v>EPA
           United States    Office of Environmental Engineering
           Environmental Protection and Technology
           Agency      Washington DC 20460
                      EPA 600/7 80 126
                      June 1980
           Research and Development
Transport and
Transformation of
Sulfur Oxides
Through the
Tennessee Valley
Region

Interagency
Energy/Environment
R&D Program
Report

-------
                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-600/7-80-126

                                               JUNE 1980
        TRANSPORT AND TRANSFORMATION
        OF SULFUR OXIDES THROUGH THE
           TENNESSEE VALLEY REGION
                     by

Timothy L. Crawford and Lawrence M.  Reisinger
        Office of Natural Resources
        Tennessee Valley Authority
         Muscle Shoals, AL  35660
 Interagency Agreement No.  EPA-IAG-D9-E721
            Project No.  81-BDL
        Program Element No. 1NE-832
              Project Officer

                C.  W. Hall
   U.S.  Environmental Protection Agency
               401  M Street
          Washington, DC  20460
               Prepared for

 OFFICE OF ENERGY,  MINERALS,  AND INDUSTRY
    OFFICE OF RESEARCH AND DEVELOPMENT
   U.S. ENVIRONMENTAL PROTECTION AGENCY
           WASHINGTON, DC  20460

-------
                               DISCLAIMER
     This report was prepared by the Tennessee Valley Authority and has
been reviewed by the Office of Energy, Minerals, and Industry, U.S.
Environmental Protection Agency, and approved for publication.  Approval
does not signify that the contents necessarily reflect the views and
policies of the Tennessee Valley Authority or the U.S. Environmental
Protection Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.

TVA is an equal opportunity employer, and is committed to ensuring that
the benefits of programs receiving TVA financial assistance are available
to all eligible persons regardless of race, color, national origin, handicap,
or age.
                                   11

-------
                               ABSTRACT
     This report is directed to scientists interested in the long-range
atmospheric transport and transformation of sulfur compounds.

     Statistical and climatological analyses of historical data and the
results of two long-range transport studies are presented.  The two long-
range atmospheric transport field studies were conducted over a 300-km
area of the southern United States centered on the Tennessee Valley region.
The first study was conducted during the spring of 1976, and the second
was conducted during the summer of 1977.  The field study region contains
seven large coal-fired power plants and one large city.

     Results indicate that the predominant flow and mass transport
direction is from the southwest to the northeast.  Also, aerometric
measurements obtained by aircraft and ground sampling compared favorably
with results obtained with an analytical transport-transformation model
developed for this study.  Results indicate that, during prevailing
southwesterly airflow, large gaseous sulfur influxes are present.  These
influxes, which are of the same order of magnitude as the Tennessee Valley
regional emission fluxes, can only partly be explained by upwind
anthropogenic sources.  Natural source emissions are hypothesized to
account for about half of this sulfur influx.

     This report was submitted by the Tennessee Valley Authority, Office
of Natural Resources, in partial fulfillment of Energy Accomplishment
Plan 81 BDL under terras of Interagency Agreement EPA-IAG-D9-E721 with the
Environmental Protection Agency.  Work was completed as of June 1979.
                                     111

-------
                            EXECUTIVE SUMMARY
     The objective of the Tennessee Regional Atmospheric Transport Study
 (TREATS) is to develop an understanding of the characteristics and mecha-
 nisms affecting regional pollutant levels and interregional transport of
 primary and secondary sulfur pollutants.  This report not only addresses
 the measurement and modeling of the transport and transformation of sulfur
 species, but also looks at climatological variations of aerometric and
 meteorological parameters and emission rates as they impact the Tennessee
 Valley region.  Other pollutant species are dealt with primarily as they
 relate to the formation of fine particulate sulfates.*  Principal ingre-
 dients of this analysis are the results from two long-range transport field
 studies conducted during the spring of 1976 and the summer of 1977.  These
 studies are among the first ever conducted in the United States over
 distance scales of hundreds of kilometers.

     During these studies, aerometric and meteorological measurements
 defined airmass pollution levels and interregional pollutant mass trans-
 port as airmasses entered, passed through, and left the Valley.  These
 unique measurements have led to several significant findings.

     In presenting these findings, two important terms, concentration and
 flux, are used to describe pollutant levels within airmasses.  Concentra-
 tion is simply the mass or amount of pollutant per unit volume of air,
 typically reported as micrograms per cubic meter (|Jg m~3), whereas flux is
 a  less frequently used term that describes the rate of pollutant mass
 transport through a horizontal or vertical area and is reported in micro-
 grams per square meter per second (ug m~2 s"1).  As typically expressed
 in this report, flux is obtained by multiplying the concentration by the
 wind speed.  This parameter is important in describing the movement of
 pollutants.

     Many different sampling and analytical techniques were used to deter-
 mine relationships between and among the various parameters measured.  One
 of the more significant findings from these analyses describes the hori-
 zontal and vertical diurnal variations of various pollutants.  In particular,
 multiple aircraft traverses at various altitudes and at widely separate
 locations within the Valley indicate that significant pollutant variations
 often occur in the horizontal and vertical during the night and early
 morning.  However, by midday, the gradients are significantly reduced,
 except near the ground, where sulfate data obtained from high-volume
 samplers indicate that ambient air concentrations average twice the air-
 mass concentrations as measured by aircraft flights near ground level.
 Additional research is needed to determine whether this is a sampling
 artifact or a real phenomenon.
'"Prominent among the "sulfates" are sulfuric acid, ammonium bisulfate,
and ammonium sulfate.  Within this report, the terms, sulfate, particulate
sulfate and S04 are synonymous and are defined as the traditional water
soluble fraction of the total filterable particulate analyzed and
expressed as micrograms of sulfate per cubic meter of sampled air.
                                   IV

-------
     Another interesting analysis involves aircraft sampling of the same
airmass as it enters the Tennessee Valley region and again as it leaves
(known as Lagrangian sampling).  Analyses of measurements made on five
such days compared very well with an analytical model developed to simu-
late the transport and transformation processes.  By using the model
simulations on six non-Lagrangian days together with the five measure-
ment days, some unexpected and interesting results were obtained.  These
results need additional confirmation due to limits of the present data
set.  One result shows that for the days with both inflow and outflow
measurements, the change in flux as the airmass passed over the region
was negligible.

     When flow was from the south, sulfate flux, from inflow to outflow,
increased; however, sulfate concentrations were low at both boundaries.
Again, model estimates indicate that the region was a minor contributor
to this increase (only 12 percent); the remaining 88 percent increase in
sulfates resulted from conversion of gaseous sulfur compounds already in
the air before the airmass entered the region.  An attempt to identify
the significant source region(s) that might be contributing to this large
inflow flux resulted in the conclusion that upwind anthropogenic sources
account for only about half of the gaseous sulfur flux, whereas biogenic
sources (i.e., wetland areas in and around the Gulf Coast States) could,
in theory, account for the other half.  However, available data are insuf-
ficient to accurately quantify this biogenic source hypothesis.

     One reason that our findings are significant is that this southwesterly
flow direction is the principal transport avenue over not only the Tennessee
Valley region but the entire United States east of the Mississippi River.
Also, this flow direction is frequently associated with summertime high
air pollution episodes over the eastern United States.

     This study also indicates that, when wind speeds are light, TVA is
a principal source of pollution in the Tennessee Valley region.

-------
                                CONTENTS
Abstract  	
Executive Summary 	     v
List of Figures	    ix
List of Tables	    xi
Acknowledgments 	   xii
Nomenclature  	  xiii

1.   Introduction 	    1
          Background  	    1
          Scope of Research	    2
2.   Conclusions and Recommendations	    4
          Conclusions 	    4
          Recommendations 	    5
3.   Materials and Methods  	    7
          Characterization of the Region  	    7
               Physical characterization  	    7
               Emissions characterization 	    9
               Meteorological characterization  	   12
                    Synoptic weather patterns 	   12
                    Trajectory data	13
          Sampling platforms  	   15
          Meteorological measurements and support 	   23
          Sampling procedures 	   23
               1976 study	25
               1977 study	25
          Analytical methods  	   26
               Continuous gas analyzers 	   26
               Low-volume filters 	   27
               High-volume filters  	   27
          Transformation-transport model  	   29
               Model development	29
                    Analytical model  	   29
                    Simple box model	32
                    Eulerian model  	   33
               Model analysis	34
4.   Results and Discussion	38
          Sulfate Concentration and Flux Climatology  	   38
               Seasonal concentration variations  	   38
               Regional sulfate flux rose	42
          Flux calculations	48
               Estimating concentration and flux  	   48
                    Estimating U, H , and+H~	51
                    Estimating SO  and N04
                      concentrations  	   51
                    Estimating N03 and 03
                      concentrations  	   51
                    Flux measurement criteria 	   54
                    Tabular summary of flux calculations  ....   58
                                   VI

-------
          Eulerian space and time variations	58
               Eulerian space variations  	   58
                    Horizontal variations 	   61
                    Vertical variations 	   66
               Eulerian time variations 	   72
          Lagrangian flux analysis  	   72
               Field results	73
               Model results	77
               Upwind sources 	   81

References	82

Appendixes

     A.   Synoptic Weather Typing 	  A-l
     B.   Tabulation of TREATS 1976 and 1977 aircraft field
            study data	B-l
     C.   Altitude correction, Meloy model SH202 and SA285
            sulfur analyzer 	  C-l
     D.   Tabulation of TREATS 1976 and 1977 high-volume
            field study data	D-l
     E.   Inversion heights and average wind velocities
            for Lagrangian and Eulerian days	E-l
     F.  Synoptic meteorological summaries for
            Lagrangian and Eulerian days	F-l
                                vii

-------
                            LIST OF FIGURES
Figure                                                           Page

   1      Physiographic divisions of the TREATS study region. .  .   8
   2      Annual average SQ% emission density map 	  10
   3      Locations and emissions of major point sources  ....  11
   4      Favored 24-h airmass source regions for
            Nashville, Tennessee  	  14
   5      Percent occurrence of airmass movement out of the
            Tennessee Valley region (1972)	16
   6      deHavilland Beaver U6-A airplane  	  17
   7      Airplane instrument layout  	  18
   8      U6-A instrument package 	  19
   9      Bell 47A helicopter	21
  10      Helicopter instrument layout	22
  11      Location and type of meteorological measurements  ...  24
  12      Region used in the development of the long-range
            transport model 	  30
  13      Schematic of the hox model	32
  14      Comparison of Lagrangian and Eulerian boundary
            conditions and solution domains_	33
  15      Predicted variation in SC>2 and 804 concentrations
            as a function of transport time	35
  16      Effect of variation in model parameters on predicted
            response after 24 h of transport	36
  17      Regional air quality trend stations 	  39
  18      Monthly variations of TVA's 802 emissions and regional
            sulfate concentrations, 1974-1977 	  40
  19      Giles County sulfate concentration (|Jg m~3) vs the
            average concentration of Cumberland, Gallatin,
            and Giles	43
  20      Annual frequency distributions of wind speed, wind
            direction, sulfate concentration, and sulfate flux
            for Nashville, Tennessee  	  44
  21      Seasonal frequency distributions of Nashville sulfate
            flux (ng nT2 s"1)	45
  22      Relative frequency distributions  	  46
  23      Schematic of typical flux calculation procedure ....  50
  24      Typical early morning sulfur profiles 	  52
  25      Typical midday sulfur profiles  	  53
  26      Typical early morning and midday normalized ozone
            profiles	55
  27      Typical early-morning nitrate profiles  	  56
  28      Airmass transport distance vs. inflow-to-outflow
            sampling distance 	  57
  29      Temporal variation of sulfate 	  62
  30      Temporal variation of ammonium  	  63
  31      Temporal variation of normalized total sulfur 	  64
  32      Temporal variation of nitrate 	  65
  33      Comparison of ground-level to upper-air sulfate
            concentrations  	  69
                                   vin

-------
Figure                                                           Page

  34      Giles County sulfate concentration vs. Colbert ....  70
  35      Comparison of inflow to outflow sulfate flux 	  74
  36      Comparison of inflow to outflow total sulfur flux  .  .  75
  37      Inflow to outflow concentrations and mole
            ratio changes	76
  38      Observed vs predicted outflow total sulfur (X 102)
            and sulfate concentration 	 78
  39      Relative change in total sulfur flux across the TREATS
            field study region	79
  40      Sulfate flux change across the TREATS field study
            region	80

-------
                            LIST OF TABLES
Table                                                            Page

  1       Airplane Instrument Package, Beaver U6-A  	 15
  2       Helicopter Instrument Package, Bell 47A 	 20
  3       Accuracy of Analytical Methods, 37-mm Millipore
            Filters	28
  4       Percent Variation in Model Response Resulting from
            a ±50 Percent Change in a Single Parameter  	 34
  5       Median Sulfate Flux and Concentration Values
            Regardless of Sector  	 47
  6       Summary of Lagrangian and Eulerian Measurements .... 59
  7       Inflow-Outflow Concentrations (|Jg m 3) by Flow
            Direction	^	61
  8       High-Volume Concentrations (pg m 3) by Flow
            Direction	66
  9       Comparison of Ground-Level and Upper-Air Sulfate
            Concentrations  	 68
 10       Suspended Sulfate Correlation Matrix  	 71
 11       Sulfate Speciation  	 73

-------
                    LIST OF ABBREVIATIONS AND SYMBOLS





B              --width of the region, m



b              --subscript indicator of pollutant concentration above inversion,

                 (Jg m'3



C.             --concentration, pg m~3



H              --height of the capping inversion, m



i              --indicator of S02 (i = 1) or 804 (i = 2)



j              --indicator used with box formulation



k              --first-order transformation rate constant for S02 to 864, s"1



kj             	k, s"1



k2             —3k/2, s"1



K , K , K      --diffusivities along the principal axes, m2 s"1
 x   y   z


Q              --point-source S02 emission rate, |Jg s~l



t              --time, s



t              --age of airmass passing the outflow plane, s
 3


t*             --time at which the relative S04 concentration is maximum, s



U, V, W        --mean wind speed components along the principal axes, m s"1



v.             --deposition velocity, m s "-1



x, y, z        --distances along the principal axes, m





Note:  Subscript or superscript zero (0) indicates an initial value.

-------
                            ACKNOWLEDGMENTS
     This work was primarily funded by the Federal Interagency Energy/
Environment Research and Development Program administered through the
Environmental Protection Agency under contract number EPA-IAG-DG-E721.
The support of this agency is gratefully acknowledged, as is the advice
and guidance of the Project Officer, C.  W. Hall.   However, without
additional funding by the Regional Air Quality Management Program of
the Tennessee Valley Authority and without the use of TVA's extensive
facilities, this work would not have been possible.

     We wish to especially express our appreciation to Dr.  Herbert
Jones, The TVA Project Director and Drs. James F.  Meagher and Leonard
Stockburger of the Air Quality Research Section for their logistic and
design assistance before and during the two field studies and for their
thought-provoking input during data analysis.   Also, we wish to express
our appreciation to William J. Parkhurst, Air Quality Monitoring Section,
for his input to the seasonal concentrations variations subsection; to
Malcolm C. Babb, Applied Research Staff, for his assistance in generating
plots; and to Hollis E. Lindley, Computer Applications Staff, for his
assistance in processing data.  Finally, we wish to express our appreciation
to Elizabeth M. Bailey of the Air Quality Research Section for serving as
instrument operator during the 1977 study.

-------
                              SECTION 1

                            INTRODUCTION
     The Tennessee Valley Authority (TVA) has long been concerned about
regional transport and transformation of atmospheric sulfur species
(Gartrell et al.  1963).  The primary reason for this concern is that TVA
operates 12 large coal-fired power plants within the southeastern United
States, which constitute a major source of sulfur dioxide (802),  the
precursor to sulfates (SO^).

     In 1975, a program to study the regional atmospheric transport  of
coal-fired power plant emissions, the Tennessee Regional Atmospheric
Transport Study (TREATS), was initiated by TVA as part of the Federal
Interagency Energy/Environmental Research and Development program being
administered through the Environmental Protection Agency.  The primary
objective of the TREATS research is to develop an understanding of the
characteristics and mechanisms affecting regional pollutant levels and
interregional transport of primary and secondary sulfur pollutants.   With
respect to this objective, the impacts of TVA emissions on the TREATS
study region and downwind regions are of primary concern.
BACKGROUND

     Historically, when evaluating the impact of power plant emissions,
the region of concern was generally the immediate vicinity (within 20
km) of the power plant in question, and the pollutants of concern were
mainly particulates and S02.   In most scientific work relating to atmos-
pheric transport of pollutants, little or no attention was given to the
intermediate (20 to 100 km) or regional (beyond 100 km) transport of
pollutants.  Recently, however, emphasis has been placed on the evalua-
tion of intermediate and regional transport of power plant emissions.
Emphasis on the intermediate transport was changed abruptly by the
December 1974 publication of the Prevention of Significant Deterioration
(PSD) regulations (Federal Register 1978).  Emphasis on the regional
transport of pollutants was changed by the gradually evolving regional
sulfate transport-transformation theory.  This theory implies that remote
primary S02 emissions (preferentially from power plants with tall stacks)
are transformed to secondary particulate sulfates and transported over
long distances.  Within the last 10 years, interest in regional transport
has steadily increased, as indicated by numerous international publica-
tions (e.g., Bolin et al. 1971; Eliassen and Saltbones 1974; Smith and
Jeffrey 1975; Bolin and Pearson 1975; Altshuller 1976; Wilson et al. 1977;
and Wilson 1978).

     During the same period, researchers have been compiling an ever-
growing list of adverse effects that result from exposure of the human
population, biota, and materials to atmospheric sulfate particulate.  In
particular, acid sulfate particulates have been implicated in the health
damage formerly attributed to S02  (EPA 1974a; Hausknecht and Ziskind 1975).
Other adverse environmental effects of atmospheric sulfates include acidic
deposition (wet and dry), with related adverse effects on the ecology of

-------
                                -2-
lakes, rivers, soils, and forests (Braekke 1976), and corrosion of mate-
rials (Yocom and Grappone 1976).  More recently, visibility reduction
has also been related to increases in atmospheric S04 levels (Trijonus
and Kung 1978).  Reduction in visibility and possible harmful health
effects are of special concern for aerosols in the 0.1- to l-|jm size range.
Sulfates are one of the more important contributors to atmospheric aerosols
in this size range.

     The problem with controlling sulfates is that they are predomi-
nantly secondary pollutants transported to the receptor from usually
unknown source regions, and little is known about the transport-
transformation phenomenon.  Even so, EPA is expected to promulgate
National Ambient Air Quality Standards for fine particulates, which are
mostly sulfates, within one to two years (Rowe et al. 1978).  The
promulgation of fine particulate air quality standards is expected to
pose difficult problems for emission control and siting strategies for
future fossil-fired power plants—possibly more difficult than those for
PSD.  The transformation and long-range transport of sulfates from
unknown sources make the problem of equity in sulfate control strategies
particularly acute.  Thus, development of a physically realistic regional
transport-transformation modeling technology is an essential link in the
development of an equitable control strategy.

     Development of a regional-scale modeling technology will require
submodels for gas-to-particulate conversion processes and various removal
processes.  Also, detailed information on the time-dependent regional
emissions and diffusion and transport wind fields will be required to
drive the model.

     Highly accurate models are expected to be complicated because of
the possible nonlinear dependence of sulfate production on the source
strength of S02.  Another potential problem is the dependence of the
gas-to-particulate reaction on other trace atmospheric constituents
(e.g., OH radicals, cloud water pH, 03, and NH3).  Also, the question of
biogenic (natural) sources of sulfur compounds is now an open area of
research, and active efforts are underway to understand it.


SCOPE OF THE RESEARCH

     This report addresses not only the measurement and modeling of the
transport and transformation of sulfur species, but also the climatological
variations of aerometric and meteorological parameters, emission rates,
and their interrelationships.  Other pollutant species are dealt with
primarily as they relate to the formation of particulate sulfates.
Prominent among the "sulfates" are sulfuric acid, ammonium bisulfate,
and ammonium sujfate.  Within this report, the terms sulfate, particulate
sulfate, and S04 are synonyms and are defined as the traditional water-
soluble fraction of the total_filterable particulate analyzed and are
expressed as micrograms of 804 per cubic meter of sampled air.

