particulate matter ......................   78
                                      vi

-------
                             FIGURES (continued)
Number                                                                   Page




  33     Enrichment factors for species in desert background and
34
35
36
37
A-l
A-2
1975 Las Vegas particulate emission density, ton/mi2/yr . . .
Location of 25 TSP monitoring stations shown in the

Monitor locations for National Aerometric Data Bank data . .
Monitoring site locations 	
83
84
87
90
98
100
                                     vii

-------
                                   TABLES

Number                                                                   Page

   1   Primary Techniques Used by Studies Relied on in this
         Report ............................   I2

   2   TSP Concentrations at Selected Background Sites in
         New Mexico, Arizona, and Nevada  ...............   1?
   3   TSP Concentrations at Selected Background Sites in
         Arizona  ...........................   19

   4   Estimated Background TSP Concentrations for Selected
         AQCR's in Colorado and Utah  .................   19

   5   Site Location Descriptions for Las Vegas Monitors  .......   34

   6   Number of Exceedences of the Secondary Standards and
         Emission Densities at Selected Sites .............   38

   7   Site Location Descriptions for Reno Monitors ..........   40

   8   TSP Concentrations in Reno and Vicinity on Days when
         the Secondary Standard was Exceeded  .............   42

   9   Site Location Descriptions for Tucson Monitors .........   45

  10   TSP Concentrations in Tucson and Vicinity on Days when
         the Secondary Standard was Exceeded in 1977  .........   49

  11   Wind Conditions on Days with the Greatest Number of
         Violations in Tucson .....................   51

  12   Site Location Descriptions for Phoenix Monitors  ........   54

  13   Average TSP Concentrations for Six Site Environments ......   58

  14   Height Variation of Particulate Concentrations from
         Andersen Sampler Data  ....................   61

  15   Particle Size Distribution for Suspended Particulates
         Measured in Phoenix, September 27 and November 14, 1975  ...   61

  16   Particle Size Information from Phoenix Sampling Program  ....   62

                                    viii

-------
                             TABLES (continued)

Number                                                                   Pago

  17   TSP Concentrations Measured by Hl-voln ;m
-------
                                   SECTION 1

                           INTRODUCTION AND SUMMARY
     Several large cities  located  in  the  southwestern United  States have not
yet attained the primary and/or  secondary national  ambient air quality stan-
dards  (NAAQS) for total suspended  particulates.  These cities are located in
desert environments;  their major economic activities tend to  be tourism, agri-
culture, or provision of services,  rather than heavy manufacturing.  As would
be expected, studies  of several  of these  cities have shown that these areas
experience relatively little pollution  from  industrial point  sources; however,
widespread fugitive dust problems  do  exist.

     At present, there are differing  opinions concerning the  contributions of
fugitive dust from human activities and of wind-blown dust from the surrounding
desert to this nonattainment problem.   To help resolve this issue, GCA was
asked  to review existing data  and  studies and assess the impact of wind-blown
desert dust on particulate levels  within  such cities.

     Figure 1 indicates the approximate study area  considered.  It includes
the semi-arid regions of the southwestern states and extends  from the Sacra-
mento and San Joaquin valleys  in California  to central Texas  and Oklahoma.
Although information  was reviewed  for nonattainment sites throughout the
region, studies of Phoenix and Tucson,  Arizona, and Las Vegas and Reno, Nevada
contained the most useful  information for the purposes of the required assess-
ment.  For this reason, this report focuses  on these four cities.

     The report is organized into  five  sections.  This section provides an
introduction, summarizes the major findings, and presents conclusions and
recommendations.  Section  2 describes the principal analytical techniques used
in the studies reviewed.   Section  3 summarizes annual and 24-hour concentration
data from remote and  rural sites throughout  the southwest.  It also discusses
chemical and physical  properties of background particulates.  Section 4 exam-
ines the distribution of particulates in  and around the four  study cities and
compares the distributions with  patterns  of  land usage and human activities.
Section 5 discusses characteristics and sources of urban particulates, focus-
ing on the contribution of wind-blown desert dust.  It includes a summary of
particle size measurements that  emphasize the inhalable size  fraction.

MAJOR FINDINGS

     The authors of this report  have reviewed numerous published documents
and some recent air quality data in an  attempt to define the  contribution of
wind-blown desert dust, and its  inhalable fraction, to the TSP nonattainment
problem in large cities set in desert surroundings.  From this review, a

-------
Figure 1.  Approximate bounds of study area.

-------
concensus emerges that the part played by dust blown in from the undisturbed
desert is very small, and can safely be ignored when developing control strat-
egies.  With the principal exception of contributions during widespread dust
storms - one of the occasional natural phenomena the effects of which may be
disregarded in control strategy development - its impact is contained within
the low background levels of approximately 25-30 yg/m3 found throughout the
Southwest.  When major duststorms do occur, they are fed principally by emis-
sions from disturbed surfaces.  These emissions increase rapidly near urban
areas as a result of increasing construction, agriculture, and similar activ-
ities.  Severe duststorms occur perhaps once or twice a year in the study
area during periods of sustained high winds; their frequency can be much
higher in the agricultural regions of the Great Plains, particularly during
drought conditions.

     In this summary discussion, the major findings of the reviewed reports
have been organized around the four principal analytical techniques described
in Section 2.  Although the findings are sometimes taken directly from the
reports, they more often represent a condensation and integration of ideas
by the GCA reviewers.  The description of TSP and land use relationships is
based on an analysis carried out by GCA for this report.  Emphasis is placed
on the interpretation of the air quality data; information provided by the
remaining techniques is considered supportive, but not conclusive.  Through-
out the summary, reference is made to more complete discussions of the various
findings provided within the body of the report.

     The particle size data that were available were too limited and incon-
clusive to be of much help in defining the inhalable particulate levels
either in the desert or within the urban areas.

TSP and Land Use Relationships

     The analysis of TSP concentrations in and around Phoenix, Tucson, Las
Vegas and Reno presented in Section 4 shows a close relationship between the
spatial distribution of concentration and land use and its associated human
activities.  Generalizing, annual geometric mean concentrations increase
steadily as one progresses from remote and rural regions to the urban core.
The increase in concentration roughly parallels an increase in the emission
density of fugutive dust.  In leaving the remote area one first encounters
areas with increasing amounts of dust generated by traffic on unpaved roads
and by agricultural activities.  Pockets of urbanization, both residential
and commercial, are reached next with an accompanying increase in vehicular
traffic; construction activities become significant sources.  In drawing
closer to the urban core, the density of residential and commercial property
continues to increase, and with it vehicular associated emissions, including
direct tail pipe emissions and entrained street dust.

     The overall effect of changing land use patterns and increasing emission
density on concentration is shown by the following average values derived
from the concentration fields around the four cities:

-------
                 Site Characteristics          Average Concentration

           Remote or  rural                           30 pg/m3
           Suburban/residential,  commercial          81 yg/m3

           Center city/commercial                  124 yg/m3

 Considerable variation  in  concentration  does  occur within  areas  having the
 same general land use classification, however,  as  a result of  differences in
 the intensity of the associated  human activities plus differences  in small-
 scale site characteristics.   For example, a direct relationship  exists between
 the average center city concentration found within each  of the four  cities and
 the city's population,  as  shown  below:

                                  Reno    Las Vegas*   Tucson    Phoenix

      Population                   82,500    201,300     301,200   684,500

      Mean  concentration, ug/m3       77         100         133        184

 This relationship is further  evidence that the  particulate emissions  leading
 to nonattainment within the cities are generated by urban  activity.

      The overall behavior of  24-hour concentrations  in and around  the  urban
 areas agrees  closely with that of the annual averages; that is, violations of
 the secondary standard  rarely occur in the outlying  areas  surrounding  the
 cities, but  do occur with increasing frequency  as  the proportion of disturbed
 soil surfaces and the amount of  vehicular traffic  increases.   Although  details
 differ from  city to  city due  to  differences in  the  physical setting of  the
 cities and in the distribution of urban activities,  the occurrence of  high
 concentrations typically fall into two patterns.  One pattern is associated
 with light winds  and low mixing  heights which severely limit the dispersion
 of  particulates.   Under these conditions, wind-generated emissions are  absent
 and high concentrations develop within the central  city and the immediately
 adjacent area where human activity peaks.  Under prolonged stagnating  condi-
 tions, high concentrations become more widespread.

      The second pattern is associated with strong winds during which wind-
 generated fugitive dust plays a  significant and sometimes dominant role.
 Such  periods  are of relatively short duration (hours rather than days).  Maxi-
mum  concentrations on strong wind days are generally found in areas with the
 greatest amount of exposed, disturbed surface.  In  the case of Phoenix, the
 largest urban area studied, the observed maximal 24-hour concentrations (1973-
 1975)  at sites classified as rural/residential and surrounded by fugitive
 sources were approximately twice the maximal values observed within the cen-
 tral  city.

     One additional feature of the air quality data is of major significance.
Site by site  comparisons of annual and 24-hour concentrations observed  either
within or near the urban core of the larger cities disclose large differences.
 Including North Las Vegas.

-------
Some of the differences among proximate sites are attributable to differences
in monitor height; others are judged to be a reflection of the distribution
of local fugitive dust sources.  For example, the 1977 annual geometric means
at four Tucson sites located within a few miles of one another were 122, 158,
68 and 119 ug/m3; the heights of the monitors were 25, 15, 80 and 18 feet,
respectively.  The large concentration gradient shown in the vertical is in-
dicative of ground-level sources of large particulates with relatively short
horizontal travel distances.  The difference of 39 yg/m3 between two sites
whose heights differ by only 3 feet is most easily attributed to differences
among local sources.  These differences are also reflected in the 24-hour
measurements at the two sites.  Out of 57 sampling days In 1977, the number
of exceedances of the secondary standards at the site with a mean concentra-
tion of 119 yg/m3 was 12, while 35 exceedances were observed at the site with
a mean of 158 yg/m3.

     Details of these analyses are given in Section 4,

Emission Inventories and Modeling

     Emission inventories for particulates have been carried out throughout
the Southwest.  These inventories vary in detail from county- or AQCR-wide
estimates to microinventories around specific nonattainment sites.  The inven-
tories have been used in dispersion models of different degrees of complexity
in some of the urban areas to calculate the impact of specific categories of
emissions on TSP levels.  All of these studies are consistent in stressing
the dominant role played by fugitive dust emissions from unpaved roads, reen-
trained street dust, construction activity and wind erosion from agricultural
fields and other disturbed surfaces.  Although total emissions from the undis-
turbed desert are estimated to be substantial, they are distributed over large
areas outside of the urban centers and therefore have little impact on the
urban monitors.  Conclusions from several of these studies follow.

     An emission inventory and modeling study covering the Phoenix area was
carried out by TRW.  TRW concluded that nearly all of the TSP above background
was caused by emissions from unpaved roads, entrained street dust, construc-
tion activities, and wind erosion from agricultural fields, unpaved roads,
undisturbed desert, tailings piles, vacant lots and the like at 12 of the 13
sites studied.  Off-road vehicles were considered to be primarily responsible
for violations at the 13th site.  The sites most affected by wind erosion were
in rural areas under development.  The impacts of the various classes of wind-
erosion sources were not disaggregated in the model results as presented.
Emissions from disturbed soil surfaces were highly concentrated in the urban
portion of the region.  In a separate study which developed an emissions in-
ventory for Maricopa County, PEDCo found that wind erosion from agricultural
land was the biggest single contributor to county-wide emissions, followed by
unpaved roads.  From another modeling study for the Phoenix area, carried out
by the University of Texas, it was concluded that the high concentrations of
particulates experienced during late fall and winter periods of atmospheric
stability are associated with local fugitive dust sources.  Advective trans-
port of dust from the countryside to the urban areas was judged to be
unimportant.  Details of these three studies can be found in Section 5, pages
65, 75, and 64, respectively.

-------
     In a study of the Las Vegas and Reno areas, PEDCo developed gridded emis
sion inventories.  These inventories showed that emissions in both cities are
concentrated in the more urbanized portions of the study areas.  PEDCo con-
cluded that approximately 80 percent of the total emissions in both cities
comes from fugitive dust sources.  Unpaved roads were estimated to be the
largest contributor, followed by natural surfaces and paved roads.  This study
is discussed more fully in Section 5, p. 80.

     PEDCo also studied 35 nonattainment sites in Colorado and Utah, none of
which were in rural locations (see Section 5, p. 85).  PEDCo concluded that the
principal sources of particulates in large and moderate sized urban centers
are traffic related.  These include reentrained street dust, winter road
sanding, and motor vehicle exhaust.  In medium and small urban centers, how-
ever, the major sources were found to be point sources, unpaved roads, con-
struction, and wind-blown dust from agricultural fields and other open areas.

     One other related study was reviewed.  In this study, Engineering-Science
investigators developed microinventories around five nonattainment sites in
Albuquerque and modeled each site (see Section 5, p. 85).  They concluded
that from 42 to 61 percent of the particulates at these sites were from un-
paved roads or driveways, paved roads, fire and exhaust emissions, and un-
paved parking lots.

Chemical, Elemental, and Morphological Analysis

     Chemical and elemental analyses coupled with correlation and enrichment
factor techniques showed that approximately 50 percent of the suspended par-
ticulate matter in the Tucson area was composed of soil.  The values at 11
sites ranged from 48 to 83 percent, with the lowest estimate being obtained
at the background site.  It was not clear that the percentage of soil material
was significantly greater in the urban area than at the background site, but
even if it were not, the much greater absolute TSP concentrations within the
city imply major contributions from urban soil sources.  Unfortunately, the
analytical techniques cannot distinguish between wind-blown desert soil and
soil suspended as the result of human activities such as travel on unpaved
roads •  A description of this analysis begins on p. 72 of Section 5.

     Microscopic analyses performed on particulate samples collected in
Phoenix led the participants in another field study (see Section 5, pp. 63-
64) to conclude that the particles were not typical of wind-eroded materials.
It was concluded from this study that vehicular traffic, especially on un-
paved roads, was the primary generator of TSP within the Phoenix area, and
that long-range transport of aerosols by winds from the surrounding deserts
was only a minor source.
     The meteorological parameters that most often correlate significantly
with TSP levels are precipitation and wind speed, although the effect of wind
speed is confounded by two opposing phenomena-   When emissions remain con-
stant, concentrations decrease with increasing wind speed due to increased

-------
dilution; on the other hand, once wind speed reaches a critical level, fur-
ther increases lead to an increasing amount of wind-generated dust.  In one
of the reports reviewed (see Section 5, pp. 86-92) meteorological relation-
ships were used as the basis for estimating the contribution of wind-blown
dust (not disaggregated by source type) to TSP nonattainment in EPA Region
VI.  The importance of wind-blown dust was judged by the degree to which
high TSP concentrations were associated with high winds and low precipitation.
Using the AID decision-tree program as the analytical tool, .it was decided
that wind-blown dust was the primary contributor to high TSP levels at one
of the seven sites (Lubbock, Texas).  This technique leads to only very
qualitative evaluations.

CONCLUSIONS AND RECOMMENDATIONS

     The review of past studies carried out in the Southwest has led us to
the following conclusions.  First, the contribution of wind-blown dust from
the undisturbed desert floor to particulate levels in desert communities is
very small and should be considered as part of the low-concentration back-
ground when developing control strategies.  Second, if human activities, such
as the use of off-road vehicles, repeatedly break up the desert crust, local
violations of the 24-hour standards are likely.  Third, the contributions of
the various source categories of fugitive dust to the nonattainment problem
cannot be accurately disaggregated with the techniques employed in the re-
viewed reports.  However, these techniques can be used to identify the major
contributing source types, and there is now substantial agreement not only on
the major source types but also on the characteristics of the urban localities
mos.t affected by each type.

     Thus, there already appears to be sufficient information to serve as a
sound basis for the development of strategies leading to an overall improve-
ment in air quality.  The spatial application of control measures can be fine
tuned to some extent on the basis of current observations and microinventories
around problem sites.  Special monitoring programs can be used to define the
impact of individual sources if desired.

     The one area where additional field programs are needed is in defining
the level and sources of the inhalable fraction of suspended particulates.
The collection of samples over short averaging times (e.g., 1 or 2 hours)
along a line through an urban area and oriented with the prevailing wind
would be invaluable in assessing the relative impacts of locally generated
and transported inhalable particulates.  The use of instrumentation providing
respirable, inhalable, and total particulate data is recommended.  This detail
would be helpful in evaluating conformance to possible future standards as
well as in assessing source impact.

-------
                                  SECTION 2

                            ANALYTICAL TECHNIQUES
INTRODUCTION

     This section introduces the analytical techniques used in particulate
studies reviewed for this report.  Four types of analysis are discussed:
(1) analysis of temporal or spatial patterns of suspended particulates;
(2) analysis of particulate emissions;  (3)  examination of chemical, elemental,
or morphological properties of suspended particulates; and (4) examination of
the impact of meteorological conditions on suspended particulates.   The dis-
cussion focuses on techniques' strengths and weaknesses in assessing the con-
tribution of wind-blown desert dust to  urban particulate concentrations.  The
section concludes with an overview of where these techniques were applied
within the study area.

ANALYSIS OF TEMPORAL OR SPATIAL PATTERNS OF SUSPENDED PARTICULATES

     This category includes many of the most commonly used techniques in par-
ticulate analysis.  Analysis of temporal patterns encompasses techniques such
as trending and examination of seasonal and daily fluctuations in particulate
concentration.  Spatial analysis includes comparison of particulate concen-
trations by site environment, inter-site correlations, and mass flux
determinations.

     In the type of problem this report deals with,  spatial analysis is par-
ticularly important, but temporal analysis  also has  a role to play.  Several
studies reviewed for this report examined seasonal patterns of particulate
concentrations,  often in conjunction with analyses of seasonal meteorological
fluctuations.   One of these studies concluded that seasonal patterns impli-
cated fugitive dust, rather than wind-blown desert dust, as a major contribu-
tor to the particulate problem in the Phoenix area.*

     Spatial patterns play a prominent  role in the analysis presented later
in this report.   One technique frequently relied on  is comparison of particu-
late concentrations by site environment.  Comparison of urban and background
particulate concentrations, for example, reveals a great deal about the con-
tribution of wind-blown desert dust to  high particulate concentrations in
urban areas.  Other spatial analysis techniques also provide information that
is helpful in assessing the contribution of wind-blown desert dust to urban
particulate concentrations.  Inter-site correlations, which were relied on in
several studies, can indicate whether geographically disperse sites are in-
fluenced by the same source or type of  source.  In Tucson, for example,

-------
concentrations of Al, a soil indicator, at the background location correlated
very poorly with Al concentrations at urban locations.2  This evidence sug-
gests that different sources (or processes) may be responsible for introducing
soil into the atmosphere in urban and background locations.  Upwind/downwind
and mass flux analyses were also -used to assess the transport of wind-blown
dust from the desert into urban areas.

ANALYSIS OF PARTICULATE EMISSIONS

     The studies reviewed for this report relied heavily on emissions inven-
tories, microinventories, and air quality modeling.  The two inventory tech-
niques are similar in many ways; both attempt to identify sources in some
bounded geographic area, then estimate emissions from these sources.  The
major difference between the two techniques is the scale of the analysis:
microinventories focus on small geographic areas - often a 1- to 5-mile
radius around a monitor; emission inventories usually encompass a larger
geographic area.  Air quality models, which range from very simple to very
sophisticated, translate the emissions documented in inventories into ambient
concentrations.

     One drawback of emission inventories/microinventories is that they often
ignore natural emissions such as wind-blown desert dust or treat these emis-
sions in a very cursory manner.  Although many of the inventories reviewed
for this report suffered from this flaw, they do give a sense of the magni-
tude of other sources - such as paved and unpaved roads - contributing soil
material to the atmosphere in urban areas.  Air quality modeling is poten-
tially very useful in assessing the impact of wind-blown desert dust on urban
air quality since one of its purposes is to estimate source contributions to
ambient concentrations.  Its usefulness is limited by the accuracy and com-
pleteness of inventory input, however.  Several studies examined in this re-
port relied on air quality modeling.3"7

CHEMICAL, ELEMENTAL, AND MORPHOLOGICAL ANALYSIS

     Technological advances have enabled scientists to characterize the chem-
ical, elemental, and morphological properties of suspended particulate matter.
This section deals with a variety of techniques used to analyze this detailed
data including enrichment factors, inter-species correlation and factor
analysis.

