OCR error (C:\Conversion\JobRoot\00000AXE\tiff\2000XG9N.tif): Unspecified error

-------
                 RESEARCH REPORTING SERIES


Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination  of traditional  grouping  was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

    1. Environmental Health Effects Research

    2. Environmental Protection Technology

    3. Ecological Research

    4. Environmental Monitoring

    5. Socioeconomic Environmental Studies

    6. Scientific and Technical Assessment Reports (STAR)

    7. Interagency Energy-Environment Research and Development

    8. "Special" Reports

    9. Miscellaneous Reports

This report has been assigned to the  INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND  DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment  Research and
Development Program. These studies  relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to  assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations  include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects;  assessments of, and development of, control technologies  for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                             EPA-600/7-79-243
                                             November 1979
  CHARACTERIZATION OF VISIBILITY-REDUCING
          AEROSOLS IN THE SOUTHWEST

    Project VISTTA Progress Report No. 1


                     by
              Edward S. Macias
     California Institute of Technology
       Consultant to MRI on leave from
      Washington University - St. Louis

            Donald L. Blumenthal
             Jerry A. Anderson
         Meteorology Research, Inc.

              Bruce K. Cantrell
              SRI International
           Contract No. 68-02-2713
              Project Officers

 William E. Wilson, Jr., and Thomas Ellestad
        Regional Field Studies Office
 Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
  ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S.  ENVIRONMENTAL PROTECTION AGENCY
 RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711

-------
                            DISCLAIMER
Laboratory,  U. S. Environmental Protection Agency,  and approved for
publication.  Approval does not signify that the contents necessarily reflect
the views and policies of the U. S. Environmental Protection Agency, nor
does mention of trade names or commercial products constitute endorse-
ment or  recommendation for use.
                                  11

-------
                              ABSTRACT

   The atmospheric visibility-reducing aerosol in the Southwest has been
experimentally characterized  in the Fall, 1977, with respect to particle
size, composition, and contribution to light scattering.  Measurements
were taken within the mixing layer  using the MRI instrumented Beech-
craft Queen Air aircraft.  The aircraft was equipped to measure andre-
cord on magnetic tape the light-scattering coefficient,  Aitken nuclei count,
size distribution,  ozone, sulfur dioxide, nitrogen oxides,  temperature,
dew point, turbulence,  pressure (altitude),  and navigational parameters.
Multistage impactor and size-fractionated filter samples were  also col-
lected in order to determine aerosol elemental composition as  a function
of size.  Visual range estimates were obtained by viewing distant land-
marks and verified by optical  photography.
                                 111

-------
                             CONTENTS
Abstract	  iii
Figures	  vi
Tables	viii
Acknowledgments	  ix

   1.  Introduction	    1
        Background	    1
        Objectives	    3
        Responsibilities	    4
   2.  Summary and Conclusions	    6
   3.  Future Plans and Recommendations	    9
        Recommendations for Future Experiments	    9
        Fall 1978 Experimental Design	  12
   4.  Program Description	  14
   5.  Summary of Results from Fall 1977 Experiments	  19
        Characterization of the Visibility-Reducing
        Aerosol in the Southwest	  19
        Source Characterization	  37

References	  54

-------
                                 FIGURES

Number                                                               Page


  1    Map of Southwestern United States	    2
  2    Flight path of VISTTA regional flights on October 5 and 9,  1977 ...   20
  3    Frequency distribution of the light-scatter ing coefficient
      due to particles,  bsp, measured in  regional flights on October
      5 and 9,  197',	   22
  4    Photograph of Navajo mountain at a distance of HO km from
      the aircraft of October 5, 1977	   23
  5    Average  size distribution of background Southwest aerosol
      measured on the  October 9, 1977,  regional flight	   24
  6    Average  size distribution of individual elemental species
      determined from impactor samples collected on October 5
      and 9 regional flights	 .   27
  7    Size-fractioned mass balance of Southwest background aero-
      sol measured on  regional flights on October 5 and 9,  1977	   30
  8    Light scattering per unit volume of aerosol material as a
      function of particle size, integrated over all wavelengths
      for a. refractive index,  m = 1. 5	   34
  9-   Light scattering contribution as a function of size for the
      Southwest region, October 9,  1977	   36
 10    Flight map of Arizona smelter flights:
            a)  October 1, 1977	   38
            b)  October 2, 1977	   38
            c)  October 4, 1977	   38
 11    Flight map of Mohave power plant  flights on October  8, 1977	   30,
 12    Photograph of the San Manuel smelter plume 8 km downwind
      of the plant looking normal to the plume	   40
                                    VI

-------
FIGURES (Continued)

Numbe r                                                             Page
 13    Visual range calculations for the smelter and power plant
       at several distances downwind compared to the background
       visual range	 44

 14    Determination of SO2 conversion rate for the San Manuel
       smelter on October 4,  1977	 49

 15    Qualitative indicators of SO2 conversion in plumes plotted vs.
       distance from plant	 50

 16    Visual range through the San Manuel smelter plume with and
       •without SO2 conversion and through background only for two
       downwind distances	 51
                                     VII

-------
                                 TABLES
Number
   1  Beechcraft Queen Air Instrumentation	   13
   2  Ground-Based Measurements,  1978 Experiments	   14
   3  VISTTA Field Program Summary - 1977	  17
   4  Average Parameters Measured during Regional Flights	  19
   5  Average Elemental Concentration of Aerosol in the
      Southwest Region, October 5 and 9,  1977	  26
   6  Chemical Species Balance for  the Southwest Region
      October 5 and 9, 1977	  29
   7  Comparison of Aerosol Mass Determinations  from
      Volume and Composition Measurements	  31
   8  Southwest Regional Aerosol Enrichment Factors	  32
   9   Light Scattering Budget for the Southwest Region
      October 9,  1977	  35
  10  Average Parameters Measured during Orbit Plume Flights	  41
  11  Plume Impact Data	  43
  12  Plume Excess  Aerosol Elemental Concentration	  45
  13  Plume Excess  Aerosol Enrichment Factors	  46
  14  Plume Accumulation Mode Integral Size Parameters	  47
  15  Plume Excess  Visibility Budget	  52
                                    Vlll

-------
                    ACKNOWLEDGMENTS
       This work has been supported by the Federal Interagency
Energy/Environment Research and Development program through
a contract with the  EPA - ESRL Office of Regional Studies.  We
appreciate the guidance provided by William Wilson, director  of
the Office of Regional Studies,  and the field support provided by
Tom Ellestad of EPA. The analysis herein was  greatly enhanced
by consultation with and guidance from S. K. Friedlander and R.
Flagan of Caltech.  We also appreciate the field  support provided
by J. A. McDonald, the data processing support of J. Roebuck,
and the analysis support of J.  Ogren and G. Markowski.  Part  of
the data analysis has been supported by the USEPA under Grant
No. R 802160 to Caltech.

-------
                              SECTION  1

                            INTRODUCTION
BACKGROUND

      The southwestern United States desert and mountain areas, shown in
Figure  1, are generally characterized by very good visibility (—100 km).
Until recent years,  scenic vistas of natural landmarks and mountains with
a visual range of over 100 miles were common.  These vistas have been
considered a major  resource of the Southwest, and numerous national parks,
forests, and monuments have been created to preserve them.  The tourism
resulting from the southwest scenery has played a major role in the economy
of the region.

      As the population of the region has grown, there has been an apparent
reduction in the average visibility (Trijonis and Yuan,  1978).  Residents of
the Southwest are concerned about this loss and point to various  anthropo-
genic activities  as the cause.  This  concern has led to the August, 1977,
amendments to the  Clean Air Act, which contain a section stating, "as a
national goal the prevention of any future, and the remedying of any existing,
impairment of visibility in mandatory Class I Federal areas which impair-
ment results from man-made air pollution."  Previously, however, com-
prehensive studies had not been performed to quantify and determine the
cause of the current problem or  to provide the basis for a control strategy
to keep the problem from getting  worse.

      Visibility  impairment in the atmosphere is due primarily to the pre-
sence of small particles which scatter and absorb light and, secondarily,
to NO2 which absorbs blue light (Charlson, Waggoner, and Thielke, 1978).
In the southwest, a regional decrease in visibility is likely due to the long-
range transport of aerosol or gaseous  precursors into unpopulated pris-
tine areas.  A second possible cause of regional visibility impairment may
be wind-blown dust.   Both types of particle may be of anthropogenic origin.
Some of the varied activities which may contribute to the decline in visi-
bility are:  increased population  and the resulting increase in automotive
emissions; emissions from power plants; emissions from smelters; dis-
turbance of the topsoil and desert pavement due to agriculture, grazing,
and off-road vehicles; and mining activities.

-------
                                                    nit
                                               • COfffH (BELTERS
                                                      PCAIITS
Figure 1.  Map of Southwestern United States,

-------
      In addition to their scenic value, the mountain and desert regions
of the Southwest represent a major energy resource.  Coal, oil, natural
gas, oil shale,  uranium, and abundant sunlight are all found in the area.
The emerging national priority of developing domestic energy supplies
will accelerate  the development of the Southwest's energy resources. If
not planned properly,  this development could lead to greater visibility
impairment and loss of scenic resources.

      Project VISTTA, * sponsored by the EPA Environmental Science Re-
search  Laboratory in  Research Triangle Park, North  Carolina, has  been
initiated to determine  the  sources of visibility reduction in the  Southwest
and to help resolve the potential conflict between the congressional man-
date for improved visibility and the need for increased energy production.

      The experimental approach of Project VISTTA is to characterize the
visibility-reducing species in the atmosphere, namely particulate matter
and NO2.  To determine the relative contribution of the various anthropo-
genic activities  to visibility degradation, and thus to develop control strat-
egies, the VISTTA project has concentrated on the study  of the size distri-
bution and chemistry of the visibility-reducing aerosol in the  region.  This
defines  the problem.  At the same time, data have been collected to  char-
acterize the emissions from some of the activities mentioned above, i.e.,
their size distribution and chemistry,  and the transport and transforma-
tions these emissions  undergo in  the atmosphere.   Knowing both the  char-
acteristics of the atmospheric aerosol and of the emissions, the relative
contribution of each type of emission to the overall problem can be deter-
mined (Friedlander, 1977).

      It has been estimated (Trijonis and Yuan,  1978)  that in the southwest
region between  30 and 60 percent of the light extinction due to  particles
is caused by sulfates.   One of the sources of sulfur compounds in the at-
mosphere is the burning of sulfur-containing fuel for power generation.
As part of Project VISTTA, the EPA would like to quantify the contribu-
tion of the power industry to visibility impairment in the  South-west.
OBJECTIVES

      The objectives of Project VISTTA are as follows:
      VISTTA is an acronym from Visibility Impairment due to Sulfur
      Transport and Transformation in the Atmosphere

-------
1.    To characterize the visibility-reducing aerosol in the Southwest
      with respect to particle size, chemical composition as  a function
      of size, and spatial distribution.  To meet this goal, measure-
      ments will be taken on a regional scale (flight paths up  to 1000
      km) and at points near urban centers, power plants  and smelters
      in the region.

2.    To characterize the emissions which are important  in visibility
      reduction from typical power plants and smelters  in the region.
      Both primary particulate emissions and gaseous aerosol precur-
      sors will be taken into account.

3.    To determine the detailed contributions to light scattering for the
      Southwest region broken down by chemical composition and size
      (light scattering budget).

4.    To quantitatively assess the contributions of various sources,
      both anthropogenic and natural, to the visibility-reducing aerosol.
      Of particular interest is the contribution of  emissions from power
      plants.

      The first experimental portion of Project VISTTA was aimed at the
first three objectives. This first  study was a preliminary experiment, and
the data are rather limited. These results are summarized in Section 5 of
this report.  In the later stages of the  project,  the fourth  objective  will be
more fully addressed.  An outline of the future direction of the project is
given in Section 3 of this report.

RESPONSIBILITIES

      Project VISTTA is a team project involving several groups.  The
groups  involved in the initial sampling period (October 1-10, 1977) and
subsequent analysis and their responsibilities are as follows :

      Meteorology Research, Inc. (MRI) -- Project planning, design,
        and coordination, aircraft sampling, and  data analysis.

      EPA - Environmental Science Research Lab (ESRL) at Research
        Triangle Park -- Calibration of aircraft instruments.

      University of California - Davis (UCD)  --  Elemental Analysis of
        samples from a specially designed UCD airborne  impactor.