-------
                                -3-
     This report describes the results of the first three years of
studies into long-range transport of secondary sulfate pollutants.
Results from two discrete field studies that characterize long-range
transport and transformation into, within, and out of the Tennessee
Valley region, along with several other "paper" studies, are reported.
Although the results are politically and scientifically significant, the
limited resources require that they be considered only indicative and
not conclusive.

-------
                                 -4-


                             SECTION 2

                  CONCLUSIONS AND RECOMMENDATIONS
     Our conclusions and recommendations are based on a limited number
of data collected during two long-range transport field studies conducted
in the Tennessee Valley region.  Therefore, these findings, although sig-
nificant and in some cases totally unexpected, should be viewed as
preliminary and requiring confirmation by further research.


CONCLUSIONS

      1.  Although TVA and other power plants account for 90 percent of
          the regional anthropogenic sulfur emissions, aircraft flux
          measurements indicated that, during southwesterly flow, these
          emissions are of the same magnitude as those advected into the
          region.  Therefore, with respect to interregional sulfur trans-
          port, TVA emissions contribute about half of the outflow sulfur
          burden.  However, during light wind or calm conditions, TVA is
          the principal contributor to regional pollutant levels.  Within
          the Valley such episode conditions are more frequent than in
          the northeastern United States, but less severe with respect
          to pollutant levels.

      2.  During transport from the southwest, only about 40 percent of
          the measured sulfate influx can be explained by upwind anthro-
          pogenic sources.  The balance appears to be due to biogenic
          sulfur emissions from the Gulf of Mexico and the extensive
          wetland areas of the southeastern United States.

      3.  Eleven Lagrangian measurement days—fourjwith inflow and outflow
          measurements of both total sulfur and 804— indicate that during
          southwesterly flow, no significant change occurred across the
          Tennessee Valley region in total sulfur flux or concentration.
          Although the sulfate flux increased by a factor of two from
          inflow to outflow, the actual concentrations were low--5.4 |Jg/m3
          at the outflow boundary.  Also, model estimates of the percentage
          of sulfate flux attributable to regional emissions are small
          (around 10 percent).

      4.  The 24-h airmass movement across the Tennessee Valley area
          frequently originates in a broad band from southern Alabama,
          Mississippi, and central Louisiana (the wetland areas of the
          Gulf Coast).  These airmasses typically exit the Tennessee
          Valley, heading toward the northeastern part of the United
          States.

      5.  Highest sulfate fluxes occur with southwesterly flow, whereas
          highest sulfate concentrations occur with northeasterly flow.
          Also, daily variation in sulfate and ammonium concentrations
          are greater for northeasterly flow than for southwesterly flow.

-------
                                -5-
      6.   Nitrate  concentrations  show  an  opposite  trend  to the direction-
          dependent  concentration variations  of  sulfate  and  ammonium
          (i.e., higher  concentrations for  southwesterly flow).

      7.   Model  calculations  (which  did not consider  wet deposition)  indi-
          cated  that airborne SC>2 decays  exponentially with  transport
          time.  Its half-life of about 16  h  results  from the nearly  equal
          effects  of removal  by dry  deposition and transformation  to  S04.
          The decay  rate is only  a mild function of mixing height.  The
          model_also predicts that S04) aside from the initial presence
          of S04 concentrations,  increases, reaching  a maximum after  about
          three  days of  transport, after  which it  decays almost  exponen-
          tially.  Model calculations  indicate that S04  is most  sensitive
          to changes in  the transformation  rate  of S02 to S04.   It is
          relatively insensitive  to  changes in mixing height or  deposition
          rates.   Comparison  with data has  shown that the model  gives
          reasonable results  for  a transport  time  of  about 10 h.

      8.   Ground-level,  high-volume  measured  S04 concentrations  indicated
          that ambient air concentrations were higher than those of air-
          craft  by a factor of two.   This finding, which is  contrary  to
          accepted dispersion theory and  measurements, may  represent  an
          artifact of the high-volume sampling  technique.

      9.   Substantial vertical pollutant  gradients often exist under
          stable conditions  (radiation inversions), especially during
          the early morning (and  night).  However, these gradients often
          dissipate  by midday.

     10.   Total  sulfur (mostly SC^)  concentration  measurements  showed a
          trend  toward lower  values  from  morning to afternoon; however,
          the other pollutants analyzed (i.e.,  S04, NH4, and N03)   showed
          no such  trend.  This disparity  could  have resulted from differ-
          ences  in photochemistry and deposition velocities.

     Although this research has allowed us  to learn much about  the charac-
teristics and mechanisms of long-range atmospheric transport of  sulfur
pollutants, it also has  raised several significant questions.   The more
significant questions and recommended actions for answering  them are
presented below.
RECOMMENDATIONS

      1.  Is the observed large total sulfur influx from the southwest
          correct in magnitude, in speculated origin (i.e., biogenic),
          and is it persistent in time?  If so, then the TVA and national
          emission reduction efforts may be limited in their potential to
          reduce the sulfate, visibility, and acidic deposition problems
          in the eastern United States.  We strongly recommend that the
          origin and magnitude be substantiated with a spring or summer
          aircraft study over this region.  If the speculation is con-
          firmed, then additional studies will be required to define  its
          persistence in other seasons.

-------
                       -6-
Equitable regional control strategies relating to regional and
interregional formation and transport of sulfate pollutants--
the major contributors to adverse health effects, visibility
reduction, and acidic deposition--will be particularly difficult.
TVA emissions have been and will continue to be looked at closely
since they are the principle sources of regional pollution in
the Tennessee Valley and are occasionally transported toward the
northeastern United States.  Evaluation of control strategies
(e.g., NAAQS, PSD, visibility degradation, cost-benefit) is
essential for both regional industrial siting and interregional
transport.  However, these evaluations are nearly impossible
because no realistic regional transport-transformation models
relating remote emissions to regional air quality exist.  There-
fore, we recommend that the TREATS program be expanded and redi-
rected to (1) collect additional data for submodel development
and model validation, (2) modify or develop a regional scale
transport-transformation model, and (3) validate the model.

In the next few years TVA will spend about 6 billion dollars
to remove 40 percent of its regional emissions.  Considering
the magnitude of this initial control technology expenditure
and the likelihood of future expenditures, we recommend that
the subsequent extent of regional and interregional air quality
improvement and benefits be established.

-------
                                -7-


                               SECTION 3

                         MATERIALS AND METHODS
CHARACTERIZATION OF THE STUDY REGION

     Because the TREATS program is focusing attention on the geographical
region encompassing the Tennessee Valley,  unique regional characteris-
tics that will influence results must be considered.   Physical,  emission,
and meteorological characteristics are important.
Physical Characterization

     The solid rectangle of Figure 1 outlines the Tennessee Regional
Atmospheric Transport Study region.  The outer map border defines the
emissions inventory region, and the inner dashed box encompasses the
area in which field studies were carried out.  The model region (solid
rectangular box) consists of parts of Illinois, Indiana, Kentucky, North
Carolina, South Carolina, Georgia, Alabama, Mississippi, Arkansas,
Missouri, and nearly all of Tennessee.  Across this region, there are six
primary physiographic features.  From east to west, they include the
(1) Appalachian Mountains, (2) valley and ridge subregion, (3) Cumberland
Plateau, (4) Highland Rim, (5) Central Basin, and (6) Mississippi
Embayment.

     The Appalachian Mountain chain is comprised of folded and faulted
igneous and metamorphic rock.  The characteristics of these types of
rocks make them particularly susceptible to the effects of acidic precipi-
tation.  The Appalachians, which are very sparsely populated, contain
few anthropogenic pollution sources.

     The valley and ridge subregion is characterized by northeasterly-
trending narrow parallel ridges and slightly wider intervening valleys.
Two large population centers, Knoxville and Chattanooga, and five TVA
coal-fired plants are found within this subregion.  The topology and
meteorology of this subregion result in persistent up-valley, down-
valley winds.  Under stable conditions, pollution tends to be trapped
and channeled in the valleys.

     The Cumberland Plateau consists of a northeasterly-trending belt of
highlands bounded by abrupt escarpments.  An unusual feature of this
region is the Sequatchie Valley.  One TVA coal-fired power plant is
found in this subregion.  The local pollution  transport problems encoun-
tered in this region are similar to those in the valley and ridge region.

     The Highland Rim consists of a rim or bench of highlands surrounding
the Nashville Basin.  Three TVA coal-fired power plants are found in
this region.  Because of the low terrain relief, pollution does not
become entrapped by physiographic restraints.  The Highland Rim supports
one of the most fertile agricultural  regions within the southeastern
United States.

-------
                                                 Mississippi Embayment
                                                 Modeling Area
                                                 1976 & 1977 Field Study Area
Cumberland Plateau
Highland Rim
Central Basin
Piedmont Plateau
Appalachian Mountain
Valley and ridge region
                                                                                            i
                                                                                            00
                                                                                            i
Figure 1.  Physiographic division of the TREATS study regions.

-------
                                -9-
     The Central Basin is an oval, northeasterly-trending depression
developed within the Highland Rim.  One large urban center,  Nashville,
and one TVA coal-fired power plant are found within this region.

     The Mississippi Embayment is characterized by its low relief and by
the absence of consolidated rock outcroppings.  One large urban center,
Memphis, and two TVA steam plants are located within this region.

     Some of the physiographic subregions within the Tennessee Valley
exhibit topography capable of entrapping and channeling air pollutants.
For example, during the summer months, warm, moist tropical air is fre-
quently channeled from southwest to northeast between the Appalachian
Mountains and stationary cold fronts to the north (Smith and Nieraann 1977).
Most parts of the regions are heavily forested, with biota ranging from
subarctic coniferous forest biome to deciduous forest biome.  Due to humid
conditions and a dense forest canopy, pollutant removal by dry deposition
should be greater than in many other regions of the United States.  The
variety of topographical regimes and substrates has, of course, resulted
in a variety of edaphic biotic communities.
Emissions Characterization

     An essential ingredient for any regional atmospheric transport study
or modeling effort is an up-to-date emissions inventory.  For this reason,
a detailed inventory has been compiled for the TREATS region (Reisinger
and Sharma 1977).

     The annual average weight of anthropogenic S02 emissions for the
map area shown in Figure 1 is large--about 15,000 metric tons per day.
On a per-unit-area basis, this emission is nearly equivalent to the
national average; however, on a per-capita basis, it is about four times
the national average.  This high emission results from a moderate popu-
lation density combined with a regional energy supply system, which depends
strongly on regional high-sulfur coal.  Figure 2 places the regional sulfur
emissions in perspective with the rest of the nation.  Some 75 percent
of the S02 emissions occur east of the Mississippi River, with the highest
emission density occurring near the Ohio River Valley (Ohio, Pennsylvania,
and Indiana).

     Nearly 90 percent of the regional emissions result from 33 major
coal-fired power plants, which are irregularly scattered over the entire
region.  Figure 3 illustrates the locations and relative magnitude of
S02 emissions for the power plants.  TVA's 12 coal-fired power plants
account for one third of all sulfur emissions.  By 1983 various S02
control technologies will reduce TVA regional emissions by 40 percent.

     Figure 3 shows that many of the larger power plants are geographi-
cally grouped along the Ohio and Tennessee river valleys.  This can lead
to one of several "source intensification" corridors, depending on the
synoptic meteorology (Smith and Niemann 1977).  Such corridors become
significant when the airmass transport direction and the corridor direc-
tion coincide.  When this happens, emission rates per unit area (along
these corridors) increase significantly.

-------
                                                                                      I
                                                                                      I—'
                                                                                      o
                                       >20  tons/km
Figure 2.  Annual average SC>2  emission density map.

-------
              LABADIEH/SIL>lt"S •'••
                  V  CJlfBAYSHORe

                 FESTuS-^Mx-N1 LLIN
0 I S
                       ^	'HrS* BALDWIN


                            ^•^C   CARBONDALE
 PETERSBURG
       GALLAGHER,

 NO I ANA  (\^*,- LOUISVILLE
               .  LOUISVILLE
NEWB
         .'   ftz- ST! ALBANS
            V    .CHARLESTON

          WES-T V I RG I Nl A
LOUISA '                      5a.
          Ml S SOUR I
      ARKANSAS
UTICf HOCX
                                   KENTUCKY
                      PARADISE—  —  —J
                POP AR BLUFF    NEW
                            MADRID
                                                                                                             NORTH CAROL IN A

                                                                                                            AS NEVILLE
   LA
                        M I  S S I S S I  P P I
                                        COLUMBUS,
                  ALABAMA


                    Q1RGAS
                    MIN6HAM,	

                   	f  BWILSONVILLE
                                                                                      G EORGIA
0                                                                                        ATLANTA

                                                                                      LWNAN
                          100-299
                    (    J 300- 499*
                                          \ 700-899*
                                                                      LEGEND
                                                              900-1099*
                                                          ifl
                                                          MutS
                                                                                                  MODEL AREA
  500-    • TVA  STEAM PLANT
                                              	:- FIELD STUDY AREA

                                             * metric  tons/day
                                           699"
                                                     LARGE NON-TVA STEAM PLANT
                                                                                                                                COLUMBIA
                        Figure  3.   Locations  and 862 emissions of major point  sources.

-------
                                -12-
     An example of a source intensification corridor is evident in the
 locations of  the TVA Colbert, Johnsonville, Cumberland, and Paradise
 Steam Plants.  These plants form a curved corridor oriented from the
 south to the  northeast through middle Tennessee and southern Kentucky.
 Actual documentation of such a source intensification event along this
 corridor was  found during the recent Sulfur Transport and Transformation
 in the Environment (STATE) Tennessee Plume Study (Private communication
 with W. E. Wilson, Environmental Protection Agency, Research Triangle Park,
 North Carolina, 1978).

     Major population centers within the map area include Nashville,
 Knoxville, Chattanooga, and Memphis, Tennessee; Birmingham, Alabama;
 Atlanta, Georgia; and the area along the Ohio River from Evansville,
 Indiana, to Louisville, Kentucky.  Historical meteorologic records
 indicate that average airmass trajectories emanating from this map area
 have maximum  frequency toward the NNE, with a second maximum toward the
 ESE.  Thus, cities such as Chattanooga and Knoxville, Tennessee, and
 Louisville, Kentucky, are downwind of potentially high emission density
 corridors.

     The Cumberland, Paradise, Widows Creek, and Shawnee Steam Plants
 are the largest TVA coal-fired plants.  These plants, together with
 TVA's Gallatin, Colbert, and Johnsonville Steam Plants, are base-load
 plants and are all within the field study area sampled during the 1976
 and 1977 studies.  The average sulfur content of the coal burned in
 these plants, most of which comes from the southern Appalachian area, is
 3.4 percent.


 Meteorological Characterization

     In the following subsection, the climatological analyses of sig-
 nificant meteorological variables in and around the Tennessee Valley
 region are described.  Specifically, analyses of seasonal and annual
 variations in synoptic weather patterns and airmass trajectory analyses
 are presented.


 Synoptic Weather Patterns--

     In an effort to objectively quantify the significant meteorological
 variables that impact the Tennessee Valley region,  a quasi-objective
 weather typing scheme has been devised.  This scheme classifies five
 basic weather parameters identifiable on National Weather Service (NWS)
 daily weather maps.  The area of interest is defined within a 500-nautical-
 mile (nmi) radius from Nashville, Tennessee.  Daily weather maps were
 tabulated every sixth day from November 5, 1973, through October 28, 1977.
Although meteorological cycles are known to exist,  we felt that integrat-
 ing the data base over four years would eliminate this bias.   In doing
 this, a total of over 235 data points were generated for each of the five
parameters.  These five parameters describe (1) the predominant frontal
 systems and associated pressure centers; (2) the distance (range) of the
most significant pressure center from Nashville, Tennessee; (3) the direc-
tion from Nashville,  Tennessee (degrees from true north) of this pressure

-------
                                -13-
center; (4) the airmass type (e.g., maritime tropical,  continental polar);
and (5) the relative frequency of measurable precipitation (£0.01  in.).
These parameters are further divided into categories.   These categories
and their annual and seasonal distributions are presented in Appendix A.

     All five parameters have significant departures from the mean during
summer, the peak sulfate season for eastern North America.  These  depar-
tures are all correlated with the strong influence of the Bermuda  high
on weather patterns in the southeastern United States,  especially  during
the summertime.  This influence results in reduced wind flow; moist,  warm
air advection; and poor ventilation, which often leads  to air pollution
episodes, not only in the southeast but over the entire eastern United
States.

     Further work that substantiates the significance of the Bermuda  high's
ability to produce elevated air pollution potential is  presented by Korshover
(1976).  He describes the most significant stagnation,  or poor ventilation,
areas within the eastern United States on both an annual and seasonal
basis.  For the high air pollution seasons of summer and fall, his analysis
shows that the most intense areas of stagnation are located in an  arc
from West Virginia southward to central Georgia.  The average distance
and location of this stagnation arc relative to the TREATS field study
region correspond well with the directional and distance analysis  for
the high-pressure centers listed in Appendix A.  This pressure orienta-
tion produces an annual airflow toward the northeastern United States
and leads to significantly reduced transport during the summer and autumn
months.  Examples of this flow pattern producing high sulfate pollution
levels two to three times as high as the annual average are the TREATS
1977 field study days of late June and early July.  However, significant
departures in this typical high pollution flow regime can occur (Reisinger
and Crawford 1979).
Trajectory Data--

     Twice daily NWS 24-h surface and 850-mb back-trajectories for Nash-
ville, Tennessee, were analyzed for a 3-year period (1976-1978) to define
significant flow directions into the Tennessee Valley region.  Also, for
a 1-year period (1972), 850-mb (~1500 m MSL) trajectory data were analyzed
for 14 cities surrounding the Tennessee Valley region to describe the
movement of airmasses out of the region.

     Annual contour analyses of favored origination locations for 24-h
back-trajectories are presented in Figure 4.  The numbers shown are the
occurrences for the period of record, 1976-1978.  These analyses were
obtained by plotting the frequency of occurrence of 24-h back-trajectories
by 1-degree latitude-longitude boxes.  The data show that on an annual
basis most trajectories originate in a sector from southern Alabama
through northwestern Tennessee.  Seasonally, the winter pattern is quite
similar to the annual frequency pattern, whereas the spring pattern is
diffuse with numerous trajectories from the north and east.  However,
the summer pattern indicates a more compact distribution, with 24-h tra-
jectory origins frequently occurring from west Tennessee.  Fall trajec-
tories, like the spring trajectories, show a diffuse source  region, with
significant maxima around the Ohio River Valley and eastern Tennessee.

-------
Figure 4.  Favored 24-h airmass source regions for Nashville, Tennessee.

-------
                                -15-
     The frequency of occurrence of trajectories originating within a
model area (defined by 38°N, 91°W, 33°N,  and 82°W)  and passing over eight
cities surrounding the area is shown in Figure 5.   This model area roughly
defines the area within which TVA has coal-fired power plants.  From this
figure, one can see that trajectories originating within this area rarely
affect regions to the south or west, whereas frequent airmass movement
occurs toward the north through east.
SAMPLING PLATFORMS

     Two aircraft were used for the airborne sampling experiments.   A
deHavilland Beaver (U6-A)--a single-engine, fixed-wing craft--was used
in both the 1976 and 1977 regional transport studies, and a Bell 47A
helicopter was also used during the 1977 study.  The instruments installed
aboard the airplane are described in Table 1.
        TABLE 1.  AIRPLANE INSTRUMENT PACKAGE, BEAVER U6-A
  Parameter
     Detector
  Instrument
    Study
Total sulfur
Total hydro-
  carbons
Ozone

NO, NO
  '   x

 scat

Temperature
Dewpoint
S04, N03, NH4
Flame photometric     Meloy Labs SH202   1976, 1977
Flame ionization
Chemiluminescence
  (Os + C2H4)
Chemiluminescence
  (03 + NO)
Integrating
  nephelometer
Thermistor
Chilled mirror
Meloy Labs SH202
McMillan 1100

Thermo Electron
  14D
Meteorological
  Research
Custom
Cambridge 137-C3
     Filter collection system

37 m Millipore        See text
  filters (Fluoropore)
1976,  1977
1976

1976,  1977

1977

1976,  1977
1976,  1977
                   1976, 1977
     Power  for the continuous monitors was provided by the aircraft
electrical  system through two 1-kW Topaz  inverters.  The output from the
instruments was applied to Hewlett-Packard 7100B, dual-channel, strip
chart recorders.