     Many of these techniques are quite useful in identifying sources of par-
ticulate matter.  Inter-species correlations, for example, indicate whether
certain components of particulate matter cluster together; a common source is
usually hypothesized for such clusters.  In the Tucson study by Moyers et al.,2
the elements Al, Fe, Si, Ti, Li, Rb, Mg, Na, Ca, K, Mn, and Sr were correlated
with each other.  The authors suggested soil or crustal material as a possible
common source for this cluster of elements.  Factor analysis is a refinement
of inter-species correlation that serves much the same purpose.  Statistical
analysis based on inter-species correlations is used to group elements into
several clusters.   The analyst then draws on his experience and knowledge to
hypothesize about the factor - in this case, the source - that generates each
cluster of elements.

-------
     Enrichment factors try to distinguish anthropogenic and "natural" sources
by examining the concentration of elements in suspended particulate matter
relative to their concentration in soil material.  An enrichment factor of
one indicates that the element has the same relative concentration in both
air and soil, implying a "natural" soil-related source.  An enrichment factor
greater than one indicates that a particular element is more abundant in air
than in soil, suggesting contributions from anthropogenic sources.  In the
Tucson study, enrichment factors for the 12 elements suspected of being soil-
related (see list above) were similar, all very close to one.  This reinforces
the evidence provided by inter-species correlations that these elements, which
comprise a large part of the particulate matter in the Tucson study, have a
common source, soil.

     Most of these techniques are incapable of distinguishing between fugitive
(man-made) dust and natural wind-blown desert dust, although they can indicate
how much these two sources, taken together, contribute to particulate concen-
trations.  Several of the techniques discussed earlier, for example analysis
of spatial patterns, can help distinguish between these two sources when
applied to chemical and elemental data.  Spatial analysis of soil-related
elements in the Tucson study showed that concentrations of soil-related ele-
ments were much different at urban sampling sites compared to the background
site.  The authors concluded from this evidence that much of the suspended
soil material in Tucson is probably due to urban activity.  Inter-site corre-
lations in Tucson also indicate that soil concentrations in urban and back-
ground locations were dissimilar.

     Morphological analysis can also distinguish between natural and anthro-
pogenic sources of suspended dust particles.   A recent study of particulates
in Phoenix,8 for example, used morphological data to distinguish the sources
of mineral material found in particulate samples.  They found that the par-
ticle sizes and shapes of most of the mineral particles in the Phoenix sam-
ples were not typical of wind-eroded materials; instead, the particles were
often sharp and angular indicating, according to the authors, vehicle travel
over paved roadways.

METEOROLOGY AND SUSPENDED PARTICULATES

     This technique focuses on relationships  between meteorological param-
eters, such as wind speed, precipitation, and mixing height, and levels of
suspended particulates.   Most studies reviewed for this report relied on
common statistical techniques such as linear  regression or correlation co-
efficients to examine the relationship between meteorology and particulate
concentration.

     Many of these studies found the complexity of the TSP/meteorology rela-
tionship defied analysis by simple statistical techniques.  Several studies,
for example, focused on the relationship between wind speed and particulate
concentrations, hoping to distinguish the relative importance of wind-blown
dust.   Most found it difficult to sort out the conflicting influences of
increased wind speed:  (1)  increased emissions which tend to produce higher
TSP concentrations and (2)  greater dilution of pollutants, which results in


                                     10

-------
lower concentrations.  Even studies that focused exclusively on the TSP/
meteorology relationship9 emphasized the qualitative nature of their results
and the inability of the techniques available to distinguish between wind-
blown desert dust and wind-blown dust from construction sites, unpaved roads,
and agriculture sites.

OVERVIEW

     Table 1 indicates which techniques were used in the studies relied on
most heavily in the preparation of this report.  The table reveals several
noteworthy patterns.  Perhaps the most noticeable pattern is the wide variety
of techniques employed by the three studies of Phoenix; techniques from all
four categories are used in these studies.  The study of Tucson also used a
large number of different approaches to analyze particulate problems.

     The second distinctive pattern is the heavy reliance of the studies on
emissions inventory and modeling approaches.  Eight of the eleven studies
utilized these approaches, six exclusively.  While some of this information
was very useful in assessing the contribution of wind-blown desert dust to
urban particulate concentrations, many of the inventories suffered from the
flaws discussed above limiting their usefulness.

     These two patterns help explain why this report concentrates heavily on
Phoenix and Tucson:  a great deal of detailed and useful information about
the nature of the particulate problems in these two cities had already been
compiled.  Less information was available in the published reports on other
cities.  This study also examines Reno and Las Vegas because the emission
inventory done by PEDCo includes an assessment of the contribution of wind-
blown desert dust.
                                      11

-------
TABLE 1.  PRIMARY TECHNIQUES USED BY STUDIES RELIED ON IN THIS REPORT



.. 0)
4-1 J-J
01 cO
3 -H -r
Q 3
CJ
CU -H *
> 4J T
-H • M
4-1 -J CQ
•H r-«. ^ J
CO CT-
3 — < u-< t
fi O

 --H co
-H C O J=
tC t-t > CJ cv
CU
PS • •
Analysis of Spatial or Temporal Patterns
Seasonal Patterns
Comparisons by Site Environment X
Inter-Site Correlations
Mass Flux Determinations
Emissions and Modeling
Emission Inventory X
Microinventory X
Air Quality Modeling X X
Chemical, Elemental and Morphological Analysis
Interspecies Correlations
Enrichment Factors
Factor Analysis
Pattern Recognition
Particle Size Analysis
Morphological Analysis
Other Elemental Analysis
Meteorology and Suspended Particulates
Regression Analysis
Other
4-1 * CO X • >i
C C >, r- -H ^ t~i
CU O 4-1 c?> C | C O t-t

10 C C t>0 cO * d tf] O 4-i C CO C M-il
i-H njcujj w o» 3j2 o&ucu- -H



JCO HCMO rH O.CJ OJ3I-.WC ^ -W

rOm cu --HI— icur^o3 cu
3 Z. OCQJ -eC'-H1^ CC C04-)O-i-HO C3-
1 -HO ^tiO cOvH-C4J.-icnc_) ocrtjs
i— i CO-H "DXcO t— ( 3 f^XcOCU QJC-H C-i-1 C>iC[d9 -Ho*
J CTv>HQtO &OO-HU1 OcO"Cfl4JO >• E3"
» C — 'CO CXI I-1 -Hi-H^<4^a'o - -H'-H'^C'
Jtnto-HOH - COWM OOEOJ -WC-HESO X-UO-H i
jOin34-ioj c copS4JNci-i-acoccotno cj QJ-H 1-1 o




• <•• • • • Z, 9 Z

X X
X X
X
X

X XX
X
XX X

X
X
X
X
X X
X


X
X XX
00
co C
X CO -H Oi

3 *j cn o cr\
U < CO r-l — I
n cu a> a

H o <; cu o
0 C Q U
lj co C Q
u C cO C W
c aj .o -H P-I
3 AJ t-l
J C 3 CU
-H u n)
3J _« to C -H

0 *" HO 4J Vj
H >-, CU tO O
U -H CO 4J -H
rt i-H tO tO C iH
u cfl tO r~^ <3 OJ co
M 3 _] P- £ U
o o- cr. 1-1 C
J-i C -H 0 CO ^H
n) -H co - c -H -u
a •< o i-i c 4-1 cn
a o o o j: co co

IJ O OJ H -HCUOM

~ co







X
X
X X






X X
X




-------
                                  SECTION 3

                      BACKGROUND SUSPENDED PARTICULATES
     The expression background suspended particulates, as generally used,
refers to particulates found in the atmosphere in remote or rural places well
removed from obvious anthropogenic sources.  Some of these particulates come
from nearby natural sources; others are transported from distant natural and
urban sources; still others are formed within the atmosphere by chemical
action during precursor transport.  In this report, background concentration
is considered to be the concentration of particulates that would have existed
throughout the study area had that area remained undisturbed by man.  Con-
ceived of in this way, background concentrations represent baseline levels
upon which particulates generated within an urban area are superimposed.  The
difference between background and observed concentrations represents the incre-
ment that may be manipulated through the use of local control measures.  The
concept is somewhat simplisitc since the background really comprises a dynamic
particulate population continually responding to changing depletion and
replacement rates.

     Background concentrations are usually expressed in terms of annual geo-
metric mean concentrations measured in remote or rural areas unaffected by
local sources.  In addressing the influence of wind-blown desert dust on the
nonattainment problem in urban areas, however, consideration must also be
given to the magnitude of 24-hour concentrations observed in remote and rural
areas.

     This section first summarizes annual and 24-hour concentrations observed
at remote and rural sites throughout the Southwest.  It then discusses some
of the physical and chemical properties of particulates in outlying areas in
an attempt to shed light on their origins and probable transport distances.
Topics discussed include:  (1) particulate size distributions, (2) the genera-
tion of wind-blown dust, and (3) chemical constituents of suspended particu-
lates.  This information, when compared with that presented for urban areas
in Sections 4 and 5 of this report, will be used in estimating the contribution
of wind-blown dust to the TSP nonattainment problem in large cities set in
desert surroudings.

BACKGROUND CONCENTRATIONS

     Background concentration data have been assembled from three sources and
are based principally on measurements made in 1976.  Large year-to-year varia-
tions in background concentrations are not expected, however, barring extreme
differences in meteorological conditions.  Data obtained from the National

                                     13

-------
Aerometric Data Bank for 24 remote and rural sites throughout the eight south
western states provides the most widespread set of annual and 24-hour concen-
trations.  This data set is supplemented by measurements made at 20 sites in
the Four Corners area for EPA by the Ute Research Laboratories.10  In addition,
contractor reports provide estimates of background levels for several specific
urban areas.

     Appendix A lists the location of each monitor, the station type as classi-
fied by the SAROAD site identification procedure, elevation above sea level,
and the elevation of the monitor above the ground.  Annual geometric mean con-
centrations, and the highest and second highest concentrations for the year,
or period of measurement, are also given.  The approximate locations of the
monitors are indicated on maps (Figures A-l and A-2).

     Descriptions of the stations operated by the Ute Research Laboratories
can be found in Reference 10.  Typically, these stations were located on
either Indian Reservation or National Park land in arid areas with sparse
vegetation and widely scattered homes.  Nearby land use was generally restric-
ted to livestock grazing; in a few cases, limited areas were devoted to agri-
culture.  The high-volume air samplers were mounted on 4.6 m (15 ft) towers
to minimize the effects of local dust and ground-level obstacles.

Annual Geometric Mean Concentrations

     Figure 2 displays the annual geometric mean concentrations obtained from
the National Aerometric Data Bank.  At three stations, the 1976 data did not
meet the SAROAD summarization criteria for the calculation of annual means,
and 1977 means were substituted;  these three 1977 values have been placed in
parentheses.  The results from special monitoring programs conducted in north-
eastern Utah and northwestern Colorado are given as ranges.   In each case,
the results from four monitors were combined.

     Although the results show no sharply defined pattern, the lowest values
tend to be found in northern Utah and Colorado.   At lower latitudes, concentra-
tions of 50 yg/m3 were measured in central Oklahoma and central Texas, perhaps
reflecting regional differences in land use and agricultural practices.  The
only concentration that exceeded  federal ambient standards,  84 yg/m,  was
measured at the Kern Wildlife Refuge in the San Joaquin Valley airshed in
California.  Examination of long-term trends at this site11  suggests an improve-
ment in air quality since 1976; in 1978, the annual geometric mean was 62 yg/m3.
Neither of these values,  however, is representative of background concentrations.

     Figure 3 displays geometric  means observed by the Ute Research Laboratories
monitoring metwork.  Activation dates for the  monitors ranged from 12 January
1976 to 6 May 1977  (see Appendix  A,  Table A-l).   The values  plotted in Figure 3,
however,  were calculated  using 1976 data from  sites where observations were
available for at least three seasonal quarters.   As the figure shows, the moni-
tors stretched along an east-west belt 50-100  km wide that extended for a
distance of about 380 km.   Geometric means within the network ranged from 12 to
56 yg/m3.
                                      14

-------
Figure 2.
1976 annual geometric mean TSP concentrations from National
Aerometric Data Bank.  Values in parentheses are 1977
concentrations.  Ranges are shown for closely located
sites in Utah and Colorado.

-------
0 50 100 miles
1 1 1
1 1 	 1
0 80 160 km

QR1CHFIELD
UTAH
.56
• «
.26 .26
• 18 *16 290
•29
•18 .
• 22 24*
•32 .17
ARIZONA

n FLAGSTAFF

QGRAND JUNCTION

COLORADO


.23 ^ .23
•12
.24 .45
.31
NEW MEXICO
D GALLOP

Figure 3.  1976 geometric mean TSP concentrations observed by Four  Corners
           Ambient Air Monitoring Network operated by Ute Research  Laboratories.

-------
     The means shown In Figures 2 and 3 have been combined and summarized as
a frequency distribution, Figure 4, that distinguishes the data by site type;
i.e., remote or rural.  In this sample, on the average, rural sites experi-
enced a mean concentration of 38 yg/m3 in contrast to the 24 yg/m3 experienced
by remote sites.  Because of the small size of the rural-site sample, this
difference may not be significant, but it suggests that the impact from rural
activities at these sites is in the vicinity of 10-20 ug/m3-

     Estimates of background concentrations can also be obtained directly from
a number of particulate studies.  Background concentrations, taken from three
of these studies are listed in Tables 2, 3 and 4.  PEDCo used the data contained
in Table 2 to develop the following relationships between vegetative cover in
the Southwest and background particulate levels:  forest-woodland, 15-20 yg/m3;
southern desert scrub, 25-30 yg/m3; northern desert scrub, 20-30 yg/m3; and
grassland, 22-27 yg/m3.  The values in Table 3 were used by TRW to determine
a weighted average background concentration of 30 yg/m3 for Phoenix, Arizona.
The local health departments suggested 35 yg/m3 and 25 yg/m3 as appropriate
background values for Las Vegas and Reno, respectively.  In addition to these
estimates that are based directly on observations, PEDCo estimated background
concentrations for sites in Colorado and Utah from the y-intercepts of emis-
sion density versus air quality, and modeling analyses.  These esitmates are
presented in Table 4.

             TABLE 2.  TSP CONCENTRATIONS AT SELECTED BACKGROUND
                       SITES IN NEW MEXICO, ARIZONA,  AND NEVADA
            State
Sampling site location
Annual geometric
   mean, yg/m3
         New Mexico  Albuquerque - NASN
                     Bernalillo County-Radar Stn.
                     Dona Ana County
                     White Rock

         Arizona     Organ Pipe Cactus Nat'l
                       Monument
                     Grand Canyon
                     Davis Dam
                     Page

         Nevada      White Pine - NASN
                     Las Vegas - Marina
                     Boulder City
                     Las Vegas - Civil Defense
                       Building
                     Reno
                                  22
                                  32
                                  13
                                  32

                                  26
                                  21
                                  29
                                  17
                                  14
                                  35
                                  30

                                  34
                                  31
         Source:   Table 3-7,  pg.  3-26,  Reference 4.
                                      17

-------
zo

1 C
CO
z
O 14
H
OBSERVA
O t\>
NUMBER OF
P> Cft CD



o




-
M
-

























:£:'?:•••







- •' • '.
'•: • '
.' •' ' • '
•'.'•' : •

(AVE.= 24)

ri.'v-y:.*:.: RURAL
ti-'-vV-'i (AVE. s 3fl)



kFRM RPPIIftP
1
sra; v
m^i(m \ i
     0-9 10-1920-2930-3940-4950-5960-6970-7980-8990-99

             TSP  CONCENTRATION,
Figure 4.   Frequency distribution of annual geometric
           mean TSP  concentrations at 45 rural and
           remote sites  in  Southwestern U.S.
                         18

-------
        TABLE 3.  TSP CONCENTRATIONS
                  AT SELECTED BACK-
                  GROUND SITES IN
                  ARIZONA
                       Annual geometric
                          mean, yg/m3
Site
Grand Canyon
Petrified Forest
Organ Pipe
Montezuma
Average
1973
22
26
34
28
28
1974
17
23
23
27
23
1975
N/A
N/A
31
34
32
     Source:  Table 4-1, pg. 4-1,
              Reference 3.
TABLE 4.  ESTIMATED BACKGROUND TSP CONCENTRA-
          TIONS FOR SELECTED AQCR's IN
          COLORADO AND UTAH
Estimated annual
State
Colorado





Utah
geometric mean
AQCR yg/m3
Denver
Pawnee
San Isabel
Colorado Springs
Pueblo
Other sites
Yampa
Grand Mesa
Wasatch Front
40-45
45

30-35
45
40-45
20
40
40
Source:  Table 2-2, pg. 2-8, Reference 12.
                     19

-------
 Twenty-four Hour Concentrations

      Figures 5 and 6 display the second  highest  24-hour  TSP concentrations for
 1976 in the same fashion used for the  geometric  means.   Again,  substituted 1977
 concentrations are indicated by  parentheses,  and a range of values is given
 for samplers in geographic  proximity.  In  Figure 5,  one  concentration exceeds
 the primary standard and five others exceed  the  secondary standard,  with two
 of the five being observed  in northwestern Colorado.  In Figure 6, two concen-
 trations exceed the primary standard and nine others exceed the secondary
 standard.

      These 24-hour, second-highest, concentrations have  been combined as a
 single frequency distribution in Figure  7a.   In  this  case,  there appears to be
 no significant difference between the  averages shown at  rural and remote sta-
 tions.  A similar frequency distribution of  the  maximum  24-hour concentrations
 has been plotted In Figure  7b, and again there appears to be no tendency for
 the highest concentrations  to occur more frequently  at rural stations than at
 remote stations.

      Finally,  inspection of the  two sets of maps reveals no obvious  relation-
 ship between geometric means and  the magnitude of  the second-highest 24-hour
 concentration.   The lack of strong relationship  was  confirmed by a linear
 correlation between the  two sets  of data of  0.42.  Although this is  significant
 at the 0.01 level with 45 pairs,  only  17 percent  (R2) of the variation is
 common.

 CHARACTERISTICS  AND SOURCES

      An  understanding of  the physical and chemical nature of  background  partic-
 ulates provides  insight  into their probable sources and likely  transport dis-
 tances.   This discussion  briefly  reviews particle-size distribution  information
 and some of the  generally accepted concepts pertaining to the generation of
 wind-blown  desert  dust.   It  also  summarizes the  results of  recent  studies that
 have  investigated  the chemical species present in desert dust and  interrela-
 tionships among  these species.  Much of this  latter work was  carried  out on
 samples  collected  at the  Research Ranch approximately 60 miles  to  the  south-
 east  of  Tucson,  Arizona.

 Particle Size Distribution

     Very few measurements covering the complete size range  of  particles
 collected by the standard high-volume sampler have been reported  in  the  litera-
 ture for desert atmospheres.  Furthermore,  it is rather widely  conceded  that
 the details of observed distributions are dependent upon the  collection  devices
 and experimental techniques  employed.   (See Lundgren and Paulus,13 and Farmer
 and Hornkohl,14 for example).  Nevertheless,   recent studies  have contributed
 substantially to knowledge of airborne particulates in a variety of  environ-
ments, and  this  information can be used to supplement the limited desert
measurements to obtain the most significant features of particle size  distri-
 butions in background, desert atmospheres.
                                      20

-------
Figure 5.
1976 second-highest 24-hour TSP concentrations from National
Aerometric Data Bank.  Values in parentheses are 1977 con-
centrations.  Ranges are shown for closely located sites in
Utah and Colorado.

-------
                     0
                     h
 50
H-
                             80
 100 miles
-H
 160 km
                                   QRICHFIELD
                     UTAH
                                         360
                                     196
so
                      86
                                                             138
                                           • 90
                                          '353
                                          • 124
                                                 ie
                                                 155
                       ARIZONA
                                        FLAGSTAFF
                                        QGRAND  JUNCTION



                                        COLORADO
                                                                     8I
                                                                          70
                                                   171
                                                                                     171
                                       NEW MEXICO

                                       DGALLOP
                Figure  6.  1976 second-highest 24-hour TSP concentrations observed at Four Corners
                          Ambient Air Monitor Network operated by Ute Research Laboratories.