      University of Washington (UW) -- Ground-based measurements of
        size distribution,  lignt scattering, and light absorption  coefficients,

      California Institute of Technology (CIT) -- Data analysis.

-------
      The individuals involved in the
affiliations, and responsibilities are

      Project Coordinator

      Senior Data Analyst
     Field Manager

     Data Analysis

     Data Analysis

     Data Analysis

     Impactor Design and
       Sample Analysis

     Sulfate  Analysis

     EPA Project Officer
MRI portion of the program, their
given below:

 D. L. Blumenthal (MRI)

 E. S. Macias (CIT, Consultant to
   MRI on leave from Washington
   University,  St. Louis)

 J. A. Anderson (MRI)

 B. K. Cantrell (SRI International)

 S. K. Friedlander (UCLA)

 J.  A. Ogren (MRI and UW)


 D. L. Shadoan, T. Cahill (UCD)

 J.  D. Husar and Associates

 W. E. Wilson (EPA)
 T. Ellestad (EPA)

-------
                             SECTION 2
                   SUMMARY AND CONCLUSIONS

   The atmospheric visibility-reducing aerosol in the Southwest has been
experimentally characterized with respect to particle size, composition,
and contribution to light scattering.  Measurements were taken within
the mixing layer using the MRI instrumented Beechcraft Queen Air air-
craft.  The aircraft was equipped to measure  and record on magnetic tape
the light-scattering coefficient, Aitken nuclei  count,  size distribution,
ozone, sulfur dioxide, nitrogen oxides, temperature, dew point, turbu-
lence, pressure (altitude),  and navigational parameters.  Multistage
impactor and size-fractionated filter samples were also collected  in order
to determine aerosol elemental composition as a function of size.  Visual
range estimates were obtained by viewing distant landmarks and verified
by optical photography.

   The multistage impactor was mounted in the nose of the aircraft.  The
impactor stages had  50 percent collection efficiency for aerosol aero-
dynamic diameters of 4,  2, 1 and 0. 5  microns.  Impactor samples were
collected on greased  Mylar film. A Nuclepore final filter was used to
collect all particles not collected on the four impaction stages.  The im-
pactor samples were analyzed by particle-induced X-ray emission.  The
fine .particle samples from a TWOMASS two-stage filter sampler were
analyzed by flash vaporization - flame photometric detection.

   The 1977 field program  consisted of seven sampling days.  The ex-
perimental data included  flights of the  region on three full days and parts
of one other,  flights  in the  plume of a  copper smelter (San Manuel smelter)
on three days, and flights in the plume of a coal-fired power plant (Mohave
power plant) on one day.

   The regional flights were designed  to characterize the visibility-re-
ducing aerosol within 500 m of the surface on flight paths over 250 km.
On two days,  the total flight paths were over 1000 km.  The visibility
along these paths was documented with a high-sensitivity nephelometer,
a camera,  and visual range measurements from aircraft to prominent
landmarks.

   A description of the 1978 field program and recommendations for future
studies  are given.

-------
   This study is the first of several similar studies.  It should be con-
sidered as a pilot study and all results should be considered preliminary
and limited.   The results of the Fall, 1977, study are summarized below:

   1.  On two regional flights over  large parts of the Southwest, the
visibility-reducing aerosol was quite homogeneous throughout the entire
region, indicating that the visibility impairment was of regional extent.

   2.  The aerosol size distribution throughout the region on October 5
and 9, 1977, was bimodal,  with a geometric volume mean size for the
fine  particles of ~ 0.25 microns,  which is  slightly smaller than the con-
tinental background aerosol average mean  size (0.3 microns).  The
measured coarse particle mode volumetric mean size was ~ 5. 5 microns,
which is equal, within error, to the continental background.

   3.  The elemental size distribution from both  regional and plume im-
pactor samples indicated that aluminum,  calcium, and iron are present
predominantly in coarse particles;  sulfur and  titanium are present pre-
dominantly in fine particles; and silicon and potassium have substantial
concentrations in both modes.

   4.  Sulfur and silicon were found in nearly equal concentrations in
the Southwest background data and were the elements present in the highest
contrations  in the fine particles (Dp^l/lm).  The detailed aerosol mass
balance determined for the Southwest region showed that sulfur (expressed
as ammonium sulfate) and silicon (expressed  as SiO2) amounted to ~ 50
percent and ~ 30  percent of the fine particle mass, respectively.

   5.  On the days  sampled, a total of ~ 90 percent of the measured coarse
particle  (D>l£lm) mass in the Southwest region was composed of elements
which were  present in the same abundance relative to aluminum as in the
earth's crust.  This indicates that the source  of these particles is either
wind-blown dust or material with an elemental composition nearly that
of crustal material, such as flyash.

   6.  The light-scatter ing budget for Southwest background aerosol on
October  9, 1977, indicated  that ~ 50 percent of the light scattering was due
to fine particles, ~ 40 percent was  due to Rayleigh scattering from gases,
and ~ 4 was due to coarse particles.  Considering  only the light scattering
due to particles,  ~ 90 percent is due to fine particles which are composed
mainly of sulfates [~ 50 percent as  (NH^SO^] and silicon compounds
~ 30 percent as SiOg).

   7.  Mie scattering calculations of the  light scattering  coefficient due
to particles using the measured average  regional size distribution were

-------
in good agreement with the values measured with a nephelometer.  This
is a quantitative indication of the internal consistency of the size distri-
bution and bscat measurements which adds credibility to the calculated
light-scattering budget.

   8.  Plume excess fine particle aerosol, i.e.,  the point source
emission plume aerosol with the background subtracted, was  composed
largely of sulfur and silicon compounds for both the smelter and power
plant plumes.  The major elemental species in the plume excess coarse
particle aerosol were  aluminum,  silicon, potassium, calcium,  and iron
in approximately crustal abundances in both plumes.  These results indi-
cate that fine particle  Si may be a good tracer for primary combustion
aerosol in smelter and power plant  plumes.

   9.  A simple semi-quantitative calculation of visual plume impact is
described which compares the visual range with and without the plume
present.  On October 4,  1977, the smelter plume caused an~90 percent
reduction of visual range relative to the background visual range (135 km)
at 8 km downwind from the plant.  As far as 127 km downwind,  increased
bscat and sulfate levels relative to background concentrations were observed
along with an ~ 40 percent reduction of visual range due to the smelter
plume.

  10.  Sulfate aerosol  was formed in Southwest power plant and  smelter
plumes.  The measured SO2 conversion rate from the San Manuel  copper
smelter on one morning between 0900 and 1230 (MST) was 0.7±0.2 per-
cent/hour between 60  and 127km downwind.

  11.  The plume excess visibility budget indicated that fine particle sul-
fur and silicon species contribute ~ 50 percent to the excess bscat  in the
San Manuel smelter plume.  In the Mohave power plant plume coarse
particles •were the major contributors  to the excess bscat, which may have
been at least partially due to measurement interferences from wind-blown
dust on the sampling day.

-------
                                SECTION  3

               FUTURE PLANS AND RECOMMENDATIONS


RECOMMENDATIONS FOR FUTURE EXPERIMENTS

      The initial exploratory experiments of Project VISTTA were produc-
tive.  The characterization of the visibility-reducing aerosol with respect
to chemical composition! particle size,  and spatial distribution in remote
areas and near sources was successful.  However, these results indicate a
number of improvements and changes needed in future experiments  in order
to meet the  VISTTA goals described in the Introduction of this  report.

      The recommended improvements and changes are described below.
Some have been incorporated into the  September 1978 VISTTA  experiments
as described later in this section.   In addition,  Systems Applications, Inc.,
has suggested a number of measurements (SAI,  1978) to validate their visi-
bility models (Latimer et al. ,  1978).  Many of these measurements are
already part of the  VISTTA experimental plan.  Those additional measure-
ments which are compatible with the goals and measurements of the VISTTA
program are also included in the following discussion:

1.    Total  aerosol mass data for at least two size ranges are needed.
      This is essential for both source identification and aerosol charac-
      terization. Aerosol mass can be determined by  gravimetric weigh-
      ing, beta absorption,  or use of a piezoelectric crystal.

2.    An improved impactor is  needed to extend the size separation of par-
      ticles  down to approximately 0. 1 Jim in diameter.  This increased
      size resolution is needed for a more accurate determination of the
      visibility budget, that is,  the  contributions of the various chemical
      species to light scattering. The impactor used in the initial studies
      has a  50 percent efficiency cutoff for the final stage of 0. 5/Ltm aero-
      dynamic diameter which is near the peak of the curve of  scattering
      efficiency per unit mass versus particle size (Figure 8).   Further-
      more, the volume distribution in the Southwest,  as  described in
      Section 3, peaks in the range of 0. 1 to 0. 3/im. Thus, data on chem-
      ical composition as a function of size are most important in the
      size range from 0.1 to 1.0/im.  An eight-stage,  low-pressure

-------
      impactor with 50 percent efficiency cutoffs of 4. 0-, 2. 0-, 1. 0-,
      0. 5-,  0. 26-,  0.11-, 0.076-, and 0. 05-jim aerodynamic diameter
      which would be suitable for aerosol collection in VISTTA has been
      developed by  Hering,  Flagan, and Friedlander at CIT (Hering et al. ,
      1978).  This instrument was  tested in the MRI aircraft and on the
      ground in the September 1978 VISTTA field program.  Particles are
      impacted on stainless steel strips for sulfate and nitrate analysis in
      the present configuration, but it could be modified to accommodate
      Mylar impaction surfaces needed for PIXE analysis.

3.     Improved data on the  concentration of the major  chemical constitu-
      ents--sulfate,  nitrate, silicon, and carbon--are needed for aerosol
      characterization.  Wherever possible,  more than one technique
      should be used  to determine  the concentration of each element in order
      to validate results. In the 1977 experiments, particulate sulfur was
      measured by  both PIXE and FV-FPD, but there were no  redundant
      measurements  of other elements.

      In future experiments, particulate nitrate in at least two size  ranges
      should  be measured.  Several methods are available which are sensitive
      enough for use  with relatively short  aircraft sampling times,  such as
      ion chromatography (1C) (Mulik et al. ,  1976) or flash vaporization and
      gas phase chemiluminescence detection with an NOX monitor (Hering,
      1978).  The 1C  analysis will also give sulfate concentration which
      can be used as  a determination of the chemical composition of the total
      aerosol sulfur as determined by PIXE analysis of impactor samples.

      Silicon was found to be a major constituent of fine  particles in remote
      areas in the 1977 VISTTA experiments  using PIXE analysis of impac-
      tor samples.   In some cases, the silicon concentration exceeded the
      sulfur  concentration.   However, the low energy of silicon x-rays
      results in poor sensitivity.  It is therefore necessary to  measure
      silicon by another,  more sensitive technique to verify these results.
      A sensitive method is flameless atomic absorption (AA), but we have
      found no reference to its use for this purpose.  A less sensitive al-
      ternative is conventional flame AA (Moyers  et al. ,  1977).

      Measurements  of total carbon, soot carbon, and aerosol absorption
      are needed for  a complete characterization of the aerosol and for com-
      pleting the Southwestern visibility budget.  Carbon is  a major consti-
      tuent of fine particles near urban areas,  with as much as 50 percent
      of the carbon in the form of soot  (Macias et  al. ,  1978b).  In this form,
      this constituent is the major  contributor to particle absorption of light
      and,  therefore, is important to visibility reduction in urban areas
      Rosen  et al. , 1978). Diesel motor vehicles are a major source of fine
      particle soot  carbon (Pierson, 1978);  coal combustion in a modern
      power  plant is thought to be a relatively minor source (Nolan, 1978).
                                   10

-------
      Measurements of particle absorption in the Southwest indicate that
      the ratio of particle absorption to total extinction is  10 percent
      in  clean background areas in the Southwest and up to 50 percent in
      urban areas (Weiss et al. ,  1978).  Total carbon can be measured by
      inelastic proton scattering followed by gamma ray analysis of light
      elements (GRALE) (Macias et  al. , 1978b) or aerosol combustion with
      detection of the evolved carbon dioxide or methane (Grosjean, 1975;
      Huntzicker and Johnson, 1978; and Macias,  1978a). Soot carbon  can
      be  measured by the GRALE technique after the volatile organic  car-
      bon compounds have been removed by heating.  An alternate method
      for estimating fine particle carbon measures the reflectance from a
      white glass filter (Macias et al. ,  1978b).  Aerosol absorption can be
      measured by the Integrating Plate Method, a technique based on the
      transparency of a thin filter containing a small amount of aerosol
      (Lin et al. ,  1973).