     The airplane, instrument layout, and instrument package  are  shown
in Figures  6, 7, and 8.  The dual-line sampling probe was constructed
from two 0.635-cm-OD stainless  steel tubes that extended 1.2  m above the
fuselage.   This arrangement allowed for separate  sampling streams for

-------
                                                                                                        0\
                                                                                                         I
Figure 5.  Percent occurrence of airmass movement out of the Tennessee Valley region (1972).

-------
                    -17-
Figure 6.  deHavilland Beaver U6-A airplane.

-------
                            -18-
H
      ELECTRIC
      INVERTERS
                         C2H4
CYLINDERS
PARTICULATE  COLLECTION
 / SYSTEM
                                                         OPERATOR
                                                         STATIONS
 PILOT      DRY TEST
STATION      METER
                                  INSTRUMENT
                                     RACKS
                       INLET
                       PROBE
               INSTRUMENT
                 RACKS
             Figure 7.  Airplane instrument layout

-------
TS
°3
                                                      DEW PT.
                                                      RECORDERS
                                                                                   c
NO  NOX
                 Figure 8.  U6-A instrument package.

-------
                                -20-
particulate and continuous gas monitors while eliminating interferences
from the propeller wash and engine exhaust.  Both probes were nearly
isokinetic--the gas sampling probe by allowing excess probe ram-air to
bleed out the end and the particulate sampling probe by matching probe
intake area and sampling flow rate to the aircraft sampling speed of 50 m
s-1.  Vacuum for the particulate collection system was obtained from the
airplane vacuum system.

     The integrating nephelometer was operated without a heater.  For
relative humidity above about 70 percent, hygroscopic or deliquescent
particles grow.  This enhances their scattering coefficient and leads to
increased b
           scat
related to b
            scat
                  Below about 70 percent, mass concentration can be
                 (Meteorological Research, Inc., 1972) by
                    Mass (g m"3) = 0.38 b
                                         scat'
     The second aircraft, a Bell 47A helicopter, was used in the 1977
regional transport study.  The instruments installed aboard the helicopter
are described in Table 2.
         TABLE 2.  HELICOPTER INSTRUMENT PACKAGE, BELL 47A
  Parameter
                           Detector
                                Instrument
Total sulfur


 scat

Temperature
Dewpoint
S04,
        , NH+
Flame photometric

Integrating nephelometer

Thermistor
Chilled mirror

  Filter collection system

37 mm Millipore filters
       (Fluoropore)
                                                  Meloy Labs Mo'del
                                                    SA-285
                                                  Meteorological
                                                    Research, Inc.
                                                  Custom
                                                  Cambridge 137-C3
See text
     Power for the continuous monitors was provided by the helicopter
electrical system through a 1200-W Deltec inverter.  The output from the
instruments was applied to Hewlett-Packard 7100B, dual-channel, strip chart
recorders.

     The helicopter and the instrument layout are shown in Figures 9 and
10.  The dual-line sampling probe was constructed from two 0.635-cm-OD
stainless steel tubes, which were mounted on the right landing strut and
extended forward about 1 m.  This arrangement allowed for separate sampling
streams for particulates and gases while eliminating rotor downwash inter-
ferences.  Airflow through the probes was nearly isokinetic.  Vacuum for
the particulate collection system was obtained from two Cast model 1550
vacuum pumps.

-------
Figure 9.  Bell 47A helicopter.

-------
                                      -22-
TEMPERATURE AND DEW POINT PROBE
 INSTRUMENT  RACKS
                          GAS  PROBE
NEPHELOMETER PROBE
 POWER
 INVERTER
                PARTICIPATE
                   PROBE
                                                                  PARTICIPATE
                                                                     FILTER
                                                                   PUMPS
                      Figure 10.   Helicopter  instrument layout.

-------
                                -23-
     In both studies, particulate samples were collected on 37-mm Millipore
(Fluoropore) membrane filters.  The 1976 study used Fluoropore filters
with 0.5-(jm pore size, whereas the 1977 study used filters with 1.0-(jm
pore size.  Liu and Lee (1976) have shown that these filters have greater
than 99 percent collection efficiency for aerosols in the 0.03- to
1.0-pm diameter range.  The volume of air sampled was measured with
Sprague dry test meters in 1976 and Matheson mass flowmeters in 1977.
The average flow rate through the filters was 50 L/min,  and the average
volume of air sampled was 1.8 m3.


METEOROLOGICAL MEASUREMENTS AND SUPPORT

     The study area for both the 1976 and 1977 aircraft measurement
programs was centered over the western Tennessee Valley region (dashed
area in Figure 11).  As shown by Figure 11, this area has one of the
most densely instrumented meteorological networks in the Nation.   TVA
operates seven upper-air stations, and the NWS operates one.  Of the
stations, four are operated on a 24-h basis, with pibal and temperature
soundings taken four times throughout the day.  Most of these soundings
are taken from early morning through midday.  The NWS upper-air station
takes temperature and wind soundings at 0600 and 1800 CST.  In addition
to these measurements, TVA measures wind and temperature at 13 meteorologi-
cal towers, and the NWS measures near-surface and cloud parameters at five
24-h weather observation stations.  This wealth of meteorological data,
particularly the eight upper-air sites, has proven useful in accurately
defining the spatial and temporal variations of the planetary boundary
layer.

     The TVA Meteorological Forecast Center  (MFC) in Muscle Shoals,
Alabama, provided planning and operational forecasts of mixing-layer
heights and wind velocities, airmass trajectories, heights of radiation
and subsidence inversions, and general weather and cloud forecasts for
the study area.  These forecasts were issued twice daily; an operational
forecast was issued on the morning of a sampling day, and an afternoon
planning forecast was issued for the following day.  Updates of actual
wind and temperature profiles near the aircraft sampling paths were
frequently obtained on a near-real-time basis to evaluate the ongoing
experiment from a meteorological standpoint.  To minimize variables and
still obtain a "typical" airflow pattern, the 1976 and 1977 studies were
designed so that sampling would be conducted during "favorable" meteorolo-
gical conditions.  These favorable conditions included ceilings greater
than 305 m, identifiable subsidence  inversions i2000 m, no measurable
precipitation within the study area, and persistent winds from the south
through west.
SAMPLING PROCEDURE

     Two full-scale long-range transport field  studies were  conducted,
one during February and March 1976 and the other during June and July
1977.  Because meteorology, chemistry, and biological activity  differ
significantly from spring to summer, and knowledge  gained  from  the  1976
study influenced the design of the 1977 study,  the  sampling  procedure
description  is divided by field study.

-------
             A  STU   J,
                                                                                                           WESTVIRGINIA
                                                    I ND I  AN A

                                          EVA|t:$VILLEOEVV
                                            f^&L.
                                             5  -NEB X
                                                                     KENTUCKY
   MISSOURI
              POPQ

          POPLAR BLUFF
                                                          T E N N E S S E|E
                                                                                                                     r inrvi
                                                                                                                CAROlINA
                                                                                                                            CHARLOTTE
ARKANSAS
                                                                                                               TH CAROL I NA
                  M I S SI S S I  PPI
                                                    AGQR
                                                     BIRMINGHAM

                                                         BHM
                                                                                                • TVRRAOB PBAL8MET TOWER 0 SFC WX STA
                                                                                                A TM» PBAL MET TOWER   A NON-TVA MET TOWER
                       Figure 11.   Location and type of meteorological  measurements.

-------
                                -25-
1976 Study

     The sampling strategy during the 1976 study was to obtain Lagrangian
airraass measurements under fairly representative airflow conditions.
This and other meteorological conditions, described in the Meteorological
Measurements and Support subsection of Section 3, could be met only  briefly
during prefrontal flow.  Also, due to the variability of the wind,  the
"window" during which the afternoon aircraft space coordinates were  similar
to the Lagrangian airmass coordinates was also limited.

     The Meteorological Characterization subsection, Section 3, indicates
that the principal flow direction is from the south through west.   With
this in mind, we decided that the aircraft sampling strategy should be
to first make representative morning inflow measurements from south-central
Tennessee through northern Alabama.  This was accomplished by flying two
horizontal traverses, one within the morning radiation inversion and one
between the tops of the radiation inversion and the subsidence inversion.
These traverses were flown at "constant altitudes" AGL; the within-inversion
measurements were typically made at an altitude of about 600 m AGL.   Sulfate
sample requirements, more than airmass characterization requirements,
dictated that the traverses last at least 1.5 h, or be about 150 km long.

     After the morning sampling traverses were completed, the aircraft
landed and refueled at the Muscle Shoals, Alabama, airport.  At that time,
the forecasted wind and weather conditions at the outflow boundary were
updated.  The aircraft then flew a "constant altitude" sampling flight
to the outflow boundary, which was typically near south-central Kentucky
and north-central Tennessee.  Immediately after reaching the outflow field
study boundary, sampling began.  The afternoon outflow sampling procedure
was similar to the morning procedure; that is, two (about 200-km) traverses
were flown at different altitudes to get representative readings within
the well-mixed layer.  Typically, these measurements were near 600 and
1200 m AGL.
1977 Study

     Three factors dictated that the single sampling strategy technique
used during the 1976 study be modified.  The first factor was a need,
identified during the 1976 study, to determine whether significant vertical
or horizontal pollutant gradients existed, especially during early daylight
hours.  The second factor was the availability of two aircraft; thus,
simultaneous measurements at two locations were possible.  The third factor,
which probably proved greater than either of the other two, was the weather.

     As with the 1976 study, favorable meteorological sampling conditions
were identified, as described in the Meteorological Measurements and Support
subsection.  However, a climatological analysis of other summers and the
initial weather conditions observed in May and early June 1977 indicated
that obtaining all the desired conditions simultaneously would be  impossible.
A combination of low wind speeds, lack of directional persistence, morning
ground fog, and airmass and frontal thunderstorm activity all  contributed
to a modified sampling strategy.  This modified strategy resulted  in two
scenarios.

-------
                                 -26-


     Scenario 1 was defined as airmass Lagrangian or Eulerian measure-
ments.  This scenario is similar to the 1976 sampling strategy, except
that separate aircraft sampled the atmosphere at the inflow or outflow
boundaries.  Measurements were obtained by multiple-altitude traverses
(typically four levels) through the inversion layers and usually for
three time periods—early morning, late morning or midday, and afternoon.
This technique allowed for both Eulerian space and time measurements in
addition to the anticipated Lagrangian measurements.

     In this report the terms "Eulerian" and "Lagrangian" refer only to
the basis of the coordinate system.  If the coordinate system is particle
attached, the term "Lagrangian" is applied; the term "Eulerian" is properly
applied to all other cases.

     Scenario 2 was defined as airmass stagnation, or blob measurements.
This scenario was similar to scenario 1 in that traverses were flown by
both aircraft.  However, because wind flow cannot be defined under stag-
nation conditions, measurements cannot be evaluated for flux, and only
concentration variations in space and time can be analyzed.

     The 1976 and 1977 data that were collected using these sampling
strategies are summarized in Appendix B.
ANALYTICAL METHODS

Continuous Gas Analyzers

     The continuous gas analyzers used in the studies were calibrated
against standardized, wet chemical methods.  The West-Gaeke (1956) pro-
cedure was used to calibrate the Meloy Labs flame photometric sulfur
analyzers before and after experiments.  During sampling, all gas moni-
tors were continually checked for malfunctions.

     The flame photometric detector is noted for its excellent sensitivity
to low background levels of sulfur and its linear logarithmic response
over several orders of magnitude.  Unfortunately, it is also sensitive—in
an instrument-specific manner— to ambient pressure, which is a function
of altitude.  The results from altitude tests on the Meloy 202 and 285
analyzers and the method used for adjusting the data to compensate for
pressure sensitivity are given in Appendix C.  Caution should be observed
when using the resulting concentration numbers in an absolute sense,
since these tests were made after the two study periods.

     The gas analyzers provide a continuous measurement in time (or dis-
tance) .  To compare these continuous measurements with integrated-traverse
low-volume filter particulate measurements, the continuous measurements
must be integrated over the sampling traverse.  The algorithm for this
integration is
                    c = i/(t2 - tx) ; fcl c(t)dt,

-------
                                -27-
where

     C      = integrated traverse equivalent concentration,
     C(t)   = time continuous concentration,
     ti* fc2 - endpoint times of the traverse.

Because of large sample times (typically, t£ - tt = 15 min)  and small
gradients of C(t) in time, instrument response time along a  traverse was
not a problem.
Low-Volume Filters

     The water-soluble fraction of particulate captured on the 37-mm
Millipore low-volume filters was extracted with Super-Q (prefiltered,
organics adsorbed, deionized, and membrane-filtered) water in an ultra-
sonic bath.  The extracts were analyzed for 864, N03, and NH4 ions.  The
S04 analyses for the 1976 study were performed by flash vaporization--
flame photometry (Roberts and Friedlander 1975; Husar et al. 1975).  The
samples from the 1977 study were analyzed for 804 and NOs by ion chroma-
tography (Mulik et al. 1976).  The NH4 analyses were performed by the
alkaline phenate method (EPA 1974b) with a Technicon autoanalyzer.   The
accuracy to be expected from these techniques is shown in Table 3.
High-Volume Filters

     Mine Safety Appliance filters were used during the 1976 field
study.  Gelman Spectrograde filters were used in all TVA high-volume
samplers and in State of Kentucky high-volume samplers operated for the
1977 study.  The Spectrograde filter was selected for the 1977 study
because of its low 864 background and lower alkalinity compared with most
other high-volume filter materials.  Coutant (1977) has shown that lower
alkalinity filters produce significantly less artifact 804 formation.

     Exposed filters were weighed for particul|te loading (Jutze and
Foster 1976) and analyzed for water-soluble 804, NOs, and NH4 ion concen-
tration.  An extract for ion analysis was obtained from a 3.4-cm strip
of the filter, which was hot water refluxed for 90 min.  Ion concen-
trations were determined with the methylthymol blue analytical finish
for SC>4, automated cadmium reduction method for nitrate-nitrite, and the
colorimetric phenate method for ammonium (analyzed as NHs).  Because of
extraction problems with the 1977 Spectrograde filters and delays in
analysis, N03 and NH4 values are considered inaccurate and are not
reported.  Checks on the accuracy of the sulfate extraction showed no
significant bias.  Based on eight triplicate measurements (from three
collocated high-volume samplers), the standard error of estimate for the
precision of the measurement and laboratory analysis was found to be
±1.4 jjg m~3 or about 10 percent.

-------
            TABLE 3.  ACCURACY  OF ANALYTICAL METHODS, 37-mm MILLIPORE FILTERS

Ion Method
S04 Flash vaporization


804 Ion chromatograph

NOg Cadmium reduction


NOs Ion chromatograph

NH4 Alkaline phenate


Typical blank
(M8/filter)
0.4


0.2

0.1


0.1

0.3


Concentration
(Mg/filter)
4.0
8.0
16.0
2.0
20.0
0.5
1.0
2.0
0.1
0.5
0.5
1.0
2.0
Recovery
91
100
98
81
93
118
104
98
100
97
84
87
92
Precision
tt)
4
4
3
25
5
40
18
10
19
20
36
24
8

                                                                                                          I
                                                                                                         M
                                                                                                         oo
Based on seven replicate determinations at the given concentration level with a single

operator.

-------
                                -29-
TRANSFORMATION-TRANSPORT MODEL

     This section presents the derivation and exploration of simple
Eulerian and Lagrangian models of the SOg-to-SC^ transformation and trans-
port processes to a receptor along a mean trajectory.   The pollutants are
considered to be confined between the ground and a capping inversion.
Diffusion in the horizontal is allowed.  Simple linear models of the
transformation and the dry deposition processes are used.  Wet removal
processes are not considered, but could be easily added to either the
analytical or numerical formulations.  A sensitivity analysis of the
Lagrangian 1-dimensional analytic formulation illustrates not only the
importance and functional relationship of various model parameters, but
also the integral characteristics of the long-range transport process.
Model response is compared with field data in Section 4.
Model Development

     The equation that describes the mean turbulent transport, diffusion,
and transformation of any conservative, gas-like, airborne substance is:
                                                                   (2)
The term on the left side of this equation is the mean rate of increase
of C. per unit time.  The first three terms on the right side define the
increase of C. resulting from 3-dimensional differential advection
transport.  Trie next three terms define the 3-dimensional turbulent
flux-convergence for C..  The last term specifies chemical transformation.
Equation (2) becomes specific to a substance when appropriate conversion
parameters and boundary conditions are specified (Monin and Yaglom
1971).  Although it is of limited utility as it stands, Equation  (2) is
a useful starting point for model development, provided appropriate
assumptions are made and applied with suitable boundary conditions.  This
apgroach was used to develop a simple regional transport model for S0£ and
864.  Variables used in this subsection are defined in the List of
Abbreviations and Symbols located at the front of this report.
Analytical Model--

      Consider  a  region  of varying width  (Figure  12),  oriented  so  that  its
length is parallel  to the mean boundary-layer  wind  direction.   It is
bounded above  by an elevated  inversion,  below  by the  ground, and  on either
end  by the vertical inflow  and outflow planes.   Sulfur,  as  S02 and S04,  is
carried into the region at  the inflow by the mean wind and  is  augmented  by
regional sources of S02, both natural and man-made.   The regional airborne
sulfur is subjected to  turbulent  transport and diffusion,  chemical
transformation,  and deposition.

-------
                                -30-
                               INVERSION
Figure 12.  Region used in the development of the long-range transport model.





     Equation (2) can be greatly simplified to describe this situation.

To do so, eight assumptions are made.
          ac.
     1.   —i = 0  (steady state).
          at

     2.   U = constant; V = W = 0.


                          ac.
     4.   C.  is  independent of y  and  z.
             ac.
               .
      5    K   i  is  independent  of  x  and  z  within the region.
             8C.
             u*-> .

      6     K   i  is  independent of x and y within the region.
      7.    K
      8.    K
             9Ci
                               ac.
                  y = B/2
                                   y = -B/2
             8Ci
                                  9C.
                                   .


                      = -      K
                 z = 0
                                z3l
                                            = 0.
                                      z = H
Letting i = 1 and integrating Equation (2) twice, once with respect to y

and once with respect to z, over the region yields



                                                                   (3)
                                                                    (4)
                            + Hk)BC!.



 Similarly, for i = 2, the resulting equation is



           UH-(BC2) = -v2BC2 + SHkBCi/2-

-------
                                -31-


Equations (3) and (4) may be considered Lagrangian by virtue of the relation
x = Ut.  These equations can be derived directly from first principles,  as
Scrivin and Fisher (1975) did to obtain an Equation similar to Equation (3).

The analytical solutions for Equations (3) and (4) are

     Ci = C° (BQ/B) exp {-[(Vl/H) + k](t - tQ)},                   (5)

and

     C2 = C?(Bo/B) exp [-(v2/H)(t - tQ)]

          +  (3/2)C?(Bo/B) {[(V! - v2)/Hk] + I}'1

          •  {exp  [-(v2/H)(t - to)] - exp {-[(vj/H) + k](t - tQ)]}.  (6)

     B /B is the  factor defining dilution due to crosswind diffusion.
For single plumes Scriven and Fisher (1975) use B = B  + 20Ut, where 0
is the (constant) angle of horizontal regional growth; Gifford (1976)
suggests that, over larger distances, B is proportional to t3 2 so that
B = bt3 2, where  b is the proportionality constant.  For regional transport
it is appropriate to set B  to a large value, and B /B = 1 since multiple
plumes exist within the region, crosswind gradients are small, and the
mass that is diffusing out is nearly balanced by that which is diffusing
in.

     In this form, Equations  (5) and  (6) do not account for sources of
S02 within the region.  This  deficiency can be  remedied by using the
principle of superposition.   If a source having emission rate Q is
encountered  at x1, x  < x1 <  x, assume that the effluent is spread uniformly
within a plane that passes through the source and  is normal to the wind.
Then the S02 concentration in this plane, which is now taken as the
initial plane, is Q/UHB1, where B = B' when x = x', and x' is taken as
the initial  distance.  The solution at the outflow plane is the sum of
the individual solutions.