-------
                          20
_
0
   16
v>

O  <4


S  12
K
Ul
£  10
O
U.  o
o  8
                        i 4
n
                                                          REMOTE
                                                          (AVE.= 138)
                                                           RURAL
                                                          (AVE. = 139)
                              0-  75- ISO- 225- 300-1375^ 450-525-600- 675-
                              74  149  224  299  374 |449 1524 |599 | 674 1749

                                       CONCENTRAT ION , yu.g An3
                                0)SECOND  HIGHEST CONCENTRATION
    20


     18


     16


     14


     12


     10
                          S 8
                                                           i  REMOTE
                                                            (AVE. = 217)
                                                       i..,....i RURAL
                                                        	    -..-- 170)
                                0-   75-  ISO-  225- 300- 375- 450- 525- 600- 675-
                                74   149  224  299  374  449  524  599 674  749
                                         CONCENTRATION,
                                    b)  MAXIMUM  CONCENTRATION
                    Figure  7-   Frequency distributions  of 24-hour TSP
                                 concentrations  at 45 rural and  remote
                                 sites in Southwestern U.S.
                                                23

-------
     The picture that emerges is of a volume, or mass, distribution that is
bimodel.  One mode is comprised of fine particles in the range from 0.1 ym to
roughly 2 ym, and the other mode is comprised of larger particles, with a peak
that depends on soil characteristics, meteorological conditions and height
above ground.  The bimodel form of the distribution over the range from 0.1 to
10 ym is illustrated in Figure 8, taken from a paper by Sverdrup, Whitby and
Clark.15  Five of the distributions are based on observations made at Goldstone
in the Mojave Desert, California.  The sixth distribution represents conditions
at Fort Collins, Colorado on a day when there was little evidence of local con-
tamination from dust.  The aerosol particle size distribution was measured with
four on-line instruments over a particle size range from approximately 0.003
to 40 ym.  A description of the instrumentation is given in Whitby
et al.16

     Particles in the fine-particulate mode remain suspended in the atmopshere
for long distances unaffected by gravitational forces.  Estimates by Esmen and
Corn,17 for example, indicate that the residence time for particles 1 ym in
diameter in the absence of precipitation is of the order of 100 hours.  At an
average transport speed of 5 m/sec (11.2 mph) this represents a travel distance
of 1800 km (1118 mi).  Thus, many of the particles in this mode reflect emis-
sions from distant sources, some of which are anthropogenic and others natural.
This fine particulate mode was named the "accumulation mode" by Willeke and
Whitby18 to reflect the fact that aerosol particles that grow into this range
from smaller sizes by coagulation or condensation tend to remain in this size
range.

     The other mode is comprised of larger particles whose primary source
is the underlying ground surface.  Under average meteorological conditions and
uniform surface conditions, TSP measurements suggest that a quasi steady state
develops in which the resuspension of soil particles from the desert floor is
roughly balanced by the removal rate over a 24-hour period.  During any 24-
hour period, however, marked variations in the resuspension rate occur as a
result of major changes in the turbulent characteristics of the wind field
and in the amount of convection.

     Evidence that resuspension occurs in puffs of short duration (less than
or equal to 2 minutes) has been presented by Porch.19  Dust devils are a
visual indication of convective adjustment in which a large size range of
particles are injected into the lower atmosphere.  The next section discusses
the generation processes for desert dust in more detail, but common experience
illustrates relationships between wind speed, soil conditions and the injec-
tion of particulate material into the atmosphere.  Once injected, the residence
time of the particles in a dry atmosphere depends upon the initital height
reached by the particles, the gravitational settling rate, atmospheric turbu-
lence,  and ground cover.

     Figure 9,  developed by MRI,20 relates particle diameter to three settling/
suspension regimes at wind speeds up to 18 mph.   The area labeled "suspension"
represents particles that have the potential for long range transport.  The
area labeled "unimpeded settling" represents particles with terminal veloci-
ties sufficient to largely overcome the effects of turbulence and horizontal
transport.   Particles in the "impeded settling" area respond to atmospheric
motions while settling.
                                     24

-------
                e
              *•
              |
              H 4
              <
                             	MITT]
                       NUN NO.   DATE*    TIME
                     O 	  Bl OCT. l»72l3'30-l6'BOb
                     0 Ktr  2 NOV. 1972 OOOO-OI«Oh
                                    I I Illlj
                OXM
                                         I
                                       D., tan
                                                     O
A comparison of  the background volume distributions  during a very clean period
                    and an evening with higher total volume.
                  so
                tf
                f«
                       O GOLOSTCNE
                       A FT. COLLINS
                     i  i  i | i ill]   i  ill

                     MM NO.  TME     MTE
                     KM   OHJO-02-OOH I NOV. «9I j
                     1-32   0»S9-O7^«Oti  (AIM. I
  A comparison of the  volume distributions of  background  aerosols sampled at
                        Goldstone and Fort Collins, Colorado.
 Figure 8.
                 !
                        I  I I  | llll|

                         GOLDSTONE
                         4 NOV. I9T2
                         O 05'000
                         A OB'40h
                         O 07<90«
                         TTTT
                   0.01
                                        D,,
Incursion of aged aerosol  from the  south coast basin.
Volume  distributions observed  in the Mojave Desert,  California,
and Fort Collins,  Colorado  (from Sverdrup,  Whitby,  and Clark,15
pp. 489 and 490).
                                        25

-------
200 i-
                   4       6       8       10       12
                           REFERENCE  WIND SPEED (mph)
14        16
18
             Figure 9.  Particle settling/suspension regimes.

                                     26

-------
     Because of the complicated relationships alluded to above, the modal peak
due to ground level sources can be expected to vary with desert location and
sampling height.  An example of the shift in the peak toward greater particle
diameter as the sampling height decreased from 30 m to 0.3 m is shown in
Figure 10 taken from Schmel.21  (Derivations of the lines "A" and "B" are given
in Schmel's paper).  The observations used in preparing this figure were made
during a dry, spring period in a  semi-arid and sparsely vegetated region of the
Atomic Energy Commission's Hanford reservation.  The upwind terrain was undis-
turbed for several miles and included a region without sage bush that extended
about 200 feet upwind.  Particles in the respirable size range were determined
using 20 cfm cascade particle impactors; the larger particles were collected
by deposition within a wind-direction self-orienting attachment to the
impactors.

Generation of Wind-blown Dust

     The generation of dust begins at a wind speed great enough for aerodynamic
forces to overcome the forces holding individual particles in the soil.
Although a considerable amount of theoretical and experimental work has been
carried out on idealized systems  Csee for example, Chepil,22 Greeley et al,23
Marshall2** and Lyles and Allison25), few studies of wind-generated fugitive
dust have addressed natural soil  surfaces and man's impact upon these surfaces.
As a result, current understanding of the mechanisms by which desert soil
becomes entrained is largely qualitative.

     Two important aspects of soil structure are the distribution of particle
sizes and the aggregate structure of the soil.  The undisturbed desert soil
surfaces in much of the southwest are composed of a silt aggregate and a thin
crust of gravel and sand, (.see Richard et al.1 for a description of soil charac-
teristics in the Phoenix area).   This surface crust is sometimes referred to
as the desert pavement.  It is easily broken up by human activities and is
particularly vulnerable to vehicular travel.  Where loose sand particles exist,
moderate wind speeds may be sufficient to initiate sandblasting of the sur-
face soil, resulting in the release of soil fines and the suspension of
particulates.

     Under sufficiently strong winds, severe dust storms may develop, particu-
larly in regions where the soil surface has been disturbed as a result of
man's activities.  For example, on February 23, 1977 central Oklahoma experi-
enced the worst dust storm in more than 20 years with 24-hour average TSP con-
centrations up to 5800 yg/m3.26   The paths of dust were clearly evident in
photographs prepared from satellite data.  Reported surface winds were 30 to
40 knots in the dust swaths and 20 to 30 knots in the clear area.  Oklahoma
had another severe storm on March 2, 1977.  The photographs show very strikingly
that in each case the western limit of the dust coincides approximately with the
Texas-New Mexico border, apparently because of changing soil characteristics
and agricultural practice in the vicinity of the border.  Rangeland is pre-
dominant on the New Mexico side of the border, and cropland on the Texas side.

     Dust storms are typically associated with sustained wind speeds of 20 to
30 knots produced by major weather systems.  In a study of dust storms in the


                                      27

-------
              AIR SAMPLING

           t-   HEIGHT, m

              O	0.3
         105
      |
         103
                              	'I

                               APRIL 4-26,1972
              Q	3

              O	10
              A	30
                                        °0
                            //»  \
                           ///w \
                         10              102
                      PARTICLE DIAMETER, (Dl, |ini
Figure  10.   Volume distribution  as a function of height.21
                              28

-------
 Great  Plains,27  Hagen and  Woodruff  found that  the average duststorm  lasted  6.6
 hours  and  the median concentration  during the  period of the  storm was  4850  yg/m3,
 In this  study, particulate concentrations were calculated from National  Weather
 Service  visibility data using the relationship


                              C6  "  TTZF  mg m~3
 developed  by Chepil  and  Woodruff.28   In this expression GS  is  the  dust  concen-
 tration 6  feet  above the surface and V is the horizontal visibility  in  kilo-
 meters.  The data  analyzed  for each  station were from 10-year  period beginning
 in either  1949  or  1950.   In a related study, Hagen and Woodruff29  roughly esti-
 mated  that the  median size  of the duststorm covered an area of about 188 *  188
 miles  square.   The average  number of days per year with dust reported at 19
 stations distributed throughout Texas, New Mexico, Oklahoma and Colorado ranged
 from 1.3 at Grand  Junction, Colorado to 47.5 at Lubbock, Texas. The mass
 median particle diameter during duststorms has been reported by Whitby  et al30
 as 50  ym.

 Chemical Constitutents

     The information provided here has been taken largely from a paper  by Korte
 and Moyers31  that  describes the results of a measurement program conducted
 between September  1975 and  October 1976 at the Research Ranch.  The Research
 Ranch  is a protected land laboratory comprising 7,840 acres of private, state
 and federal land,  generally high desert grassland.   It is situated in the
 transition zone between  the Sonoita  Plains and the Hauchuca Mountains.  The
 nearest town is Elgin (population 50)  approximately 5 miles to the north.

     Thirty-two size fractionated samples were collected during the sampling
 period as  part  of  a  larger  program described by Moyers et al.2 Sampling was
 performed  using Sierra 230  two-stage,  high-volume slotted Cascade  Impactors.
 Particle separation  using this system was about 2 ym aerodynamic diameter.
 Elemental  analysis was performed on  digested filters by atomic absorption
 spectrometry.   Analyses  for ammonium, nitrate and sulfate ions were  also
 carried  out.

     The data were evaluated using statistical correlations, particle size
 distributions,  and comparison of particle composition with  the composition of
 suspected  source materials.   In calculating the crustal enrichment factor,
 aluminum was chosen  as the  reference element since soil is  considered to be
 the sole source of aluminum in this  region.

     It  is  concluded  on  the basis of this and the previous  work by Moyers et al
 that wind blown soil  materials dominate the mass concentration of  large parti-
 cles (greater than 2  ym)  and all measured species except Cd, Zn, Pb, Cu, 80^=,
 N03- and NHtt+.   On the other hand, the smaller particles are composed of soil
material, gaseous  oxidation products,  and species apparently from distant
 anthropogenic or natural  sources.  The small particle sulfate,  nitrate, and
ammonium ions are  the  result of  gas-to-particle conversion  from gaseous SOa ,
NOx, and NHs released  into  the atmopshere from natural  sources such as  decaying
vegetation, and anthropogenic  sources  such as smelters,  agriculture, power

                                     29

-------
plants and urban areas.  Previous work by Moyers et al32 estimated that the
contribution of soil material to the total atmospheric burden of particulate
matter at the Research Ranch is 50 percent.

     Figures 11 and 12 taken from the paper by Korte and Moyers,31 and the
linear correlation coefficients calculated between chemical species that are
presented in the paper, support these conclusions.  Figure 11 shows the per-
centage distribution of the individual species on the respirable (less than
2 ym) particles, while Figure 12 is a plot of the crustal enrichment factors
for the respirable particles.  Species with enrichment factors near unity
(from 0.5 to 5) are attributed to airborne soil material, while species with
enrichment factors greater than about 50 are attributed to sources other than
the injection of crustal weathering products into the atmosphere.

     More general information concerning the magnitude and distribution.of
sulfates and nitrates is being obtained from an expanded monitoring network
of high-volume samplers within the Western Energy Resource Development Area.33
Preliminary summaries  of data,  collected before major expansion of the net-
work occurred,  show sulfate concentrations in the 1  to 3 yg/m3 range in north-
ern Arizona, southern Utah, northwestern New Mexico,  and southeastern Colorado.
With the exception of southern Arizona where values  from 3 yg/m3 to greater
than 15 yg/m3 were reported, the region lies in the  range from 1 to 6 yg/m3.
This is in agreement with the mean value of 2.6 yg/m3  reported for rural  areas
in the West by Greeley et al34 and Altshuller,35  Better definition of regional
differences will be available from the expanded network.   Nitrate data from
the early network is less complete and shows greater variability than the
sulfate data.   Mean values generally range from 1  to  5 yg/m3  throughout the
region,  with the lower values being found in the rural areas.
                                    30

-------
                 100-
                5
                s -
                                   RESEARCH
                                 SEPT 1975 - OCT 1976
                    Ct Si Cl II III CD Ml Ti Ft K M| li A I, A nlj

    Figure  11.   Percentage distribution of measured  chemical  species on
                 respirable particles  in the atmosphere at the Research
                 Ranch.   (Korte and Moyers31)
                 10'-
                                  Ruiti $mu PuriciE
                                   ElllCIIWIIT FtCTOIS
                     Ft Sr Ci fi  Hi m Si Mg K Rb Ni Li Cr 2n Pb Cu S0,= NO, NH|
Figure  12.   Crustal enrichment factors of  species measured on respirable
             particles in the  atmosphere  at the Research Ranch.   (Korte
             and Moyers31)
                                       31

-------
                                  SECTION 4

                          URBAN PARTICULATES LEVELS


     This section examines the distribution of particulates in and around four
major urban areas located in dry, arid regions of the southwest.  The review
is based in large part on published reports summarizing 1975 TSP data, but is
supplemented in some cases by more recent observations.  The objective is to
analyze the relative contributions made by particulates arriving from outside
the urban area, and those that are generated or reentrained within the area
to violations of the standards by comparing spatial particulate patterns with
land usage and associated human activities.  The four urban areas studied are
Las Vegas and Reno, Nevada, and Tucson and Phoenix, Arizona.

PARTICULATE LEVELS IN AND AROUND LAS VEGAS

     Las Vegas is situated near the center of a broad desert valley which runs
from northwest to southeast.  The valley floor slopes gradually upward on each
side towards the surrounding mountains.  The climate is charactrized by low
humidity, abundant sunshine, mild winters and hot dry summers.  The prevailing
winds at McCarran Airport, 7 miles south of downtown Las Vegas are from the
west through south-southwest.  Precipitation averages about 4 inches per year.
Las Vegas has grown very rapidly over the last 20 years, with the population
of the metropolitan area reaching 361,100 in 1977. 6  The tourist industry
dominates.  Agriculture is carried out on a small scale.

     The following discussion is based on information provided in a 1977
report by PEDCo5 prepared for the U.S. Environmental Protection Agency.  In
this report, TSP data from 12 sites are presented and analyzed.  Figure 13
shows the sampling locations and their relationship to Las Vegas.  In terms
of physical setting, the sites range from center city/commercial to rural/
recreational.  Table 5 lists site location information provided in the PEDCo
report, plus monitor heights.

     The concentration data analyzed covers the 12-month period from October
1974 through September 1975.  Mean concentrations for this period are plotted
against distance from a central point in the Las Vegas business district (CBD),
Fire Station No. 1, Site E.  The data cover three distance intervals,  The
first interval contains 5 monitors centrally located within the heart of the
urban area.   (Sites C, I, F, H, and E).  The second interval contains 4 moni-
tors that ring the urban area from the McCarran International Airport to the
south to the Nellis Air Force Base to the northeast; the distances of these
monitors from Site E range from about 10 to 14 km.  The third interval contains
3 monitors located from 20 to 24 km from Site E.  As shown in Figure 14, two
of these monitors (B and D) are located within the city of Henderson (1977

                                      32

-------
                  KEY
                  A  LAS VEGAS AIRPORT
                  B  BASIC SCHOOL
                    CLARK COUNTY HEALTH DEPT.
                    HENDERSON POST OFFICE
                    LAS VEGAS FIRE DEPT. No.
                    LAS VEGAS FIRE DEPT. No.
                    NELLIS AIR FORCE BASE
                    NORTH LAS VEGAS
                    SAHARA HOTEL
                    LAS VEGAS STADIUM
                    SUNRISE POWER PLANT
                    LAS VEGAS WASH
Figure 13.   The  Clark County APCD sampling  network for  TSP  (after PEDCo).5
                                         33

-------
TABLE 5.  SITE LOCATION DESCRIPTIONS FOR LAS VEGAS
          MONITORS
Site
designator
A
B
C
D
E
F
G
H
I
J
K
L
Site
Las Vegas Airport
Basic School
Clark County Health
Department
Henderson Post Office
Las Vegas Fire
Department No. 1
Las Vegas Fire
Department No. 2
Nellis Air Force Base
North Las Vegas
Sahara Hotel
Las Vegas Stadium
Sunrise Power Plant
Las Vegas Wash
Site type
Suburban/
commercial
Suburban/
commercial
Center city/
commercial
City/industrial/
commercial
Center city/
commercial
Center city/
commercial
Suburban/
commercial
City/commercial
Center city/
commercial
Suburban/
commercial
Suburban/
industrial
Rural/
recreational
Monitor
height
(ft)
25
30
25
30
35
25
25
25
30
10
25
25
                       34

-------
120


no


100


90
*»E  80
       •E
z   ?0
o

|   60

z
uj
«   50
O
o

a.   40
     30



     20


     10
              •c
                                 \

                                          •G
                                              X
                        INCREASING  URBANIZATION

                                                                HENDERSON

                                                                   0
                                                                 \°
                                                                POST  OFFICE
                                                                  BASIC
                                                                 SCHOOL
                  Figure 14.  Annual mean TSP concentrations
                             versus distance from urban core
                              (Las Vegas).
                                                                    X  L
                        6    8     10   12    14    16     18    20   22   24

                         DISTANCE  FROM URBAN  CORE, km
                                     35

-------
population of 20,100)36 southeast of Las Vegas.  These monitors are therefore
subject to particulates generated by activities within that city.  Eliminating
these two monitors, and averaging the concentrations within the other two
intervals gives the following results as one approaches the Las Vegas CBD from
the largely undisturbed rural/recreational area to the east:  (1) background
concentration - 32 yg/m3, (2) outlying concentration - 69 yg/m3, and (3) down-
town concentration - 100 yg/m3-  The dashed line in Figure 14 connects the
central values (open circles) of the three groups.

     Plots of the 24-hour concentrations at each of the 12 sites are shown in
Figure 15.  The individual frequency distributions have been arranged from
left to right in order of distance from the city center (Site E).  Several
features of this figure are of note.  Most striking are the very high concen-
trations joined by the solid lines.   These were observed on June 17, 1975 and
clearly show the effect of regionwide blowing dust.  This was the only TSP
observation day during the 12-month period when blowing dust was reported in
the 3-hourly local climatological data summaries for the McCarran International
Airport.  On this date, visibility at the airport dropped from 20 mi at 1300
PST to 2 mi at 1600 PST, was still only 2 mi at 1900 PST, and had only
increased to 4 mi by 2200 PST.  During this dusty period winds blew from the
southwest at about 20 mph.  The only site not affected was the topographically
protected Las Vegas Wash site in the National Recreational Area east of the
city.