4.    More sensitive and extensive data of trace elements are needed for
      source characterization studies.  PIXE analysis for elements such as
      silicon, sulfur, iron, etc. should be combined with neutron activa-
      tion analysis  for elements such as selenium, arsenic,  antimony, etc. ,
      seen in earlier studies (Ragaini and Gndov,  1977).  Some of these
      elements can be detected by both methods, which will act as a check
      on  the analysis technique,

5.    Sampling should be carried out when the scattering due to particles
      is  greater than 40 x lO'^rrr1 in large parts  of the region.   Observa-
      tions from aircraft indicate that  occasionally areas of  reduced visi-
      bility ~1 00 km across exist in  pristine areas of the Southwest sur-
      rounded by areas of very good  visibility (Beil, 1978).  However, it
      has been estimated by visual range observations that in pristine areas
      of the Southwest these low visibility conditions  exist only ~1 5 per-
      cent of the time (Latimer et al. , 1978).  Aircraft sampling under
      these conditions would lead to an estimate of the spatial distribution
      of the visibility-reducing aerosol.  In addition, this -would result in
      much more detailed elemental  data and would greatly facilitate
      source characterization studies.

6.    Sampling should be carried out at different times of the year, par-
      ticularly at times when the visibility impairment in the Southwest
      is  predicted to be greatest.  Examination  of seasonal visibility
      patterns (Roberts et al. ,  1975) shows  that the lowest values of
      visual range in the Southwest are observed in the summer months.
      It has been suggested that, in the Southwest,  stable conditions occur
      most frequently in the winter,  which should  result in the greatest
      plume visibility impairment (SAI, 1978).  However, background
      visual range is greatest during the winter.
                                 11

-------
FALL 1978 EXPERIMENTAL DESIGN

      The second set of field experiments in Project VISTTA was carried
out September 12-27, 1978.  The instrumented MRI Beechcraft Queen Air
aircraft was  again used for airborne sampling.  This was augmented by
an expanded ground-based measurement program.  The field measurements
were based in Farmington, New Mexico,  during the first 10 days of the
program.  Flights in the plume of the Four Corners  Power Plant were
carried out during six days; regional sampling was performed on two days.
The last six days of the program were based in Tucson, Arizona, •where
the aircraft Was flown in smelter and urban plumes and on flights through-
out the region.
      The instrumentation used in the 1977 experiments was flown in the
aircraft (Table 1).  However, a number of improvements, recommended
earlier in Section 4, -were made in this second experimental period.  A
Washington University St.  Louis aerosol  charger (Husar,  Macias,  and
Dannevick, 1976) was added to the  instrumentation package.  This instru-
ment gives an on-line measurement which is most sensitive to particles
in the size range of  0. 01-  to 0. 2- /im diameter (Sverdrup, 1977).  A cy-
clone separator with backup filters was operated on most  1978 flights.
One filter was Teflon-coated glass used for ion chromatography analysis
of sulfate and nitrate. Occasionally a second backup Nuclepore filter was
used for determining aerosol absorption by the Integrating Plate Method.
A Caltech low-pressure impactor was installed on some of the flights in
place of the Nuclepore filter to  determine the detailed sulfur size distri-
bution.  The  impactor stages were analyzed for sulfur using flash vapor-
ization followed by gas phase flame photometric detection (Roberts  and
Friedlander, 1976).   Fine particle aerosol mass was determined gravi-
metrically from the  filters.  The final filter from some of the UCD im-
pactor samples will be analyzed for trace elements by neutron activation
analysis.   A  few of the TWOMASS sampler fine particle filter samples
will be analyzed for  carbon, sulfur,  and nitrogen before FV - FPD sulfur
analysis.

      Ground-based measurements were  greatly expanded in the 1978
experiments.  The measurements carried out  in the Four Corners  area
and at Mesa Verde National Park are outlined  in Table 2.  With this
more comprehensive set of aircraft and ground-based measurements,
it is expected that a  more  detailed  characterization of the visibility-
reducing aerosol will be possible.
                                  12

-------
     TABLE 1.  BEECHCRAFT QUEEN AIR INSTRUMENTATION
  Measured Parameter
        Instrument
Aerosols  - Integral Size:

    Light Scattering

    Aitken Nuclei


Aerosols  - Differential Size:

   Particle Diameter -

    0. 01  -1.0 |jm


    0. 5 - 4. 0 |jm


    2.0- 30. 0 (am


Aerosol Samples;

    Two Stage Sampler

    Multistage Impactor

Gases:

    Ozone

    Sulfur Dioxide

    Nitrogen Oxides
Other:
    Dew Point
    Temperature
    Turbulence
    Altitude
    Indicated Air Speed

    Position - VOR
               DME

    Visual Range
MRI 1562 Nephelometer (modified by
  UW for high sensitivity)
Environment One Condensation Nuclei
  Monito r
TSI 3030 Electrical Aerosol Size
  Analyzer (EAA)

Royco 218 Optical Particle Counter
  (OPC)

Particle Measurement Systems Axial
  Scattering Probe  (ASP)
MRI TWOMASS Sampler

UCD Impactor System



REM 612 03

Meloy 285 Total Sulfur Monitor

Monitor Labs 8440 NO - NOX



Cambridge Systems 137

MRI Airborne Instrument Package



Aircraft  Navigation System

Optical Photography
                                  13

-------
TABLE 2.  GROUND-BASED MEASUREMENTS,  1978 EXPERIMENTS
Parameter
Sampling Device/
Analysis Technique
       Analysis
    Responsibility
Four Corners Area;

 Size Distribution

 Light Scattering
 Sulfate Size
 Distribution

 Nitrate Size
 Distribution

 Mass1
 Sulfate2
 Nitrate2

 Carbon

 NH3

 Plume Optics

 Wind Speed
 Wind Direction
EAA, OPC
Nephelometer

LPI - FV/FPD
LPI - FV/
Chemiluminescence
Cyclone Separator with
Teflon Filter
1 Gravimetric Weighing
2 Ion Chromatography
        CIT
        CIT

        CIT

        CIT
        CIT
        ERT
TWOMASS Sampler/GRALE           WU
Tandem Filter                Rockwell Internat'l
Telephotometer
Pibal
     EPA - LV
New Mexico Health

        MRI
Mesa Verde National Park:
  Size Distribution
  Light Scattering

  Aerosol Absorption

  Silicon
EAA, OPC
Nephelometer

IPM

AA
        UW
        UW

        UW

  Univ.  of Arizona
                                  14

-------
                              SECTION  4

                       PROGRAM DESCRIPTION
      The initial exploratory field program of Project VISTTA was carried
out October 1-10, 1977. This initial phase was designed to outline the ex-
tent and nature of the visibility impairment in the Southwest.  The empha-
sis was placed on obtaining as complete a characterization of the aerosol
as possible in order to aid in planning a more optimal set of experiments
in the later stages of the VISTTA program.

      Measurements were taken within the mixing layer (<500 m above
ground level) using the instrumented MRI Beechcraft Queen Air  aircraft.
This twin-engine aircraft has minimum and maximum sampling  speeds of
180 and 290 km/hr, respectively.   With a full instrument complement, the
maximum sampling time is over four hours.  The aircraft was equipped
to to continuously measure and record on magnetic tape light-scattering
coefficient, Aitken nuclei count, ozone, sulfur dioxide,  nitrogen oxides,
temperature, dew point,  turbulence,  pressure (altitude),  and navigational
parameters.  Aerosol size distribution measurements were made over a
range of 0. 01 to 30^im. using three size-selective aerosol  sensors.  An
electrical aerosol analyzer and optical particle  counter  were used in con-
junction with an automated "grab bag" sampling system  designed to col-
lect and hold a discrete  static air  sample for analysis (Blumenthal et al. ,
1978).   An axial scattering probe was operated in the free stream near
the nose of the aircraft for particle size  measurements  in the range of
2 to 30pim.  Multistage impactor and two-stage  filter sampler samples
were also collected in order to determine the elemental composition as
a function of size.  The details of the aircraft instrumentation are given
in Table 1.  Visual range measurements were carried out by viewing  dis-
tant landmarks and verified by optical photography.

      Gas instrument calibration was carried out by personnel from the
EPA - ESRL.  The nephelometer was calibrated with Freon-12 (bgcat =
235 x lO^rn"1 at 525 nm, sea level) and filter air (bscat =  1 5  x  1 O"6 nr1
at 525 nm, sea level).  A gas phase titration was used to calibrate the C-3
and NO monitors. The SO2 monitor was calibrated with a permeation
tube.
                                  15

-------
      The multistage impactor, a Lundgren-type impactor specially mod-
ified for aircraft sampling by investigators at University of California at
Davis (UCD), was mounted in the nose assembly of the aircraft.  The im-
pactor stages had 50 percent collection efficiency for particle diameters
of 4,  2,  1, and 0. 5j*m.  Impactor samples were collected on greased My-
lar film.  A Nuclepore final filter was used to collect all particles not col-
lected on the four impaction stages.  The impactor samples were analyzed
by particle-induced X-ray emission (PIXE) (Cahill, 1975) at the Crocker
Nuclear Lab cyclotron at UCD.  This analysis gives  elemental concentra-
tion of elements with atomic numbers of 13 (Al) or greater.

      The fine particle samples from the TWOMASS  two-stage sampler
(Macias and Husar,  1976) were analyzed by flash vaporization-flame
photometric detection (Husar  et al., 1975) by J. D. Husar and Associates
(St. Louis, MO).

      The field program, summarized in Table 3,  consisted of seven sam-
pling  days. The experimental plan included flights of the region on three
full days and parts of one other, flights in the plume of a copper smelter
(San Manuel smelter) on three days, and flights in the plume of a coal-fired
power plant (Mohave power plant) on one day.

      The regional flights were designed to characterize the visibility-
reducing aerosol on flight paths of over 300 km.  On two days the total flight
paths were over 1000 km. All flights were within  the mixing layer;  how-
ever, occasional spirals were made to define the mixing layer structure.
The visibility along these paths was documented with a high-sensitivity
nephelometer,  a camera, and with visual range measurements from the
aircraft to prominent landmarks.  Ground measurements were made by
UW at Canyonlands National Park with a nephelometer and sampler for
aerosol absorption.

      Smelter plume flights were carried out at several downwind distances
up to  127 km from the San Manuel smelter. Flights in the 1580 MW Mohave
power plant plume were  performed at distances as far as 62 km downwind
of the plant.

      The plume flight patterns were designed to provide a detailed char-
acterization of the plume at discrete distances downwind from the source.
At each distance horizontal traverses were made through the plume  normal
to the plume axis  at several elevations.  Traverses were long enough to
include background air on either side of the plume.  Vertical  spirals were
also performed to help characterize the plume structure.  In  addition to
cross-sectional plume traverses,  orbital flights within the plume of up to
                               16

-------









r^-
r^
o

i
>^
Pi
<
1^-1
S
S

CO
S
<
Pi
O
\M/
O
Pi
ft
p
j
H
H- 1
[T.
H^
<
H
H
CO
I-H
>
K^

TABLE 3.