     Equations (5) and  (6) likewise do not apply to a  region having a
variable mixing depth.  This  deficiency can be  overcome either by
(1) deriving Equations  (5) and  (6) by assuming H is a  function of x,  or
(2) applying Equations  (5) and  (6)  in their present form in a stepwise
manner, requiring H  to be constant  during any step, but allowing H to
change from  step  to  step.  With  this  approach,  the region  is modeled  by
stacking "boxes"  of  possibly  different dimensions  end-to-end  along the
trajectory and modeling  transport,  diffusion, chemical transformation,
and deposition through  each box  in  a  piecewise-continuous  manner.
Additional details are  given  in  the following subsection.   For  complex
situations,  the  latter  method is  more versatile because obtaining a
closed-form  solution by method  1  may  not  be possible  when  H is  variable.
The box method can also  be used  to  handle  the presence of  S02  sources
within the region.

-------
                                -32-
Simple Box Model--

     With the  "box"  approach,  the  region  is  modeled  by stacking "boxes"
end-to-end along  the trajectory  and modeling,  in  a piecewise-continuous
manner, transport, diffusion,  chemical  transformation,  and  deposition
through each box.  Figure  13 illustrates  how boxes are  stacked to  vary
mixing height  and handle multiple  sources.

     When Equations  (4) and (5)  are applied  across a "box," they become
for S02
C    = (1 - H   /H )C    + (H   /H )C      exp [-(v /H  +
 1 >J         J -1  J  I)"     *• x  J  ^JJ •*•         -1  J
                                                                k)At]
       + Q./H.U ;
          J  J
                                                                    (7)
                       BOX FOR THE
                        PREVIOUS
                       TIME  STEP
                               BOX FOR
                               CURRENT
                               TIME STEP
LID
                Figure 13.  Schematic of the box model.
and for S04
              v2/H.k) + I]"1 exp [-
                                             -i + 3/2
                                             .. + k)]At
                 (8)
     Equations (7) and (8) are simple algebraic equations, which are
Lagrangian in nature and allow for varying mixing height and sources
along the trajectory.  Again, the dilution term for crosswind spread has
been neglected in this formulation; the solution process proceeds simply
by solving Equation  (7) for the J box S02 concentration and then using
this concentration in Equation (8).  This solution procedure is repeated
as "boxes" are stacked end-to-end along the trajectory.

-------
                                -33-
Eulerian Model—
     Equation (2) could be solved  (as is) with  implicit  finite-difference
techniques since they demand nothing about the  flow  or diffusion  situation.
Unfortunately, if assumptions are not imposed,  such  procedures make  signi-
ficant demands on computer resources and knowledge of boundary conditions
(Crawford 1977).  If the same assumptions used  to obtain Equations  (7)
and (8) are imposed, except that X ^ Ut, the  following Eulerian finite-
difference model is obtained:

                    [I/At + U/Ax +
       (l/At)C°   + U/Ax(l-H  ,/H )C
              *->j           J A  J
                                               +  k]
                                              +  H    /H C   + 2/3 kC
                                                J  •"•  J *•
                                                                    (9)
        >J                      [I/At + U/Ax  +  v2/H.]

     Equations  (9) and  (10) are  similar  to  the  previous  box model,  and
 oo
C   . and C0  . are the box  concentrations of S02 and  S04  at the previous
 i >J      ^>J
time step.

     Lagrangian finite-difference  equations similar  to Equations (7) and
(8) can also be obtained:
                                                                         (10)
       2/At [(1-H._1/H.)C1 b + H._1/H.C1  .]
                [2/At + Vi/H. + k]          +
       2/At
                                                    2/3
                                       v2/H
                                                                    (11)
                                                                    (12)
 *'J                      [2/At

These finite-difference equations follow  the  analytic  ones  remarkably
well.  For  example,  Equations  (11)  and (7)  agree within a few percent
for a time  step  of  1 h  for typical  values of model parameters.  But for
Lagrangian  modeling, the  truly analytic Equations (5) and (6) or (7) and
(8) are preferred because they are  exact within model assumptions and
functionally tell more  about the transport process.  Figure 14 compares
the boundary conditions required and solution region obtained with
Lagrangian  vs Eulerian  models.  As  can be seen, the Lagrangian model
requires much less  input  information, but also yields less predictive
information.
  \
    POINT
    BOUNDARY
    CONDITION
SOLUTION ONLY
ALONG LAGRANGIAN
TRAJECTORY
t
                   SLOPE s
\
                             SOLUTION  FOR.
                             THE ENTIRE
                             TIME /DISTANCE
                              DOMAIN
                                               LINE BOUNDARY
                                               CONDITION
 Figure  14.   Comparison of Lagrangian and Eulerian boundary conditions and
             solution domains.

-------
                                -34-
Model Analysis

     A sensitivity analysis of Equations (5) and (6) is useful because
it reveals the functional behavior of the model relative to its various
parameters and the theoretical behavior of the transport, transformation,
and deposition processes insofar as they are properly described by the
assumptions made to derive the model.  In Equations (5) and (6), t (or
x IT1) is the independent variable, and H, V^, V2, and k, along with the
initial  conditions Cx and €2, control the model response, Cj and C2.
Following the usual approach to sensitivity analysis, we varied each of
the  four parameters and two initial conditions independently over a
range of possible values, centered about their normal values, while the
other parameters and initial conditions were held fixed at their nominal
values.  The effect on model response with B /B = 1 is presented in
Table 4  and Figures 15 and 16.  For these results, typical midday parameter
values were used:  H = 1500 m, Vj = 1 cms'1; v2 = 0.1 cms"1; and k = 5.6
x  10~6 s"1 (2 percent per hour).  Table 4 shows the effect of a 50 percent
change from the nominal value of each parameter.
         TABLE 4.  PERCENT VARIATION IN MODEL RESPONSE RESULTING
            FROM A ±50 PERCENT CHANGE IN A SINGLE PARAMETER

Relative response (%) due to
parameter variation
Model parameter Typical value
k 2% h
v1 I cm s"1
v2 0.1 cm s"1
H 1500 m
Plus 50%
C j C2
-21 +36
-25 -11
-2
+23 +10
Minus 50%
C/">
1 ^2
+27 -45
+33 +13
+2
-44 -23

     Figure 15 illustrates the effect of transport time on model response.

The Cj/Cj curve is exponential and is characterized by a time constant of
[(vj/H) + k]"1.  For typical midday parameter values, dry deposition of
S02 is slightly more important than transformation of S02 to 804.  These
processes together imply a half-life for SQ2 of about 16 h.
On the other hand, the
                                  curve of Figure 15 is not exponential.
In fact, for B /B = 1, C2/C1 reaches a maximum value at
t* = to + in
                                            - v2)/H] + k},
                                                              (13)
or about three days for typical parameter values.  For Figure 15, we
assumed no initial sulfate (i.e., C2 = 0) and that t  =0.  Beyond t =
t*  Cg/Ci decays more slowly—approximately exponentially at a rate

-------
                                                                                     10'
                              TRANSPORT TIME (s)
Figure 15.  Predicted variation in S02 and SO, concentrations as a function of transport time.

-------
                                                   -36-
                2 _       4         6         8

        S02 TO  SOi CONVERSION RATE  CX/h)

«) EFFECT OF VARIATIONS OF K>THE REACTION RATE CONSTANT,
   ON SOZ AND SO? CONCENTRATIONS
   100
                1         2

              SO, DEPOSITION VELOCITY  lcm/6)
 C) EFFECT OF VARIATIONS OF V,, S02 DEPOSITION VELOCITY,
    ON S02 AND SO; CONCENTRATIONS
                                                            100


                                                            90
                                                           Ife/C*
         V|-t.O m/dVrO.t
         K *• UC/H    T •
                                                                                 I
                                                                                             _L
    800       12OO      1 BOO      2OOO

                MIXING DEPTH  (m)
                                                                                                     240O
b)  EFFECT OF  VARIATIONS OF H, THE MIXING DEPTH, ON S02
   AND SO; CONCENTRATIONS
                                                            too
               0.5        1.0        1.5
               SO^" DEPOSITION VELOCITY (cnv/s)

 d) EFFECT OF VARIATIONS OF V2, SQ~ DEPOSITION VELOCITY,
   ON SO2 AND SO7 CONCENTRATIONS
        Figure  16.  Effect of variations  in  model parameters  on  predicted response
                       after 24  h of transport.

-------
                                -37-
characterized by a half -life, due to dry deposition, of around 12 days.
Therefore, as shown in the meteorological characterization subsection,
precipitation occurs frequently enough to indicate that wet-sulfate
removal processes are probably dominant.

     Without wet removal, the model confirms that long-range transport
over great distance is possible.  Because C2/Ci is unique for each value
of t, the average age of an airmass passing the outflow plane can be
computed when this ratio is known.  Assuming C2 = 0, the age is

          ln[l + 2/3{[(Vl-v2)/Hk] + lUCa/Ct)]
     t  = -   - t
      3
     Figure 16 illustrates the effect of individual parameter variation
on model response after 24 h of transport.  This figure shows that k and
V! have the greatest control over response and v2 has the least.  The
figure also shows that increases in either H or k will increase the
C2/Ct ratio.

     Finally, the form of any of the models presented is such that the
effect of a single phenomenon cannot be separated or therefore estimated.
For example, the simplified form of Equation (5) is C/C  = exp {(-vjYH
+ k) t}.  It is apparent (based on the present best guess at the parame-
ters vx and k) that the effect of deposition is equivalent in magnitude,
direction, and form to that of chemical transformation.  The sulfate
equations are even more transcendental in nature.  This illustrates the
need for physical understanding and study of separate deposition and
transformation phenomena.  Comparisons with data are presented in
Section 4.

     In summary, the model predicts that airborne S02 decays exponentially
with transport time.   Its half-life of about 16 h results from the yearly
equal effects of removal by dry deposition and transformation to 804.  The
decay r^te is only a mild function of mixing height.  The model also shows
that 864, aside from the initial presence of S04 concentrations increases
until it reaches a maximum after about three days of transport, after which
it decays almost exponentially. The model indicates tha| S04 is most sensi-
tive to changes in the transformation rate of S02 to S04.  It is relatively
insensitive to changes in mixing height or deposition rates.

-------
                                -38-


                                SECTION  4

                         RESULTS AND  DISCUSSIONS


SUL¥ATE COHCHWRMIOH MTO TLUX CLIMATOLOGY

     Although the regional field studies supplied unique and informative
field data, these data are limited to only a few specific spring and summer
days.  Long-term climatological data do not suffer from this deficiency,
and much can be learned from •variations in and interrelationships between
aerometric and meteorological parameters.


Seasonal Concentration Variations

      The seasonal  fluctuation of a pollutant is usually  the most prominent
feature of a  long-term data  set.  The analysis of four years of sixth-day
suspended  sulfate  data taken from the five  rural Tennessee Valley monitor-
ing  sites  (Figure  17) reveals the strongly  seasonal  configuration shown
in Figure  18.   Suspended sulfate  concentrations  are  lowest  in  the winter,
highest  in the summer, and intermediate in  the spring and fall.  The mean
summer concentration of  10,4 ^g m~3  is  more than double  the mean winter
 concentration of 4.5 pg  m"3. This pattern  of  summer suspended sulfate
maxima is  consistent with the observations  found in  other sulfate research
 conducted in the eastern United States  and  Canada (Garvey 1975;  Hitchcock
 1976; ERT 1976; Lioy et  al.  1977; Tony  and  Batchelder  1978; and Melo  1978).

      Although the seasonal patterns  found in these studies  are similar,
 significant differences  in magnitude do exist,  with the highest sulfate
 levels occurring at urban sites in the northeastern United  States
 (Altshuller 1973; Frank 1974; ERT 1976).

      Many factors may be related to the seasonal pattern of suspended
 sulfate values.  These factors fall into three major categories—
 emissions, transformation,  and transport-related phenomena.

      The  anthropogenic  contribution to sulfur in the atmosphere is well
 quantified, particularly  in the industrialized  regions  of  the world.   In
 the  past, the  release of  anthropogenic sulfur from  these regions was strongly
 seasonal—high emissions  in the winter when much fossil fuel  (particularly
 coal) was used for  space  heating and low emissions  in the  summer months
 when space  heating  was  not  required.   However,  the  advent  of  cleaner fuels,
 such as natural gas and low-sulfur  coal, for  space  heating and  industrial
 processes and the promulgation of  strict air  quality  regulations resulted
  in an overall reduction in total sulfur  emissions per  kilowatt.  During
  this same period, however,  total sulfur  emissions were not reduced,  pri-
 marily because of the  increased demand for fossil-fuel-generated electrical
  power.   Across the country and particularly in the  Tennessee  Valley,  coal-
  fired generating units  provide the main source of electrical  power.   These
  units provide base load generating capacity,  and although  peak demands
  occur in heating and cooling seasons, the relative seasonal variation in
  the rate of sulfur emissions can explain only a small percentage of the

-------
       LBL
       TRIGG COUNTY
       KENTUCKY
              LOVES  MILL
              WASHINGTON COUNTY
              VIRGINIA
GILES COUNTY
GILES COUNTY
TENNESSEE
HYTOP
JACKSON COUNTY
ALABAMA
LOUDON
MONROE  COUNTY
TENNESSEE
        Figure 17.  Regional air quality trend stations.

-------
     12.0
•°E  10.0

 ^
 o»
 P-
 <
 cc
 LJ
 O


 O
 O
 O
 CO
      7.5
     5.0
     39 -
              v>

              o
                  28-1
                  2.6 -i
                  2.4-
 £   2.2
ID
 g

 2   2.0-
 (O

 g
 CO
 CO
                 1.8-
            LJ

             CJ

            O
            CO
                  1.6-
                   1.4-
                   1.2-
                   1.0
                                           -f-
                                       M
  	1	(-

A     M    J


    MONTHS
                                                                                                 •15
                                                                                                -14
                                                                                                -12
                                                                                               -10
                                                                                                -9
                                                   -8   UJ
                                                         O
                                                         o:

                                                     f   w
                                                   -7   Q.
                                                                                                -6
                                                                                                -5
                                                                                   N
                                                                   o
                                                                    I
Figure  18.   Monthly variations of TVA's  S02 emissions  and regional sulfate concentrations, 1974-1977,

-------
                                -41-
seasonal sulfate variation (Figure 18).   Therefore,  anthropogenic  sulfur
emissions are not significantly related  to the seasonal pattern of sus-
pended sulfates in the Tennessee Valley  region.

     The natural contribution of sulfur  to the atmosphere is not well
quantified.  Several researchers (e.g.,  Robinson and Robins 1972;  Lovelock
et al. 1972; Friend 1973; Hitchcock 1975) have estimated natural emissions
on a global scale, but little additional information exists.

     Biological emissions of atmospheric sulfur, thought to be mostly
hydrogen sulfide (H2S), are undoubtedly  related to seasonal fluctuations
in temperature.  Biological activity and emissions of H2S are greater
with increasing temperature.  This source of emissions may be partly
responsible for the seasonal sulfate pattern—particularly in the  TREATS
region, where typical summertime southwesterly transport allows sulfur
input from the potentially large natural sulfur-producing regions  of low-
lying marshes and swamps found in east Texas, Louisiana, and Mississippi.
Natural sulfur sources and their contribution to regional airsheds are
of paramount concern.

     Various transformation processes and their respective  rates may also
be partly responsible for the high summer levels of suspended sulfate.
The conversion rate for S02 to S04 is greater in the summer (about 1.4
percent per hour) than in the winter (about 0.3 percent per hour)  (Meagher
1977).  The increase in summer conversion rate is believed  to result from
increased photochemical activity.  The difference in conversion_rates
and the relative deposition velocities of S02 (1 cm s"1) and S04  (0.1
to 0.5 cm s"1) indicates that high summer sulfate levels result at least
partly from seasonal variations in transformation processes.

     High sulfate concentrations are usually associated with stagnating
anticyclonic airmasses (ERT 1976; Teknekron 1977).  Within  the TREATS
region these stagnating conditions are more prevalent in the summer and
fall months than in the other seasons (Korshover 1976).  Of either con-
tinental or maritime origin, these conditions are associated with high
temperature, humidity, and insolation and with low precipitation and
ventilation.  Meteorological factors present in these airmasses seem to
provide optimal conditions for both sulfate generation and  its atmospheric
buildup.  Some airmass characteristics are thus related to  optimization
of the transformation process (i.e., high temperature, insolation, and
humidity), whereas others relate to concentration of the end product
(i.e., low precipitation and poor ventilation).  The strong frontal
activity associated with the winter months exhibits contrary factors,
which result in less transformation and  concentration buildup.

      In summary, the factors resulting in the seasonal summer  sulfate
concentration maxima may be relegated to three major categories:
(1) increased natural and anthropogenic  emissions;  (2) enhanced gaseous
sulfur to particulate sulfate transformation; and (3) poor  transport  con-
ditions, resulting in atmospheric sulfate buildup.

-------
                                -42-


Regional Sulfate Flux Rose

     To better describe the flux of sulfate pollution through the TREATS
region and the significant variables that influence it, a sulfate trajec-
tory climatology analysis is presented.  This analysis summarizes three
years of sulfate flux data, gathered every sixth day beginning on January 4,
1976.  It incorporates (1) sulfate data obtained from three TVA high-volume
sampler sites centered around Nashville, Tennessee; (2) boundary-layer
meteorologic parameters obtained from the NWS site at Nashville, Tennessee;
and (3) synoptic weather typing (as described in the Meteorological Charac-
terization subsection of Section 3).  These parameters have been analyzed
seasonally and annually.   Correlations between and among variables are
described, and significant findings are discussed.

     The sulfate concentration numbers used are the averages from three
TVA monitoring locations—Cumberland and Gallatin Steam Plants and Giles
County trend station—that surround Nashville, Tennessee.  A logical
question is whether the proximity of the power plants to the high-volume
samplers leads to spurious sulfate readings.  However, as described in
more detail in the Eulerian Space and Time subsection of Section 4, the
sulfate measurements obtained at the plants are thought to be represen-
tative of regional levels.  Also, a plot (Figure 19) of the concentration
values from the Giles County trend station vs. the average of all the
sites supports this conclusion.  Figure 19 shows a slope of near one and
a high correlation coefficient.

     The meteorological parameters include average wind velocity and dew-
point temperature through the first 1500 m AGL averaged from twice daily
NWS radiosondes.  Also included are the weather typing parameters as
described in the Meteorological Characterization subsection, Section 3.
Analysis of the data (Figures 20 and 21) graphically shows that the south-
west sector dominates as the favored sector for mass transport through
the Tennessee Valley region.  The dominance of this sector for transport
is also supported by the trajectory and meteorological analyses presented
earlier.  An analysis of 24-h back-trajectory data shows that the favored
source origination region for this southwesterly flow sector is located
in southern Louisiana and Mississippi.  Implications of this region as a
potentially major biogenic source region are discussed in the Lagrangian
flux analysis subsection.

     The relative distribution of the range of aerometric and meteoro-
logical variables within sectors is also of interest.  The frequency of
occurrences of a given parameter by sector have been divided into three
ranges.

     Intercomparisons between sectors show the relative importance of
each in producing elevated pollutant and wind speed levels.  Due to a
limited data base, the individual 22.5-degree sectors have been grouped
into 90-degree sectors for analysis purposes.  Even when this is done,
the southeast sector still has insufficient data (8 observations) for
inclusion in the analysis.  The results of these annual relative fre-
quency distributions are shown in Figure 22.

-------
                                -43-
       25,
        15
CO

CO
CJD
                             10                  20
                                  TOTAL  SO  ;
Figure 19.  Giles County sulfate  concentration  (|Jg m"3) vs. the average
            concentration of Cumberland, Gallatin, and Giles.

-------
                                   -44-
                                 -E    W-
  (A)  WIND SPEED  (m/s)
(C) WIND DIRECTION  (degrees]
                                 -E     W-
                  s                                    s
(B) SULFATE CONCENTRATION  (Mg/m3)      (D)  SULFATE FLUX (pg  m2 s'1)
       Figure 20.  Annual frequency distributions of wind speed, wind direction,
                 sulfate concentration, and  sulfate flux for Nashville,
                 Tennessee.