     The two other most extreme concentrations occurred at the Sahara Hotel
site on April 24, 1975 (895 yg/m3) and at the Power Plant site on January 12,
1975 (793 yg/m3) .  Both of these were isolated violations and must be attri-
buted to local and perhaps non-reoccurring sources.  On April 24th,  the next
highest concentration observed was 129 yg/m3 at Nellis Air Force Base;  on
January 12, 1975, the next highest concentration was 87 yg/m3 at the Las Vegas
Fire Department No. 2.

     Also of interest is the changing nature of the distributions as one pro-
gresses from the relatively undisturbed area represented by Site L west to
the center of Las Vegas (Site E).   The distributions appear to reflect the
influence of the city of Henderson,  particularly at Site D.  At the Stadium
(Site J), concentrations are generally quite low,  but five values exceed the
standards.  (No attempt has been made to relate these exeedances to stadium
events, but a relationship could be expected).  Concentrations are low at the
Nellis Air Force Base (Site G), but the number of  exceedances of the secondary
standard increase thereafter in moving toward the  city center until the Sahara
Hotel is reached.   After that, the number of exceedances decrease rather
steadily, reaching 6 at Sites C and  E.

     It is tempting to associate the improvement in air quality as one pro-
gresses inward toward the CBD from the Sahara Hotel with a decrease in fugitive
dust emissions from disturbed soil surfaces, unpaved roads, and the like, and
this may well be the case.  Such a hypothesis does not agree well with the
emission densities determined by PEDCo for the various sites, as shown in
Table 6, however.
                                     36

-------
   1200
   1100
   1000
    900
    BOO
10
 E •
 * 700
 4
 o
 Q.
 at
   600
   400
   300
   200
   100
                       6/IT/7S
       FIRE
       OEPT
C      H
HEALTH  NORTH
OEPT.    LAS
       VAGAS
FI      KAGJDBL
FIRE    SAHARA  POWER  AIRPORT NELLIS  STADIUM HENOER- BASIC   LAS VEGAS
OEPT    HOTEL   PLANT         AFB          P.O.    SCHOOL  WASH
  Figure  15.   Plots of  24-hour TSP concentrations at  12 sites ordered
                 by distance from city center  (Las  Vegas).
                                           37

-------
         TABLE 6.  NUMBER OF EXCEEDENCES OF THE SECONDARY STANDARDS
                   AND EMISSION DENSITIES AT SELECTED SITES
Site Number of
designator Site exceedences
I
F
H
C
E
Sahara Hotel
Fire Department No. 2
North Las Vegas
Health Department
Fire Department No. 1
15
11
9
6
6
a
Emission density
(tons/mi2 /day)
251
75
85
210
163
      •a
       Contributed by fugitive dust.

PARTICULATE LEVELS IN AND AROUND RENO

     Reno is located on the semiarid eastern slope of the Sierra Nevada Moun-
tain Range.  The Truckee River flows from the Sierras in California eastward
through Reno to Wadsworth on the eastern edge of the study area, and then
turns northward to drain into Pyramid Lake.  Temperatures are mild and sun-
shine is abundant throughout the year.  Annual precipitation averages around
7 inches, more than half of which occurs from December through March largely
as mixed rain and snow.  The prevailing winds at the International Airport
are from the west.  The approximate 1977 population of Reno is 82,500, and
that of Sparks roughly three miles to the east of the Reno CBD, is 33,800.36

     This discussion is also based on information provided in the 1977 PEDCo
report5 in which data from 12 sites are presented for the period from October
1974 through 1975.  The site types range from city/commercial in downtown
Reno and Sparks to rural sites on the western and eastern edges of the study
area.  Figure 16 shows the approximate locations of the monitors; Table 7
lists the site location information provided in the PEDCo report, plus moni-
tor heights when available.

     Figure 17 shows the similarity in the spatial patterns of annual mean con-
centration and number of observations exceeding 150 yg/m3 in the vicinity of
Reno and Sparks.  Comparison of these patterns with the site-type map at the
top of the figure indicates a strong relationship between the level of parti-
culates and the pattern of land use and its associated human activities.  The
highest concentrations occur in the busiest, most commercial areas, while the
lowest concentrations occur in the suburban/residential areas to the west.
Farther to the west at Lake Tahoe and Verdi, the mean concentrations were 22
and 18 yg/m3, respectively.  Concentrations for different site types can be
generalized as follows:  background concentration (Sites I and J), 20 yg/m3;
suburban/residential and rural (Sites C, E, F, and L), 46 yg/m3; suburban/
residential/commercial (Site G), 58 yg/m3; and city/commercial and suburban/
commercial (Sites B, H and A, D), 77 yg/m3.  On 15 of the 61 sampling days
during the 12-month period, concentration exceeded 150 yg/m  at one or more
of the Reno-Sparks sites.   Table 8 lists the concentrations observed at the
9 sites on each of these days.   On 8 of the 15 days, the concentration


                                      38

-------
                                             KEY
                                                                                         10
                                                                                                 15km
vo
                                               WASHOE COUNTY HEALTH DEPT.
                                               CAL-NEVA CLUB
                                               JESSIE BECK SCHOOL
                                               RENO AIRPORT
                                               NEVADA FISH AND  GAME
                                               MAMIE TOWLES SCHOOL
                                               GREENBRAE SCHOOL
                                               SPARKS NUGGETT
                                               LAKE TAHOE
                                               VERDI SCHOOL
                                               WADSWORTH FIRE DEPT.
                                               SEWER PLANT
10 ml
                               Figure  16.  The Washoe County DEP sampling network for TSP
                                             (after PEDCo5).
                \
                /

-------
TABLE 7.  SITE LOCATION DESCRIPTIONS FOR RENO MONITORS
Site
designator Site
A Health Department
B Cal-Neva Club
C Jessie Beck School
D International Airport
E Nevada Fish and Game
F Mamie Towles School
G Greenbrae School
H Sparks Nugget
I Lake Tahoe
J Verdi School
K Wadsworth Fire Department
L Sewer Plant
Site type
Suburban/
commercial
City/commercial
Suburban/
residential
Suburban/
commercial
Suburban/
residential
Suburban/
residential
Suburban/
residential/
commercial
City/commercial
Suburban/
commercial
Rural/residential
Rural/agricultural
Rural/agricultural
Height of
monitor
Cft)
15
50
20
15
20
20
15
35
___
15
—
35
                          40

-------
                                     SUB/RES/COM
                                      SITE TYPE
                  0    I   2   3 mi
                                       58*6     \
                 0123
                                   ANNUAL GEOM WEAK
                                        NUMBER
                                    OF OBSERVATIONS
Figure 17.  Spatial  distribution of TSP concentrations  and
            land  use in and around Reno.
                             41

-------
    TABLE 8.  TSP CONCENTRATIONS IN RENO AND VICINITY ON DAYS WHEN
              THE SECONDARY STANDARD WAS EXCEEDED  (10/74-9/75)
Site
Date
H
A
B
G
D
Concentration
10/02/74
10/20/74
10/26/74
11/07/74
11/13/74
11/19/74
12/19/74
01/12/75
01/18/75
01/24/75
01/30/75
02/11/75
03/01/75
04/24/75
06/17/75
No. >
150 yg/m3
134
174a
100
154a
197a
152a
137
170a
152a
206a
95
142
298a
161a
71
9
-
186a
94
113
182a
107
204a
144
166a
234a
165a
200a
156a
73
69
8
134
121
82
86
110
84
162a
139
140
154a
98
202a
132
87
66
3
128
366a
187a
71
161a
125
135
120
105
139
60
77
96
74
53
3
173a
181a
99
121
128
100
100
118
124
124
58
-
114
86
200a
3
E
yg/m3
130
153a
86
81
189a
111
149
105
108
149
68
124
87
73
41
2
C F

103
87
54
54
65
56
96
78
85 -
64
52
64
49 -
36 38
38 2
0 0
L

88
106
64
52
96
48
-
52
55
59
26
39
50
-
82
0
No. >
150 yg/m3

1
5
1
1
4
1
2
1
2
3
1
2
2
1
1
28
a
 Concentrations greater than the secondary standard.
                                 42

-------
exceeded the standard at only one  site.  The  greatest number of  simultaneous
exceedances was 5 on October 20, 1974, followed  by 4 on  November 13,  1974.
Thus on the occasions when high concentrations did occur,  they were almost
always confined to  small geographic areas.

     On October 20th, wind speeds  were light  but increased  to about 15 mph
early in the afternoon and remained near this level for  the balance of the
day.  The maximum wind speed reached during the  day was  24 mph.   In contrast,
on November 13th the winds were very light throughout the day with the fastest
mile of wind reported at the airport being 3  mph.  Although high wind speed may
have contributed to the high concentrations observed on  October  20th, this is
not the case on November 13th.  It is of interest to note  that a large high
pressure system that had dominated the Great  Basin for several days was just
beginning to weaken on the 13th.   This suggests  a regionwide accumulation of
the finer particulates, a possibility supported  by the fact that five of the
Las Vegas sites also exceeded the  standard on that day.  Using a Chi Square
analysis PEDCo designated November 13th as a  high regionwide concentration
day in both Reno and Las Vegas.

PARTICULATE LEVELS  IN AND AROUND TUCSON

     Tucson is a moderately large  city (1977  population  of 301,200)36 located
in a desert valley  that runs northwest-southeast between the Santa Catalina
and Rincon Mountain ranges to the  northeast and  east and the Tucson Mountains
to the west.  The climate is characterized by abundant sunshine  and a long
hot summer extending from April to October.   Precipitation averages about 11
inches per year.  Roughly half of  the precipitation occurs during summertime
showers (July-September); the balance is distributed throughout  the year,
with a secondary maximum during December and  January.  Surface winds are
generally light and frequently undergo a diurnal change  in direction as a
result of the daily heating and cooling cycle of the mountain slopes and chan-
neling by the local topography.  Early morning winds are usually from the
southeast quadrant; during the day the winds  become northwesterly.  Tucson has
no heavy industry and is supported primarily  by  tourism, agriculture and gov-
ernment services.   Copper mining and refining is extensive, but  is removed
from the immediate  metropolitan area.

     The air quality data reviewed in this section come  from two sources.
The first, a report by the University of Arizona (Moyers et al2), contains
averages of 24-hour concentrations measured at 11 sites  from September 1975
to October 1976 under a program sponsored by  the Electric Power Research
Institute (EPRI).   The second is the National Aerometric Data Bank (NADB)
which provides 1977 annual geometric means and individual 24-hour observations.
The monitoring network used in obtaining the  two sets of data is shown in
Figure 18 and described in Table 9.  The EPRI network comprised  the sites
numbered from 1 through 11, as designated in  the report; the NADB network
comprised all sites from 1 through 16 except  for the remote site at the
Research Ranch (Site 11).  Note that the monitor at Sites 10 and 14 are at
heights of 47  and  80 feet respectively,  while the remaining heights range
from 6 to 25 feet.
                                     43

-------
                  • 4
                           115
          '16
                               CORONA  DE  TUCSON
       GREEN VALLEY



       • 5
 km

 mi
                10
 20
	i
                      10
RESEARCH

RANCH
Figure 18.  TSP monitoring network in Tucson and vicinity.
                            44

-------
TABLE 9.  SITE LOCATION DESCRIPTIONS FOR TUCSON MONITORS
Designator Site address
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
7920 E. Tanque Verde
7901 E. Scarlett
Route 7 (Corona
De Tucson)
Davis Mountain Air
Force Base
245 West Esperanza
(Green Valley)
1810 S 6th Ave.
(S. Tucson)
1970 W Ajo Rd.
1019 Prince Rd.
Magee Rd.
Florence Highway
University of Arizona,
2nd and Palm
Research Ranch
Orge Gr Rd. and
El Camino De Terra
(Tucson)
3915 E Ft. Lowell Rd.
151 W Congress
2400 Harrison Rd.
Nogales Highway and
Hughes Access Rd .
Height of
monitor
Site type (ft)
Suburban/residential
Suburban/residential
Remote
Suburban/ indus tr ial
Suburban/ commer c ia 1
Center City/commercial
Suburban/ res id ent ial
Center City/commercial
Rural/near urban
Center City/commercial
Remote
Suburban/residential
Suburban/ commer c ial
Center City/commercial
Suburban/residential
Rural/near urban
16
18
13
14
13
18
18
25
6
47

6
15
80
6
6
                           45

-------
     Figure 19 displays the 1977 annual geometric means on the site map.  Two
features of the spatial distribution are of special interest.  The first is a
progressive increase in mean concentration as the urban core is approached;
the second is the large variation in concentration observed within the urban
core itself.  A significant part of the latter variation appears to be related
to differences in the height of the monitors.  In particular, the low concen-
trations that appear to be nonrepresentative of the urban core at Sites 10
and 14 are quite likely due to the greater measurement heights at those loca-
tions.  Reports reviewed for this study do not contain sufficient siting
information to speculate upon the reasons for other large differences between
proximate monitors (Sites 1, 2, and 15, for example).

     Some comment should be made, however, on the difference in concentration
observed at Sites 3 and 5, located more than 40 kilometers from the center of
Tucson.  Site 3 is located in a retirement community with a population of
about 1,000 and no through traffic.  Site 5 is in a somewhat larger retirement
community (population of about 5,000) located on an interstate highway with
moderate amounts of through traffic.  There are also two large copper mines
within five miles to the north.

     Figure 20 is a plot of annual concentration versus distance from Site 14
in downtown Tucson.  The EPRI data are represented by open circles and the
NADB data by filled circles.  Sites 10 and 14 have been omitted from the
figure.  If, as suspected, the relatively high concentration at a distance of
52 km is due to the impact of local sources, then the figure suggests that
background levels are reached within 40 km of the center of Tucson.

     Averaging mean concentrations for similar site types provides the follow-
ing relationships:  background concentration, (Sites 3 and 11),  26 yg/m3;
rural/near urban, (Sites 9 and 16), 61 yg/m3; suburban/residential (Sites 12,
1, 15, and 7),  72 yg/m3; suburban/industrial, (Site 4),  76 yg/m3;  and center
city/commercial (Sites 6,  8, and 13), 133 yg/m3.   In calculating these aver-
ages, Sites 10 and 14 were omitted because of their greater monitoring heights,
Site 2 was omitted because data were available only from Januray through July,
Site 13 was considered to  be center city rather than suburban because of its
proximity to Sites 8 and 10 both of which were classed as center city in the
NADB, and Site 5 was omitted because it was considered to be a special situa-
tion outside of the urban  influence of Tucson, but with local source impact.
Except for the inclusion of Site 11 data, the averages were calculated from
1977 NADB concentrations.

     Examination of the 24-hour observations for  1977 shows that the secondary
standard of  150 yg/m3 was  exceeded at one or more of the sites on 39 of the 61
sampling days;  Table 10 lists the concentrations  at all of the sites on these
39 days.   The sites have been arranged in the table so that the number of
exceedances  per site (and  degree of urbanization)  decreases from left to right.
Site 13,  at  the left of the table, experienced the greatest number of exceed-
ances by far.   There were,  in fact, only two days out of the 39 when the con-
centration at 'Site 13 was  observed to be below the standard;  on the  two other
days,  concentration was not reported.   Furthermore, on 19 of the days when
the concentration exceeded the standard at Site 13, all other reported values
throughout the  monitoring  network were below the  standard.  Clearly,  a local

                                     46

-------
                       •76
                               • (63)
              • 50
                                   CORONA DE TUCSON

                                 • (23)
           GREEN VALLEY
          o
        kini
        mi o
  10
    20
5
10
                                           29
                                              RESEARCH-
                                               RANCH
                                         (9/75-1
                                         10/76)
Figure 19.  Annual geometric mean TSP concentrations (yg/m3) in
            and around Tucson in 1977.  Values in parentheses are
            derived from incomplete record.
                              47

-------
   160
   140
                                           o 1977  GM (NADB)


                                          (  ) INCOMPLETE  RECORD


                                           O EPRI  STUDY AVERAGE
   120
         •

         O
*>  '00
o

I-

ir
t-
z
UJ
o
z
o
o

Q.

H
    80
       .  O
    60
    40
    20
                20         40         60        80

                        DISTANCE  FROM URBAN CORE, km
                                                           100
120
         Figure 20.   Average TSP concentration versus distance  from

                     the urban core.   (Site  14, Tucson).
                                   48

-------
TABLE 10.  TSP CONCENTRATIONS  IN TUCSON AND VICINITY ON DAYS WHEN THE
           SECONDARY  STANDARD  WAS EXCEEDED IN 1977
Site
Date
13
8
6
9
12
2
7
4
Concentration
01/13/77
01/19/77
02/06/77
02/12/77
02/18/77
02/24/77
03/02/77
03/08/77
03/14/77
03/20/77
04/01/77
04/07/77
04/13/66
04/25/77
05/07/77
05/13/77
05/19/77
05/25/77
05/31/77
06/18/77
06/24/77
06/30/77
07/06/77
07/12/77
182a
239a
250a
28 3a
301a
-
225a
227a
204a
179a
174a
202a
189a
191a
157a
-
186a
177a
170a
231a
161a
169a
158a
117
122
144
143
257a
211a
204a
156a
155a
146
133
76
129
98
111
93
124
-
108
112
132
126
108
97
123
101
103
111
206a
168a
2233
136
141
135
98
66
104
140
80
92
160a
124
112
117
132
120
110
91
154a
39
57
55
82
108
111
-
113
92
99
39
114
89
71
81
191a
104
98
109
68
124
105
86
77
66
-
-
-
152a
2153
112
177a
103
89
52
89
76
79
87
-
84
72
103
115
96
124
76
129
127
135
110
-
154a
1853
123
89
117
-
90
88
93
92
121
210a
98
86
96
100
75
116
-
71
47
79
171a
122
113
122
104
113
95
119
39
65
85
58
65
64
73
65
105
77
67
77
71
107
75
74
71
102
99
I82a
116
96
111
62
58
69
88
76
49
74
83
57
68
66
74
72
61
147
14
(Ug/m3)
67
69
69
108
109
135
109
114
110
66
38
63
74
64
54
82
75
58
82
72
83
77
63
92
1

65
-
89
93
130
60
101
115
85
70
47
84
90
53
58
87
83
67
72
93
73
85
75
63
16

41
62
52
66
69
95
85
107
80
59
33
52
48
53
48
88
55
43
70
61
59
63
-
-
10

65
83
69
111
124
89
99
113
101
78
52
81
77
86
51
115
71
65
83
82
83
85
80
65
15

-
-
-
-
-
121
110
58
85
73
48
64
64
46
68
102
69
66
86
84
70
74
61
55
5

56
66
62
90
72
73
119
78
92
59
50
56
44
57
43
66
59
58
81
75
12
88
57
47
3

-
-
-
-
-
-
-
-
-
-
23
28
24
30
24
40
35
25
41
43
40
38
-
3.1
No. >
150 ug/m3

1
1
2
3
5
5
2
3
1
1
1
1
1
1
1
3
1
1
1
1
1
1
1
1
                            (continued)

-------
                                                       TABLE  10 (continued)
Ul
o
Site
Date
13
8
6
9
12
2 7
4
Concentration
08/05/77
08/29/77
09/28/77
10/04/77
10/16/77
10/22/77
10/28/77
11/03/77
11/15/77
11/21/77
11/27/77
12/03/77
12/09/77
12/15/77
12/21/77
No. >
150 yg/m3
179a
161a
191a
160a
102
216a
202s
19la
2613
2793
1833
222a
413a
322a
204a

35
108
153a
118
100
208a
-
198a
204a
177a
175a
1703
247a
331a
228a
209a

16
110
-
84
87
119
131
-
-
407&
159a
177a
305a
661a
352a
624a

12
61
110
119
162a
81
86
131
184a
119
145
56
107
2903
158a
189a

6
88
-
82
67
85
93
132
90
-
Ill
102
149
1923
133
196a

5
63
73
50
60
-
70
91
84
84
83
147
119
112
121
- 5913

3 2
74
93
-
-
75
84
80
107
112
114
55
70
149
109
308a

2
14
(Pg/m3)
68
82
62
54
79
75
-
86
96
102
73
17
141
86
223a

1
1

-
-
76
52
67
-
78
100
35
104.
_
-
1783
130
117

1
16

63
50
44
39
61
52
64
47
75
58
38
63
81
51
261a

1
10 15

72 67
89 64
72
-
61
92
61
_
79
91
64
90
99
107
66

0 0
5

64
71
51
41
58
62
75
78
96
-
46
63
120
143
214a

1
3

27
29
26
14
42
28
13
20
32
27
17
20
23
22
84

0
No. >
150 ug/m3

1
2
1
2
1
1
2
3
3
3
3
3
6
4
10

85
             Concentrations greater than the  secondary standard.

-------
problem  exists at this  site.   Note also  that the two sites with  the next
highest  number of exceedances, (16 and 12,  respectively) are  city center
sites No.  8 and No. 6.