CD
-s
cu
8
0
U


M 0)
CD .5 0 ^
a i-H Pi C1
- a rt S
S £ »*
^P
•y *
-£ n
.2° ^
£ 0
ft ffi
» a
OJ *^
d °
c -^ +*
•S CH CO
HQ
SS*>
bo £
•IH rt
i— i -P
ft w




0
o
ght Descripti
• r-t
•— (
ft
CU
-p
rt
Q

Tt
C
rt
3 -0
ume Wee
Scattere
r-H
ft
•tf
i~H

PO


(M
PO

O
PO
o

t^-
r— t
v£>

_^
NJ
5
CO
JH
Manuel and
ayden Smelte
C ffi
rt
CO
r~
t^
I-H

-^
0
• — i
h
'rt
ft
fi
o
• H
.•s
£
•»-<
WH
CU
Q
CD
1
i-H
OH
O
vO
*
fM
PO
•t
00

s£>
•
^

in
, — i
•*
,— (
(M
•*
• *
O



^(
CD
-P
Manuel Smel
£
rt
CO
r^
r^
(M

— .
o
i — i

T3
CD
fi
tri
*+H
CU
Q
i-H
^H
CD
^
CD
S
0
!-H
ft
t^
CM
^•4

*.
o
xO
"^
CM r-
•-H
Manuel Smel
enix Urban P
s|
CO ft
h-
r^-
^f

^>
o
i-H














/•*
PO
•
PO

CM
PO
PO
^H
i-H
— 1
• •
O
1 — 1


T)
1
CO
I-H
1
N
<
a
rt
-P
CO
bo
rt
i— i
ft
r~
r>-
m

>^
0
i-H


bo
.3
U
^M
g
rt
CO
i-H
rt
o
• H
bo
CD
Pi





-^- v
r^-
•
CM

0
PO
*•
r*-
r-H
m
•*
• •
^F
i— (
to
c
rt
JH
U
1 1
<1— ^
si
^ «
CO £
,vs
fl s
•s s
ft





                              bo
                              Pi
                              CD
                              CO
                              cu
                         ft
                           (M
                           CM
                           ro
                          00
                          oo


                          O
           bO
           C
          •H
          U
           g
           rt
           rt
           PI
           0
          • |H
           bo
           cu
          Pi

CM  m    T$<
o  o    -*

O  vO    sD
                                                   bo
                                                   c
CO
i-H
 rt

 0
• tH
 bo
 cu
                                             PO      in

                                             CM      t^-
                                             Tl*      PO
                                             O      O
               bo
               PI
               • |H
               i-H
               (X
               g
               rt
 rt
 C
 O
•H
 bo
 cu
Pi
                                                                 0
                                                                 o
                                                                 fO
ro
i-H
ft
Power
CD
>
rt
A

S
<*H
rt
-4->
CO
bo
rt
i-H
h
cu
rt
X

S
CO
J_»
fi
rt
^
O
<
bo
.3
g
^
rt
h


i
^
rM
CO
CU
43
•p
C
•|H
T3
rt
^H
CO
1 — I


«H
»4H
rt
-M
CO
bo
rt
>— i
h


CD
i-H
•o
CD
CD
z
ia
rt
-4->
CO
bo
rt
i— 1
h
\«/
Tl
w
c
rt
i-H
O,
D
1
^^
<
U


                                        o
                                        ^H

                                        O
17

-------
one hour were carried out at each distance to gather heavily loaded impac-
tor and filter samples.   Orbits upwind of the source were flown to  gather
background samples.

      Single theodolite pilot balloon (pibal) releases were carried out at
various times on every sampling day to determine wind direction and speed
as a function of altitude in the sampling area.
                                   18

-------
                               SECTION 5

                     SUMMARY OF RESULTS FROM
                       FALL 1977 EXPERIMENTS
CHARACTERIZATION OF THE VISIBILITY-REDUCING
AEROSOL IN THE SOUTHWEST

Spatial Distribution

       The character of visibility-reducing aerosols in the Southwest
was determined in long regional flights in order to measure a regional
background.  These flights were made between or upwind of large point
sources of pollutants, such as urban areas and power plants.  All flights
were  at a constant altitude to within ±25 m.  On October 5, 1977, flights
were  made  between  Flagstaff (AZ) -  Tuba City (AZ) Island in the Sky -
Farmington (NM)  - Grants  (NM) Flagstaff, as shown in Figure 2.  The
flights on October 9 followed the same 1080-km path in the  reverse di-
rection.   Shorter flights in the region (~260 km) were flown on October 8
and 10.   On all days, the visibility and air quality were quite good (aver-
age visual range = 140 km).  This resulted in very  low values for all mea-
sured aerosol and gas parameters except ozone, which was present in
concentrations representative of clean air.  Therefore, only data from
the longer flights on October 5 and 9 which have more meaningful aver-
ages, given in Table 4, are discussed in this report.  The averages for
each day were determined by averaging over each segment  of the flight
and then averaging the flight segment averages. The NO, NOX, and SO2
concentrations were below the sensitivity of the monitors throughout the
region.   The bscat values, although quite low, were detectable with the
aircraft instrumentation.  The light scattering coefficient values bscat,
given in Table 3, are from the nephelometer output with the contribution
of Rayleigh scattering of air molecules included.


  TABLE 4.  AVERAGE PARAMETERS MEASURED DURING REGIONAL
             FLIGHTS
Date
10/5/77
10/9/77
Altitude
(MSL)
2428i23
2289A21
Flight
Length bscat SO2
(km) (xlO"6^!"1) (ppb)
1080 28i3a <1
1080 25±1 
-------
 m
in

 s-,
 o
 -p
 y
 o

 00
•+J
X

 bJD
 bC )
  c
 1-1 m
   •1-1
i-
   bo
(M

0)
}-,

S
bD

-------
      The spatial distribution of the visibility-reducing aerosol is most
easily determined from the frequency distribution of the light-scattering
coefficient due to particles,  bsp, as shown in Figure  3.  These values are
determined from the 10-sec average nephelometer readings with the Ray-
leigh scattering contribution to the  scattering coefficient at the flight al-
titude subtracted.

      The average value of bsp on both  sampling days was rather low,  rang-
ing from (25 ±1) x 1 O^m'1  on October  9 to (28±3) x lO^rrf1  on October 5.
Although these values are  low,  they are above the minimum sensitivity of
the instrument [bs_ (min)~ 5 x 10"6m"1].  The Rayleigh scattering or clean
                                                 /i
air contribution at the sampling altitude (~11 x 10m) was, on the aver-
age, 42 percent of the total light scattering coefficient, during the very
clear days on which these  regional flights were made.  The average mea-
sured visual range in-Mesa Verde and other national parks in the region
in 1976 (Latimer et  al. , 1976) was about equal to the average visual range
calculated from the  scattering coefficient on the VISTTA sampling days
(~150 km).  Therefore, the VISTTA data  can be considered typical for the
region.  The calculated visual range can be verified from photographs  such
as shown in Figure 4.  Navajo Mountain, which is at  a distance of 11 0 km,
can be seen near the horizon  in the  photo.  This photo was taken with a 50-
mm Nikon lens,  Kodachrome 64 film,  and a UV filter.  The ground eleva-
tion is ~2000 m msl, the aircraft is at 2429 m msl, and the mountain top
is at 3166 m msl.  The visual range calculated from  bscat measured along
the sight path shown in Figure 4 was ~140 km.

      The coefficient of variation of the bsp frequency distribution was  7
percent on October 9 and 18  percent on October 5 over the entire 1080-km
flight path,  which indicates that the aerosol •was homogeneous on both days.
Size Distribution

      The average size distribution of background Southwest aerosol mea-
sured on the aircraft during the October 9 regional flight is shown in Fig-
ure 5.   The distribution is bimodal.  Reduction of the aerosol size distri-
bution data was accomplished using reduction procedures described by
Cantrell and Whitby, 1978.  The size distribution data were collected every
five minutes during the entire flight.  All measured size distributions were
averaged before extracting integral parameters using a lognormal approxi-
mation technique.   There is some question about altitude effects on these in-
struments,  which may lead to errors in the size  distribution measurements.
However, we do not think these errors are large enough to alter the conclu-
sions of this report.  We are presently looking into the magnitude of this
                                 21

-------
3£
oe o —



— z n

= 5
                                                                                ,.*•
                 L
                                         I              f •
                       o
                    •*  VO
                    
                 r^  fl  _i

                 5?  ~  §
                 ^,   Q.1^
                 O   i/l —
                   O
                   a
o
o
                                                                                              to
                                                  a
                                                  n

                                                S
 cu  o
 rO  t~{

 T(  CO
 J-i  nj
 rt  >

 ^^
 Rj  0
 •P  CO
 CO  o


 £  cu
 •H  nj

 *  cu
_  45
                                                        rt
                                                       45
                                                    O -^
                                                                                                O S  -H
                                           bo 'S
                                           a  S
                                          .•-i  rt

                                           cu  ^
                                          S  ^
                                           (ti  CU
                                           O  42
                                           m  o
                                          +*  u

                                          •a°

                                          -  §
                                           2  o
                                                                                             °  ^H  —1
-g

"H 'S
 fl» -S
 cu U

 cu •£

 £ "o

^ ex
T3 co
 (5 cu
 Ctf £-(
                                                                                             a
                                                                                             o
                                                                                                       !H •

                                                                                                       CU
                                                                                                      42
                                           C  «»S
                                          -u  CU
                                           CO  (H  h
                                          ^     CU


                                           >>"^  O
                                           O T3 TH
                                           .-i  QJ  U
                                           Q  ^  cu
     > CO


      rt

      T)

      CU
      CO
                                                                                                      £ o cu
                                                                                                      U ^H j-.
                                                                                                      S
                                                                                                      O
      £5
      o
g
                                           Hi
                                          h  a  &
                                                                                            CO
CO
CU  T3  CO
bC C  j3
ctt  ni  o


S^  g
>•  m  cu

ctf  O  bO
                           3DN3aano:o
                                                                                             50
                                               22

-------
Figure 4.  Photograph of Navajo mountain at a distance of
           110 km from the aircraft of October 5, 1977.  The
           visual range  calculated from measured bscat values
           along the sight path was ~140 km.
                             23

-------
                                                                                                 Q)
           QL
           V)
  u

n
  E  E
  a. a.

 in in c\j

 CM in csj

  u  u  u

  (J  L) D>
 ^> c^ b
     o
     O.
     O
<
<
LU
                    \O CM CM


                    CM O CM
                                                                                                 O

                                                                                                T3
                                                                                                 0)
                                                                                                 at
                                                                                                 rt
                                                                                                 0)
                                                                                                 o
                                                                                                 CO

                                                                                                 O
                                                                                                 ^

                                                                                                 0)

                                                                                                 rt

                                                                                                 4->
                                                                                                 CO
                                                                                                 0)
                                                                                                •g
                                                                                                 bo
                                                                               C   rt
                                                                               o   C
                                                                              '•S   2


                                                                              5   «
                                                                               ^   "-1
                                                                              -<->

                                                                              .2  t-^
                                                                              •^3  t—
                                                                                                 0)
                                                                                                un

                                                                                                 a>
                                         901
                                                                                                 bo
                                                24

-------
flow rate effect.  The fine particle or accumulation mode had a geometric
volume mean size,  D_ , of 0. 25jim and a geometric standard deviation, Og,
of 2.0.  The coarse mode geometric volume mean size is 5. 5pm, with a 0"
of 2. 2.  The volume concentration of particles is 2.6  mVcm3 and 2. 5
mVcrr^for the fine and coarse modes, respectively.  The accumulation
mode mean size is slightly smaller than the continental average of 0.
described by Whitby (1978).  However, the coarse particle mean size is
the same as the continental average of 6 jzm, within experimental error,
expected for particles from similar sources such as windblown dust.
Elemental Composition

      The elemental composition of the visibility-reducing aerosol in the
Southwest was determined from impactor and filter samples  collected on
regional flights October 5 and 9, 1977.  As discussed previously, the  aer-
osol was  quite homogeneous throughout the region during this sampling,
making a regional average meaningful.  The elemental concentrations
divided into fine particles (Dp £l /am) and coarse particles (Dp 21 /im) aver-
aged for all samples collected on October  5 and 9 are given in Table 5.
The rationale for the separation at 1 pm is based on the high  scattering ef-
ficiency for particles with diameters less than 1 ^im.  Furthermore, parti-
cles in each of these size ranges generally have different  chemical compo-
sition.  All other elements not listed were below the minimum detectable
limits.  The elemental concentration of fine particle sulfur was measured
by FV-FPD analysis of TWOMASS  samples.  All other data are from PIXE
analysis of impactor samples. There was some discrepancy between the
sulfur  concentration values from the two techniques with the PIXE values
being generally higher than the FV-FPD results.  The explanation of this
discrepancy is made more difficult because different samplers as well as
different analysis techniques were used. In the second VISTTA study in
Fall,  1978, additional samplers and sulfur analysis methods  were used to
get a more detailed  understanding of this discrepancy and  to obtain more
reliable sulfur concentration values.  In this report, the FV-FPD results
were used in order to obtain preliminary conservative sulfur  values. The
detailed explanation of this problem will be given in a later VISTTA report.
The coarse particle fraction is dominated by crustal elements such as
silicon, aluminum,  calcium,  and iron.  Sulfur, silicon, and potassium
are predominant elemental species  in the fine particles.