-------
                            -45
                          -E     W
     :AJ  WINTER
(B) SPRING
    (C)  SUMMER
  (D)  FALL
Figure 21.  Seasonal frequency distributions of Nashville sulfate flux
           (|Jg m~2 s""1).

-------
                                    -46-
ctr
ct:

CJ
CJ
CD
CD

>-
CJ
C3
LU
CrT
         201
ctr
                                        LEGEND

                COMCENTRAION
                                   ->      s«
                                             :0-32
                                          «?c
                WIND SPEED  (m  sl)

                NUMBER OF OBSERVATIONS PER SJCja? IN PARENTHESIS
                           CONC

                       NE  SV  NV

iViV
:»T*i»r<
                 Figure  22.   Relative  frequency  distributions.

-------
                                -47-
     An analysis of the data presented in Figure 22 shows that,  although
most sulfate is transported into the Valley from the southwest,  the
greatest concentration maxima occur with northeasterly flow.   Reisinger
and Crawford (1979) have shown that this trend can result from the
transport of modified continental polar air from the high emission
density region along the Ohio River valley into the study area.   Also,
the southwest and northwest sectors have almost identical concentration
distributions, probably indicating a lesser, more diffuse source region
than is associated with northeasterly flow.  As will be shown subse-
quently, aircraft data from the two field studies also support this
analysis.

     The relative annual flux distribution by sector is also shown in
Figure 22.  Contrary to the concentration analysis, this figure indicates
that the cumulative flux distribution shown in Figure 20 is a good
indicator for the relative magnitude of flux by sector.  An analysis of
the relative flux distribution shows that, within the southwest sector,
flux values are greater than 66 (Jg m"2 s"1 on 30 percent of the days.
This is almost twice the relative occurrence of the other two sectors
analyzed.  Naturally, since pollutant flux is inherently tied to wind
speed, the close similarity between the relative flux plot and the wind
speed plot shown in Figure 22 is not surprising.  Also, the wind speed
plot is well correlated with the other pollutant frequency distributions,
especially for the southwest sector.

     An analysis of the seasonal variation in both flux and concentra-
tion, regardless of sector, is shown in Table 5.
TABLE 5.  MEDIAN SULFATE FLUX AND CONCENTRATION VALUES REGARDLESS OF SECTOR


                                 Flux               Concentration

Season                       (ug m~2 s'1)              (|Jg m~3)
Winter
Spring
Summer
Fall
42
50
44
38
5
6
11
7

     This table shows  that  sulfate  flux peaks  during  the  spring, whereas
 sulfate  concentration  peaks  during  the summer.  Variations  in these  varia-
 bles depend at least partly  on variations  in airmass  type,  wind velocity,
 solar  radiation, and biogenic and anthropogenic emission.   In an attempt
 to  determine  the relative importance  of one of these  parameters, airmass
 type,  a  joint frequency  distribution  analysis  was  performed.   This analy-
 sis indicates that  concentration and  flux  levels differ only  slightly when
 the airmass type is either  maritime tropical or modified  continental polar.
 However, when modified maritime polar airmasses are present,  high pollutant

-------
                                -48-
levels rarely occur.  This, no doubt, results partly from low natural and
man-made sulfur emission rates in the upwind areas of the Great Plains.
Analysis of the relationship between high sulfate flux levels and meteoro-
logical characteristics indicates that the highest sulfate flux values
(2  100 |Jg m"2 s"1) occur during southwesterly flow of maritime tropical air.
This flow occurs, almost exclusively, in response to gradients related to
prefrental or Bermuda high pressure patterns.
FLUX CALCULATIONS

     Measurements of regional mass transport (or flux) of a pollutant
are very useful for assessing not only the impact of the studied area on
itself and adjacent regions, but also the impact of adjacent regions on
the studied region.  Ideally, pollutant flux can be defined at any
location, but measurements at inflow and outflow locations relative to
the study region are the most informative.  An inflow measurement defines
the impact of an upwind region on the study area, whereas an outflow
measurement defines the combined impact of the upwind and study regions
on the downwind region.  The net impact of a studied area on a downwind
region is defined by differences in these measurements (i.e., outflow
minus inflow).  This difference, when complemented with meteorological
and source information, is also useful in studying characteristics and
mechanisms of the long-range transport phenomenon.

     Few researchers have attempted regional flux measurements because
of the measurement difficulties imposed by the large space and time
scales.  This section presents the methods and results of two modest
field studies that attempted such measurements within the Tennessee
Valley region.  Our intent is not that these measurements be considered
exact (no measurement is), but that they are reasonable estimates from
which several significant conclusions can be made.
Estimating Concentration and Flux

     The variable that describes net pollutant mass transport is pollutant
flux.  Pollutant flux can be defined at a point, over a line, or through
a plane.  The appropriate method for assessing regional impact is the
pollutant flux through a plane.  This is defined as

     Ft(x,t) =  /;c.Udyd2/(H-L),                                   (15)

where

     C. = concentration of pollutant species i,
     U  = transporting wind speed (ideally normal to the measurement
          plane),
     H  = height of the plane,
     L  = length of the plane,
     F. = transported mass of pollutant species i per unit area per unit
      1   time.

-------
                                -A9-
Unfortunately, Equation (15) is impossible to evaluate rigorously because
C. and U vary, not only as a complex unknown function of position coordinates
x, y, and z, but also as a function of time t.  Therefore, assumptions
must be made about the spatial and temporal variations of C.  and U.

     By assuming that variation in time is "slow" compared with the  time
scale required for the measurement (about 2 h) and the horizontal
inhomogeneity can be integrated out by long horizontal sampling traverses
(about 150 km), then Equation (15) can be approximated as
HR +  SUS _
H    was used in the F. approximation.  Values of C~ for SOz ar*d 864 were
 max                  i                            D
obtained by averaging the limited number of actual measurements made above
the subsidence inversions with the  10 percentile values determined from an
analysis of all measurements made by the two aircraft.  This correction
allows  for mass conservation when comparing mass flux during the diurnal
cycle and is especially important when comparing early morning  to midday
measurements.  When the morning  subsidence inversion height is  greater
than the afternoon mixing height, mass concentration is assumed to result
from meteorological subsidence of superior air.  Thus, no  corrections are
made under these conditions.  Also, so that intercomparison between days
can be  made, the sulfur flux values for all days have been normalized to
the grand mean subsidence inversion height measured  on all field study

-------
MORNING


     Figure 23.
                                 AFTERNOON


Schematic of typical flux calculation procedure.
                                                                    I
                                                                    (^
                                                                    o

-------
                                -51-


sampling days, H (1600 m).   In the following subsections,  the methods  of
approximating the variables listed in Equation (16)  are discussed.


Estimating U, HR, and H_--

     Any accurate calculation of mass flux inherently depends on a  good
estimate of wind speed.  In like manner, an accurate estimate of the
mass balance depends on a good estimate of mixing height.   For both the
1976 and 1977 studies, the  TVA and NWS upper-air stations  closest to and
most representative of the  airmass sampled were used to estimate actual
mixing height and resultant wind speeds.  These stations and their
representative wind speed and mixing height values for the Eulerian and
Lagrangian days are listed  in Appendix C.   HR was simply defined as any
surface-based inversion with 9T/3z ^ 0.  To determine the  best estimate
of the actual average mixing height for a group of measurements, the
representative sounding was plotted, and the near-surface  temperature
measured at midsampling time was used to find the dry adiabatic inter-
section of the sounding.  Variation can possibly occur when estimating
the actual mixing height, especially during midmorning, which is the time
when the radiation inversions usually dissipate.  However, on the average,
this method is a good approximation to the true mixing height (Holzworth
1972).

     After the mixing height was determined, the layer resultant wind
velocity was determined.  This velocity multiplied by the mean layer
concentrations and normalized for H produces the flux values presented
in the following analyses.


Estimating SO  and NH4 Concentrations--
             A.

     Sampling limitations necessitated estimation of CR, €„, and C, the
mean transport layer concentration, from a few  (two to four) discrete
horizontal integrated average values and "typical" pollutant profile
shapes.  Figures 24 and 25 show the typical profiles of sulfate and total
sulfur, as determined from actual measurements  during the TREATS 1976 and
1977 field studies.  The observed sulfur values were interpreted by con-
sidering the appropriate profile and were averaged both within and between
inversion layers to obtain the most representative estimates for C-, Cc,
                              +                                   Kb
and C.  Because most of the NH4 is assumed to be in the form of  (NH4)2S04,
the typical  sulfate profiles were also used to  estimate the average for

KH4.  CR for NH4 was assumed to be the  stoichiometric amount needed to  react
with the background S04  (i.e., 0.3 |Jg m"3 of NH4).


Estimating N03 and 03 Concentrations--

     In a manner similar to the estimation  for  sulfur and  ammonium
salts, discrete horizontal integrated values were obtained for  nitrate
and ozone.   The most  significant differences between  these pollutant
profiles and the typical sulfur profiles were  their high  scatter and

-------
  1.5


  1.4


  1.3


  1.2


X 1.1
                                                   tEGEND
Z
o
  1.0




   .8


   .7
2
OL
O  -5
Z
               PREVIOUS _DA_YS
                 MIXED LAYER
           TYPICAL  RADIATION
           I NVERSION   HEIGHT
                          D
                                       SYM   DATE
                                                        H      C 3
                                                       (m)    Qig/m )
                                       D    6/03/77   2800
                                       -K-    6/04/77   1520
                                        •    6/05/77   1480
                                        +    6/08/77   2010
                                        X    7/06/77   1570
                                        A    7/07/77   1160
                                        H    PREVIOUS BAYS MIXED
                                             IAYER DEPTH
                                                                2
                                                                4
                                                                8
                                                                2
                                                               15
                                                               12
                                             AVERAGE CONCEHTRATIOH
                                             VJITHIH H
                                                 C;C^ 1.55 -I.24IZ/H)
                                                    F : to.*
                                                   n
            .8
LLj

LLJ
   .7
Q
uj
hJ
Z  -6


OC
O
2
   .4



   .3


   .2
                 P R_E VJOUS _DA^S
                  MlXED  LAYE R
                                                                                                                          SYM   DATE
                                             6/04/77
                                             6/05/77
                                             6/08/77
                                             7/06/77
                                             7/07/77
X
A
H
I£gEND

    H
   (m)
  1520
  1480
  2010
  1570
  1160
   C  3
Qig/m )
  55
  24
  57
  46
  40
                                                                                                                              PREVIOUS DAYS
                                                                                                                              IAYER DEPTH
                                                                                                                                            HOED
                                                                                                                              AVERAGE CONCENTRATION
                                                                                                                              WITHIN H
                                                                                                                                   C/C =1.98-1.65 (I/HI
                                                                                           TYPICAL  RADIATION
                                                                                           i NVERSION  HEIGHT
                                                                                                                                      C/C=0.2J + S.4HZ/H)
                                                                                                                                                i   i   i   i
                                                                                           0.2   0.4   0.6    0.8    1.0   1.2    1.4    1.6    1.8   2.0    2.2
                                                                                         NORMALIZED  TOTAL  SULFUR   PROFILES (C/C~)
                                          Figure  24.    Typical  early  morning  sulfur  profiles.

-------

2.0
1.8
1.6
I
LEGEHD
SYM DATE H
	 	 £m)
• 12/19/75 762
+ 2/24/76 1060
O 3/11/76 1200
X 3/23/76 1750
• 3/24/76 1800
# 6/03/77 1220
Oil *- 6/04/77 1430
r D 6/24/77 2000
A 7/07/77 1900

c" ,
Oifi/ni )
1.8
2.8
7.0
3.2
4.7
3.7
7.2
1.9
17.1
-^ j H MIXED LAYER DEPTH
— 1.4 L C AVERAGE MIXED LAYER
Z \±t COHCEMTRATIOH
P C
< L
> U •
u, 1 +
UJ
Q
i: i.o
IVJ
<
I 0.8
O
Z

0.6


0-4

0.2




• SUBSIDENCE
u -° ' 1 N VERSION 1
I
A Di-
•xi
• *-
i
T
#-*-

» i f I D ' A . , , , . , , , , 1 ,• i + i i i i lO j i tf [ ! KED /ER A* , , .. 2.0 1.8 1.6 I IN ~ 1.4 Z O \ u UJ UJ 2 i.o M < oc 0.8 O Z o.« 0.4 0.2 LEGEHD SYM DATE -EL ,03. # (£) Qag/m ) • 3/11/76 1200 69 X 3/23/76 1750 56 O 3/24/76 1800 60 + 6/24/76 1800 60 # 6/03/77 1220 38 *- 6/04/77 1433 25 H 6/24/77 2000 216 A 7/07/77 1910 80 H MIXED LAYER DEPTH C AVERAGE MIXED LAYER CONCERTRATIOlf # 1 1 S U B SI DE NCE I 1 N VERSION 1 : A t i # ! «x i * i MIXED 1 I 0 LAYER j i * ' *x * i j, * t 1 E o , i O i ' o'? ' n'^ ' r>V ' «'. ' .'o ' .' ' '. ' -T^-^-A- ' J. '.'.'. t W I 0.8 1.0 1.2 1.4 1.6 1.8 7.0 2.2 NORMALIZED SULFATE C ONC E NTRA T ION (C/C) 1A «... w.u W.B i.u i.* 1.4 1.6 1.8 2.0 2.2 NORMALIZED TOTAL SULFUR CONCENTRATION (C/C) Figure 25. Typical midday sulfur profiles.


-------
                                 -54-
 lack  of  shape,  as  illustrated  by  Figures  26  and  27.   Because  of this  scatter,
 a  vertical  weighing  factor,  based on  actual  measurement  location within
 the mixed layer, was  considered to be more representative of  the mean
 layer concentrations  than  any  other estimation technique.  No attempt
 was made to correct  for background concentrations above  the subsidence
 inversion when  calculating C.
Flux  Measurement  Criteria--

      For  the  flux analysis to be  reasonably accurate and useful, certain
criteria  are  required:

       1.  The direction of airflow has to be relatively steady; that is,
          the directional variation of the mean transport wind in both
          space and time must be  less than or equal to ±45°.

       2.  Both inflow and outflow measurements must be made to obtain
          daily Eulerian or Lagrangian concentration values.  These
          measurements must be made over a sufficiently wide spatial
          extent,  in both the horizontal and vertical, to be represen-
          tative  of the airmass sampled.

       3.  The meteorological conditions specified in the Meteorological
          Measurements and Support subsection of Section 3 must be met.

      These criteria were met on five Lagrangian days and six Eulerian
days  during the two studies.  Only one Lagrangian day was sampled during
the 1977  summer study, whereas four Eulerian days were sampled.  Three
of the inflow-outflow Eulerian measurement days in 1977 were made during
northerly wind flow.

     A Lagrangian event (day) is  defined as outflow aircraft measurement
of the same airmass that was previously sampled at the inflow end,  plus
or minus  1 h.  All other airmass  measurements were defined as Eulerian.
The average time  of a set of aircraft traverses at both inflow and outflow
is the time used  for Lagrangian-Eulerian calculations.

     For the Lagrangian days, the second criterion usually imposed an
airmass age of 7 h to the outflow sampled airmass.  This occurred for
two reasons:  (1) The aircraft had limited performance capabilities, which
restricted them to daytime hours  and limited cruise speeds (<100 knots);
and (2) the large distance (~250 km) between inflow and outflow measure-
ment points required a fairly rapid airmass inflow-to-outflow traverse
time so that darkness would not curtail sampling operations.  Because
inflow measurement times usually  averaged 0800 h these requirements
limited Lagrangian operations to  a fairly narrow "window" of favorable
wind flow speeds  of from 8 to 12 m s"1 (i.e.,  airmass age of 6 to 8 h).
Figure 28 graphically shows both  the Lagrangian and Eulerian measurement
regimes.  Here the earliest measurement defines the relative origin for
other measurements for the same day.'  Only days with both inflow and
outflow measurements are plotted.

-------
                                 =Y180L  DATE
I
\
N
Z
a
LL
J
UJ

D
U
O
Z
     6 —
    I •» -
                                   J
                                   »
                                   O
                                   _>
                                   G
                                       2/20/7P
                                       5/1 i/7f
                                       V18/7C
                                       V24/7C
                                       e/O^/77
                                       C/18/77
                                       fV^tt/77
                                       f/TO/77
                                       7/OCV77
                                       7/C.7/77
1 981
1219
                                                    (y q/m 5
                                                     IBb
                                                     177
       m
       ea
       9=;
       42
             MI_XED_
             LA YER
     8 -
         TYPICAL RADIATION
         INVERSION HEICMT
     2 -

1-2!-
f SUBSIDENCE
1 D
^ i INVERSION
0 '
( 	 . 1 —
1

8 i—
LU i- 	 	 	
Cu f-
|_
a . i
LU 0 t—
r^Kj —
5 h

21 i
a: •* 1-
i L

i—
K
LEGEND
DATE H
t> 2/19/76 1900
O 3/11/76 12BO
» 3/18/76 1372
O 3/23/76 1767
O 3/2V76 1 829
a 6/02/77 280S
•+- 6/28/77 1372
* 6/30/77 1629
• 7/06/77 1067
H-MIXED LAYER DEPTH
«
O

o




c.
Z07
163
180
170
196
20S
114
162
151
(ml
C-AVERACE M;XED LAYER
CONCENTRATION tug

/m3)

D
0 o» +

1 •
D
* 00

•

o °
* * D




Ln
C/i
1












                                             ! 2
                 MORNING OZONE PROFILES  (C/C)
                                                                               2      1      6      8     I 0     I 2

                                                                                  MID  DAY OZONE PROFILES  (C/C)
                      Figure 26.   Typical early morning and  midday  normalized ozone profiles.

-------
   2.0
o
i—i
i—

LU
_J
LU

Q
LU
rM
i—i
_J

4- D
r o
**
i i ill
LEGEND
DATE H C DATE H C
t> 6/03/77 1676 .35 0 6/18/77 25OO .30
O 6/0^/77 1219 .^0 X 6/28/77 152^ .18
•ft 6/CH/77 128O .^8 ^ 6/ZO/77 1220 .33
<> 6/05/77 1159 .25 • 6/30/77 1220 .95
O 6/08/77 18Z9 .Z3 * 7/06/77 990 .30
D 8/08/77 1*28 ,*S • 7/06/77 137Z .33
+ 6/18/77 2SOO .1Q * 7/O7/77 1676 .25
H-MIXED-LAYER DEPTH Cm)
C-AVERA^GE MIXED-LAYER CONCENTRATION (Mg/m3)
SUBSIDENCE
INVERSION
O
»
t>
*K * *
x o
<-> D
m t> TYPICAL RADIATION
w <-> INVERSION HEIGHT
o
^ •%
• 21 x£<> ^01 +
i i i i ili^itili
                                                                         I
                                                                         Ul
     0
            .4       .8       1.2      1.6      2.0

           NORMALIZED NITRATE PROFILES  (C/C)
            Figure 27.  Typical early morning nitrate profiles.

-------
   LU
   U
   z
   <
   h-
   O)
   1—I
   Q

   H
   a:
   o
   Q_
   U)
   z
   <
   cr
   D
   Z
   CJ)
        3OO
250
2OD
150
        1 DO
         50  -
           0
                                 LEGEND
                - Z/ 10/76
             O - 2/ 1 9/76
             O ~ 3/11 /76
             [> - 3/18/76
             ^ - 2/22/76
             O - 2/24/76

             NOTE-DAYS OF  INFLOW AND OUTFLOW
             MEASUREMENTS  PLOTTED WITH EARLIEST
             INFLOW POINT  AT ORIGIN
> - 6/03/77
O - 6/0^/77
O - 6/Z8/77
D - 6/30/77
+ - 7/06/77
     LANCRANCIAN
     MEASUREMENT
        REGION
                         EULERIAN
                       MEASUREMENT
                          REGION
O
O
                                                                  O
                      50      1OO      150     200      250     3OO


                   OUTFLOW  MEASUREMENT  DISTANCE   Ckm)
Figure  28.   Airmass transport  distance vs. inflow-to-outflow sampling distance.