     Table 11 contains  wind speed and direction data for the  four days with
the greatest number of  exceedances.  The day with the most widespread problem
was December 21, 1977,  when 10 of the 13 sites reporting exceeded the stan-
dard.  The sustained high winds observed throughout the day suggest that this
is a classic case of wind-generated  fugitive dust  (see Section 3, p. 27).
Winds were from the southeast and exceeded  20 mph for much of the day; the
highest  wind speed reported at the airport  was 42 mph.  Based on afternoon
and evening wind speeds,  it is also  likely  that wind-generated dust contri-
buted heavily to the high levels observed on February 24, 1977;  of the 61
sampling days, February 24th experienced the second highest average wind
speed  (13.5 mph).  On the other hand, airport wind speeds suggest that wind-
generated dust was probably not a major  factor on either February 18, 1977,
or December 9, 1977.  On December 9th, two  concentrations in  excess of
400 yg/m3  were observed but the highest  wind speed was only 13 mph and the
wind speed for the day  averaged only 6.5 mph.
       TABLE  11.   WIND CONDITIONS ON DAYS WITH THE  GREATEST NUMBER
                   OF VIOLATIONS IN TUCSON
No. > Hour
150 ug/m3 Date (MST)
5 02/18/77 02
05
08
11
14
17
20
23
5 02/24/77 02
05
08
11
14
17
20
23
6 12/09/77 02
05
08
11
14
17
20
23
10 12/21/77 02
05
08
11
14
17
20
23
Wind
direction
(deg)
140
140
140
160
320
360
250
160
160
140
00
320
250
220
240
300
140
140
140
150
00
40
160
140
130
120
120
120
110
120
120
120
Fastest mile
LM nH ^ — —
winu ^^^^^^^^^^^^^^^^^
speed Speed
(knots) (mph) lrecti°n
7 14 SE
10
9
3
5
6
5
6
6 34 SW
6
0
8
20
25
15
14
8 13 S
9
8
5
0
4
5
6
25 42 SE
16
23
23
18
16
16
13
Average
speed
9.4







13.5







6.5







21.6







              Notes:  1 knot - 1.15 mph - 0.515 m/sec
                    Wind speed tabulated under "fastest mile" is a measure of the max-
                    imum wind speed during the 24-hr period.  It Is calculated from
                    the length of time required for one mile of wind to pass the anemo-
                    meter.  The direction tabulated under "fastest mile" is the direc-
                    tion measured during this period.

-------
PARTICULATE LEVELS IN AND AROUND PHOENIX

     Phoenix is located on the broad,  nearly flat plain of the Salt River Valley.
The Salt River runs from east to west  through the valley, but is almost always
dry.  Mountains rise in nearly every direction, the closest being the South
Mountains to the south and the larger  Phoenix Mountain Range to the north and
northeast.  The metropolitan area of Phoenix, with a population of over one  >
million, is the largest of the four urban areas studied.  In addition to the
city of Phoenix, it includes the cities of Scottsdale, Tempe and Mesa to the
east, and Glendale to the northwest.  In 1977 these cities had the following
populations:  Phoenix - 684,516; Scottsdale - 81,458; Tempe - 98,146; Mesa -
110,079; and Glendale - 73,730.36  The unincorporated area of Sun City lies
farther to the northwest, roughly 10 miles beyond Glendale.  The centralized
urban area is surrounded by pockets of urban development, particularly in the
extensive agricultural areas to the west and southeast.  These include Avondale,
Goodyear, Litchfield Park, Tolleson and Cashion to the west, and Chandler and
Mesa to the southeast.  Manufacturing  and tourism also contribute substantially
to the economy of the area.

     The climate is desert-like with low annual rainfall (about 7 inches),
high summertime daytime temperatures,  and low humidity.  Rainfall is centered
on two seasons of the year.  The first is from November to March when occa-
sional storms from the Pacific reach the area, and the second is during July
and August when Arizona is subjected to widespread thunderstorm activity.
May and June are the driest months and frequently experience no significant
precipitation.  The valley floor is rather free of wind.  During the spring
months southwest and west winds associated with the passage of low pressure
troughs predominate.  During the thunderstorms season local gusty winds often
occur, usually from an easterly direction.  Because of the topography and pre-  .
dominance of weak pressure gradients,  mountain-valley wind circulations are
common.  Typically, downslope east-to-west flow is established about 8 p.m.
and lasts until early morning; the flow reverses about noon and lasts until
after sunset.

     The principal basis for the following discussion is Volume I of the series
of four support documents prepared for EPA by TRW Environmental Engineering
Division entitled "An Implementation Plan for Suspended Particulate Matter in
the Phoenix Area."1  The air quality data analyzed in this document came from
17 monitoring sites operating throughout all or part of 1973 through 1975.
Figure 21 locates these sites with respect to the center of Phoenix, and Table
12 provides site type and additional location data.

     The duration of the sampling periods at the various sites was inconsis-
tent.  However, TRW used the available distributions to calculate the annual
geometric means and maximum 24-hour concentrations that would have been expec-
ted under a 60-sample per year measurement program.  This was usually done by
fitting a log-normal distribution to the data; in three cases the actual dis-
tribution appeared to seriously over predict concentrations at the high end,
and an adjustment to the log-normal fit was made.  Comparison of expected with
actual concentrations showed good agreement in the case of annual means; the ex-
pected maximal 24-hour values were generally higher than measured maximal val-
ues, however.  This would be expected for those cases where fewer than 60

                                     52

-------
                                                                               Carefrte Alrocrt
Ul
CO
                                                                   Paradise Valley  north Sefltt«d«l*/P«r«iHM v«n»v
                                              l St..Johns

                          Figure  21.   Hi-vol monitoring sites within the Phoenix area (TRW).x

-------
                            TABLE  12.   SITE LOCATION DESCRIPTIONS  FOR PHOENIX MONITORS
Ul
Site
designator
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Site
St. Johns
Litchfield Park
Sun City
Glendale
West Phoenix
North Phoenix
Paradise
North Scottsdale/
Paradise Valley
Carefree
Scottsdale
South Phoenix
Guadalupe
Chandler
Mesa
Downtown Phoenix
Arizona State
Central Phoenix
Site type
Rural/residential
(Indian Reservation)
Suburban/residential
(growth in progress)
Suburban/residential
Suburban-rural/
Residential-Agri-
cultural
Suburban/residential
Suburban/residential
Suburban/residential
(development)
Rural/residential-
commercial
Remote
Suburban/ residential
Urban/residential-
commercial
Rural/residential
Suburban/residential
Suburban/residential-
commercial
Urban/ commercial
Suburban/residential
Urban/residential
Representativeness
Site specific
Site specific
Area wide
Area wide
Area wide
Area wide
Site specific
Area wide
Area wide
Area wide
-
Area wide
Area wide
Area wide
Site specific
Site specific
Area wide
Area wide
Monitor
height
(ft)
15
10
25
20
5
5
5
5
5
15
36
5
21
5
23
15
22

-------
observations were made per year.  In part, the differences may be due to a
failure of the log-normal distribution to properly fit actual TSP data at the
high concentration end of the distribution.

     The expected annual geometric means and expected maximal 24-hour TSP con-
centrations are shown on site maps in Figures 22 and 23, respectively.  TRW
concluded that the lack of a consistent spatial pattern in both figures was due
to the influence of local fugitive dust sources and hence related to the site
environment.  For example, sites in rural/residential areas are frequently
surrounded by numerous vacant fields and unpaved roads.  Also, construction
activities, frequent in such areas, disturb the soil surface, making it sus-
ceptible to suspension by wind.  Other rural/residential sites were surrounded
by improved property lacking obvious fugitive dust sources.

     Table 13 presents average concentrations for the site categories developed
by TRW.  Data for the individual sites within each category can be found in
Table 5-1 of the TRW report.  Because of the influence of fugitive dust, it is
very likely that monitor height is a complicating factor.  Note in Table 12
that height ranges from 5 to 36 feet above the ground.  No attempt was made
to factor in monitor height in preparing Table 13, however.  Although the
highest annual average concentration occurs in the central city/residential-
commercial environment, the highest 24-hour concentrations occur outside the
city in rural/residential environments surrounded by fugitive dust sources.

     Closer examination of the data reveal that meteorological conditions and
the ambient particulate distributions associated with high TSP concentrations
fall into two patterns.  One is primarily a wintertime phenomenon characterized
by low wind speeds and low mixing heights.  Under these conditions,  high con-
centration centers around the city monitoring sites and often at Paradise
Valley.  Concentrations also increase at other stations throughout the region,
although to a lesser degree.  The other pattern occurs most frequently from
March to August during periods with strong wind gusts from the southeast or
west and above average wind speeds during the day.  Under these circumstances,
concentrations are highest at the rural and suburban residential sites.  These
responses of the particulate field to meteorological conditions support the
following two-part hypothesis put forth by TRW:

     1.   "Human activity, which is most densely focused in the
          city area, is responsible for suspension of substan-
          tial fugitive emissions.  These emissions are of higher
          density than those released at the rural sites and this
          is reflected by the higher concentrations produced dur-
          the stable atmospheric conditions of winter."

     2.   "Because vast expanses of agricultural land, unpaved
          roads,  and unimproved (but disturbed) soil surfaces
          surround the rural sites, suspension of dust by soil
          wind erosion is very likely a dominant factor affect-
          ing high particulate levels during gusty winds in the
          rural areas.   Soil erosion by wind is of less conse-
          quence  in the more developed areas."


                                     55

-------
                                                                                       41) Cartfr** Airport
t_n
                                                                  Paradise Valley
                                                                          9?
[143JNorth Scottsdale/Paradlse Valley
                                                                                                              U.Srg
                                                                                                       >tesa  U.S.-60. U.S.-80
                                                                                                       (124)
                           U.S.- 89
                                                                                                         indltr
                                Figure 22.   Expected  annual geometric means in  ug/m3  (TRW).1

-------
                                                                                  69) Carefree Airport
Ul
                                                                                528)North Scottsdale/Paradise Valley
                                                State  \~/|Downtown Phoenix
                         Figure  23.   Expected maximal 24-hour  concentrations in  pg/m3  (TRW).1

-------
                       TABLE 13.  AVERAGE TSP CONCENTRATIONS FOR SIX SITE ENVIRONMENTS (AFTER TRW) x
Ul
oo





Site environment category
Central City/Residential-Commercial
(Surrounded by fugitive dust sources)
Central City/Residential
(No sources)
Rural/Residential
(Surrounded by fugitive dust sources)
Suburban/Residential
(Surrounded by fugitive dust sources)
Rural/Residential
(No sources)
Remote

Expected
annual
geometric
mean
Sites (ug/m3)

11, 15 184

5, 16, 17 168

1, 7, 8, 12 160

6, 13, 14 129

3, 4, 10 97
9 41

Expected
maximal
24 -hour
level
(pg/m3)

438

487

638

362

250
169
Observed
maximal
24 -hour
level ,
1973-1975
(Ug/m3)

481

436

1099

380

256
277

-------
The relationships between wind  speed and concentration  leading  to  the  two par-
ticulate patterns are illustrated qualitatively  in  Figure  24.   Human-activity-
related emissions result in high concentrations  when dilution is minimal as a
result of light winds;  in contrast, the contribution from  wind  erosion is
insignificant during light winds, but  can be  of  overiding  importance in areas
with disturbed soil surfaces when wind speeds are high.
  c
  o
  in
  in
 O
  O)
                    Suspension by Wind
                                                Suspension  by Human Activity
                               Wind Speed
  c
  o
  re
  i-
  •»-»
  C
  01
  o
c

_o
                                       Dust  Storm
                       Resultant
                     Concentration
        ontribution
          from
       Human Activi
                                                    Contribution
                                                        from
                                                     Wind Erosion
                                 Wind speed
      Figure 24.  Effect  of wind  speed  on  ambient  suspended  particulate
                  levels  (TRW).:
                                       59

-------
                                  SECTION 5

               URBAN PARTICULATES:   CHARACTERISTICS AND SOURCES


     This final section discusses the characteristics and sources of urban
particulates in the Southwest, focusing on the contribution of wind-blown desert
dust.  It begins with a summary of particle size measurements that emphasizes
the inhalable size fraction.   Next, the discussion turns to the sources of par-
ticulates in several urban areas including Phoenix and Tucson, Arizona, and Las
Vegas and Reno, Nevada.  The  section concludes with a brief review of the con-
tribution of wind-blown desert dust to high TSP levels in Colorado, Utah,
Western Arizona, Texas and New Mexico.

PARTICLE SIZE DISTRIBUTIONS

     Particle size distribution data from the reviewed reports have been
assembled in this section in  an attempt to estimate the impact of dust from
desert areas on levels of inhalable particulates in urban areas.  The data
are very limited and unfortunately are somewhat contradictory.  The most use-
ful data appear to be those obtained in Phoenix in 1975 (IITRI).8

     In the Phoenix study, particle size information was acquired by three
techniques:  (1) the use of fractionating samplers, (2) sizing by microscopy,
and (3) the use of dichotomous samplers.  Data were obtained on five sampling
days at several heights and at upwind, downwind, and centrally located sites.
Table 14, derived from material presented in the TRW report3 but based on the
IITRI study, presents cross-city averages for heights of 3, 10, and 30 meters
for two wind speed conditions.  These concentrations were measured by Andersen
samplers.  Two features of data are of interest.  First, on the day with a
wind speed of 4 m/sec the total concentration decreases only slightly with
height, while on the very light wind or calm days the concentration at 30 meters
has dropped to about half of  the value at 3 meters.  This drop is also accom-
panied by a decrease in the proportion of particles greater than 15 ym.  Second,
the percentage of the particulates that are inhalable (defined as less than
15 ym in aerodynamic diameter) is higher on the light wind days than on
November 18th.  The average values for the 3-m and 10-m heights, for example,
are 44 percent inhalable on November 18th, and 59 percent inhalable on the
other three days.  These results agree qualitatively with expectations since
mixing is enhanced by increasing wind speed and the principal sources of the
larger particles subject to gravitational setting are at ground level.

     Particle size analysis of hi-vol filters from three heights were also
carried out by IITRI by microscopy.  Table 15, taken directly from Volume 3
of the TRW report,3 shows the results for one windy day (average wind speed
of 4.4 m/sec) and one calm day.  Over this smaller height interval, no

                                      60

-------
       TABLE 14.  HEIGHT VARIATION OF PARTICIPATE
                  CONCENTRATIONS FROM ANDERSEN
                  SAMPLER DATA (TRW,3 IITRI8')
             November 18,
            wind = 4 ra/sec
November 17, 21, 25,
   wind = 1 m/sec
Height  	  	
above       Total                   Total
ground  concentration  Percent  concentration  Percent
 (m)       (yg/m3)     <15 ym       (yg/m3)     <15 ym
3
10
30
112
107
105
46
42
38
149
90
79
52
66
66
Note:  Data from five sites were used to compile these
       averages.
       TABLE 15.   PARTICLE SIZE DISTRIBUTION FOR
                  SUSPENDED PARTICULATES MEASURED
                  IN PHOENIX,  SEPTEMBER 27  AND
                  NOVEMBER 14,  1975 (TRW3)
                             Percent of  particles
                           (by weight)  in size range

                         2 ym  2-8  ym  8-20 ym >20  ym
September 27 (windy)
Monitors
Monitors
Monitors
November
Monitors
Monitors
Monitors
at
at
at
14
at
at
at
20
15
5
feet
feet
feet
.07
.06
.05
2
2
2
.8
.2
.0
32
33
28
.2
.0
.3
65
64
69
.0
.6
.7
(calm)
20
15
5
feet
feet
feet
.14
.14
.13
2
2
2
.8
.5
.7
32
29
34
.4
.5
.9
64
67
62
.7
.9
.6
                           61

-------
significant differences were found in the particle size distributions with
either height or wind speed.  On both occasions, approximately 65 percent of
the particulates were greater than 20 ym in diameter.  If, as a rough approxi-
mation, half of the particulates in the 8-20 pm size range were less than
15 ym, only 18 percent of the particulates found on these filters would have
been classed as inhalable.  Note that optical micorscopy generally does not
account for particles smaller than about 1-2 ym; consequently, this technique
may severly underestimate the percent of particulates that are inhalable.

     The third summary prepared from the IITRI data, Table 16, compares results
obtained from dichotomous samplers with results obtained by the microscopic
analysis of hi-vol filters.   The data for each day listed in Table 16 are
average values from all available measurements; however, only 11 pairs of
dichotomous sampler — hi-vol observations from the same location (site and
height) are contained in the data set.   Table 17 lists the concentrations from
the collocated samplers.   The average concentrations measured by the two devices
on these 11 occasions were almost identical:   232 yg/m3 by dichotomous sampler
and 228 yg/m3 by hi-vol.   There were some very large discrepancies,  however,
but no consistent bias.   Linear correlation between the two sets of  data,
shown in Table 17, results in a correlation coefficient of + 0.66, which is
significant at the 5 percent level.
      TABLE 16.   PARTICLE SIZE  INFORMATION  FROM PHOENIX  SAMPLING  PROGRAM
                 (IITRI8)
                   Dichotomous sampler
High-volume sampler*
Number
of
Date samples
Number
of
<3.5 ym >3.5 ym samples
<5 ym 5-15 ym >15 ym
(Mass Percent)
11/17/75
11/18/75
11/21/75
11/23-24/75
11/25/75
Average
5
6
4
5
5
-
35
28
32
45
32
34
65
72
68
55
68
66
5
4
6
6
6
-
11
5
17
15
13
12
42
35
41
52
45
43
47
60
42
33
42
45
   * Estimated by optical microscopy.
                                     62

-------
     TABLE  17.   TSP  CONCENTRATIONS  MEASURED BY HI-VOLS  AND DICHOTOMOUS
                 SAMPLERS
                Concentration,  yg/m3
Concentration, yg/m3
Dichotomous
Date sampler
11/17/75
11/17/75
11/18/75
11/18/75
11/18/75
11/21/75
253
210
460
136
499
149
Hi-vol
221
206
441
287
257
108
Dichotomous
Date sampler
11/21/75
11/23-24/75
11/23-24/75
11/25/75
11/25/75
Average
250
122
106
170
197
232
Hi-vol
315
142
116
258
157
228
     Although a direct comparison between the results obtained by the two
techniques, is impossible, it is instructive to note that the percent found
to be less than 5 ym by microscopy is consistently much less than the percent
found to be less than 3.5 ym by use of the dichotomous sampler.  Also note
that microscopy indicates that, on average, 55 percent (12 plus 43)  of the
particulates are inhalable.

     A small amount of information is available from the PEDCo study of Las
Vegas and Reno.  In this report results are presented for three sites in or
near each city where size distributions were determined by microscopy.  When
averaged within each city, 30 percent by weight of the particulates were found
to be less than 15 ym.  In this analysis, the particles collected on sections
of the glass-fiber, hi-vol filters were resuspended in distilled water with the
use of an ultrasonic bath and refiltered onto a membrane filter prior to sizing.
In IITRI's analysis, particles were counted as collected on the filter through
the use of immersion oil.

     The data summarized above are too limited and the estimates cover too
broad a range to be of much help in defining inhalable particulate levels
within the urban areas.  Further, no information was found on the proportion
of incoming particulates that is less than 15 ym.  Thus, it appears that a
comprehensive measurement program will be required before the impact of dust
from desert areas on the levels of inhalable particulates within the cities
can be reliably ascertained.

PARTICULATE SOURCES

Phoenix

     Several studies of Phoenix reviewed for this report examined the contri-
bution of wind-blown dust from the desert  to high levels of TSP in urban loca-
tions.  Two of these studies concluded that high particulate  levels  should not
be attributed to wind-blown dust from the undisturbed desert  region  surrounding


                                     63

-------
Phoenix and its suburbs.  Instead they implicated anthropogenic sources as  the
prime contributors to the TSP problem.  Several of the other studies present
evidence that supports this conclusion.