      The  size distribution of individual elemental species determined  from
regional impactor samples collected on the October 5 and  9 regional flights
is shown in Figure  6.  Aluminum, calcium, and iron all show very similar
size distributions, with the highest concentration of each element in parti-
cles with  diameters greater than 4fim.  The concentration on the final  fil-
                                   25

-------
ter (DpSO. 5^m) was below the minimum detectable limit for all these ele-
ments.  Silicon has a size distribution similar to aluminum down to Ijim;
below that diameter the concentration of silicon increases.  The concen-
tration of small particle silicon (DpSO. 5^m) is nearly equal to the large
particle value (Dp>4jzm).   These  data  indicate that  silicon is  a major con-
tributor to the fine  particle mass  and,  as discussed in a later section of
this report, to visibility reduction.  This is an unexpected but quite impor-
tant result of this work.  From the source characterization determined in
this work, fine particle silicon is likely to be the result of condensation of
volatile silicon emitted from high-temperature combustion sources.  It is
important to verify the presence of small particle silicon by further chem-
ical measurements.
    TABLE 5.   AVERAGE ELEMENTAL, CONCENTRATION OF AEROSOL
                 IN THE SOUTHWEST REGION, OCTOBER 5 AND 9, 1977
                                 Fine Particles1"           Coarse Particles0
           Element'	fcg/m3)
Al
Si
S
Cl
K
Ca
Ti
V
Cr
Fe
Zn
Zr
Pd
Ba
Pt
Pb
0.01
0.72
D.68
0.11
0.21
0. 03
0.08
0.04
_
0. 004
0. 03
-
0.02
0.10
0.07
0.09
0.28
1.12
0.06
0.01
0.09
0.24
0.01
0.01
0.01
0.16
0.003
0.02
0. 01
0.01
0.01
0.004
           *The following elements were detected in only a few samples and, there-
           fore, it is not possible to determine a meaningful average concentration:
           Fine Particles - Cu, Mn, Ni, Rb, Co, Hg,  Mo.
           Coarse Particles - Cu, Mn,  Se, Ni, Rb,  Co, Mo, Br.
            Fine particle (Dp £ 1pm) concentrations were determined from the sum
            of impactor stage 4 and the final filter.  Sulfur values were determined
            from TWOMASS samples.
           "Coarse particle (Dp ii^im) concentrations were determined Irom the
            sum on impactor stages 1, 2, and 3.
                                     26

-------
1
CO

1 1

r f

-
-
CO
1 I
in
1

.f : ....

I



	 I
If)
6
1 1
^

i 1

l 1



I 1
m
o










1 1
P

1 1

I l



o>
LL
1 1
m
6
7, »
^" Q^ --!->
s ^ •-
1 ;?s
» " *
« 8 o
""^ o
2 « ^
'u ^ ^ •
^ a ^ a; «
O CO 73 LLJ M
1 1 S^s
^ (-< tn ^ u
r-l CU _1
(U ^ fi
K s s «:-
Lu cu ^ n c/5
| 	 1-4 O W
' CU 4J  £ prt
O ^ cu tti i
~~ LU c "o ^ P.
-J iH ^ S u
p rs;§
5 -2 co .0 „
^i, "^""' ^ i ' f~!
Q_ 3 ^H 3 ^3
CM T* C ^2 ^
JH c r cu
-^ n5 43 -u
•" M 2 d
T^ ^| rj^ C
« 5 cu-2
_N o ^ T3
co rt 'r* -u
10 « C ^
O.T ^ CtJ ""^
tlj .5 •"
(u CO s.
«- C ttf >
0) C rj O
> ° ^
< ^ ^ n
^s 'i i 0 j , "
(y
fcr
.LU/&nr'dQ 901
 NOIiVdlN3DNOO
27

-------
      Particulate sulfur is concentrated mainly in small oarticles
(Dp  £ 0. 5/im).  This is reasonable considering that SOi" i-s a large frac-
tion  of the accumulation mode which peaks at 0. 25/^m.  Titanium and
potassium show slightly increased concentration for small partricles,
but the variations with size are not dramatic.
Chemical Composition

      The total aerosol mass was not measured, but it can be estimated
from the measured elemental concentrations, and the oxygen concentra-
tion estimated with assumptions of the chemical form of the measured
elements.  The  sulfur can be expressed as sulfate because the FV-FPD
method is specific for water-soluble sulfate (Husar  et al. ,  1975).  It is
also assumed that ammonium is the only ion associated with the sulfate.
Other elements  are assumed to exist as oxides.  This leads to the mass
balance listed in Table 6 and Figure 7.  The mass estimates may be low
due to unmeasured elements, particularly those with  Z < 13.  The total
mass in each size range can also be estimated from the aerosol volume
determined by integrating each mode in the  differential size distribution
and by estimating the  aerosol density from the chemical composition.  A
comparison of these two approaches,  given  in Table 7 indicates that the
elemental mass balance accounts for ~93 percent and ~64 percent of the
fine and coarse  particle mass  derived from volume, respectively.  The
assumptions  of chemical form, although somewhat arbitrary, lead to a
conservative estimate of the total mass.  For example, iron has been  as-
sumed to be  present exclusively as  Fe2O3 when  in reality it is likely that
FeO is  present as well.  This  is the case for crustal material (Mason,
1966).   Comparison of fine particle mass and bsp allows a  further check
on the mass  estimate.  Extensive measurements in  the western  United
States indicate that the fine particle mass /bsp ratio is normally 0.32 to
0. 3i g/m2 (Waggoner,  1978).  The ratio for the VISTTA regional data  is
0. 34 g/m2, -which is a further indication that the fine particle mass esti-
mate is reasonable.

      The total measured aerosol mass in both  size ranges (9. 23/ig/m3)
determined from the sum of the concentration of each compound is lower
than the average total suspended particulate value (~20pig/m3) measured
in several National Parks in the region (Trijonis and Yuan,  1978).  It
appears from, volume  and bscat measurements that this discrepancy is  not
primarily due to fine particle mass estimates.  It is more likely that the
coarse  particle  mass  estimate is low due to unmeasured constituents.
Furthermore, coarse particles may be  diminished due to settling at 300
to 500 m above the ground where the samples were collected.
                                  28

-------



1
1
2
O
t— t
O
H
«
w
W
£>
|J~
H
D
O
t/3
W
ffi
H
05
O
fa
H
U
j^
<
<; £
CQ c>
W ~

*™H ^N
n
M ^
£ 5
w ^
in
3«
u w
*""* w
S o
H H
ffi U
U O
•
NO

H
CQ

(C
rt
o
U




c
^o
D ^
T^
^ *
c
o
U



c
o
*^ Jl^
rt r
i E
c -•»
v be
u a.
c ^
0
U












CR
0)
U
'•*J
w
rt
0.
OJ
c
fa




rt
C
?~.
*n "^^
^>
c
o
U





c
entratio
K/m3)
y "3
C "-
o
U





•o
c
a
0
U


^sD^PONOmoO-- ^
mOoroooc*in°o ro
^^ >,O








rf Tf
• •-•••••• *
^^ ^^ f^ "T!*1 ^^ ^5 ^^ ^5 <^* tT*i











^J* ^^  *^j ^^ f^
N m









OmooiMo — o o — • f^^^
O_'NOOO'OOC5 oE
IH
o
**H
0)
'x
o
l»(
cc
£
o S
to ^
^ nn 5 O o" w O O S.5
.JJ'U^H tflhioc^ i C-



o
o
o
i-H







sO
CT-
•
m











0
0
o
^4








fsj
in











rt
O
H

29

-------
                                       so:
           COARSE PARTICLES.
           MASS =4.0
                                                            SiO,
                                                            61%
           FINE PARTICLES  .
           MASS = 5.3  4/g/mJ
                      39%
                                                        SiO,
                                                        —A10, CaO,
                                                                   78-404/1
Figure 7.  Size-fractioned mass balance of Southwest background aerosol mea-
           sured on regional flights on October 5 and 9, 1977.  The concentration
           of individual compounds was determined from the measured elemental
           concentration and assumed chemical form.  The total aerosol mass in
           each size fraction was  determined from the  sum of the individual chem-
           ical components.  Comparison with  aerosol  volume measurements indi-
           cates that the mass  estimated from  composition measurements repre-
           sents 93% and 64% of the fine and coarse particle mass,  respectively.
                                        30

-------
f













NATIONS
CEMENTS
§5
I?
w 3
H 5
W *
Q Z
S2
;2 H
< KH
^ CO
^ 0
ROSOL
COMP
H Q
< Z
r 
C
a;
a
0)
h
s
to
rt
flj
W
S-H
S
C
O
• «-)
. Compos it
erosol
 o J5
.£ is 5
^ OJ
<" ^ T3 —
Q .^ aJ *J
I *J > C
1 n) ^i 
& - a
c s £
o •"*
U ,0
S §
2-1^
TO ^ •£ fO_,
« TS g a
rt  H" 60
£ C 3-
^ o ~
Q U
U)
"T3  <
w a a
" 23
S^


0)
t) tJ
01 rt
•" ^
•^ .B
be 4J -^
|wraa
^ ^^
to .tJ &C
03 03 -^
rt C
•5!  3







0)
, — 1
o
'43 0)
M OH
n) >.
OH H






CO ^*
0s vD






ro O
in •*




r- IM
• •
ID sO





-------
      Probably the most important species unaccounted for are nitrates,
carbon compounds, and water.  The average annual concentration of ni-
trate in the nonurban Southwest region was about 20 percent of the sulfate
value in the early 1970's as measured in Grand Canyon and other remote
Southwestern national parks by the National Air Surveillance Network
(Latimer et al. ,  1978).  These values may be  subject to error due to arti-
fact sulfate and nitrate effects on filters.  Very few data exist on the con-
centration  of carbon in Southwest ambient aerosols, but it is a significant
component of eastern urban aerosols (Lewis and Macias, 1978).
Enrichment Factors^

      The predominant species in the coarse particles are elements found
in high concentration in soil and crustal material.  Elemental enrichment
factors,  given in Table 8, help to characterize the source of these particles.
These factors are calculated as follows:
           Enrichment Factor (EF)   =
(X/A1) Aerosol
 (X/A1) Crust
  TABLE 8.  SOUTHWEST REGIONAL AEROSOL ENRICHMENT FACTORS
     Element
                                Enrichment Factor
                     Fine Particles             Coarse Particles
Al
Fe
Si
K
Ca
Ti
S
Cl
V
Cr
Zn
Zr
Pd
Ba
Pt
Pb
1.0 ) Crustal 1
0. 7 /Abundance 0
21 ^ 1
66 \ 1
7
IxlO2
2xl04
7xl03
2xl03
_
4xl03
_
2xl07
2xl03
6xl07
1
0
67
Enriched 23
Relative 29
to Crust 1 3
35
3
7
3
88
6 x 1 O4 J
• ° ^\
Q
. 2 Crustal
. 0 f Abundance
Q
.7 _J
^\
\

Enriched
> Relative
xlO5 to Crust

xlO5 1
y

                                 32

-------
Aluminum was chosen as the reference element because its major source
is likely to be soil.  However, aluminum is also present in flyash from
coal combustion (Lyon,  1977).  Crustal abundances were taken from Mason
(1976).  Enrichment factors >1 indicate an enrichment of that element
relative to crustal abundances; values <1 indicate a depletion relative
to crust.  Because of the uncertainty in the data,  EF values between 0. 5
and 2 indicate no significant difference from crustal abundance.  It can be
seen from Table 8, that Si, K, Ca, Ti, V, and Fe are all near the crustal
abundance in the coarse particles relative to At.  These elements repre-
sent 93 percent of the measured coarse particle mass.  The other coarse
particle species are enriched, indicating an anthropogenic source. This
analysis is complicated by the fact that many of these elements are pre-
sent in coal flyash with abundances  similar to crustal material.