-------
                                -58-


     The bias toward  sampling during southwesterly flow conditions results
 from three  factors:

       1.  As stated previously, "typical" meteorological conditions were
          desired.  As evidenced from Section 3, flow from the southwest
          is typical.

       2.  High sulfate pollution episodes have been documented with southwest-
          to-northeast flow of maritime tropical air from the Gulf of
          Mexico  (Perhac 1977; Smith and Niemann 1977).

       3.  Wind velocity sufficiently steady and strong enough to allow
          for Lagrangian measurements occurred almost exclusively with
          flow from the southwest.
Tabular Summary of Flux Calculations--

     The Lagrangian and Eulerian aircraft sampling data for both 1976
and  1977 are summarized in Table 6.  All days when Valley-wide measure-
ments were made are listed in this table.  The complete data set for all
aircraft measurements is given in Appendix D.  Due to instrument problems,
only relative changes can be analyzed for the 1977 measurements made by
the Meloy 202 total sulfur instrument.  This instrument was almost exclu-
sively used for measurements at the northern end of the Valley (i.e.,
Scottsville, Kentucky, area), and its readings are identified by a super-
script "i" after the total sulfur reading.  In the following sections,
the data from Table 6 are analyzed by measurement type.  A detailed
synoptic meteorological summary for the Lagrangian and Eulerian days is
given in Appendix E.
EULERIAN SPACE AND TIME VARIATIONS

     The Eylerian space and time variations of four pollutants--total
sulfur, S04, N03, and NH4--are explored.  Eulerian space variations are
explored by grouping measurements that were made at similar times, but
in different horizontal or vertical space locations.  Similarly, Eulerian
time variations are explored by grouping measurements that were made at
the same locations, but at different times.  Aircraft measurements fitting
into these two groups occurred exclusively during the 1977 summer study.
The aircraft data used in these analyses are taken from Table 6.

     High-volume measurements of total suspended particulates and 864
are taken from both study periods.  The high-volume sulfate is analyzed
separately from the aircraft sulfate because of an interesting discon-
tinuity between the two measurements.  This discontinuity is discussed
further in the Vertical Variations subsection.
Eulerian Space Variations

     Eulerian measurements were separated in both horizontal and vertical
space.  Typically, measurements were made at two locations separated by
an average horizontal distance of over 200 km.  These sampling locations

-------
TABLE 6.  SUMMARY OF LAGRANGIAN AND EULERIAN MEASUREMENTS
 Sampling Days With Both Inflow and Outflow Measurements
Pollutant concentration
Date
2/10/76
2/19/76

3/1J/76
3/18/76
3/23/76
3/24/76

6/3/77




6/4/77




6/28/77





6/30/77





7/6/77





Time3
0916
1526
0927
1455
0908
1425
083S
1418
0834
1345
0800
1332
0743
1030
1110
1440
1505
0726
0717
1030
1453
1518
0723
1025
1025
1345
1335
1651
0637
0713
1033
1000
1345
1406
0646
0815
1030
1240
1425
1533
b c d 	
Sampling WD WS DST
methodology (•) (m/s) (tern) SO,
L 217 13.6 293 1.8
L 274 9.8 200 1.2

E 200 3.6 278 4.5
L 226 10.5 234 3.2
E 210 2.5 250 2.9
I 198 9.5 242 2.8

E 021 3.5 190 1.5
3.6

1.6

E 004 1.3 190 3.8

4.1
6.6

E 240 8.7 234
3.4

3.9

3.4
L 233 9.7 234 1.9

2.6

3.2

E 357 2.0 234
19.3

11.7

16.3
Inflow
WH,
0.5
0.4

1.3
0.8
1.8
0.9

0.4
0.9

LR

1.1

1.5
1.8


0.7

0.5

0.5
0.4

0.5

0.9


2.1

2.0

2.6
rat/so; TS' ml o3
1.5
1.9

1.6
1.4
3.2
1.7

1.3
1.4

LR

1.6

2.0
1.4


1.0

0.7

0.8
:. i

0.9

1.4


0.6

0.91

0.84
42
67

49
71
85
52

LR
37

LR

LR

31
22


237

134

93
63

76

62


40

51

45

-------
TABLE 6  (continued)
                                                     Other Eulerian Days



Date
6/5/77





6/8/77



7/7/77





Time3
0702
1137
1500
0658
1045
1430
0716
1105
0645
1015
0721
1115
1100
1428

Sampling
class
0
0
0
N
N
N
N
N
B
B
N
N
0
0

WD WS
(°) (m/s)
327 5.0


285 2.9


287 4.4

VRB

274 4.9

317 3.2

Pollutant concentration (\ig m 3)e

S04
22.7
7.8
10.8
5.9
7.4
13.1
1.3
2.0
3.6
1.8
8.5
9.2
12.1
22.8
-f.

NH4 NH4/S04
3.0
2.9
1.6
1.3
2.2
3.7
0.4
0.9
1.3
0.7
1.3
2.1
3.2
2.9
0.7
2.0
0.8
1.2
1.6
1.5
1.7
2.4
2.0
2.2
0.8
1.2
1.4
0.7




TS1 N03 03 H (m)*
186
57
52
21
42
30
45
34
89
61
25
31
103
80
0.3
0.6
0.5
0.2
0.7
0.4
0.2
0.3
0.4
0.3
0.2
0.1
0.5
0.5
_
-
-
215
246
267
LR
141
-
-
98
111
-
—
1677
1677
1677
1677
1677
1677
1829
1829
1829
1829
1900
1900
1900
1900
Pollutant
fluxE
(Mg m'2 s"1)

• S04
114.2
41.7
61.9
16.6
31.8
28.8
9.0
9.6
NA
NA
40.9
48.9
41.7
105.7
.
TS1
934
307
300
60
180
66
323
161
NA
NA
160
167
355
369
Center point
of traverse
Threet, AL
Center Star,
Anderson, AL
Scottsville,
Scottsville,
Portland, TN
Franklin, KY
Franklin, KY
Red Bank, AL
Wolf Springs



AL

KY
KY




, AL
Lewisburg , KY
Central City
Waterloo, AL
Red Bank, AL
, KY



               Midtime of sampling traverses.
               L = Lagrangian;  E = Eulerian; B = blob; 0 = outflow;  N = neither.
               Mean wind direction and speed for the sampling day.
               Distance in kilometers between average midpoint of inflow and outflow sampling traverses.
              Abbreviations:  - = pollutant not measured; LR = lost record; NA = not applicable;  
-------
                                -61-
were typically in south-central Kentucky and northwestern Alabama.   These
locations characterized either inflow, outflow, neither, or blob (stag-
nation) conditions, depending on meteorology.  Such measurements were
also made at various elevations.
Horizontal Variations--

     When winds were steady enough, inflow and outflow measurements were
made.  Five such days were sampled during the 1977 summer study.  Again,
the limitation of this data base_to just five summer days allows only
tentative findings.  Plots of 864, NH4, total sulfur, and NOg aircraft
values, listed by measurement type and flow direction, are presented in
Figures 29 through 32.  Intercompari|on by pollutant shows that, with
northerly flow, average values of S04 concentrations (Figure 29) vary
significantly from inflow to outflow and from day to day.  However, with
southwesterly flow the situation is reversed.  Little spatial or temporal
variation is evident.  Although this difference may be due only to the
limited number of days sampled, it could also be due to the variation in
SOg emission density.  Large sulfur sources are immediately upwind from
the inflow (northern) boundary during northerly wind flow, whereas just
the opposite is the case for the inflow (southern) boundary with south-
westerly flow.  Similarly, for ammonium (Figure 30) the northerly inflow
boundary shows greater variability and higher concentrations vs. those
at the southern inflow boundary.  Nitrate concentrations  (Figure 32)
show an opposite correlation, indicating that its major source  regions
differ significantly  (possibly natural vs man-made).  The average
inflow and outflow concentrations and standard deviations by pollutant
for three days with northerly flow and for the two days with south-
westerly flow are shown in Table 7.
TABLE 7.   INFLOW-OUTFLOW CONCENTRATIONS   (pg m"3) BY FLOW DIRECTION

Flow direction
North-northeast
Pollutant
S04
NH4
NO;
Inflow

7.6 (6.6)
1.5 (0.7)
0.3 (0.2)
Outflow

11.2 (5.0)
2.6 (1.3)
0.6 (0.2)
South- southwest
Inflow

3.1 (0.7)
0.6 (0.2)
2.0 (1.0)
Outflow

2.6 (1.0)
0.6 (0.5)
0.5 (0.2)

 Q
 Standard  deviations  in  parentheses.


      Significant  increases  occur in the  average concentrations of all
 three pollutants  from inflow to  outflow  with northerly winds, whereas
 no  significant  change occurs from inflow to outflow with southwesterly
 flow, except for  the  nitrates.   The nitrates show a large inflow concen-
 tration with southerly flow and  a small  outflow concentration.

-------
     \
                                                                                                             ON
                                                                                                             10
                                                                                                      7/07/77




                                                                                                       O NW

Figure 29.  Temporal variation of sulfate (I = inflow, 0 = outflow, N = neither, B = blob; wind
            direction  in upper right corner of each axis).

-------
                                                                                        X
                                                                                                            I
                                                                                                           o
                                                                                                           U>
Figure 30.  Temporal variation of ammonium (I = inflow, 0 = outflow, N = neither, B = blob; wind
            direction in upper right corner of each axis).

-------
                                                                      6/08/77

                                                                       N • w
                                                                       I VB6
                                                                                        7/07/77


                                                                                         O- Nw
Figure 31.  Temporal variation of normalized total  sulfur (I = inflow, 0 = outflow, N =  neither,
            B = blob;  wind direction in upper  right corner of each axis).

-------
                                                                                                    N- w
Figure 32.  Temporal variation of nitrate (I = inflow,  0 = outflow,  N = neither,  B = blob;  wind
            direction in upper right corner of each axis).

-------
                                 -66-
      Traditionally,  total  filterable  sulfates have been measured as water-
 soluble  sulfates  from  glass-fiber, high-volume  filters.  Although serious
 questions  as  to the  adequacy  of  this  sampling method have often been raised,
 governmental  agencies  and  private industry  continue to use the high-volume
 method.  Therefore,  establishment of  a  relationship between high-volume
 concentrations, upper  air  concentrations, and interregional transport
 would be useful.

      High-volume  filter data, mostly  for daytime periods, from the two
 studies  are presented  in Appendix D.  These data are obtained from fixed
 monitors,  most of which are located near TVA power plants (Figure 3).
 Depending  on  wind flow conditions, these sites  are designated as being
 representative of inflow,  outflow, or neither (in between).  A presenta-
 tion  of  inflow-outflow-neither high-volume measurements by wind direction
 is  shown in Table 8.
TABLE 8.  HIGH-VOLUME CONCENTRATIONS  (ug m~3) BY FLOW DIRECTION

Number
of days
7
3
Average
4
12
Average
Weighted
Wind
direction
WNW (270-329°)
N (330-029°)

S (150-209°)
WSW (210-269°)

grand average
Inflow

TSP
79
92
85
62
76
69
75

_
S04
12
15
13
6
8
7
9
.1
.1
.6
.6
.0
.3
.7
Outflow

TSP
93
86
89
86
60
73
79

-~
S04
17
17
17
8
6
7
11
.5
.3
.4
.0
.6
.3
.2
Neither

TSP
79
85
82
55
50
53
65
^
S04
15.4
17.5
16.5
7.2
6.4
6.8
10.5
Average

TSP
84
88
85
68
62
65
71

so;
15.0
16.6
15.8
7.3
7.0
7.2
10.5

     Analysis of these data shows that, similar to the aircraft data,
sulfate concentrations differ significantly with various wind flows.
West-northwesterly through northerly flow is accompanied by high sulfate
concentrations at both inflow and outflow, whereas the opposite is true
for southerly or west-southwesterly flow.  Total suspended particulates
follow the same pattern; only the relative magnitude of the changes are
less.  For all days, the average increase in sulfate from inflow to
outflow is 15 percent, whereas total suspended particulates show no
significant change.
Vertical Variations—

     Typical early morning and midday sulfur profiles are illustrated in
Figures 24 and 25 respectively.  To generate these plots, data were norma-
lized and pooled, regardless of location.  Both plots illustrate a somewhat
unexpected behavior.  The early morning profiles scale well in the vertical
when the previous day's mixed-layer depth is used.  At the lower boundary
(Z/H = 0), the data suggest that through the night SO?, (the main component

-------
                                -67-
of the total sulfur plot) is selectively removed, whereas S04 shows no
apparent tr.end.  This could indicate a significant difference in nighttime
SC-2 vs. S04 deposition velocities.   The midday sulfur profiles of Figure
25 illustrate jyst what is expected—a uniformly mixed layer—except for
ground-level S04, which was determined by high-volume sampling.

     The difference between similar spatial and temporal measurements of
aircraft and ground-level sulfate is graphically shown in Table 9 and
Figure 33.  A paired-t test shows that this difference—aircraft concen-
trations nearly half the high-volume concentrations—is statistically
significant (p = 0.99).  Because most monitoring sites are located at
the power plants, a bias was at first believed to have been introduced
by local source effects.  This bias might result from primary sulfate
emissions, artifact sylfate formation, or a more frequent exposure to
secondarily formed SO..  However, Bailey and Ruddock (1978) have shown
that primary sulfate emissions from TVA power plants are typically low
(~1 percent by molar ratio), and artifact sulfate formation on Gelman
Spectrograde filters, as discussed in the Analytical Methods_subsection,
is also low.  Also, the average conversion rate of 862 to 804 [<2 percent
per hour (Meagher et al. 1977)] and the relatively low deposition velocity
of S04 indicate that this pollutant is long-lived and therefore widely
dispersed.  This is supported by an analysis of high-volume data from
the two field  studies.

     A high-volume sampler located at the Giles County trend station was
compared with  the average of four samplers located at the nearest TVA
coal-fired power plant, the Colbert Steam Plant.  These two measurement
sites are separated by over 100 km.  A plot of the Giles sulfate data
vs. the Colbert data for 18 simultaneous measurements obtained during
the two studies is shown in Figure 34.  A slope statistically not dif-
ferent from one and a  high correlation coefficient (r2 = 0.77) indicate
the regional nature of the sulfate pollution and the representativeness
of the power plant sulfate values in indicating regional sulfate levels.

     Additional  correlations among five trend stations and between  Giles
County and the average of Cumberland and Gallatin Steam Plants,  as  tabu-
lated  in Table 10, show  that,  although all correlations are  good,  the
correlation between Giles and  the average of the two power plants  is better
than the  correlations  between  trend stations.  This  result is probably
due to the variabilities inherent in the measurement technique  itself.
Thus, because  two or more high-volume  samplers exist at each power  plant,
the resultant  smoothed or averaged measurement concentration may be more
representative of regional sulfate levels.

     Terra  and Hilst  (1978) of the Electric  Power Research Institute's
SURE program  reported  midday  ozone bulges  near  the  ground  (surface to
200 m  AGL)  during  their  summer measurement program.   These ozone bulges
are an indication  of  photochemical  reactions and may be  tied to hydroxyl
radical production,  a  favored  reactant with  S02  that probably leads to
sulfate formation.   Unfortunately,  our study had no  ground-level or near-
ground-level  ozone measurements; thus,  this  hypothesis for explaining
the high  S04  readings  measured near  ground  level for the TREATS data
could  not be  checked.

-------
                                -68-
               TABLE  9.   COMPARISON OF  GROUND-LEVEL AND
                   UPPER-AIR  SULFATE CONCENTRATIONS

Date
6/03/77
6/05/77
6/08/77
6/24/77
6/28/77
6/30/77
7/06/77
7/06/77
7/07/77
Sample
class
0
N
B
I
I
I
N
0
N
Aircraft measured Closest High-volume
concentration high-volume concentration
(pg m~3) station(s) (|jg m"3)
5.5
9.3
4.0
2.8
2.8
3.0
19.8
17.9
14.4
Colbert
Colbert
Colbert
Colbert
Johnsonville
Colbert
Paradise and
Gallatin
OACD and
Colbert
Cumberland and
Paradise
17.0
17.0
9.0
4.5
13.0
5.3
27.1
28.7
27.2

I = inflow, 0 = outflow, B = blob, and N = neither.

Each air concentration is an average of at least two measurements
spanning at least 4 h and within 50 to 305 m AGL.

Results from at least two high-volume monitors were averaged.  Both
9- and 24-h results are presented; values for 6/24/77, 6/30/77, and
7/6/77 are 24-h samples.

-------
E
\
O)
   20
   16
CO

h-
LL


CJ
    8
    0
           AIRCRAFT S0^=0.57(HIGH-VOLUME
                    r2= 0.86
     0
                   8      12     16    20     24     28     32     36

                             HIGH-VOLUME SO^(Mg/rn3)
                                                                                 I
                                                                                 CT>
                                                                                 VO
      Figure 33.  Comparison of ground-level to upper-air sulfate concentrations.

-------
 /^v
 «•

 E


 U)
 X
II
 o
 O)
 o
 CJ
 UJ
 -J
                 in
                                COLBERT  S04  CMg/ma)
                 Figure 34.  Giles Couaty sulfate concentration vs Colbert.

-------
                               -71-
           TABLE 10.  SUSPENDED SULFATE CORRELATION MATRIX

Air quality trend

Loves Mill
Loudon
Hytop
Giles County
Land Between
the Lakes
Valley-wide

Giles County
Cumberland
Gallatin
Average
Loves
Mill

86b
92
82
93
97

Giles
County

112b
112
112
Loudon Hytop
0.58a 0.55
(250) (320)
0.69
(185)
89
80 217
89 227
93 98
Selected trend
Giles
County
0.54
(450)
0.58
(225)
0.66
(80)

222
88
and power
Cumberland
0.713
(125)

113
112




stations
Land Between
the Lakes
0.47
(520)
0.45
(350)
0.51
(260)
0.46
(175)

100
plant stations
Gallatin
0.78
(115)
0.85
(110)

112

Valley-
wide
0.73
0.83
0.83
0.83
0.67


Average
0.89
0.93
0.95


 Based on high-volume data collected simultaneously every sixth day;
 parenthesized numbers are the distance between sites in kilometers.

 Number of data points in correlation.
£
 Data from around a power plant were averaged.

-------
                                 -72-


      These analyses  do  not indicate  the reason for the large  discontinuity
 evident in the aircraft vs.  ground-level sulfate  measurements.   However,
 dispersion and deposition theory would indicate that this  phenomenon  is
 an aberration; therefore, we speculate that  this  phenomenon results from
 artifact 804  formation  on the Gelman  Spectrograde filters.  Until  such
 time as an explanation  is forthcoming, the two data sets will be analyzed
 separately.


 Eulerian Time Variations

      Analyses of  the time dependence  of the  various pollutant concentra-
 tions are presented  in  this  section.   The data shown in Figures  29 through
 32 also show  temporal variations for  the various  pollutants measured  at
 different locations  throughout the Valley.   A  temporal analysis  of the
 data indicates diurnal  variations between and  within pollutants.  For
 instance,  the sulfate concentration plots show a  fairly flat profile with
 southwesterly wind flow,  whereas large variations  usually  occur  with westerly
 through northerly flow.   However, the  flux of  sulfate generally  increases
 throughout the day,  with significant departures evident on June  3 and
 5,  1977.   Ammonium,  an  ion thought to  be strongly  tied to  the sulfates,
 also shows a  fairly  flat  profile on the  southwesterly flow days, but has
 significant departures  from  the  sulfate  profile shapes  on  other  days.
 The  molar ratio (JO^/SO^)  presents no  identifiable  temporal trend.  How-
 ever,  a  relatively high ratio  occurred on the  stagnate  or  blob day, June 8,
 1977,  indicating  that a much more aged airmass was  sampled.  The normalized
 measurements  of total sulfur concentration definitely show a trend toward
 lower  values  from morning to afternoon,  whereas nitrate values show no
 discernible pattern.  However, as previously mentioned, the inflow nitrate
 values measured during  southwesterly flow are  significantly higher than
 any  other  inflow  or  outflow  nitrate values measured.