     The first two studies relied on different techniques to reach very similar
conclusions.  IITRI's Aerosol Sampling and Analysis8 used several analytical
techniques to characterize Phoenix particulates including mass flux determina-
tions, particle size and morphological analysis, and elemental analysis.  The
second report, Dust Transport in Maricopa County,7 relied exclusively on the
emissions inventory/modeling approach.

     The purpose of the IITRI study was to determine whether high TSP levels
in Phoenix are due to human activities within the Phoenix area or to long-range
transport from the surrounding desert.  To answer this question, authors of
the study designed a sampling program to collect and measure aerosols both
upwind and downwind of Phoenix.  This sampling program took into account the
typical diurnal pattern of wind direction in Phoenix:  day wind from the west,
night wind from the east.  Microscopic and elemental analysis was used to
determine the physical and chemical characteristics of the particulate samples
collected.

     The study found that the sizes and morphologies of the majority of mineral
particles found in the Phoenix samples were not typical of wind-eroded materials.
Instead of being frosted, indicating tumbling and collision with finer parti-
cles, particle surfaces tended to be scratched with fine particles embedded
suggesting mechanical wear.  In addition, TSP concentrations were not signifi-
cantly higher in a day when the wind-speed increased to 5 m/sec compared to
other sampling days when wind speeds averaged 1 m/sec.  The report concluded
that "wind is not the primary source for suspension of particles" and "long-
range transport of aerosols by winds from the surrounding deserts is only a
minor source for suspended particulates."

     The report went on to state that "vehicular traffic, especially on unpaved
roads is the primary generator of suspended particles."  According to the
authors, the appearance of smooth-surfaced, worm-shaped,  rubber tire particles,
fine carbonaceous particles, and sharp, angular mineral fragments indicate
vehicle travel over paved roadways while torn, rough-surfaced rubber tire par-
ticles and fine carbonaceous particles, together with rounded clay coated
minerals indicated vehicle travel over unpaved roads and  shoulders.

     The report cautions that agricultural activity could also produce the
round, clay-coated mineral fragments attributed to vehicle travel over unpaved
roads.  The authors felt, however, that the relative absence of plant or fer-
tilizer materials in the samples suggested that agricultural activity generated
less suspended particulate matter than vehicular traffic.

     The second report,  Dust Transport in Maricopa County, Arizona relies on
a behavioral model of suspended particulate dynamics to analyze the contribu-
tion of dust,  including  wind-blown desert dust, to high urban particulate
levels in and  around Phoenix.   Three major categories of  dust emissions were
considered:   (1)  traffic dust from unpaved roadways and off-roadway vehicle


                                     64

-------
emissions, (2) "diurnal dust" - windblown desert dust and agricultural and
construction emissions and  (3) "constant dust" consisting of tailing piles,
cattle feed lots, and vacant soil  emissions.  The  analysis  is limited to the
late fall, a period characterized  by fairly light  winds, when - according to
the study - high ambient particulate concentrations are usually observed.
The report concluded that:  "observed high particulate concentrations in
Maricopa County urban areas during late fall and winter periods of atmospheric
stability are associated with local fugitive dust  sources.  Because of light
drainage winds prevalent during these periods, advective transport of dust
from countryside to the urban areas is not an important contribution to urban
supermicrometric particulate concentrations,"

     Several other studies also examined TSP nonattainment  in the Phoenix
metropolitan area.  TRW Environmental Engineering  Division  produced a study
for EPA that focused on the development of control strategies for total sus-
pended particulates in the Phoenix area.  As part  of this study, TRW developed
an emission inventory37 that included estimates of emissions from wind-blown
desert dust, then used this inventory to model air quality  in the Phoenix area.

     The TRW study relied on both  the Climatological Dispersion Model (COM)
and the Linear Rollback model (LR) to simulate the relationship between emis-
sions and ambient air quality in the Phoenix area.  The study assumed that
particulates smaller than 20 pm aerodynamic diameter could  be adequately
modeled with the COM model, and those greater than 20 ym could be adequately
treated in the context of the LR modeling concept.  Particle size distributions
for emissions were estimated from  a number of different sources.

     The model predicts that nearly all the TSP level (excluding background)
at 12 of the 13 sites monitoring in 1975 was caused by emissions from unpaved
roads, entrained street dust, construction activities, or wind erosion.  Wind-
erosion includes emissions from agricultural fields, unpaved roads, undisturbed
desert, tailings piles, and vacant lots.  Table 18 indicates that undisturbed
desert and disturbed soil are by far the largest sources of these emissions -
roughly equal in magnitude.  At the Sun City site  however off-road vehicles
were responsible for most of the TSP levels.  Table 19 indicates the relative
importance of these four types of  sources at each  of the 13 sites; in all but
a few cases, the contribution of wind-erosion is modest relative to the contri-
bution from unpaved roads and entrained street dust.  "Monitors which were
most dramatically affected by wind-erosion emissions tended to be located in
the rural areas under development, such as the Paradise Valley and North Scotts-
dale/Paradise Valley sites.  (Figure 21 indicates  the location of the 13 moni-
toring sites while Table 12 provides a brief description of the site environ-
ments.)  Unfortunately, the impacts of the various classes  of wind-erosion
emissions on TSP levels were not disaggregated in  the report.

     Figures 25 through 28 display 1975 emissions  from the  four major sources.
Although these figures provide a vivid picture of  the spatial distribution of
emissions, the reader is cautioned not to use them to compare emissions from
different types of sources since all the figures use slightly different scales
to measure emission magnitudes.  These figures show that emissions from all
four sources, even wind erosion, are much greater  in the more urbanized portion
of the study area.  In Figure 28,  wind erosion emissions include those from


                                     65

-------
agricultural fields,  unpaved roads,  undisturbed desert, tailings piles, and
disturbed soil.   (The monitors roughly define the urbanized portion of the
study area; Figure 21 shows the location of each monitor.)


    TABLE 18.   1975 TOTAL PARTICULATE EMISSION INVENTORY FOR PHOENIX STUDY
               AREA (TONS/DAY) (AFTER TRW3)
Source category
Stationary Sources
Area Sources
Mobile Sources
Aircraft
Agricultural Tilling
Venicles on Unpaved Roads
Aggregate Piles
Cattle Feed Lots
Off-Road Vehicles
Construction
Paved Streets
Wind Blown Agriculture
Wind Blown Unpaved Roads
Wind Blown Undisturbed Desert
Wind Blown Disturbed Soil
Tailing Piles
Total

1st
quarter
22
1.4
11
.4
22
1281
.10
6
71
100
248
4.1
1.4
160
161
1.4
2091

2nd
quarter
22
1.4
11
.4
30
1365
.10
6
71
100
248
4.0
2.1
244
248
1.4
2354
1975
3rd
quarter
22
1.4
11
.4
1
1365
.10
6
71
100
248
3.8
2.7
321
323
1.4
2478

4th
quarter
22
1.4
11
.4
17
1086
.10
6
71
100
248
4.0
3.8
450
456
1.4
2478

Annual
average
22
1.4
11
.4
20
1281
.10
6
71
100
248
4.0
2.5
294
297
1.4
2360
                                     66

-------
TABLE 19.  IMPACT OF MAJOR SOURCES ON TSP LEVELS (AFTER TRW3)
Contribution of
major
Monitor site
Central Phoenix
South Phoenix
Arizona State
Glendale
North Phoenix
North Scottsdale/Paradise Valley
Scottsdale
Mesa
Downtown
St. Johns
Sun City
Paradise Valley
Chandler
TSP
in 1975
112
144
169
101
121
. 149
115
117
200
145
88
184
119
Unpaved
roads
25
75
35
30
26
24
27
32
42
93
15
42
64
suspended particulates from four
sources (vg/m3)
Entrained
dust
31
20
59
17
28
8
33
35
70
2
12
14
10
Construction
activities
4
2
7
7
7
14
6
8
8
0
3
17
7
Wind
erosion
19
15
33
15
28
71
16
10
40
18
2
78
5
Percentage of TSP
level contributed
from four major
sources & background
96.3
98.2
96.4
97.2
97.8
98.3
96.5
97.7
94.1
98.3
55.2
98.1
96.6

-------
Figure 25.  Average daily dust emissions from unpaved roads,  1975.   (TRW37)

-------
Figure 26.  Average daily street dust emissions entrained by motor vehicles on paved streets,  1975.   (TRW37)

-------
Figure 27,  Average dally dust emissions from construction activities, 1975.  (.TRW37)

-------
X, o


   o
    •

   o
 Figure 28.   Emissions of fugitive dust  arising  from wind erosion, average daily emissions, 1975.  (TRW37)

-------
     Figures 29 and 30 depict emissions from the two most significant wind-
erosion sources:  disturbed soil surfaces and undisturbed desert.  Figure 29
illustrates that emissions from disturbed soil surfaces are highly concentrated
in the urban portion of the region.  For these calculations, the vulnerable
soil surfaces were considered to be vacant lots, parking lots, and dirt resi-
dence yards.  Emissions from the undisturbed desert, which are similar in aver-
age daily magnitude, are distributed over a much larger area to the north and
south of Phoenix.

     PEDCo Environmental Inc. examined major sources of fugitive dust in the
Phoenix - Tucson AOCR, as well as in five other AQCR's in New Mexico, Arizona,
and California.1*  Table 20 shows PEDCo's emission estimates for the Phoenix-
Tucson AOCR.  In Maricopa County, where Phoenix is located, agriculture is by
far the largest source of emissions, followed by unpaved roads.  The report does
not estimate emissions from paved roads or undisturbed desert.

     PEDCors conclusion that wind-erosion from agricultural lands is the pri-
mary source of emissions in Maricopa County differs from the conclusions of
other studies examined for this report that unpaved roads are the principal
source of emissions.  Although it is difficult to resolve this discrepancy,
the TRW report provides some evidence that might explain the magnitude of agri-
cultural emissions in the PEDCo study.  TRW notes that PEDCo assumed that agri-
cultural soils experienced the same low moisture levels as native soils;  they
conclude that neglecting the effect of irrigation on agricultural soil losses
has probably resulted in emissions estimates which were substantially over-
stated in the PEDCo inventory.

Tucson

     A study done by the University of Arizona for the Electric Power Research
Institute (EPRI),2 discussed earlier, utilizes many techniques to examine
sources of particulates in the Tucson area.  Some of these techniques, chemical/
elemental analysis and examination of spatial patterns for example, were used
in the studies of Phoenix discussed above.  Others, such as enrichment factors,
inter-species correlations, and pattern recognition are used almost exclusively
by the Tucson study.

     The evidence compiled in the study indicates that a great deal of the par-
ticulate matter in Tucson is soil.   Table 21 shows that more than 50 percent
of the particulate matter was composed of soil at 9. of the 11 sites studied.
Site locations may be found in Figure 18 of Section 4.  The contribution of
soil to particulate concentrations ranges from 82.7 percent at a suburban/
commercial site to 48.2 percent at Site 11, the "background" location.  Examin-
ing all the sites, however, the report concluded that there appears to be per-
haps no significant differences in the relative amounts of soil-like material
in Tucson and the background location.

     Since the soil is composed of many elements, it cannot be measured direc-
tly; the analysts had to rely on several indirect techniques to determine the
soil content of their particulate samples.  They used correlation analysis to
discover which elements appeared to cluster together in each sample; they found
that the elements Al, Fe, Si, Ti, Li, Rb, K, Ca, Mg, Na, Mn, and Sr were corre-
lated with each other at both urban and background locations*.  (Tables in the
original report list the inter-species correlation coefficients, r, for the

                                     72

-------
Figure 29.  Fugitive dust emissions arising from wind erosion of
            disturbed soil surfaces, first quarter, daily average,
            1975.  (TRW37)
-*?

-------
Figure 30.  Particulate fugitive emissions arising from wind erosion of undisturbed desert in Phoenix
                , first quarter, daily average, 1975.  (TRW37)
area

-------
                   TABLE  20.   PHOENIX-TUCSON AQCR SUMMARY SHEET, ESTIMATED ANNUAL EMISSIONS FROM
                                FUGITIVE DUST SOURCES  (PEDCo1*)

                                                                                    Aggregated       Cattle       County
            Unpaved roads       Agriculture       Construction   Tailings piles       storage        feedlots     emission
           Vehicle  Emission           Emission  Acres   Emission         Emission  103    Emission  103   Emission  Total,
 County    mi/day   tons/yr    Acres   tons/yr   per/yr  tons/yr   Acres  tons/yr   tons   tons/yr   head  tons/yr   tons/yr

Gila        13,266              1,300             -              1,785              30
                     9,200                50              -              5,430             90             -      14,770

Maricopa   121,758            408,500           3,775              -               552             235
                    82,200            175,000            62,440              -             1,590             250   321,470

Pima        45,530             50,700           1,440            2,680             212              13
                    34,910              8,900            24,160            9,430            540              20    77,960

Final       58,936            238,000             -              1,100             120             230
                    39,750            126,500              -              7,100            340           1,010   174,700


SCruz        9'258              1'400             -                -                75              -
                     6,250                50              -                -              220             -       6,520
AQCR
Activity
Total      249,104            699,900           5,215            5,565            989             451

AQCR
Emissions
Total               172,310            310,500            86,590           21,960           2,780           1,280    595,420

-------
             TABLE 21.  SOUTHERN ARIZONA ATMOSPHERIC PARTICULATE
                        MATTER, PERCENT COMPOSITION OF MAJOR
                        COMPONENTS AT EACH LOCATION (EPRI2)

             Location  Soil  SO^ - NH^  NOa   Pb   Zn - Cu - Cd
1
2
3
4
5
6
7
8
9
10
11
81
52
61
60
82
65
57
49
54
61
48
.8
.8
.7
.9
.7
.2
.9
.6
.2
.4
.2
4
2
5
4
4
2
4
3
4
3
16
.1
.7
.9
.4
.2
.9
.1
.5
.1
.5
.4
2.
1.
2.
1.
1.
1.
2.
1.
1.
1.
2.
2
7
6
8
6
5
0
8
8
4
9
0
0
0
0
0
0
0
0
0
0
0
.74
.38
.10
.48
.32
.61
.58
.45
.28
.51
.08
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
38
15
30
26
33
26
30
22
18
55
47
background location as well as a "typical" urban site."  Figures 31 and 32 use
cluster analysis, a form of pattern recognition, to display the information
conveyed by the correlation coefficients more vividly.  The height of the lines
connecting elements is proportional to 1-r; thus short vertical lines designate
element pairs that are highly correlated, long lines, pairs that are more weak-
ly correlated.

     The authors observe that the clustering of these elements suggests a com-
mon source (or type of source) for these species and that a likely source for
many of these elements would be soil material.  To test this hypothesis they
examined the enrichment factors of each element.  Enrichment factors, E, are
defined as:

                  Air Concentration of Element
              E _ Air Concentration of Reference Element
                  Crustal Concentration of Element
                  Crustal Concentration of Reference Element

An enrichment factor of one indicates that the relative concentration of a
given element is the same in both air and soil, supporting the hypothesis that
soil is the primary source.  An enrichment factor significantly greater than
one provides tentative evidence that another source besides soil is contribut-
ing the element to the atmosphere.  In this study the authors use Al, the sec-
ond most abundant crustal element, as the reference element.

     As Figure 33 shows, enrichment factors for Fe, Si, Ti, Li, Rb, K, Ca, Mg,
Na, Mn, and Sr are all close to one lending credibility to the hypothesis that
soil is the source of these elements.  Unfortunately, this technique does not
                                     76

-------
             HIERARCHICAL CLUSTERING - LOCATION  2
NH^SC^CU CD ZN  PB NO^ Nl CO CS CA K  FE MG NA CR RB Tl  AL  SI SR LI MN MS
  Figure  31.   Dendogram of feature clustering for desert urban
               particulate matter, (EPRI2)-
                                77

-------
             HIERARCHICAL  CLUSTERING - LOCATION  11
NH*SO^CU CD  PB ZN NOj MG SR LI  MN  NA  Nl CO CA AL FE  RB  SI  Tl CR CS  K  MS
Figure 32.  Dendogram of feature clustering  for  desert background
            particulate matter, (EPRI2).
                               78

-------
                              Enrichment Factors
10-
      • Tucson
      A Background
                                                           •         A
                                             •  •
         -1	f	1	1	1	T-
    Mg  Sr Fe Na  Mn  Cr Si  K  Co Ti  Ni  Rb  Ca  Li Cs  Zn Cu SO; Tl  Cd NO^ Bi  In NH4 Pb
 Figure 33.   Enrichment factors for  species  in desert background
               and urban particulate matter, (EPRI2).
                                   79

-------
distinguish between wind-blown desert soil and soil material suspended as the
result of human activities such as travel on unpaved roads.

     Urban-background concentration ratios shed some light on the possible
sources of soil-related particulate matter.  As Table 22 shows, urban-background
ratios exceed one for all the soil-related elements at all urban and suburban
sites.  These ratios typically are higher for the sites within Tucson, lower
for sites 3 and 5 which lie between Tucson and Research Ranch, the background
site.  Although this evidence does not reveal specific urban sources, the
authors of the study conclude that it does implicate urban activity as an
important source of soil and other particulate matter in the Tucson area.

Las Vegas and Reno

     In a report prepared for the U.S. Environmental Protection Agency,5 PEDCo
Environmental Inc. estimated emissions, including emissions from natural sur-
faces for both the Las Vegas and Reno areas.  Tables 23 and 24 summarize their
emission estimates for 1975.  Approximately 80 percent of the total emissions
in both cities come from fugitive dust sources.  The data indicates that natural
surfaces are an important but not predominant source of fugitive dust emissions.
In Las Vegas, unpaved roads are the largest source of emissions, followed by
natural surfaces; paved streets and cleared areas also generate large amounts
of particulate emissions.  In Reno, unpaved roads contribute more than a third
of all emissions.  Natural surfaces and paved roads are also major sources of
fugitive dust.

     Figures 34 and 35 show the distribution of 1975 particulate emissions in
Las Vegas and Reno; these figures do not include emissions from natural sur-
faces.  As was the case with Phoenix, emissions in both cities are concentrated
in the more urbanized portion of the study area.  (Note, for example, the mag-
nitude of emissions at points within the city boundaries where interstate high-
ways intersect).  In Las Vegas, many of the outlying grids with high emission
densities are the location of large point sources.  In Reno, the three grids
north of the city with high emission densities all experience a large contri-
bution to emissions from unpaved roads.

OTHER STUDIES

     The conclusion that emerges from the studies of Phoenix, Tucson, Las Vegas,
and Reno is that fugitive dust from urban activities, rather than wind-blown
desert dust, appears to be responsible for the high particulate concentrations
observed in those cities.  Evidence from studies of other areas in the southwest
provide some support for this conclusion, although this evidence is not as
complete or detailed as that available for the four cities already discussed.