      All fine particle species except iron are enriched relative to the
crust (normalized to aluminum) with enrichment factors ranging from 7
to 2 x  107.  The enriched fine particle species represent 99.5 percent of
the measured fine particle mass, which indicates that the vast majority
of fine particles are not due to wind-blown dust of crustal origin.
Light Scattering Budget

      The detailed contributions to light scattering as a function of par-
ticle size can be determined from Mie calculations (Mie, 1908) of the
light scattering efficiency and the measured particle  size distribution.
The sum of the calculated bsp  over all particle sizes  can be compared to
the measured light scattering  coefficient from the integrating nephelo-
meter.  Mie scattering calculations combined with particle size and com-
position measurements permit the determination of the individual contri-
bution to visibility reduction for particles of a given composition and size
(light scattering budget).  This approach can also serve as a test of the
consistency of the various measurements. Ensor et  al. (1972) have de-
scribed the calculations used in this work in an  earlier paper.  The fol-
lowing expression was evaluated in the  scattering coefficient calculation:
      bscat ~

where G is the light scattering per unit volume of aerosol.  The quantity
G, calculated using Mie scattering functions,  is plotted in Figure 8 as a
function of particle size for a refractive index m = 1. 54 and a standard
solar radiation distribution at sea level.  It can be seen from the figure
that particles with diameters between 0.1 and 1 jim have the greatest scat-
tering efficiency.
                                   33

-------
Figure 8.  Light s;. i-; ^ring per unit volume of aerosol material
           as a function of particle size, integrated over all
           wavelengths for a refractive index,   m = 1.5.  The
           incident radiation is assumed to have the standard
           distribution of solar radiation at sea level (Bolz,
           R, E. , and Tuve,  G. E.  (Eds.),  1970, Handbook of
           Tables /or Applied Engineering Science, Chemical
           Rubber  Co. Cleveland, Ohio, p. 159.)  The limits
           of integration on wavelength were 0.  36 to 0. 680jjm.
           The limits of visible light are approximately 0. 350
           toO.TOO^m.  (Friedlander,  1977)
                           34

-------
      The contributions to the light scattering coefficient as a function of
particle  size for the average Southwest regional aerosol (October 9, 1977)
calculated from the average regional aerosol size distribution are given
in Figure 9.  A refractive index, m = 1.54, was used in this calculation,
which is slightly lower than the mass-weighted values of 1. 57 and 1.65
for fine and coarse particles, respectively, calculated from aerosol com-
position  measurements.  The calculation was done for light of 525 nm
wavelength.  The sum over  all calculated contributions yields

     ]£bsp   =    (16±2) x I0-6m-1  ,

which agrees well with the average value of light scattering due to parti-
cles measured with the integrating nephelometer (with the contribution of
Rayleigh scattering due to gases removed) of bsp(measured) = 14±1 xlO"6m"

      The visibility budget for the Southwest background aerosol on Octo-
ber 9,  1977, determined from the Mie calculations, indicates that 52 per-
cent of the light scattering was due to fine  particles (Dp sl^im),  44 per-
cent was due to Rayleigh scattering from gases, and 4 percent was due to
coarse particles.  Considering only the light scattering due to particles,
93 percent is due to fine particles.   The detailed visibility budget, sum-
marized in Table 9, was constructed using the results of the fine particle
mass balance and assumes  that sulfur and  silicon compounds have the
same size distribution as the total fine particle  aerosol.  This is reason-
able, since these compounds represent  82  percent of the measured fine
particle  mass.
TABLE 9.  LIGHT SCATTERING BUDGET FOR THE SOUTHWEST REGION
            OCTOBER 9, 1977  (Visual Range ~160 km)
Component
Dp  (Aim)
                                         scat
Contribution
to Total bscat
 (percent)
Air Molecules
(NH4)2S04 a
SiO2b
Other Compounds
Coarse Paricles
(Given in Table 5)

0.
0.
0.
1.


1
1
1
0


to
to
to
to


1.
1.
1.
20.


0
0
0
0

11
7
4
2
1
25 XlO^m-1
44
28
16
8
4
100
 aAssumes that all fine particle sulfate exists as ammonium sulfate.
 Assumes that all fine particle silicon exists as SiO2.
                                  35

-------
                                                                  E
                                                   •I

                                                    •
                                                                           TO  «



                                                                           I  «
                                                                           .£  TO

                                                                           tJ  
-------
      The ratio of fine particle sulfate mass to the total bsp value for the
Southwest region measured on October 9, 1977, is 0.15 g(SO4=)/m2.  A
similar calculation for the fine particle SiO2 mass / bsp ratio yields
0.11 g(SiO2)/m2.  Perhaps a more interesting calculation is the ratio
of fine particle  sulfate or SiO2 mass to the contribution of that species to
bsp, as determined in the visibility budget.  This ratio is 0.4 g/m2 for
for both fine particle sulfate and SiO2. *  This calculation assumes that
sulfur and silicon have the same size distribution.
SOURCE CHARACTERIZATION

      In order to characterize the emissions from specific sources in the
region, measurements were made in plumes from the San Manuel  copper
smelter on October 1,  2, and 4, 1977, and the Mohave power plant Octo-
ber 8, 1977.  The flight paths  on these days are shown in Figures  10 and
11.  Separate orbital flights in the plume and in the background, as well
as plume traverses, were carried out at each distance downwind of the
source  as outlined in Table 2.   Analog gas and aerosol parameters mea-
sured during these flights averaged over each flight segment are summa-
rized in Table 10. Data from October 1 have not been included in Table  10
or in any subsequent data analysis because of the poor spatial resolution
of the plume on that day. A photo of the San Manuel smelter plume 8 km
downwind of the plant looking normal to the plume is shown in Figure 12.
The plume is  easily visible and well defined against the sky and mountains.

      To help evaluate the data presented  in this section on source char-
acterization it is instructive to compare the emission  rates of the San
Manuel smelter and Mohave power plant.  In 1975 the  Mohave power plant
emitted an estimated average of 66 tons/days of SOX and 77 tons/day of NOX
(Marians and  Trijonis, 1978).   In 1977 the San Manuel smelter emitted  557
tons  per day of sulfur,  expressed as SO2 (Larson and Billings, 1978).
      The inverse of this ratio, i.e., the contribution of SO4~ or
      SiO2 to bgp divided by the fine particle SO4~ or SiO2 mass is
      2. 5 x 10-^m
                    for SO4  and for SiC2.
                                 37

-------
                                  u

                                 O   .
                                     O
                                     4->



                                     o
                                     rj-j
                                 o

                                 !—I



                                  0>





                                  QD
38

-------
Figure 11.  Flight map of Mohave power plant flights
            on October 8,  1977.
                         39

-------
Figure 12.  Photograph of the San Manuel smelter plume
            8 km downwind of the plant looking normal to
            the plume.
                        40

-------
 in
 H
 ffi
 O
 H
 H
 (—4
 CQ

 O

 O
D
Q
Q
H
w
2
w
C4
w
H
W
H
O
W










SMELTER
J
H
D
2
"S
i^
2
5















XfcL
O n
2 —
^-.
f~\ r^
23
3"

^•"•*

0N5



•^
4J C
m i
J3 O
^H

 o r* o m co
CO O CO CO ON Tf O
rO T3
00 CO O £ O I- £
CO vD QJO vD CO yj
CQ «
r^ r^
CO -^
*^^ • —
0 0










'ER PLANT
£
0
PH
W
>
 v£)

CO
00 ^H O
-H
M Tl QS
m co .
0


CO CO CO
CO -4
•U -H •«
^4 ^N v^
in \o co
m co co
r-4 ^H CO
-H -H -ti
O O Tj4
r^ r- co
vO "^ ON

CO CO "g
CO vO >-(
«
£
00

o
t—4
                                                                              e
                                                                              o

                                                                              0°
C
0)

g
                                                                              CO
                                                                              a
                                                                              0)
                                                                              o
                                                                              n
                                    41

-------
Plume Impact

      The impact of a plume on visibility can be estimated by comparing
the visual range with the plume present to the visual range through the
background without the plume.  The parameters used in this semi-quanti-
tative calculation are  shown in the schematic diagram below.
OBSERVER
PLUME
BACKGROUND
                                                         BLACK OBJECT
                                                                    78-*::
The subscripts p and b indicate plume and background parameters, re-
spectively. The light scattering coefficient in the plume, b , includes
the background contribution.

      The  visual range, V, is defined as "the distance,  under daylight
conditions, at which the apparent contrast between a specified type of
target and its background becomes  just equal to the threshold contrast
of an observer" (Huschke, 1959).  For the following analysis, the target
and background are a black object viewed against the horizon sky, and
the threshold contrast of a typical observer,  0. 02,  is used (Middleton,
1952).  In  this case, the visual range is given by

             -In 0. 02    _  3.9
                °ext        ""fext  »

where bext is the average extinction coefficient along the sight path.  This
equation assumes uniform illumination along the sight path, and neglects
multiple scattering.

      The  following calculations assume that the contribution of absorp-
tion to extinction is small, so that the equation for visual range can be
rewritten  as  V bscat =  3.9.  For the case depicted above of a plume im-
bedded in  background air, the corresponding equation is
Xpbp + Xbbb  = 3.9, where bb and bp are  the average scattering coef-
ficients in the background and plume, respectively,  and Xp is the mea-
sured plume width.  The quantity  Xpbp, termed the optical depth of the
plume (Tp),  is the  integral of scattering across the plume, i.e.,
                                  42

-------
       Tp   =  /bp(x)dx.

 The visual range looking through the plume was calculated using the equation
       V  =  XT
                                 3.9 - T
*b   =
 The values for Xp, bp, and b^ were determined from measurements  at the
 the lowest altitude at which plume traverses were made.  The plume  opac-
 ity is defined in terms of the plume optical depth as   1  - exp(-T).

       The visual range and plume optics calculations are summarized
 in Table 11 and Figure 13.  The visual range through the copper smelter
 plume is reduced at 8 km downwind of the  stack from 135 km in the back-
 ground to 1 3 km with the plume present (90 percent reduction).  The visi-
 bility reduction decreases  further downwind of the  plant.  At 60 km down-
 wind, the visual range is 82 km with the plume  present  (39 percent visi-
 bility reduction relative to the background).  The visibility reduction
 with  the plume present is nearly equal  at 60 and 127 km downwind of the
 smelter.  The power plant plume exhibits  quite different visual behavior.
                  TABLE 11.   PLUME IMPACT DATA
Visual Range
Downwind


San Manuel
Smelter


Mohave
Power Plant
Distance
(km)
8
32a
60
127
32
60
Plume
Width
(km)
16
20
25
54
8
27
Plume
Optical
Depth
4.9
2.8
2.2
3.1
0.66
1.92

Plume
Opacity
0.99
0.94
0.89
0.95
0.48
0.85
With
Plume
(km)
12
54
82
78
98
82

Back-
ground
135
135
135
135
110
110
Visual
Range
Reduc-
tion due
to Plume
(%)
90
60
39
42
11
25
aPlume width at 32 km downwind interpolated from measurements at
 8 and 60 km downwind.
                                 43

-------
DISTANCE
DOWNWIND, km
8

32
60
127
BACKGROUND
32
60
BACKGROUND
PLUME IMPACT
SMELTER



;- :

^^^^§i^^^^:iSs^^^




POWER
X"x'x"X'X'X'x'"X'X-X*X*l'X*X-X'X'x"X***i










*X* X vX'XvX*X*X • X v.%

;!^^^^^^^^^^^^
PLANT







•X'X'X- *X*X*X*X*X *; ;X* •">



                               50                100
                                  VISUAL RANGE, km
150
                                                                  76-;::
Figure  13.  Visual range calculations for the smelter and
            power plant at several distances downwind
            compared to the background visual range.
                         44

-------
The reduction of the background visual range is less at 32 km than at 60
km downwind (11 percent versus  25 percent,  respectively).   However,
the power plant plume was measured under meteorological conditions
which varied during the day,  resulting in erratic  plume behavior.  The
difference in the reduction of the background visual range between these
plumes may be the  result of greater emissions released from the smel-
ter stack and of greater dispersion of the Mohave plume on October  8.

Elemental Composition in Plumes

       The elemental composition of the plume excess aerosol,  i.e., the
plume aerosol  with the background subtracted, was determined from the
analysis of samples collected in orbital plume flights.  The background
concentrations were determined in  orbital  flights upwind of the plume.
The aerosol elemental composition for the San Manuel smelter plume  at
60 km downwind and the Mohave power plant  at 32 km downwind are  given
in Table 12.


TABLE 12.   PLUME  EXCESS* AEROSOL ELEMENTAL CONCENTRATION
                       San Manuel Smelter
                    60 km downwind (10/4/77)
                       Fine     Coarse
                     Particles*5 Particlesc
           Element3
  Mohave Power Plant
32 km downwind (10/8/77)
   Fine      Coarse
  Particles   Particles
             (mg/m3)
Al
Si
S
K
Ca
Ti
Mn
Cr
Fe
Cu
Ni
Se
Pb
<0. 01
1.08
2.29
0.43
-
0.02
_
0. 20
0. 02
.
_
0.26
-
1.91
6.12
_
0.60
1.58
0.03
0.02
0.06
1.02
0.06
-
-
-
0.05 1.46
0.34 3.67
0.30
0.25
1.16
0.04
.
0.05
0.55
.
0.20 0.01
-
0.03
           aMolybdenum was detected in concentrations above background but at
            the minimum detectable limit in the Mohave power plant plume.
           "Fine particle (Dp fil^m) concentrations were determined from the
            sum of impactor stage 4 and the final filter.  Sulfur values were de-
            termined from TWOMASS samples.
           cCoarse particle (EL zl/im) concentrations were determined from the
            sum of impactor stages 1, 2, and 3.