     Some  of  these phenomena a*e explainable.  For  instance, the decrease
 in total  sulfur concentration with time  may be related  to  an increase in
 turbulent  mixing, thus  leading to increased removal.   Conversely, the
 sulfate concentration data do not follow any pattern;  due  to its longevity,
 it should  be  a  more  regional and uniformly distributed  pollutant.  The
 high inflow concentrations of nitrate  during southwesterly flow  indicate
 a significantly different source region  than that within or north of the
 field study area.   Because no large anthropogenic NO  sources exist within
 500 km upwind  from the southwestern field study boundary,  it seems probable
 that a natural  area  type source  region is responsible for  this anomaly.


 LAGRANGIAN FLUX ANALYSIS

     This  section presents (1) the results of Lagrangian field study data,
 (2) comparisons of model predictions vs. actual data, and  (3) additional
predictions based on measured inflow boundary conditions and model simula-
 tion of transport to  the outflow.  Both the field study data and model
predictions support  several unexpected conclusions concerning TVA's
 regional and  interregional impact and the significance of upwind sources.

-------
                                  -73-


Field Results

     Five Lagrangian days (Figure 28) were sampled during the two field
studies—four during the 1976 spring study and one during the 1977 summer
study.  A breakdown of the Lagrangian days and the flux calculations is
shown in Table 6.  This analysis shows that all Lagrangian measurements
were made under relatively brisk southwesterly through westerly flow.
Four of the five Lagrangian measurement days occurred during prefrontal
flow of maritime tropical air.  The only exception was on February 19,
1976, when a weak front moved through the area early in the morning and
westerly flow occurred thereafter.  No significant rainfall occurred on
any of these days; however, cloud ceilings at <2000 m were present on
February 10, 1976, and June 30, 1977.  Also, solar radiation was consider-
ably limited on February 10, 1976, March 24, 1976, and June 30, 1977,
due to high cloud cover.

     Analysis of these limited field data shows that, for the four Lag-
rangian days with both inflow and outflow total sulfur and sulfate mea-
surements, the average daily sulfate flux from inflow to outflow increased
by 47 percent, with a standard deviation of 35 percent, whereas the total
sulfur (gaseous and particulate) flux decreased.  Plots of both the Lagran-
gian and Eulerian inflow-outflow sulfate and total sulfur flux measurements
are shown in Figures 35 and 36.  Although both the inflow and outflow
sulfate concentrations are relatively low, there appears to be a significant
percentage increase in outflow sulfate flux.  Also, total sulfur flux
decreases slightly from inflow to outflow.  However, neither of these
conclusions is statistically significant (due to a high standard deviation
and a limited number of data points).

     The average measured mole ratios of NH4/S04 are 1.5±0.3 at the inflow
vs 1.910.7 at the outflow.  This is a 27 percent increase from inflow to
outflow.  Although this difference may be real, statistically it is not
significant.  The scatter  (Figure 37) for the composite of Lagrangian
and Eulerian sulfate and ammonium measurement comparisons is large.

     If the measured sulfate is assumed to be derived only from ammonium
sulfate [(NH4)2S04], ammonium acid sulfate  (NH4HS04), and sulfuric acid
(H2S04) and if all ammonium is associated with sulfate, then limits can
be placed on the relative abundance of these compounds.  The minimum and
maximum percentages of each compound that could yield the observed ratios
are given in Table 11.  The sulfur acid data includes all sulfate compounds
other than the ammonium salts.

                      TABLE 11.  SULFATE SPECIATION


Compound
(NH4)2S04
NH4HS04
H2S04
Inflow
(%)
50-75
0-50
0-25
Outflow
(%)
91-96
0-9
0-5

-------
 CO
 o   60
(M
 ?  50 i-
X
D
_J
LL
UJ
I—
<
LL
_l
13
LL
h-
D
O
30 -
    20
10
          OUTFLOV  =  1.3(INFLOW)
                                    INFLOW  = OUTFLOW
                                     LEGEND
                                   FLOW

                                2/1O/7Q

                             O  2/1S/7O

                                3/18/76


                             D  6/28/77

                                0/30/77
                                              N
 NE FLOW

O/O3/77

6/04/77

7/06/77
     0
         10     20      30      40      50      60

         INFLOW  SULFATE FLUX  (pg/(m2°s))
                         I
                        -vl
     Figure 35.  Comparison of inflow to outflow sulfate flux.

-------
 CO
 O)
 1
X
CO
O
    1 OOO
     800
     OOO
     4OO
     ZOO -
                           O
           INFLOW = OUTFLOW
                                   LEGEND

                            t> 2/10/76 - 1_ O 3/1 8/76 - 1_
                            n 2/ie/7e - L *• 3/23/70 - E
                            + 3/11/70 - E O 3/2t/76 - L

                            L - L^GRANGIAN  E - EULERIAN
       O
 2OO    4OO    6OO   8OO   1 OOO  1 2OO  1 4OO


INFLOW-TOTAL  SULFUR  FLUX   (ng/ (mz•s ) )
                                                                                   Ln
                                                                                   I
          Figure 36.  Comparison of inflow to outflow total sulfur flux.

-------
-1  2.5
                                             1- 2/10/76
                                             2- 2/19/76
                                             3- 3/11/76
                                               3/18/76
                       6- 3/24/78
                       7- e/04/77
                       8- 0/28/77
                       9- 6/30/77
                                            5- 3/23/7B  10- 7/06/77
                                      MOLE RATIO = 0  (hLSO.J
                          6
8
10    12    14    16    18
                      CONCENTRATION   (pg/m3  )
     Figure 37.  Inflow to outflow concentrations and mole ratio changes.

-------
                                 -77-


     Nitrate (N03) concentrations are all below the minimum detectable
limit for the 1976 samples, whereas for the one Lagrangian day in 1977,
the nitrate averaged 1.1 |jg m 3 at inflow and 0.6 Mg m 3 at outflow.


Model Results

     Equations (5) and (6) of the Transformation-Transport Model subsec-
tion of Section 3 are the analytical models used for the comparisons  and
predictions of this subsection.  Given a set of inflow boundary conditions,
superposition was used to account for multiple sources along a trajectory
path.  The parameters used were specified as B /B = 0.0, V: = 1.0 cm/s,
V2 = 0.5 cm s  ,  K = 0.3 percent per hour; and H was specified as observed
for the given day being simulated.   The deposition velocities are as  sug-
gested by Hicks and Wesley (1978).   Although their sulfate deposition
velocity is high in comparison with other reported values, it has minor
influence on model simulations (see Model Analysis subsection). The S02
to 864 transformation rate of 0.3 percent per hour is intentionally low
with respect to the recommended value of 2 percent per hour (Husar et al.
1977).  This lower rate was selected because it is more representative of
actual measurements made within TVA power plant plumes at lower atmospheric
temperatures (5 to 10°C) and higher plume dilution ratios (Meagher et al.
1977).  Also, chemical transformations within the Tennessee Valley region
should proceed at a somewhat slower rate because of the relatively low
levels of urban pollutants.

     Figure 38 illustrates observed vs. predicted outflow total sulfur
(symbols) and sulfate concentrations (small letters).  The five days pre-
sented were characterized by wind speeds around 8.5 m s 1, mixing heights
of 1530 m, and trajectory lengths of 300 km.  The predictions are surpris-
ingly good considering the simplicity of the model and model input.

     The relative change in total sulfur flux across the TREATS field study
region is illustrated in Figure  39.  Here the observed data points are
represented by circled letters,  and the predicted data points  (using
inflow measured values as boundary conditions) are shown by noncircled
letters.  Although some of these predicted points are noted as Eulerian,
this is only because  of the data tagging system used.  That is, the model
can  only yield Lagrangian  results.  When the modeled results are added,
the  range of inflow-to-outflow flux values is significantly increased.
If the outflow total  sulfur flux decreases, as indicated by the limited
observed data, then the removal  by deposition exceeds TVA's rate of
emission, a conclusion supported by the model.

     Both observed and predicted inflow-to-outflow  sulfate  flux  are  illus-
trated in Figure  40.  This plot  indicates  that,  although  the  outflow  sulfate
flux is about twice the inflow flux, the  average  outflow  sulfate concentra-
tion is only 5.4  pg m 3.

     Model  results indicate that sulfur  emissions from upwind sources
are  primarily responsible  for  the  gaseous  and  particulate flux values
measured.   The following  subsection integrates  the results from this  and
previous  subsections  in an attempt to  identify emission rates and  locations
of  significant upwind sulfur  sources.

-------
                              -78-
ro
 en    7
CD
LJ
CD
LJ
CD
CD

CD
LU
CJ
h—t
CD
or
Q_
2/10/76
2/19/76
3/18/76
3/24/76
6/30/77
                                          k-l/4*/h   V,-1.0cm/s
                                             Vo=0.4cm/s
                                                          oo
         OBSERVED  OUTFLOW  CONCENTRATION   (ug/V)

     Figure  38.   Observed vs. predicted outflow total sulfur (X 102) and
               sulfate concentrations.

-------
                                   -79-
    2000.
    1500 ,
CN
 I
 CO
 D
X
    1000
CO
CD
           SPRING 1976 TREATS DATA
                 E EULERIAN
                 L LAGRANGIAN
                                                          (X-454?, Y=
           SUMMER 1977 TREATS DATA
                 e  EULERIAN
                 1  LAGRANGIAN
             CIRCLED LETTERS ARE
             MEASURED DATA
                           INFLOW
      Figure 39.
Relative change  in total sulfur flux  across the TREATS
field study region.

-------
 CO
CM
 CD
 D
X
o
en
o
_i
U_
h-
Z)
o
       175
       150 ..
125 ..
       100 ."
 75 ..
 50 ..
 25 ..
         0
      SPRING 1976 TREATS DATA

           E  EULERIAN

           L  LAGRANGIAN

      SUMMER 1977 TREATS DATA

           e  EULERIAN

           1  LAGRANGIAN
         CIRCLED LETTERS

        ARE MEASURED DATA
                                                                                                  00
                                                                                                  o
                                                                                                  i
                                            (9
                                            O)
                                                            (9
                                                            in
(O
                               INFLOW  S0=   FLUX
                                              4
             Figure 40.  Sulfate flux changes  across the TREATS field study region.

-------
                                 -81-


Upwind Sources

     A large influx of total sulfur pollutants occurs when the flow is
southwesterly.  This influx is evident from both aircraft (spring and
summer studies) and seasonal and annual ground-level sampling (high-volume
sulfate data).  These inflow (and outflow) fluxes are typically of the same
order of magnitude as the regional emission flux contribution.  The obvious
question is, "What is the location(s) and magnitude of the upwind source
region(s)?"  In trying to answer this question, we look first at the upwind
anthropogenic source regions and second at possible biogenic sources.

     An analysis of man-made SC>2 emissions from the southwest quadrant (Fig-
ures 2 and 3) shows that no large sulfur sources exist within 500 km of the
inflow field study boundary.  Further analysis (private communications with
B. Gilbert, EPA, Region IV, 1979) reveals that the area from central Texas
eastward to the Alabama-Mississippi border, including the States of Arkansas,
Louisiana, and Mississippi, has about the same daily S02 emission rate as all
12 TVA coal-fired power plants (i.e., ~5000 metric tons/day).  To evaluate
the significance of these emissions on the inflow measurements made for the
four Lagrangian days, the transformation-transport model was used.  Average
meteorologic and aerometric parameters were used, and all emissions were
assumed to originate at the approximate center (i.e., central Louisiana) of
the land area included in the upwind source region.  Using typical deposi-
tion velocities (Vj = 1.0 cm s *, V2 = 0.5 cm s 1), Equations (5) and (6)
show that, on the average, the upwind anthropogenic sources contribute
about 50 percent to the measured sulfate flux.  Other man-made sources
farther upwind also probably contributed to the measured inflow values;
however, due to the relatively short half-life of S02 (~16 h) and the low
measured inflow sulfate flux, emissions from any other sources farther
upwind would not be likely to have a very significant effect.

     The above results seem to point to a large biogenic source region as
the only possible explanation for the high influx readings observed.  Vari-
ous large wetland areas (e.g., inland and tidal marshes, bogs, etc.) exist
along the Gulf Coast region of east Texas, Mississippi, Alabama, and Loui-
siana and in Arkansas (Shaw and Fredine, undated).  However, quantification
of the relative importance of each type of wetland in producing natural
sulfur emissions is limited (Adams et al. 1979).  Thus, in a crude attempt
to determine whether these extensive wetlands  (about 8.6 x 104 km2) could
supply the needed gaseous sulfur,  the transformation-transport model was
run (using the boundary conditions used in making the anthropogenic calcu-
lations) backward in time to estimate the emission rates, airmass age, and
average source location needed to explain the  concentrations measured.
These calculations result in an estimated rate of emission of 2.02 Tg y  1
(as S02), a mean age of 10 h, and a source distance of about 300 km.  The
age and distance results  compare well with the central location and mean
transport times from the  principal swampland  areas of Louisiana  (~500 km).
Also, when the estimated  emission rate is applied evenly  across the wet-
land areas,_the average sulfur emission flux  rate from wetlands alone is
~10 g m 2 y"1.  This estimate is one to two orders of magnitude higher
than current  approximations of wetland emissions  (Adams et  al.  1979).
However, our  estimate does not include emissions  from nonwetland  soils  or
emissions from the Gulf of Mexico.  Also,  as  stated by Adams  et al.,  their
approximations likely underestimate actual natural emissions  due  to
sampling system losses.

-------
                                  -82-
                              REFERENCES
 Adams,  D.  F.;  Farwell,  S. 0.; Robinson, E.; and Pack, M. R. 1979.
      Assessment  of biogenic  sulfur emissions in the SURE area.  Final
      report  prepared  for Electric Power Research Institute, Palo Alto,
      California; Research Project, 856-1.

 Altshuller,  A. P. 1973.  Atmospheric sulfur dioxide and sulfate—
      distribution of  concentration at urban and nonurban sites in the
      United  States.   Environ. Sci. Techno1.  7:709-712.

 Altshuller,  A. P. 1976.  Regional transport and transformation of sulfur
      dioxide to  sulfates in  the U.S.  JAPCA 26:318.

 Bailey, E. M., and Ruddock,  H. A. 1979.  Measurements of sulfuric acid at
      Tennessee Valley Authority coal-fired power plants using the condenser
      method.   TVA/EP-79/05,  Division of Environmental Planning,
      Tennessee Valley Authority, Muscle Shoals, Alabama.

 Bolin,  B.; Granat, L.;  Ingelstam, L.; Johannesson, M.; Matesson, E.;
      Oden, S.; Rodhe, H.; and Tamro, C. 0. 1971.  Air pollution across
      national boundaries.  The impact of sulfur in air and precipitation.
      Sweden's Case Study for the United Nations Conference on the Human
      Environment,  P. A. Norstedt and Sons, Stockholm, Sweden.

 Bolin,  B., and  Pearson, C.  1975.  Regional dispersion and deposition of
      atmospheric pollutants with particular application to sulfur pollu-
      tion  over western Europe.  Tellus 24:281-310.

 Braekke, H., Ed. 1976.  Impact of acid precipitation on forest and fresh-
      water ecosystems in Norway.  SNSF Fagrapport FRG, Aas-NCH, Norway.

 Coutant, R. W.  1977.   Effect of environmental variables on collection of
      atmospheric sulfate.   Environ. Sci. Technol.  11:873.

 Crawford, T. L. 1977.  Numerical modeling of complex two and three
      dimensional flow and diffusion problems in the natural air environment.
      Ph.D. Thesis, University of Waterloo,  Waterloo, Ontario.

Eliassen, A., and Saltbones, J.  1974.  Decay and transformation rates of
      503, as estimated from emission data,  trajectories, and measured air
      concentrations.   Atmos. Environ.  9:425-429.

Environmental Protection Agency.  1974a.  Health consequences  of sulfur
      oxides—a report from CHESS, 1970-1971.  EPA-650/1-74-004.

Environmental Protection Agency.  1974b.  Manual of methods for chemical
      analysis of water and wastes,  pp.  168-174.

Environmental Research and Technology, Inc.  1976.   Design of  the Sulfate
      Regional Experiment (SURE).  EC-125, vol.  IX,  prepared for Electric
      Power Research Institute, Palo Alto, California.

Federal Register.  1978.  43(43):8962-9059.

-------
                                 -83-
Frank, N. H. 1974.   Temporal and spatial relationships of sulfate,  total
     suspended particulates, and sulfur dioxide.   Presented at 67th APCA
     Annual Meeting, Denver, Colorado.

Friend, J. P. 1973.  The global sulfur  cycle.   Chemistry of the Lower
     Atmosphere.  New York, pp. 177-201.

Gartrell, F. E.; Thomas, F. W.; and Carpenter,  S.  B.  1963.   Atmospheric
     oxidation of SOj in coal-burning power plant plumes.  Amer. Ind.
     Hyg. Assoc. J.  24:113.

Garvey, J. H. 1975.  Sulfate monitoring in the town of Huntington,  New
     York--statistical analysis.  Engineering Report EN-1405, Long  Island
     Lighting Company.

Gifford, F. A.  1976.  Tropospheric relative diffusion observations.
     J. Appl. Meteorol.  16:311-313.

Hansen, M. H.;  Ingvorsen, K.; and Jorgensen, B. B. 1978.  Mechanisms of
     hydrogen sulfide release from coastal marine sediments to the
     atmosphere.  Liinnol. Oceanogr.  23:68-76.

Hausknecht, D.  F., and  Ziskind, R. A.   1975.  Effects of sulfur oxides on
     the  lung:  An analytical base.  Research Project 205, prepared for
     Electric Power Research Institute, Palo Alto, California.

Hicks, B. B., and Wesley, M. L. 1978.  Recent results for particle deposition
     obtained by the Eddy correlation method.  ERC No.  78-12, Argonne
     National Laboratory.

Hitchcock,  D. R. 1975.  Dimethyl sulfide emissions to the global atmosphere.
     Chemosphere 3:137.

Hitchcock,  D. R. 1976.  Atmospheric sulfates from biological  sources.
     JAPCA   26:210-215.

Holzworth,  G. C. 1972.  Mixing  heights, wind speeds,  and potential for
     urban  air  pollution throughout the contiguous United States.  EPA Pub.
     No.  AP-101, Environmental  Protection  Agency, Office of Air Programs.

Husar, J. D.; Husar, R. B.;  and Stubuts, P. K. 1975.  Determination of
      submicron  amounts  of  atmospheric  particulate sulfur.  Anal. Chem.
      47:2062-2064.

Husar, R.  B.; Lodge, J. P.;  and Moore,  D.  J., Eds.  1977.  Sulfur in the
      atmosphere.   Dubrovnik Conference, Pergamon  Press, Oxford, England.

Jutze, G.  A., and  Foster,  K. E.   1967.  Recommended  standard method for
      atmospheric sampling  of fine  particulate matter by filter media--
      high-volume sampler.   JAPCA  17:17.

Korshover,  J.   1976.   Climatology  of stagnating  anticyclones east  of  the
      Rocky Mountains.  1936-1975.   NOAA Technical Memorandum ERL ARL-55,
      Air Resources Laboratory,  Silver  Springs, Maryland.

-------
                                 -84-
Liggett, W. S., and Parkhurst, W. J.  1977.  Suspended particulates and
     sulfates at rural locations in the Tennessee Valley.  TVA Internal
     Report I-AQ-77-14, Division of Environmenal Planning, Tennessee Valley
     Authority, Chattanooga, Tennessee.

Lioy, P. J.; Wolff, G. T.; Czachor, J. S.; Coffey, P. E.; Stasiuk, W. N.;
     and Romano, D. 1977.  Evidence of high atmospheric concentrations of
     sulfates detected at rural sites in the northeast.  J. Environ. Sci.
     Health  A12:  1-14.

Liu, B.H.Y., and Lee, K. W.  1976.  Efficiency of membrane and nucleopore
     filters for submicrometer aerosols.  Environ. Sci. Technol.
     10:345-350.

Lovelock, J. E.; Maggs, R. J.; and Rasmussen, R. A.  1972.  Atmospheric
     dimethyl sulfide and the natural sulfur cycle.  Nature 237:452.