     One study38 compiled emission inventories for the two counties that form
Arizona's western border:  Mojave and Yuma.  Table 25 summarizes the inventories
for both counties.  Although they do not contain estimates of emissions from
undisturbed desert areas, these inventories do indicate the magnitude of other
fugitive dust sources in the region.   Fugitive dust sources account for approxi-
mately 98 percent of all emissions in both counties.  In Mohave County, the
principal sources of fugitive dust are unpaved roads and vehicle travel on

                                     80

-------
TABLE 22.  URBAN/BACKGROUND CONCENTRATION RATIOS (EPRI2)
Site
Element
Pb
Ca
K
Sr
Mg
Li
Rb
Si
Al
Mn
Na
Fe
Mass
Ti
Cs
In
Tl
Co
NO;
Ni
NflJ
Zn
sol;
Cd
Cr
Cu
Bi
1
10.7
5.8
4.9
4.7
4.4
3.9
3.7
3.7
3.5
3.5
3.9
3.4
2.2
2.9
2.2
3.8
3.3
2.8
2.0
2.0
1.3
1.7
1.3
1.3
1.3
1.9
1.0
2
12.1
6.2
5.8
5.4
5.3
5.2
4.4
4.3
4.3
4.3
4.3
4.1
3.7
3.1
3.0
2.9
2.8
2.8
2.5
1.9
1.6
1.6
1.5
1.5
1.4
1.4
1.1
3
1.4
1.7
1.5
1.9
1.7
1.9
1.7
1.3
1.3
1.4
1.8
1.4
1.2
1.2
1.1
2.7
2.9
2.2
1.5
1.2
1.0
1.3
1.3
0.7
0.5
1.2
0.9
4
10.4
5.1
4.6
4.6
4.5
3.7
3.8
3.3
3.4
3.3
3.7
3.5
3.1
2.7
2.1
4.1
3.0
2.8
2.5
2.8
1.6
1.8
1.6
3.9
2.0
1.3
1.1
5
5.5
3.4
3.1
4.3
3.3
3.2
3.8
3.3
3.3
3.0
3.7
3.0
2.3
3.2
2.0
3.0
3.0
2.8
1.5
1.1
1.0
1.3
1.1
0.9
1.0
1.9
0.7

27
11
5
13
8
5
5
4
5
4
5
5
4
3
2
5
6
4
2
2
1
2
1
2
2
3
1
6
.0
.7
.9
.8
.9
.9
.0
.9
.3
.9
.5
.0
.2
.9
.7
.1
.0
.4
.7
.8
.5
.2
.6
.4
.4
.3
.3
7
10.5
5.4
2.8
4.3
3.8
3.5
2.6
2.6
2.8
3.0
2.8
2.6
2.6
2.2
1.6
3.8
3.7
2.2
2.1
1.4
1.4
1.6
1.5
0.8
1.1
1.9
1.0
8
18.0
6.2
4.4
5.7
5.2
4.0
3.8
3.6
3.5
3.5
4.5
3.7
3.7
2.7
2.3
4.7
4.2
3.1
2.3
2.3
1.8
1.5
1.8
2.4
1.6
2.0
1.0
9
10.3
2.1
4.0
2.8
2.2
2.8
3.1
2.8
2.8
2.4
3.5
2.4
2.6
2.2
1.7
2.3
2.5
2.9
1.8
1.6
1.3
1.2
1.5
1.3
1.0
0.9
1.0
10
15.7
6.0
3.5
6.0
4.5
3.7
3.5
3.5
3.5
2.9
3.2
3.0
3.2
2.5
1.9
3.5
3.2
2.4
2.0
2.7
1.2
1.6
1.2
4.4
2.0
10.9
0.8
                           81

-------
TABLE 23.  LAS VEGAS PARTICIPATE
           EMISSION INVENTORY SUM-
           MARY, TONS/YEAR (AFTER
           PEDCo5)
      Source category
1975
Fuel Combustion:
  Residential                 37.2
  Commercial/Institutional   134.4
  Industrial                  55.8

Industrial processes        2986.0

Burning                       51.9

Mobile Sources:
Aircraft
Railroad
Auto exhaust
Off-highway
Fugitive Dust Sources:
Construction
Normal paved streets
Dirty paved streets
Unpaved roads
Sand and gravel pits
Agriculture
Cleared areas
Heavy equipment storage
Playgrounds
Unpaved parking lots
Road shoulders3
Railroad right-of-ways
Horse corrals
Natural surfaces
Total
71.8
13.5
795.7
206.8

620.0
2122.7
43.7
5818.0
711.0
180.0
2309.9
27.6
5.0
38.5

202.9
51.0
5591.7
22074.3
a
              TABLE 24.  RENO PARTICULATE EMIS-
                         SION INVENTORY SUMMARY,
                         TONS/YEAR (AFTER PEDCo5)
Source category
Fuel combustion
Industrial processes
Burning
Mobile Sources:
Aircraft
Railroad
Auto exhaust
Off -highway
Fugitive Dust Sources:
Construction
Normal paved streets
Dirty paved streets
Unpaved roads
Sand and gravel pits
Agriculture
Street sanding
Cleared areas
Heavy equipment storage
Playgrounds
Unpaved parking lots
Road shoulders
Railroad right-of-ways
Horse corrals
Natural surfaces
Total
1975
388.2
23.0
1099.5

10.7
24.7
437.3
85.5

814.0
1034.1
20.6
3969.0
460.0
121.0
140.0
40.1
3.4
50.9
38.8
2.8
35.6
9.0
2323.2
11131.2
 Road shoulder emissions  included
 with unpaved  parking  lot emissions,
                                     82

-------
4020 to
4010 km
4000 k»
3990 U
3980 ta
     650 k»
                       660
                                        670 km
                                           0	L
                                                  SCALE
                                                KILOMETERS
                                                          680 km
                                                           10
                                                 MILES
                                               1  ?  ?   1
                                                                            690 bi
                                       I	I   0-49  ton/mi2/yr
                                       tg-tag*  50-99  ton/mi2/yr
                                       HIIIIIIIIIIIIIH 100-149 ton/miVyr
                                       BBI150-199 ton/mi2/yr
                                               >200  ton/mi2/yr
        Figure 34.  1975 Las Vegas participate emission density, ton/mi /yr.
                    (After PEDCo5).  Excludes  emissions from natural surfaces.
                                      83

-------
        242.5 (a EAST
        100.000 rt. lAit
HI.4M ft. mni
' 43M 4 » HOOTH
43M.Z If (OCX
IU.M n. Htm
        uo.oot rv IMT
        24!.3 IK CASr
                                                                          4353.7 t. MOTH
                                                                          tM.OM ft. Mm
                                                                   1M.4t» ft. IAJT
                                                                   HI.» tt (AST
                                           0-49 ton/mi2/yr

                                          50-99 ton/mi2/yr

                                         100-149 ton/mi2/yr

                                         150-199 ton/mi2/yr

                                           >200 ton/mi2/yr
      Figure 35.   1975  Reno  participate emission density,  ton/mi2/yr.

                   (After PEDCo5).   Excludes emissions  from natural surfaces.
                                        84

-------
off-road surfaces.  In Yuma County, travel on off-road surfaces is the primary
source of emissions, followed by unpaved roads and exposed surfaces.  The
latter category consists of agricultural land and land cleared for development.

                  TABLE 25.  1976 TOTAL EMISSIONS BY SOURCE
                             CATEGORY, (TONS/YR)  (AFTER
                             PEDCo38)
                                             Mohave    Yuma
                    Source category          county   county

                Point Sources                    631      924

                Traditional Area Sources         736    3,968

                Fugitive Dust Sources:

                  Paved Roads                  2,134    4,374
                  Unpaved Roads              105,830   66,045
                  Off-road Vehicles           99,375  183,425
                  Construction Activity          986    1,422
                  Agricultural Activity           26    8,172
                  Exposed Surfaces3              360   14,414

                Total                        210,078  282,744

                 alncludes emissions  from agricultural sur-
                  faces and land cleared for development.
                  Does not include emissions from undistrubed
                  desert.

     Other  studies  relied on  the microinventory  technique to analyze particu-
 late problems.  One such study prepared by  Engineering Science for the U.S.
 Environmental Protection Agency,6 analyzed  particulate problems at five non-
 attainment  sites in the Albuquerque,  New Mexico  area.  The study inventoried
 all sources within  a one-mile radius  of the nonattainment monitors including
 fugitive dust sources such as paved and unpaved  roads, areas subject to wind-
 erosion, and aggregate storage piles. The  study utilized the Air Quality Dis-
 play Model  (AQDM) to relate emissions to ambient air  quality.  The report
 concluded that  four anthropogenic fugitive  dust  sources - unpaved roads or
 driveways,  paved roads, tire  and exhaust emissions, and unpaved parking lots-
 contributed 42  to 61 percent  of  the TSP concentration at  these five  sites.

     Another study12 compiled microinventories for  35 nonattainment  sites  in
 Colorado and Utah as well as  8 nonattainment  sites  in Montana, Wyoming, and
 North  Dakota.   The  authors concluded  that  "in general,  the  sources associated
 with nonattainment  areas in the large  urban  centers  of Region VIII  such as
 Denver, Salt Lake City, Colorado Springs,  and Pueblo  were traffic-related
 reentrained dust from streets, winter road  sanding, and motor-vehicle  exhaust).
 In some moderate sized cities such as Grand Junction  and  Rapid City,  traffic
 related emissions also appeared  to be a major cause of  high concentrations.
 However, in most of the medium and small cities, the  major  sources were point
 sources, unpaved roads, construction, and/or  wind-blown dust  from agricultural
 and open areas. None of the  nonattainment  sites were in rural  locations."

                                      85

-------
     Finally, a study by Technology Service Corporation9  analyzed TSP/meteor-
ology relationships in order to examine the contribution of wind-blown dust
to TSP nonattainment in EPA Region VI.   The study covered 25 nonattainment
sites in four states:  Arkansas, Oklahoma,  Texas, and New Mexico.  Seven of
the sites, located in New Mexico as well as Western Texas and Oklahoma, fall
into the study area for this report which is designated by the cross-hatched
area in Figure 36.  The discussion which follows focuses on these seven sites
for which site descriptions are provided in Table 26.

     The authors of the study used a variety of techniques to relate meteor-
ological variables and TSP concentrations.   One of the simpler analyses in
the study assesses the contribution of  wind-blown dust on days when measured
TSP concentrations exceed the 24-hour primary or secondary NAAQS.  In this
analysis, the authors classify TSP concentrations according to the wind speed
and precipitation observed on or before the day of the measurement.  If viola-
tions of the 24-hour NAAQS at a site are associated with high winds and low
precipitation, the authors conclude that wind-blown dust is a major contribu-
tor to high TSP concentrations at that  site; if violations are usually asso-
ciated with low winds and moderate to high  precipitation, the authors conclude
that wind-blown dust is only a minor contributor.  Table 27 lists the classi-
fication of each of the seven sites in  the  study area.


                        TABLE 26.  CLASSIFICATION OF
                                   SITES ACCORDING
                                   TO IMPORTANCE OF
                                   WIND-BLOWN DUST
                              Location            Class

                       Albuquerque (No.  1),  NM     III

                       Albuquerque (No.  2),  NM     III
                       Dona Ana,  NM                 II
                       Las Cruces, NM               II

                       Raton,  NM                    II
                       Roger Mills Co.,  OK          II
                       Lubbock,  TX                 III

                       Class II  - wind-blown dust is a
                       significant but not the major
                       contributor, or there is uncer-
                       tainty as to wind-blown dust
                       contributions

                       Class III - wind-blown dust is the
                       major contributor.
                                     86

-------
oo
                   AQCR Boundary

                   State Boundary

                   Particulate Data Site

                   Surface Weather Site
orpus Christi
                                                                       Brownsville
        Figure 36.  Location of 25 TSP monitoring stations shown in the regional map of AQCRs.  (Numerals
                    in parentheses indicate the number of stations in a same city.   After Tech. Ser. Corp9)

-------
                             TABLE 27.   SUMMARY OF SITE CHARACTERISTICS (.AFTER TECH. SER. CORP.9)
00
00
AQCR
152

153

154
187
211
Site
Albuquerque, NM
Albuquerque, NM
Dona Ana, NM
Las Curces, NM
Raton, NM
Roger Mills, OK
Lubbock, TX
SAROAD
site
type
CC/C
S/C
R/C
CC/C
S/C
R/A
S/C
Countrywide
emission
density I
High
High
Low
Low
Moderately Low
Low
Moderate
Climatology
'recipitation Humidity
Low
Low
Low
Low
M. Low
M. Low
M. Low
M. Low
M. Low
Low
Low
Moderate
Moderate
M. Low
Wind
speed
Moderate
Moderate
M. High
M. High
Moderate
High
High
Elevation
Heating of monitor
degree above ground
days (ft)
M. High
M. High
Moderate
Moderate
M. High
Moderate
Moderate
29
13
12
50
15
12
14

-------
     Although this analysis implicates wind-blown dust as a significant con-
tributor at all seven sites in the study area, the authors of the report had
some reservations about this conclusion.  Results of a more detailed analysis,
described below, indicated that man-made sources played a greater role in
high TSP concentrations than wind-blown dust at most of the sites listed in
Table 27.  The authors suggested two possible explanations for this discrep-
ancy.  (1)  Both conclusions may be correct.  The simpler analysis deals
only with days when TSP concentrations exceed primary or secondary standards.
The authors speculated that wind-blown dust from desert pavement, agricultural
lands, construction sites, etc., may play a greater role on such days than on
more typical days.   (2)  The results of one of the analyses may be incorrect.
The authors expressed greater confidence in the more detailed analysis
described below, noting that it relied on all the sampling days rather than
just a few and used many more variables to explain variations in TSP concen-
trations.  The detailed analysis used the AID (Automated Interactive Decision)
decision-tree program developed at the University of Michigan Institute for
Social Research.  This program accounts for the variance in the dependent
variable (TSP) by splitting the data according to ranges in the independent
(meteorological) variables, each time choosing the split that maximizes the
variance explained in the dependent variable.  The study examined the 18
meteorological variables listed in Table 28 to explain the variance in
24-hour TSP concentrations at the 25 study sites.

     Figure 37 illustrates the output from the AID decision-tree program for
a site in Raton, New Mexico.  The output indicates that seasonal fluctuations
account for most of the variance in TSP concentrations at this site; TSP
concentrations average 27.9 yg/m3 during January and February compared with
52.5 vig/m3 during the rest oi the year.  Other important explanatory variables
at this site include temperature, wind direction, and wind speed.   Higher
TSP concentrations are associated with high temperatures (greater than 40°F),
winds from the east, and low average wind speed (less than 7 knots).

     To assess the contribution of wind-blown dust, the analysts drew con-
clusions based on their interpretation of the meteorological variables that
are important at each site.  At the Raton, New Mexico, site, for example,  they
concluded that man-made sources, mostly to the east, were the primary contri-
butors to high TSP concentrations.  As Table 29 indicates,  the authors of  the
study concluded that man-made sources - including some fugitive dust sources
- were the principal contributors to high TSP concentrations at all five of
the sites in New Mexico and at the Roger Mills site in Oklahoma.   Wind-blown
dust, which includes dust from construction activities and agricultural lands
as well as dust from undisturbed desert areas, was the primary contributor
at Lubbock, Texas, and was  listed as a potential contributor at the other
six sites as well.
                                     89

-------
    TABLE 28.  SUMMARY OF METEOROLOGICAL
               VALUES
                  Variable
Month of the Year
Daytime Average Visibility
Number Observations Blowing Dust
Arithmetic Average Wind Speed
Vector Average Wind Direction
Average Relative Humidity
Wind Variability
A.M. Mixing Height
A.M. Average Wind Speed Through Mixing Layer
P.M. Mixing Height
P.M. Average Wind Speed Through Mixing Layer
Maximum Temperature
Minimum Temperature
Amount of 1-Day Precipitation
Maximum Wind Speed
3-Day Accumulated Precipitation
Number of Days Since Last Precipitation or
Snow Cover
Number of 1-Day Precipitation Observations
                     90

-------
         1 < Month of the Year  52    3  < Month of the Year < 12
                      Maximum Temperature < 1(0    Maximum Temperature > 1(0

                                                         \         '
                   ISO  ? Vector Average Wind Direction  < 360    0 < Vector Average Wind Direction < 180
                                   /                                        X
                                Y   1(8.2
                                N   38
                                    Y   59.*
                                    N = 36
                            ^     \
                      5 < Month of    35 Month of
                      the Year  S 12    the Year 5 It
                     '_                    \
                    Arithmetic Average    Arithmetic Average
                      Wind  Speed > 7        Wind Speed  < 7
              Y   <(1(.
              N = 33
Y   72.6
N =  5
                                                                    Y =  55.1
                                                                    N -  33
Y   106.3
N =   3
Average Relative    Average Relative
 Humidity > 65       Humidity  < 65

  Y" = 25.0
  N =  3
                                                P.M. Average Wind Speed    P.M. Average Wind Speed
                                                thru Mixing Layer < 3-5    thru Mixing Layer > 3.5
                                                         Y - 29.0
                                                         N =  k
             Wind Variability < 2    Wind Variability  > 2
Y
N
31.2
It
                                            Y   1(8.8
                                            N - 26
                                         Y   53.7
                                         N = 29
                                                                    0 < Vector  Average     135 S Vector Average
                                                                   Wind  Direction <  135    Wind Direction < 180
Y
N
53. *
20
Y - Average  TSP  Level
N   Sample Size
                            Figure  37.    Example  of AID output.

                                                    91

-------
TABLE 29.  SUMMARY OF CONCLUSIONS FROM THE INTERPRETATION OF THE AID DECI-
           SION TREES
          Location
                  Conclusion
Albuquerque (No. 1), NM
Albuquerque (No. 2), NM
Dona Ana, NM
Las Cruces, NM
Raton, NM
Roger Mills Co., OK
Lubbock, TX
Man-made sources  (mostly dust related).  Also,
a major contribution from wind-blown dust.

Man-made sources, mostly dust related.  Also
wind-blown dust.  Wind-blown dust sources and/
or man-made dust  sources may be concentrated
toward the southeast.

Man-made dust sources, possibly concentrated
toward the northeast.  Possibly some minor
contributions from wind-blown dust.

Man made sources.  Also, significant contribu-
tions from wind-blown dust.  Dust sources may
be most intense during winter and spring
(possibly agriculture related).

Man-made sources, mostly to the east.  Also a
possibility of wind-blown dust.  Dust sources
may be most intense in spring and least intense
in winter (possibly agriculture related).

Man-made sources, mostly dust related.  Also a
possibility of major contributions from wind-
blown dust.

Substantial contributions from wind-blown dust.
Significant dust sources (wind-blown or man-
made) toward the west.
                                   92

-------
                                 REFERENCES


 1.  Richard G., and R. Tan.  An Implementation Plan for Suspended Particulate
     Matter in the Phoenix Area.  Vol.  1, Air Quality Analysis.   EPA-450/3-77-
     021a.  TRW Environmental Engineering Division.   Redondo Beach, California.
     November 1977.

 2.  Moyers, J.L.  Identification and Analysis of Urban Air Quality Patterns.
     Final Report.  EPR1 EA 487, Project 438-1.  University Analytical Center.
     University of Arizona.  Tucson, Arizona.  December 1977.

 3.  Richard, G., J. Avery, and L. Baboolal.   An Implementation Plan for
     Suspended Particulate Matter in the Phoenix Area.   Vol. 3,  Model Simula-
     tion of Total Suspended Particulate Levels.  EPA-450/3-77-021c.  TRW
     Environmental Engineering Division.   Redondo Beach, California.
     August 1977.

 4.  Jutze, G.,  and K.  Axetell.  Investigation of Fugitive Dust.   Vol. 1,
     Sources, Emissions,  and Control.   EPA-450/3-74-036a.   Vol.  2, Control
     Strategy and Regulatory Approach.   EPA-450/3-74-036b.   PEDCo Environmental
     Specialists, Inc.   Cincinnati,  Ohio.  June 1974.

 5.  Jutze, G.A., J.M.  Zoller,  K.  Axetell,  and R. Livingston.   Neveda Particu-
     late Control Study for Air Qaulity Maintenance  Areas.   Reno and Las Vegas
     Urban Areas.  EPA-909/9-77-001.  PEDCo Enviornmental,  Inc.   Cincinnati,
     Ohio.  March 1977.

 6.  Engineering-Science.   Arcadia,  California.  Fugitive Dust Emission Study
     for the City of Albuquerque.   EPA-906/9-79-002.   February 1979.

 7.  Suck, S.H.,  B.C. Upchurch, and  J.R.  Brock.  Dust  Transport  in Maricopa
     County.   Arizona.   Atmos.  Environ.   Vol  12, No.  12, 2265-71.  1978.

 8.  Graf, J.,  R.H.  Snow,  and R.G.  Draftz.  Aerosol  Sampling and Analysis.
     Phoenix, Arizona.   EPA-600/3-77-015.  ITT Research Institute.  Chicago,
     Illinois.   February 1977.

 9.  Trijonis,  J.,  Y. Horie, and D.  Bicker.  Statistical Analysis of TSP and
     Meteorological Data in EPA Region  VI.  Preliminary Draft-Final Report.
     EPA Contract No. 68-02-2828.   Technology Service  Corp.   Santa Monica,
     California.   May 1978.