           * -  Plume excess values were determined by subtracting background
               concentrations from plume concentrations of each element.
                                    45

-------
      The largest plume excess elemental constituents of the fine particle
aerosol in both plumes are sulfur and silicon.  Silicon,  aluminum, cal-
cium,  and iron are  major constituents of the plume excess coarse parti-
cles.   These data indicate that coal-fired power plants and copper smel-
ters in the Southwest may be major sources of the fine particle silicon as
as well as sulfur, the two major consitutents of the Southwest visibility-
reducing aerosol.

      Plume excess  aerosol enrichment factors for the San Manuel smel-
ter and Mohave power  plant are given in Table 13.  Coarse particle Si,
K,  Ca, Ti, and Fe in both plumes and coarse particle Mn in the smelter
plume are present in crustal abundances relative  to Al.  Thus,  the crus-
tal abundances of these elements in Southwest background coarse particles
may be an indication that the source of these particles is  crustal weather-
ing or flyash.  Plume excess coarse particle Cr,  Ni, Cu, and  Pb are all
enriched relative to  crustal abundances.  All plume excess fine particle
species are enriched relative to the crust except for fine  particle Si  in the
Mohave plume.

      The  size distribution of plume excess elemental constituents shows
many similarities with the regional data.  Aluminum, calcium, and iron all
exhibit the highest concentrations for particles with diameters >4pim.   Sili-
con has a size distribution similar to aluminum down to 1pm but increases
below that diameter. Sulfur is  mainly concentrated in fine particles.


    TABLE  13.   PLUME EXCESS AEROSOL ENRICHMENT FACTORS

                        San Manuel Smelter         Mohave Power Plant
                     60 km downwind (10/4/77)    32km downwind (10/8/77)
            Element     Fine      Coarse         Fine       Coarse
                     Particles    Particles      Particles     Particles
Elements in crustal abundances in coarse particles, enriched in fine particles
Al
Si
K
Ca
Ti
Mn
Fe
1.0
32
IxlO2
-
37
-
3
1.0
0.9
1.0
2
0.3
0.9
0.9
1.0
2
.
.
-
.
-
1.0
0.7
0.5
2
0.5
-
0.6
                      Elements enriched in both size fractions
              S        7xl04      -             2xl03
              Cr       2x1 0<      26             -           28
              Ni       -          -             4x1 O3      7
              Cu       -         46
              Se       4xl07
              Pb       -          -             -          128
                                  46

-------
      The total aerosol mass was not measured but can be estimated from
the elemental concentrations and mean aerosol volume as described pre-
viously.  A comparison of these two approaches  shows that in the San
Manuel smelter on October 4,  1977, at 62 km downwind, 100 percent of
the plume excess fine particle  mass determined  from plume excess  aer-
osol volume (assuming a mass weighted density of 2 g/cm2) can be ac-
counted for from the  measured constituents.  The coarse particle volume
was not measured in  this  plume.  Approximately 73 percent of the plume
excess fine particle mass  is siilfate (expressed as ammonium sulfate),
and 17 percent is silicon (expressed as SiO2).   In the Mohave power plant
plume on October 8,  1977,  only about 50 percent of the fine particle mass
determined from aerosol  volume can be accounted for  by the  measured
constituents.  Of this 50 percent at 32 km downwind, plume excess  sulfate
[expressed as (NH4)2SO4] accounts for 52 percent of the mass inferred from
measured aerosol composition and plume excess fine particle silicon (ex-
pressed  as SiO2) accounts  for 31  percent. A summary of accumulation
mode aerosol integral size parameters is given in  Table 14.
       TABLE 14.  PLUME ACCUMULATION MODE INTEGRAL
                   SIZE PARAMETERS
    Downwind   Accumulation Mode   Accumulation Mode
    Distance      Volume, Va          Mean Size, Da
       (km) _ (^tm3/ cm3) _ (ptm)
                      San Manuel Smelter (10/4/77)
       60              9.9                   0.18            2.0
      127             12.9                   0.14            1.9
    Background        3.5                   0.24            2.2

                      Mohave Power Plant (10/8/77)
       32              5.0                   0.23            2.1
       62              4.7                   0.19            2.1
    Background        2. 8                   0. 20            2. 1
Sulfur Transformation

      The conversion of SO2 to particulate sulfate in plumes can be esti-
mated from the measurements of gaseous SO2, particulate  sulfate, light
scattering coefficient, and mean accumulation mode volume.  A number
of methods to determine SO2 conversion rates in power plant plumes have
                                47

-------
been used in the past (Wilson,  1978).  The approach used here compares
the fraction of particulate  sulfur of the total sulfur, SD/Sj,  to plume age,
estimated from analysis of winds aloft.  Plume excess sulfur values with
background subtracted were used in this calculation.  The SO2 conversion
rate measured in the San Manuel smelter  plume on October 4,  1977,  from
0900 to 1230 MST was measured to be  0. 7±0. 2 percent/hour between 60
127 km downwind from the plant (Figure 14). This calculation assumes no
SO2 deposition between measurements.  However, even as much as 20 per-
cent SO2 deposition would not lower the conversion rate below the  stated
uncertainty.   The SO2 conversion rate was not calculated for the Mohave
power plant plume because of poor plume  resolution and erratic wind be
havior encountered on October 8, 1977.

      This single measurement of the  SO2—> SO4=  conversion rate should
not be considered typical without many additional measurements under  a
variety of conditions.  It is interesting to  note,  however, that this value
is within the range  of conversion rates measured in plants in the midwest
(Husar et al. , 1978).

      A number  of other parameters can be  used as qualitative indicators
of SO2 conversion,  such as plume excess bsp/Sg, SD/Sg,  and mean accu-
mulation mode volume (Va)/Sg ratios and accumulation mode aerosol  size,
Dy.  These indicators are shown graphically in Figure 15 as a function  of
distance downwind from the plant for the San Manuel smelter and Mohave
power plant plumes on October  4, 1977, and October 8, 1977, respectively.
The three plume excess  parameter ratios all increase downwind in both
plumes,  indicating that sulfur aerosol is being formed.  The mean accu-
mulation mode size decreases  downwind from both plants which also  in-
dicates fresh  aerosol is  being formed.

      One method of assessing the  effect of  sulfur transformation  on
visib;lity is to determine the improvement of visual range if no  SO2 was
converted to sulfate.   This can be  estimated for the San Manuel smelter
at 20 and 32 km downwind  on October 2, 1977,  assuming that the bsp/SO2
ratio remains constant beyond 8 km downwind.   The visual range with the
plume present is then calculated using bscat  and assuming no conversion,
as described previously.  This  approach will slightly overestimate the
effect of aerosol dilution (and therefore visual range) due to loss of SO2
from the plume because of deposition and  transformation. The  results of
this calculation are shown in Figure 16.  At 32 km downwind from the
smelter on October 2 the visual range would increase by nearly a  factor
of two if no SO2 were converted to  sulfate.  At 60 km downwind the visual
range would be increased by 36 percent if there were no conversion.
Thus sulfur transformation has a large effect on visibility impairment
within at least 60 km of  the shelter.
                                 48

-------
        15
         10
     o:
     •a:
     Q-
     O

     z
     o

     t—
     (_>

     s
                SM MANUEL  SMELTER

                OCTOBER 4,  1977
     0.7 %/hr

          \x
                                               X
4.1
        x
                             -©-
                             2.5
                              I
                    I
                              5                  10


                                 PLUME AGE, hours
                                      15



                                     76-424/1
Figure 14.   Determination of SO2  conversion rate for the San Manuel

             smelter on October 4,  1977, from the fraction of particu-

             late sulfur vs. plume  age.
                                    49

-------
                    SAN MANUEL SMELTER
                         10/4/77
MOHAVE POWER PLANT
     10/8/77
           0.2
           0.1
       o -o
           0.5
           0.25
       £
           0.2
           0.1
                                                I
                      60       127              32

                              DISTANCE FROM PLANT, km
           62
                                                                 0.2
                     0.1
                                                                 20
                                                                 10
                     0.5
                                                                 0.2
                     0.1
                                                               78-423
Figure  15.  Qualitative indicators of SO2 conversion in plumes
             plotted vs. distance from plant.
                                   50

-------
                 SAN MANUEL SMELTER
                      10/2/77
                                            PLUME (Measured)
                                            PLUME (No Conversion,
                                            Scaled to S02)


                                            BACKGROUND ONLY
                                            (Measured)
   150

                                                  60
                           DISTANCE DOWNWIND,  km
Figure 16.  Visual range through San Manuel smelter plume
             with and without SO^  conversion and through back
             ground only for two downwind distances.
                              51

-------
Light Scattering Budget in Plumes

      The  detailed contributions to visibility impairment as a function of
particle size due to a smelter and power plant plume can be determined
using Mie  calculations and measured plume excess aerosol size distribu-
tions in an analogous  manner to the treatment of regional data.  However,
there are  several problems with the plume data.   During the San Manuel
smelter plume flights only the fine particle size distribution was measured
(Dp £l pm).  During the Mohave power plant flights on October  8,  1977,  the
entire size distribution was measured,  but wind-blown dust was occasion-
ally mixing into the plume.   Therefore, the plume visibility budgets must
be considered as much  more  tentative than the  regional visibility budget.

      In the San Manuel smelter plume at 62 km downwind on October 4,
1977, the plume excess bsp value  synthesized from only the measured fine
particle size distribution (D_ sl/jm),  using Mie scattering functions, was
Ibsp  =  56 x lO"6^!'1 .   This  is 59 percent of measured value:
bsp (measured) =  95 x lO^rrf1 .  The detailed fine particle visibility bud-
get given in  Table  15  was determined from chemical composition and size
distribution, using Mie scattering calculations.  Sulfate [assumed to be in
the chemical form of  (NH4)2SO4] accounted for 43 percent of the total bsp in
the smelter plume of  10/4/77; SiC2 accounted for  10 percent.
         TABLE 15.    PLUME EXCESS VISIBILITY BUDGET
Component 1
Dp b.
(^m) (x 10'" m'
SAN MANUEL SMELTER (62 km downwind)
(NH4)2S04
SiO2
Other Compounds
Coarse Particles

0. 1 to 1.0
0. 1 to 1.0
0.1 to 1.0
1.0 to 20.0

MOHAVE POWER PLANT (32 km downwind)
(NH4)2SO4
SiO,
Other Compounds
Coarse Particles1

0.1 to 1.0
0.1 to 1.0
0.1 to 1.0
1.0 to 20.0

Contribution
to total bsp
-1 ) (percent)
10/4/77
41
9.5
5.5
39
95 xlO-fcm-'
10/8/77
3
2
1
21
27 x!0-6m-'

43
10
6
41
100

11
7
4
78
100
           1 Assumes that all fine particle sulfate exists as ammonium sulfate.
           b Assumes that all fine particle silicon exists as SiO2.
           'Determined from b.cat(total) - b.cat(fine particles).
                                   52

-------
      The Mohave power plant plume excess bsp calculation on October 8,
1977, at 32 km downwind was   EbSD(calc)  =  27xlO'6m'1 ,  while the
                             /   i    "
measured value was 18x10" m  .  This difference may be due to the loss
of large particles in the nephelometer sampling line.

      Calculations of the light scattering budget for the Mohave power plant
are complicated by the fact that aerosol composition calculations account for
only about fifty  percent of the fine  particle aerosol mass as estimated
from aerosol volume measurements.  Therefore, the estimated contribu-
tions of individual chemical special to bsp may be somewhat high.  The
visibility budget for the Mohave power plant plume at 32 km downwind on
October 8,  1977,  is quite different from the smelter results.   Coarse
particles account  for 78 percent of the calculated bsp.  This high value
is at least partially due to wind-blown dust mixing into the power plant
plume at the time of the measurement.  Fine particle  ammonium sulfate
accounted for 11 percent of bsp (bsp [ (NH4)2SO4 ]  -  3 x  lO^rrr1). Fine
particle silicon accounted for 7 percent of bsp in the Mohave plant.
                                  53

-------
                         REFERENCES
Beil,  B.  Personal communication to E. S.  Macias. 1978.

Blumenthal, D. L. ,  J. A. Ogren,  and J. A. Anderson.  Airborne
Sampling System for Plume Monitoring.  Atmos.  Environ., 1 2,
613 -  620.   1978.

Cahill,  T. A.  In New Uses  for Ion Accelerators,  J. Ziegler, ed.
pp. 1  - 72. 1975.

Cantrell,  B. K. ,  and K.  T.  Whitby.  Aerosol Size Distributions and
Aerosol Volume Formation  for a Coal-Fired Power Plant Plume.
Atmos.  Environ. , 12,  323  - 333. 1978.

Charlson, R. , A. Waggoner,  and J. Thielke.  Visibility  Protec-
tion for Class I Areas:  The Theoretical Basis. (To be published).

Ensor,  D. S. , R. J. Charlson, N. C. Ahlquist,  K. T. Whitby,
R. B.  Husar, and B. Y.  H.  Liu.  Multiwavelength Nephelometer
Measurements in Los Angeles Smog Aerosol. I. Comparison of
Calculated and Measured Light Scattering.  J.  Colloid and Inter-
face Sci. , 39, 242 - 250.  1972.

Friedlander, S.  K. Smoke, Dust and Haze.  John Wiley and Sons,
N.Y., Chapter 11.  1977.   317pp.

Grosjean, D.  Solvent Extraction and Organic Carbon Determina-
tion in Atmospheric Particulate Matter: The Organic Extraction -
Organic Carbon  Analyzer (OE - OCA) Technique.  Anal. Chem. , 47,
797 -  805.  1975.

Hering, S. ,  R. C. Flagan,  and S. K. Friedlander. Design  and
Evaluation of New Low-Pressure Impactor. I.   Environ.  Sci. and
Technol. , J_2, 667 - 673.  1978.

Huntzicker,  J. J.,  and R. L.  Johnson.  Analysis of Volatizable and
Elemental Carbon in Ambient Aerosols.  Proceedings, Conference
on Carbonaceous Particles  in the Atmosphere, Berkeley, California.
March 20  - 22,  1978.
                                 54

-------
Husar,  J. D. , R. B. Husar,  and P. K.  Stubits.  Determination of
Submicrogram Amounts of Atmospheric Particulate Sulfur.  Anal.
Chem. , £7,  2062 - 2064.  1975.

Husar,  R. B., E. S. Macias, and W. P. Dannevick. Measurement
of Dispersion with a Fast Response Aerosol Detector.  AMS Third
Symposium on Atmospheric Turbidity,  Diffusion,  and Air Quality,
Raleigh, North Carolina.   October 19-22, 1976.

Husar,  R. B., D. E.  Patterson, J. D. Husar, N.  V.  Gillani,  and
W. E. Wilson, Jr.  Sulfur Budget of a Power Plant Plume. Atmos.
Environ., 12., 549  - 568.  1978.

Huschke, R. E. ,  ed. Glossary of Meteorology.  American Meteor-
ological Society,  Boston,  Massachusetts.  1959.

Larson, N. I., and P. E.  Billings.  Second Annual Report on Ari-
zona Copper Smelter  Air Pollution Control Technology.  Arizona
Department of Health Services,  Phoenix, Arizona.  1978.

Latimer, D. A. , R. W. Bergstron, S. R. Hayes, M. -K. Liu, J. H.
Seinfeld, G. Z. Whitten,  M. A. Wojcik, and M. J. Hillyer.   The
Development of Mathematical Models for the Prediction of Anthro-
pogenic Visibility Impairment.  SAI Report EF 78-63.  1978.

Lewis,  C. W. , and E. S. Macias.  Composition of Size-Fraction-
ated Aerosol in Charleston,  West Virginia.  (To be published)

Lin, C. , M. Baker, and R. J. Charlson.  Absorption Coefficient
of Atmospheric Aerosol:  A  Method for Measurement.  Applied
Optics,  j_2,  1356.  1973.

Lyon, W. S.  Trace  Element Measurements at  the Coal-Fired
Steam Plant.  CRC Press, Cleveland, Ohio.  1977.  136  pp.

Macias, E. S. , and R. B. Husar.  Atmospheric Particulate Mass
Measurement with Beta Attenuation Mass Monitor.  Environ Sci.
and Technol., 10,  904 - 907.  1976.

Macias, E. S. , C. D.  Radcliffe, C. W.  Lewis, and C. R. Sawicki.
Proton-Induced y-Ray Analysis of Atmospheric Aerosols for  Car-
bon, Nitrogen, and Sulfur Composition.  Anal. Chem. , 50, 1120 -
1124.  1978a.
                                 55

-------
Macias,  E. S., R. Delumyea, L.-C. Chu,  H. R. Appleman, C. D.
Radcliffe, and L. Staley.  The Determination, Speciation and Be-
haviour of Particulate Carbon.  Proceedings, Conference on  Car-
bonaceous Particles in the Atmosphere, Berkeley, California.
March 20 - 22.  1978b.

Marians, M. , and J. Trijonis.  Empirical Studies of the Relation-
ship between Emissions  and Visiblity in the Southwest.  Preliminary
draft of report to USEPA.

Mason, B.  Principles of Geochemistry.   J. Wiley and Sons,  Inc.,
New York.  1966.   329 pp.

Middleton, W. E. K.  Vision Through the Atmosphere.  University
of Toronto Press, Toronto, Canada.   1952.

Mie, G.   Beitrage zur Optik truber Medien,  speziell Kolloidaler
Metallosungen.  Ann.  Physik, 25,  377 - 445.  1908.

Moyers, J. L. ,  L.  E.  Ranweiler, S. B. Hops, and N. E.  Korte.
Evaluation of Particle Trace Species in Southwest Desert Atmo-
sphere.  Env. Sci.  and Technol. , 11,  798  -  795.  1977.

Mulik, J. , R. Puckett, D. Williams,  and  E.  Sawicki.  Ion Chro-
matographic  Analysis  of Sulfate and Nitrate in Ambient Aerosols.
Anal. Letters, 9.,  653 - 663.  1976.

Nolan, J. L.  Measurement of Light-Absorbing Aerosols from Com-
bustion Sources.  Proceedings,  Conference on Carbonaceous  Parti-
cles in the Atmosphere,  Berkeley, California. March 20-22, 1978.

Pierson,  W.  R.  Particulate Organic Matter and Total Carbon from
Vehicles on the Road.   Proceedings,  Conference on Carbonaceous
Particles in the Atmosphere,  Berkeley, California.  March 20 - 22,
1978.

Ragaini, R. C. , and J. M. Cndov.  Trace-Element Emissions from
Western U.S. Coal-Fired Power Plants.   J. Radioanalytical Chem. ,
.37,  679  - 691.  1977.

Roberts,  F.M. , J. L.  Gordon, D. L.  Haase,  R.E. Kary, and J. R.
Weiss.   Visibility Measurements in the Painted Desert.  Proceed-
ings, 68th Annual Meeting of the Air Pollution Control Association.
June 15 - 20, 1975. Paper No.  75-26.1
                                 56

-------
Roberts, P. T. , and S. K. Friedlander.  Analysis of Sulfur in De-
posited Aerosol Particles by Vaporization and Flame Photometric
Detection.  Atmos. Environ. , 10, 403  - 408.  1976.

Rosen, H. ,  A. D.  A.  Hansen, L. Gundel, and T. Novakov.  Identi-
fication of the Graphitic Carbon Component of Source and Ambient
Particulates By Raman Spectroscopy and an Optical Attenuation
Technique.  Proceedings,  Conference on Carbonaceous Particles
in the Atmosphere, Berkeley,  California.  March 20 - 22, 1978.

SAI.  Preliminary Suggestions for Validating SAI's Visibility Models,
Systems Applications, Incorporated, Report No. ES 78-92,  August.
1978.

Sverdrup, G. M.  Parametric Measurement of Submicron Atmos-
pheric Aerosol Size Distributions.  Ph. D.  Thesis, Particle Tech-
nology Laboratory, University of Minnesota, Minneapolis  55455.
1977.

Trijonis, J. ,  and  K.  Yuan.  Visibility  in the Southwest:  An Explor-
ation of the Historical Data Base.  EPA - 600/3-78-039.  1978.

Waggoner, A. P.  Personal communication to E. S. Macias. 1978.

Weiss,  R. E..  A.  P.  Waggoner,  R.  J., Charlson,  D. L.  Thorsell,
J. S. Hall, and L. A.  Riley.   Studies of the  Optical, Physical,  and
Chemical Properties  of Light-Absorbing  Aerosol.   Proceedings,
Conference on Carbonaceous Particles  in the Atmosphere, Berke-
ley,  California. March 20 - 22, 1978.

Whitby,  K. T.   The Physical Characterization of Sulfur  Aerosols.
Atmos. Environ. , 12, 135 - 159.  1978.

Wilson,  W. E.   Sulfates in the Atmosphere: A Progress Report on
Project MISTT. Atmos. Environ. ,  12, 537 - 547. 1978.
                                57

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing]
1. REPORT NO.

  EPA-600/7-79-243
4. TITLE AND SUBTITLE
 CHARACTERIZATION OF VISIBILITY-REDUCING AEROSOLS IN
 THE  SOUTHWEST
 Project  VISTTA Progress Report No.  1
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
 E.S. Macias,  D.L.  Blumenthal, J.A.  Anderson and
 B.K. Cantrell
             8. PERFORMING ORGANIZATION REPORT NO.


              MRI 78IR-1585
                                                           3. RECIPIENT'S ACCESSION NO.
             5. REPORT DATE
               November  1979
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Meteorology  Research, Inc.
 Box  637,  464 West Woodbury Rd.
 Altadena, CA  91001
             1O. PROGRAM ELEMENT NO.

              1NE625  EA-13   (FY-77)
             11. CONTRACT/GRANT NO.
              68-02-2713
12. SPONSORING AGENCY NAME AND ADDRESS
 Environmental  Sciences Research Laboratory - RTP, NC
 Office of Research and Development
 U.S.  Environmental Protection Agency
 Research  Triangle Park, N.C.  27711
             13. TYPE OF REPORT AND PERIOD COVERED
              Final  10/1/77 - 10/10/77
             14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT

       The  atmospheric visibility-reducing  aerosol in the Southwest has been ex-
 perimentally characterized with  respect to particle size, composition, and contri-
 bution  to light scattering.  Measurements were taken within  the  mixing layer using
 the MRI  instrumented Beechcraft  Queen  Air aircraft.  The aircraft was equipped to
 measure  and  record on magnetic tape  the light-scattering coefficient, Aitken nuclei
 count,  size  distribution, ozone,  sulfur dioxide, nitrogen oxides, temperature, dew
 point,  turbulence, pressure (altitude), and navigational parameters.   Multistage
 impactor  and size-fractionated filter  samples were also collected in  order to de-
 termine  aerosol elemental composition  as  a function of size.   Visual  range estimates
 were  obtained by viewing distant landmarks and verified by optical  photography.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                           c. COS AT I Field/Group
  *Air  pollution
  *Aerosols
  *Sulfur oxides
  Visibility
  *Light scattering
  *Plumes
  *Measurement aircraft
 Project VISTTA
 Southwest
13B
07D
07B
18. DISTRIBUTION STATEMENT
•  RELEASE TO PUBLIC
                                               19 SECURITY CLASS (This Report)

                                                IINCLASSIEIED	
                           21  NO. OF PAGES
                               68
20 SECURITY CLASS (This page!

  UNCLASSIFIED
                                                                          22. PRICE
       2220-1 (Rev. 4-77)    PREVIOUS E.CIT'CN SOBSOL.FTE
                                             58

-------
CD X
Z3 —
Q) O
=•*

??
  c/>
75
? *
Is
CD
m
or
d
O
O
 I
M
-P-
U)
                                                                                    2 5
                                                                                    Q; V)


                                                                                    T>

                                                                                    O
                                                                                   nom
                                                                                   = CO 3
                                                                                   23|.
                                                                                   3 » o
                                                                                   33
                                                                                   £   3
                                                                                   -   CD
                                                                                   o   2
                                                                                   I   a?


                                                                                   8   I
                                                                                   s   *
                                                                                   (X   0)
                                                                                      3

                                                                                      O


                                                                                      0)

                                                                                      o
                                                                                      3
                                                                            m> 73 m TI -D
                                                                            >° 1 i»s
                                                                            all I J|

-------