10.  Smith, D.E., O.K.  Spencer, J.  Richards,  and G.  Peterson.   The Environ-
     mental Protection  Agency Four Corners  Ambient Air Monitoring Network.
     EPA-600/7-79-135.   ute Research Laboratories.  Ft.  Duchesne, Utah.
     June 1979.
                                     93

-------
11.  PEDCo Environmental,  Inc.   Kansas City,  Missouri.   Technical Assistance
     in Developing Nonattainment Plans for Selected Areas in California.
     Vols. 1 and 2.  Draft Final Report.   EPA Contract  No.  68-02-2535, Task
     Order No. 9.

12.  PEDCo Environmental Specialists,  Inc.  Cincinnati, Ohio.  Characterization
     of Particulate Sources Influencing Monitoirng Sites in Region VIII Non-
     attainment Areas.   EPA Contract No.  68-02-1375, Task Order No. 30.

13.  Lundgren, D.A., and H.J.  Paulus.   The Mass Distribution of Large Atmos-
     pheric Particles.   JAPCA Vol.  25, No. 12.   December 1975.

14.  Farmer, W.M., and  J.O. Hornkohl.   Environmental Aerosol Measurements
     Using an Airborne  Particle Morphokinetometer.  EPA-600/3-76-087.  Spec-
     tron Development Labs.  August 1976.

15.  Sverdrup, G.M., K.T.  Whitby, and  W.E. Clark.  Characterization of Cali-
     fornia Aerosols -  II.  Aerosol Size  Distribution Measurements in the
     Mojave Desert.  Atmos. Environ. Vol.  9,  pp.  483-494.

16.  Whitby, K.T., W.E. Clark,  V.A. Marple, G.M.  Sverdrup,  G.J. Sem, K. Willeke,
     B.Y.H. Liu, and D.Y.H. Pui.  Characterization of California Aerosols - I.
     Size Distributions of Freeway Aerosol.  Atmos.  Environ.  Vol.  9, pp. 463-
     482.

17.  Esmen, N.A., and M. Corn.   Residence Time  of Particles in  Urban Air.
     Atmso. Environ. Vol.  5, pp. 571-578.

18.  Willeke K., and K.T.  Whitby.  Atmospheric  Aerosols:   Size  Distribution
     Interpretation. JAPCA.  Vol.  25, No. 5.   May 1975.

19.  Porch, W.M.  Recent Studies of the Suspension of Desert  Dust  and Resuspen-
     sion of Toxic Aerosol Due  to Wind.   University of  California.   Livermore.
     ASME Paper.  July  1974.

20.  Cowherd, Jr. C., K. Axetell, Jr., C.M. Guenther, and G.A.  Jutze.  Develop-
     ment of Emission Factors  for Fugitive Dust Sources.   EPA-450/3-74-037.
     Midwest Research Insitute.   Kansas City, Missouri.  June 1974.

21.  Sehmel, G.A.  Influence of Soil Erosion on the Airborne  Particle Size
     Distribution Function.  Battelle, Pacific  Northwest  Laboratories.
     Richland, Washington.  APCA Paper No. 73-162.   June  1973.

22.  Chepil, W.W.  Properties  of Soil  which Influence Wind  Erosion, 4, State
     of Dry Aggregate Structure.  Soil Sci. 72, pp.  387-401.

23.  Greeley, R., J.D.  Iverson,  J.B, Pollack, N,  Udovich, and B.  White.  Wind
     Tunnel Studies of  Martian  Aeolian Processes:  NASA.   Tech. Mem. 62.  1973.

24.  Marshall, J.  Drag Measurements in Roughness Arrays  of Varying Density
     and Distribution.   Agr. Meteor. Vol.  8,  pp.  269-292.
                                     94

-------
25.   Lyles,  L.  and B.  Allison.   Wind Erosion:   The Protective  Role  of  Simulated
     Standing Stubble.  Trans,  or the ASAE,  Vol.  19,  pp.  61-64.   1976.

26.   Kessler, E.,  D.Y. Alexander, and J.F.  Rarick. Duststorms from the U.S.
     High Plains in Later Winter 1977 - Search for Cause  and Implications.
     Proc. Okla. Acad. Sci.,  Vol. 58, pp.  116-128.  1978.

27.   Hagen,  L.J. and N.P. Woodruff.   Air Pollution from Duststorms  in  the
     Great Plains.  Atmos. Environ.  Vol. 7,  pp.  323-332.   1973.

28.   Chepil, W.S.  and  N.P- Woodruff.   Sedimentary Characteristics of Dust-
     storms - II.   Visibility and Dust Concentration.  Am.  J.  Sci.  Vol.  255,
     pp. 104-114.   1957.

29.   Hagen,  L.J. and N.P- Woodruff.   Particulate Loads Caused  by Wind  Erosion
     in the Great Plains.  U.S.  Department  of  Agriculture.  Manhattan, Kansas.
     APCA Paper No. 73-102.  June 1973.

30.   Whitby, K.T., A.B. Algren,  R.C.  Jordan, and J.S. Annis.   The ASHAE Air-
     borne Dust Survey.  Heating, Piping, Air-conditioning, pp.  185-192.
     November 1957.

31.   Korte,  N.E.,  and  J.L. Moyers.  The Concentration of  Inorganic  Species  in
     Airborne Respirable Particulate Matter  in Rural  Southern  Arizona.  J.  of
     Ariz-Nev.  Acad. of Sci., Vol. 13, pp.  79-83.  October  1978.

32.   Moyers, J.L., L.E. Ranweiller,  S.B. Hopf, and N.E. Korte.  Evaluation
     of Particulate Trace Species in Southwest Desert Atmosphere.  Environ.
     Sci. and Tech., Vol. II, pp. 789-795.   August 1977.

33.   Pearson, M.J., M. Pitchford, and R. Snelling. Western Energy  Sulfate/
     Nitrate Monitoring Network Progress Report.   EPA-600/7-79-074.  Environ-
     mental Monitoring and Support Laboratory.  Las Vegas,  Neveda.   March 1979.

34.   Greeley, R.S., R.P.  Ouellette,  J.T. Stone,  S. Wilcox.  Sulfates and the
     Environment — A Review.  Mitre Corporation.   MTR-6895. March  1975.

35.   Altshuller, A.P.   Atmospheric Sulfur Dioxide and Sulfate.  Distribution
     of Concentration  at Urban  and Nonurban Sites in  United States.  Environ.
     Sci. and Tech., Vol. 7,  pp. 709-712.   August 1973.

36.   Current Population Reports.  Population Estimates and Projections.
     Individual State Reports.   U.S. Department of Commerce.   Bureau of the
     Census.

37.   Richard, G., R. Tan, and J. Avery.  An Implementation Plan for Suspended
     Particulate Matter  in the Phoenix Area.  Vol. 2, Emission Inventory.
     EPA-450/3-77-021b.  TRW Environmental Engineering Division.  Redondo Beach,
     California.  December 1977.

38.  Ungers, L.H.   Comprehensive  Emission Inventory for Mohave and Yuma Counties,
     Arizona.   Vols.  I and II.  EPA-909/9-78-003.  PEDCo Environmental.
     Cincinnati,  Ohio.   July 1978.
                                       95

-------
          APPENDIX A




BACKGROUND STATION INFORMATION
              96

-------
                        TABLE A-l.   1976, 1977, AND 1978 TSP CONCENTRATIONS AT 33
                                    (FROM NATIONAL AEROMETRIC DATA BANK)
SITES
VO
Elevation
(ft)
State
Arizona



Nevada
California
Oklahoma

Texas

New Mexico

Colorado





Utah





Denotes a
County
Yavapai
Pirns
Coconino
Plraa
Clark
Kern
Tulsa
Oklahoma
Jeff. Davis
Tom Green
Los Alamos
Bernallllo
La Plata
Montezuma
Rio Blanco
Rio Blanco
Rio Blanco
Rio Blanco
Uintah
Uintah
Uintah
Uintah
Kane
Kane
value derived
Site location
Camp Verde
Org. Pipe Cact. Nat. Hon.
Grand Canyon Village
Corona de Tucson
Las Vegas Wash-Marina
Kern Wildlife Refuge
Bixby
Draper Lake
Mt. Locke
Fish Hatchery No. 1
White Rock
West Mesa Radar Station
Red Mesa
Mesa Verde
Tract C-A, Site 1
Tract C-A, Site 2
Tract C-A, Site 3
Tract C-A, Site 4
Tracts UA+UB, Site A7
Tracts UA+UB, Site A6
Tracts UA+UB, Site A4
Tracts UA+UB, Site A3
Glen Canyon
Glen Canyon
Site code
030940002F03
030620005F03
030370001703
030620008G03
29008001 1G03
053480001103
373020133F03
372200020F03
452740003F03
455200001F03
3207 20001 F03
320140031H03
061 300001 F03
061530003F03
061860001J03
061860002J03
061860003J03
061860004J03
461 200007 J03
461200006J03
461 200004 J03
461200003J03
460400002F03
4604 00001 F03
from data which do not meet SAROAD sum
Note: Concentrations for 1976, 1977, and 1978 are
tabulated In
Abov«
MSL
3381
1661
6955
3225
1300
0215
0918
1290
6791
1864
6265
6003
6520
7060
7360
7360
6600
6300
5360
5250
5717
5320
3890
3750
Abovs
ground
4
4
4
13
25
4
18
15
4
15
15
29
10
15
197
33
33
33
20
20
20
20
5
5
Ststion type
Remote
Remote
Remote
Remote
Rural -Unqualified
Remote
Rural-Near Urban
Rural -Agricultural
Rural -Agricultural
Rural-Near Urban
Rural -Unqualified
Rural-Unqualified
Rural -Agricultural
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
arlzatlon criteria of OAQPS Guideline
Number of OBS
52/30/33
48/42/31
36/37/9
52/45/29
52/52/20
48/36/34
19/62/61
59/56/61
42/51/40
21/53/32
59/61/25
60/57/58
61/53/—
77/57/77
103/6/—
121/10/—
124/10/—
121/10/—
60/3/~
61/2/~
61/3/—
61/3/—
202/340/156
272/326/—
1.2-040, Vol.
Highest
64/68/49
139/83/98
69/82/27
51/84/40
115/79/92
390/196/450
87/110/98
147/3094/116
221/99/66
391/191/154
147/134/172
238/314/235
189/280/ 	
132/125/38
81/6/-
57/10/-
281/14/—
182/39/ —
7S/22/—
101/31/—
64/15/—
39/20/—
118/661/156
120/423/—
Second
highest
59/65/49
83/78/49
58/42/27
42/45/33
91/78/71
222/157/203
76/95/91
138/162/105
131/71/44
94/127/98
123/104/38
236/116/132
172/188/ 	
102/92/38
80/5/—
57/6/—
210/8/—
173/9/—
42/20/—
72/23/-
60/12/--
36/16/--
114/322/143
US/258/—
Geometric
mean
30/29a/24a
33/34a/3l"
14/20a/14a
5/25a/17a
37/38/Z43
84/76a/62
33a/38/40
50/51/47
20/18/18
56a/50/48a
30/32/223
37/27/19
43/34a/—
13/15/12
11/43/-
ll/4B/~
17/3a/—
14/5a/—
14/17a/—
20/273/—
15/3l"/—
12/15a/~
23a/24/ll
IS/17/—
3, Sec. 2.3.0.
chronological order, separated by slashes.

-------
vo
00
                       Monitor Locations

   1.   Kern Wildlife Refuge (near Corcoran Rd)
   2.   Park Service Bldg Las Vegas Wash-Marina
   3.   Glen Canyon Nat. Rec. Area (Maintenance  Garage)
   4.   Glen Canoyn Nat. Rec. Area (Fish 6 Game  Office)
   5.   Grand Canyon Village (Visitor Center)
   6.   Hontezuma Castle Nat. Park (Camp Verde)
   7-   Organ Pipe Cactus Nat.  Monument (Pump  Station)
   8.   Corona de Tucson
   9.   White River Shale Project - Tracts UA+UB (Sites 3,  *. 6, 7)
  10.   RBOSC - Tract C-A (Sites 1, 2, 3, It)
  11.   Mesa Verde Nat. Park (Fire Lookout Station)
  12.   Red Mesa CGI lien Dairy)
  13.   White Rock (Treatment Plant)
  14.   West Mesa (Radar Station)
  15.   Mt. Locke (McDonald Observatory)
  16.   Tom Green County (Fish  Hatchery, No. 1)
  17.   Draper Lake (Ranger Station)
  18.   Bixby (Water Plant)
                              Figure A-l.   Monitor locations  for  National Aerometric  Data Bank data

-------
             TABLE A-2.  1976 TSP CONCENTRATIONS AT  26 SITES  OPERATED  BY  UTE  RESEARCH LABORATORIES
1C
VO
Elevation
(ft)
State County
Arizona Navajo
Conconino
Coconino
Coconino
Coconino
Mohave
Apache
Apache
Coconino
Coconino
Colorado La Plata
La Plata
Montezuma
New Mexico San Juan
San Juan
Rio Arriba
San Juan
Utah San Juan
Washington
Garfield
Kane
Garfield
San Juan
San Juan
Washington
Site Location
Bacobi
Bodaway
Coppermine
Kaibito
Lee's Ferry
Piute
Redrock
Teec Nos Pos
Tsa Schizzi
Tuba City
Ignacio
Redmesa
Towaoc
Burnham
Chaco Canyon
Dulce
Huerfano
Aneth
Boom ing ton
Escalante
Glen Canyon
Kenrieville
Navajo Mtn.
Oljato
St. George
Site code
030520003K03
03020001 2K03
03 020001 1K03
03020001 OK03
030200007K03
030500009K03
030040000K03
03004 0001K03
63020001 3K03
03020001 4K03
061300003K03
061300002K03
061600004K03
321 00001 2K03
32100001 1K03
320920003K03
321000007K03
46096003K03
46128001K03
460300002K03
460400003K03
460300003K03
460960001K03
460960002K03
461280002K03
Above Above Station
MSL ground type
6300
5960
6050
6000
3190
4880
6110
4990
5550
5170
6550 ,
6460
5740
5610
6190
7220
6140
4590
2520
5755
4040
5905
6050
4810
2880
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Remote
Number
of
OBS
137
76
93
152
61
151
155
141
67
99
131
121
113
83
75
153
118
157
152
141
59
143
13
150
62
Highest
208
370
663
253
481
193
128
473
102
305
188
80
110
569
113
207
174
144
123
592
36
308
99
207
57
Second
highest
155
353
90
216
256
134
125
171
57
124
171
70
81
185
111
171
171
138
86
360
34
196
56
193
46
Geometric
mean
17
22
18
29
21
18
24
29
16
32
23
24
23
24
31
12
45
30
26
56
15
42
22
26
16

-------
        UTAH
                                         • Gre*n Rlwr
                        Capitol Reef \
                                       Canyonlands,'
          .C^cit,  Bryce Canyon
                       \'      :"
                       \>" "^ f
                                               . BU^n.
                                          20
                                           ..
                                           *28
                                                       COLORADO
                                                 • Grind Junction
                                                          Cortei           Pagoia Spr«.
                                                          '       " Du"no°
                  IT-
                                                  9*
                                *4
                                                          '"'W •
                                                           19
                        \
ARIZONA

                                           Canyon de C'helly
                                                              Chaco Canyon
                   FUoMilT
                                                              NEW MEXICO
 •^•indicates Ute Research Laboratories' monitoring  stations:
 1  Babcobi, AZ
 2  Bodaway, AZ
 3  Coppermine, AZ
 4  Kaibito, AZ
 5  Lechee, AZ
 6  Lee's Ferry, AZ
 7  Piute, AZ
 8  Red rock, AZ
 3  Teec Nos Pos, AZ
10  Tsa Schizzi,
11 Tuba City, AZ
12 Ignacio, CO
13 Redmesa, CO
14 Towaoc, CO
15 Burnham, NM
16 Chaco Canyon, NM
17 Dulce, NM
18 Huerfano, NM
19 Navajo Farm Proj., NM
20 Aneth, UT
12 Bloomington, UT
22 Escalante, UT
23 Glen Canyon, UT
2k Henrieville, UT
25 Huntington Canyon,  No.  1.
26 Huntington Canyon,  NO.  2.
27 Navajo Mountain,  UT
28 Oljato, UT
29 St. George, UT
                                                                      UT
                                                                      UT
       Figure A-2.  Monitoring site locations  (from EPA-600/7-79-135,
                    p. 5, Reference 10).
                                      100

-------
  TABLE A-3.  ACTIVATION DATES FOR SITES OPERATED BY UTE RESEARCH
              LABORATORIES
Monitoring
site
Bacobi, AZ
Bodaway , AZ
Coppermine , AZ
Kaibito, AZ
Lechee, AZ
Lee's Ferry, AZ
Piute, AZ
Redrock, AZ
Teec Nos Pos, AZ
Tsa Schizzi, AZ
Tuba City, AZ
Ignacio, CO
Redmesa , CO
Towaoc, CO
Burnham, MM
Activation
date
02/27/76
06/21/76
02/02/76
01/22/76
05/22/77
10/05/76
01/27/76
01/12/76
01/20/76
04/02/76
01/16/76
01/19/76
04/19/76
04/29/76
05/17/76
Monitoring
site
Chaco Canyon, NM
Dulce, NM
Huerfano, NM
Navajo Farm Project, NM
Aneth, UT
Blooming ton, UT
Escalante, UT
GLen Canyon, UT
Henrieville, UT
Huntlngton Canyon, No. 1,
Hunt ing ton Canyon, NO. 2,
Navajo Mountain, UT
01 j a to, UT
St. George, UT

Activation
date
06/30/76
05/17/76
04/13/76
02/06/77
02/11/76
02/02/77
01/24/76
10/13/76
01/24/76
UT 01/18/77
UT 05/06/77
01/30/77
01/23/76
07/14/76

Note:  No uniform sampling schedule was in use in 1976, but sampling
       was required for three 24-hour periods each week.
                                 101

-------
                                    TECHNICAL REPORT DATA
                             (Please read Instructions on the reverse before completing)
 \. REPORT NO.
  EPA 450/2-80-078
              3. RECIPIENT'S ACCESSION NO.
 4. TITLE AND SUBTITLE
  Evaluation  of Contribution  of Wind Blown Dust  From
  the  Desert  to Levels of Particulate Matter in
  Desert  Communities
 7. AUTHOR(S)               ~          ~~
              5. REPORT DATE
                August 1980
              6. PERFORMING ORGANIZATION CODE
                                                            8. PERFORMING ORGANIZATION REPORT NO.
  Frank A.  Record and Lisa A.  Baci
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
  GCA Corporation
  Bedford, Massachusetts
                                                            10. PROGRAM ELEMENT NO.
              11. CONTRACT/GRANT NO.
                                                              68-02-2607
                                                              (Work assignment no. 41)
 12. SPONSORING AGENCY NAME AND ADDRESS

  Environmental  Protection Agency
  Research  Triangle Park, N.C.  27711
              13. TYPE OF REPORT AND PERIOD COVERED

                Final	
              14. SPONSORING AGENCY CODE
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT
           This report uses existing data and studies  to assess the  impact of
     windblown desert dust on the attainment of TSP standards in major cities
     situated in desert environments in the Southwestern U.S.  Primary emphasis
     is  placed on four cities:   Phoenix and Tucson, Arizona; and Las  Vegas and
     Reno, Nevada.  It is concluded that:  (1) the contribution of  windblown
     dust from the undisturbed  desert floor to particulate levels in  desert
     communities is very small  and should be considered as part of  the back-
     ground;  (2) if human activities repeatedly break up the desert crust,
     local violations of the 24-hour standards are likely; (3) there  is
     substantial agreement on the principal source categories of fugitive
     dust contributing to the nonattainment problem and the characteristics
     of  urban areas most affected by each category.
 7.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
^.IDENTIFIERS/OPEN ENDED TERMS  C. COSATI Field/Group
      Fugitive Dust
      TSP
      Desert
 8. DISTRIBUTION STATEMENT

      Limited Availability
           :LASS (This Report)
21. NO. OF PAGES
   108
                                                         :LASS (This page)
                                                                          22. PRICE
EPA Form 2220-1 (Rev. 4-77)    PREVIOUS EDITION is OBSOLETE

-------