Meagher, J. F.; Stockburger, L., III; Bailey, E. M.; and Huff, 0.  1977.
          Chemical interaction in coal-fired power plant plumes.   II.
          Sulfur dioxide oxidation - a progress report.  TVA Internal
          Report I-AQ-77-20.  Division of Environmental Planning, Tennessee
          Valley Authority, Muscle Shoals, Alabama.

Melo, 0. T.  1978.  Aerosol and precipitation sampling in southern Ontario -
     1976 results.  Internal Report No. 78-125-K, Ontario Hydro.

Meteorology Research, Inc.  1972.  Integrating nephelometer and recorder -
     instrument manual.  Altadena, California.

Monin, R. S., and Yaglom, A. M.  1971.  Statistical fluid mechanics:
     Mechanics of turbulence, vol. 1.  MIT Press, Cambridge, Massachusetts.

Mulik, J.; Puckett, R.; and Sawicki, E.  1976.  Ion chromatographic analysis
     of sulfate and nitrate in ambient aerosols.  Anal. Lett. 9:653-663.

Perhac, R. M.  1978.   Sulfate regional experiment in Northeastern United
     States:  The "SURE" program.  Atmos.  Environ. 12:641-647.

Reisinger, L. M., and Crawford, T. L. 1979.  August 1976 sulfate  episodes
     in the Tennessee Valley region.  TVA/EP-79/04, Division of
     Environmental Planning, Tennessee Valley Authority, Muscle Shoals,
     Alabama.

Reisinger, L. M., and Sharma, V.   1977.  Regional sulfur dioxide  emission
     inventory.   TVA Internal Report I-AQ-77-18, Division of Environmental
     Planning,  Tennessee Valley Authority, Muscle Shoals, Alabama.

Roberts, P.  T.,  and  Friedlander, S. K.  1975.  Conversion of S02 to sulfur
     particulate in the Los Angeles atmosphere.   Environ. Health  Perspect.
     10:103-108.

Rowe, M. D.; Morris,  S. C.; and Hamilton,  L.  D.   1978.   Potential ambient
     standards for atmospheric sulfates:  An account of a workshop.
     JAPCA 28:772.

-------
                                 -85-
Robinson, E., and Robbins, R.  C.   1972.   Emissions,  concentrations and fate
     of gaseous atmospheric pollutants.   Air pollution control,  vol.  II,
     Strauss, W., Ed., pp. 1-93.   Wiley (Interscience),  New York.

Scriven, R. A., and Fisher, B.E.A.  1975.  The long-range transport of air-
     borne material and its removal by deposition and washout.   I. General
     considerations.  Atmos. Environ.   9:49-58.

Shaw, S. P., and Fredine, C. G.  Undated.  Wetlands  of the United States.
     Circular 39, Fish and Wildlife Service, U.S. Dept.  of the Interior.

Smith, F. B., and Jeffrey, G.  H.   1975.   Airborne transport of sulphur
     dioxide from the United Kingdom.   Atroos. Environ. 9:643-659.

Smith, L. F., and Niemann, B.  L.   1977.   The Ohio River Basin Energy Study:
     The Future of Air Resources and Other Factors Affecting Energy
     Development.  Paper presented at the Third International Conference
     on Environmental Problems of the Extractive Industries, Dayton, Ohio.

Teknekron.   1978.  An integrated technology assessment of electric utility
     energy  systems.  Vol. Ill:  Air quality impact model and results.
     Teknekron, Inc., Berkeley, California.

Terra, S., and Hilst, G.   1978.  SURE takes to the air.  EPRI J.  12:14-17.

Tony, E. Y., and Batchelder, R. B.  1978.   Compilation and analysis of data
     sets  for the evaluation of regional sulfate models.  Teknekron, Inc.,
     Berkeley, California.

Trijonis,  J., and Yvan, K.  1978.  Visibility in the Northeast—long-term
     visibility trends and visibility/pollutant  relationships.  EPA
     600/3-78-075, Environmental Protection Agency, Research Triangle Park,
     North Carolina.

West, P. W., and Gaeke,  G.  C.  1956.  Fixation of sulfur  dioxide  as sulfi-
     tomercurate III  and  subsequent colorimetric determination.   Anal. Chem.
     28:1816.

Yocom,  J.  E.,  and Grappone, N.  1976.  Effects of power plant emissions on
     materials.  EC-139,  Summary  report  prepared for Electric Power Research
     Institute,  Palo  Alto, California.

-------
        A-l
       APPEKDIX A



SYNOPTIC WEATHER TYPING

-------
                                  A-2


                               APPENDIX A


                          SYNOPTIC WEATHER TYPING
      A quasi-objective  weather  typing  scheme has been  devised.  The area
 of  interest  is  defined  by  a  500-nautical-raile  (nmi)  radius from Nashville,
 Tennessee.   Data  for  five  parameters were  obtained from NWS daily weather
 maps  for  every  sixth  day for a  5-year  period beginning on November 5,
 1973.   Thus,  over 235 data points were generated for each parameter.  These
 five  parameters describe (1) predominant frontal systems and associated
 pressure  centers;  (2) distance  (range) of  the most significant pressure
 center from  Nashville,  Tennessee; (3)  direction from Nashville, Tennessee
 (degrees  from true north)  of this pressure center; (4) airmass type (e.g.,
 maritime  tropical, continental  polar); and (5) relative frequency of
 measurable precipitation (>0.01 in.).

      These parameters are  further divided  into categories.  These
 categories and their annual  and seasonal distributions are presented in
 Tables  A.I through A.5.

     Annually, 27  percent  of the frontal system-pressure center parameter
 can be  described by the high-pressure  center (HPC) without front category,
 while  the next most frequent occurrence is divided almost equally between
 HPC associated with cold fronts west of Nashville or with stationary
 fronts  (~16 percent each).   Seasonally, the only significant variation
 occurs  during the  summer months, when  stationary fronts account for 36
 percent of the weather occurrences.

     Joint frequency distributions (JFD) between some of the parameters
were analyzed.  First, comparisons between the distance and pressure-
 frontal parameters indicate that, although HPC without cold fronts occur
most frequently, their relative distance from Nashville is often within a
 100- to 399-nmi range (53 percent).   Directional considerations indicate
that most of the pressure centers within this range  (28 percent) occur in
the 045- to 089-degree sector.   Also, for pressure centers greater than
500 nmi, 46 percent occur within sectors 2 and 3 (045 to 134 degrees),
whereas an additional 31 percent occur within sectors 7 and 8 (270 to 359
degrees).   Seasonally, the  annual JFD varies  only slightly for the pressure
system defining parameters  1, 2, and 3.

     An analysis of the direction and airmass parameters shows that,  when
pressure centers are southeast of Nashville,  there is an 81 percent
chance that the airmass type affecting the Tennessee  Valley region will
be of maritime tropical origin.   This,  no doubt,  is in response to the
summertime Bermuda high (48 percent  occurrence).   For the approximately
50 percent of the time when pressure centers  are within a 500-nmi  radius
of Nashville, Tennessee, the most favored range is from 300 to 399 nmi
 (16 percent).  Except for the range  from 0 to 99 nmi  (3 percent),  the
other three categories all  occur at  about the same frequency (11 percent).
Seasonally, summer has the  largest frequency  (23 percent)  for the  300- to
399-nmi range.

-------
                                A-3



TABLE A.I.  PERCENT FREQUENCY DISTRIBUTION FOR PRESSURE-FRONTAL PARAMETER

a
Weather type
Cold front west with HPC
Cold front east with HPC
Cyclone with or without front
HPC without front
Stationary front with HPC
Stationary front with LPC
Stationary front without
pressure center
East-west cold front south
of BNA with HPC
East-west cold front north
of BNA with HPC
Season
Winter
16
15
5
38
11
8

0

3

3
Spring
20
8
8
26
16
7

0

7

8
Summer
11
2
7
20
23
13

7

8

10
Fall
17
16
0
26
17
6

0

6

12
Annual
16
10
5
27
17
8

2

6

8

 a
 Abbreviations:  HPC  = high-pressure  center;  LPC  =  low-pressure  center;
 BNA = Nashville station.

-------
                            A-4
TABLE A.2.  PERCENT FREQUENCY DISTRIBUTION FOR RANGE PARAMETER

Season
Distance (nmi)
0-99
100-199
200-299
300-399
400-499
>500
Winter
2
11
8
13
16
49
Spring
3
18
7
13
13
46
Summer
2
9
11
23
14
41
Fall
3
7
12
17
9
52
Annual
2
11
9
17
13
47

-------
                        A-5






TABLE A.3.  PERCENT DISTRIBUTION FOR DIRECTION PARAMETER

Direction
(degrees from
true north)
0-44
45-89
90-134
135-179
180-224
225-269
270-314
315-359
Season
Winter
5
23
13
7
10
13
15
15
Spring
7
25
16
8
10
5
20
10
Summer
4
29
23
13
4
5
9
14
Fall
3
28
16
3
3
10
14
23
Annual
4
26
17
7
6
9
15
16

-------
                          A-6






TABLE A.4.  PERCENT DISTRIBUTION FOR AIRMASS TYPE PARAMETER

Season
Airmass type
Continental polar
Modified maritime polar
Maritime tropical
Modified continental polar
Modified maritime tropical
Winter
21
25
25
25
5
Spring
8
15
44
30
3
Summer
0
3
75
18
3
Fall
10
20
28
36
6
Annual
10
16
42
27
4

-------
                             A-7






TABLE A.5.  PERCENT FREQUENCY DISTRIBUTION FOR RAINFALL PARAMETER

Number of stations
recording precipitation
0
1
2
3
4
Season
Winter
48
15
10
11
16
Spring
41
10
20
11
18
Summer
39
23
18
16
3
Fall
57
12
16
6
10
Annual
46
15
16
11
12

-------
                                A-8
     Pressure center  locations occur most  frequently  (26 percent) in the
045- to 089-degree sector from Nashville;  three other sectors  (090 to
134 degrees, 270 to 314 degrees, and 315 to 359 degrees) all have occur-
rences at about the same frequency  (16 percent).  Again, the most signi-
ficant departure occurs during the  summer, when 52 percent of all pressure
centers are located in the 045- to  089-degree and 090- to 134-degree sectors.
Also, during the autumn, 51 percent of the directional location is defined
by two sectors, the 045- to 089-degree and the 315- to 359-degree sectors.
These results indicate the strong influence of the Bermuda high on weather
patterns in the southeastern United States.

     Annually, the airmass typing parameter is dominated by maritime tropical
airmasses (42 percent), with the next most frequent type being modified
continental polar (27 percent).  Wide variations occur seasonally.  Winter
is almost equally divided among all the airmass types; spring shows a
bias toward maritime tropical (44 percent) and modified continental polar
(30 percent); summer is strongly dominated by maritime tropical (75 percent);
and autumn has a slight bias toward modified continental polar (36 percent).

     Annually, an analysis of the rainfall parameter indicates that 46
percent of the time no rainfall occurs at any of the four stations, and
that an almost equal probability exists of having either one or two
stations recording measurable precipitation (16 percent).

     JFD analysis of the frequency of precipitation vs.  airmass type indi-
cates that,  when maritime tropical airmasses are affecting the area, a
70 percent chance exists that at least one station will receive measura-
ble precipitation.   Conversely, when other airmass types are influencing
the area,  only a 40-percent chance exists of at least one station receiving
measurable precipitation.

-------
               B-l
            APPENDIX B




TABULATION OF TREATS 1976 and 1977




     AIRCRAFT FIELD STUDY DATA

-------
                                   B-2
                      ABBREVIATIONS AND UNITS FOR APPENDIX B
 DATE      Date of sample,  month,  day,  year

 TIME      Midpoint time of sample,  h and  min

 SHIP      Aircraft used for sample,  B-deHavilland  Beaver U6-A, H-Bell
           47A helicopter

 XFROM      X-coordinate* at start  of sample,  km

 YFROM      Y-coordinate  at  start of  sample, km

 XTO        X-coordinate  at  end of  sample,  km

 YTO        Y-coordinate  at  end of  sample,  km

 ELEV      Typical  elevation of sample, m  AGL

 STYPE      Sample  type,  T (traverse), S (spiral)

 SCLASS     Sample  classification,  I  (inflow), 0 (outflow), N  (neither I or 0),
           B  [blob  (no significant airmass transport)], T (travel from inflow
           location  to outflow and vice versa)

 804        Sulfate ion concentrations, pg m 3

 N03        Nitrate ion concentrations, pg m 3

 NH4       Ammonium  ion concentrations,  |Jg m 3

 TS        Total sulfur concentrations,  pg m 3 as SOg

 Oa        Ozone concentrations, |Jg m 3

 BSCAT     Atmospheric extinction coefficientjdue to light scattering by
          both gases and particulate, E-4 m

 TEMP      Ambient temperature, °C

 DP        Dewpoint temperature, °C
*The origin of the coordinate system is Nashville, Tennessee.  The Y-axis
 is rotated 1.8 degrees east of true north.

-------
                                               THEATS AIHCHAFT FIELD STUDY DATA
DATE  TIME SMP XFROM YFRCM   XTO  YTu   tLEV STYPt SCLAiS  SO* N03 NH4  TS  03 BbCAT  TF.MP   DP    COMMENT
10676 1214
10676 1511
21076 915
21076 917
21076 1312
21076 14*2
21076 1611
21176 1205
? i Q7 A R A A
c 1 ** r O O * II
21976 1015
21976 1*07
21976 150*
21976 1544
22076 754
22076 927
22*76 1307
22*76 1307
31176 901
31176 915
31176 122?
31176 13*0
31176 1510
31176 163B
31876 819
31876 857
31876 1213
31876 133*
31876 1502
31876 1636
31976 709
31976 739
31976 1125
32376 823
32376 845
32376 1119
32376 1239
32376 1452
32376 1630
32476 751
32*76 806
32*76 1111
32*76 1221
32*76 1*36
32*76 1610
52077 13*1
52077 142?
5?077 1451
52077 15!5
52077 1541
60177 1650
60177 1750
60277 1*15
60?77 1505
60277 1535
e
B
B
a
B
B
B
B
Q
B
H
B
ft
B
B
B
B
B
R
R
B
B
B
Q
R
B
B
B
B
P
e
R
B
B
P
R
H
B
H
R
P
B
B
P
p
a
P
H
H
B
e
B
p
p
-204
-HP
-e
-182
-76
30
126
30
l n i
— 1 u 1
-147
7V
-If
75
-76
92
-6
-18?
-6
-182
-76
-13
126
-13
-US
-7H
-76
113
-14
113
-119
-76
-76
-p
-1?2
-76
-14
126
- 14
93
-119
-76
-8
126
-14
-76
-31
-76
-61
-27
-78
40
30
101
76
-16.6
-184
-219
-60
-156
92
146
9'
"'ts
-151
-156
93
-156
-160
-?1S
-60
-21S
-60
-156
6e
146
be
-156
-156
-156
-12
140
2
-156
-156
-156
-21S
-60
-156
75
14t
75
-Ind
-156
-156
75
146
75
-156
-Ib3
-156
-15C
-171
-156
70
92
167
1Z4
-«B
-204
-182
-£
30
126
30
-7e
-101
75
T~i
7V
92
-76
-1»2
-tf
-1S2
-e
-13
12e
-13
-7e
V j
VJ
113
-14
113
-7e
83
93
13
-1N2
-e
-14
126
-14
-7e
-m
93
-14
1 ,14
*3 V14
-151 1676
-151 l»-7b
-lr>0 152
-iDb 914
-CO MO
-219 ItlS
-60 152
-21V 6.10
6" \i 1*
14*1 1E19
DM M 0
- lab 610
-leO \^

-------
                                              TREATS AIKCKAFT FIELD STUDY DATA




DATE  TIKE SHIP XFROC  YFKOM   XTO  YTO ELEY STYPE SCLASS SO*  N03 NH*  TS 03  BSCAT  TEwH
                                                                                             DP  COMMENT
60377 1615
60377 705
60377 730
60377 755
60377 625
60377 1015
60377 1045
60377 1115
60377 1425
60377 1*55
60377 1525
60377 1040
60377 1115
60377 1135
60377 1425
60377 1455
60377 1530
60377 1555
60477 640
60477 715
60477 740
60477 810
60477 1015
60477 1045
60477 1115
60477 1145
60477 1425
60477 1450
60477 1525
60477 1550
60477 610
60477 645
60477 725
60477 755
60477 1455
60477 1540
60577 630
60577 655
60577 730
60577 800
60577 1015
60577 1045
60577 1115
60577 1145
60577 1415
60577 1*45
60577 1515
60577 615
60577 645
60577 720
60577 750
60577 1125
60S77 1150
60577 1*30
B
B
B
B
e
B
B
B
8
B
B
H
H
K
H
H
H
H
B
B
B
B
e
a
B
B
B
8
B
B
H
h
H
H
h
H
B
B
e
B
B
B
e
a
B
B
B
H
H
H
f
H
H
H
101
30
83
30
83
30
76
30
30
64
30
-If
-131
-US
-78
-99
-132
-105
30
27
51
?7
56
28
56
28
51
27
51
27
-78
-112
-78
-112
-77
-29
30
99
30
99
30
99
30
99
30
2P
30
-87
-124
-87
-111
-76
-*fi
-78
167
92
46
92
46
92
7
92
92
23
92
-156
-105
-37
-156
-117
-72
-114
92
25
101
25
101
20
101
20
101
2S
101
25
-1S6
-104
-156
-104
-165
-162
92
23
92
23
92
23
92
23
92
20
92
-152
-115
-152
-115
-156
-125
-156
30
«3
30
*3
JO
76
30
76
64
30
53
-131
-144
-114
-99
-1J2
-105
-B4
27
51
<;7
51
£8
56
28
56
27
bl
27
51
-112
-78
-112
-78
-29
-78
99
30
99
30
99
30
99
JO
26
30
16
-124
-b7
-\e<>
-a l
-46
-78
-15
92 3048
46 152
92 457
46 1067
92 Ib24
7 610
92 1214
7 lfj«b
23 610
92 1128
27 Ib46
-117 305
-87 305
-13* 1067
-117 305
-72 305
-114 762
-152 762
25 152
101 *57
25 1067
101 1524
20 457
101 91*
20 1372
101 1629
2b 457
101 91*
25 1372
101 1829
-10* 122
-156 305
-10* 610
-156 914
-162 762
-156 305
23 122
92 2**
23 *57
92 1372
23 *57
92 91*
23 1372
92 213*
20 610
92 121V
40 1829
-115 152
-152 305
-115 762
-152 152*
-125 305
-156 762
-123 1219
S
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
r
T
T
T
T
T
T
T
T
7
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T








I
I
1
0
0
o
0
0
0
0
N
fc
N
N






I
I
I
I
0
0
0
0
N
N
N
N
N
N
H
N
N
N
N
0
0
0
0
0
0
0
6.4 0.* 1.7
1.* 0.* 0.3
1.* 0.* 0.5
1.8 O.o 0.3
3.1 0.2 0.6
3.2 0.* 0.0
4.2 0.3 1.1
0.2 0.2 0.0
0.7 0.6 0.3
i.5 0.9 3.0
1.8 0.8 0.*
. O.I 0.0


S.6 0.8 2.1
4.3 0.6 1.5
. 0.1 1.6
•..3 0.5 1.5
5.8 0.6 1.8
2.9 0.2 0.9
3.0 O.I 1.0
0.3 0.1 0.0
O.h 0.* 2.1
4.3 0.1 1.3
1.9 0.1 0.5
0.1 0.1 0.0
7.1 0.3 1.9
7.9 0.2 2.2
6.1 0.3 1.6
0.0 0.2 0.0
6.5 0.7 1.8
2.5 0.3 O.b
*.5 0.5 1.2
*.2 0.4 1.1
7.0 0.6 2.2
6.3 O.b 2.0
6.9 0.3 1.8
7.7 0.3 1.8
11.9 0.3 2.3
0.5 0.1 0.0
3.9 0.8 0.9
11.9 1.1 3.6
9.6 0.2 3.0
0.9 0.1 0.0
14.3 0.5 4.1
11.9 0.3 3.3
1.4 0.0 0.4
7.6 0.6 1.8
11,7 0.1 1.6
. 0.0 0.0
1.1 0.7 1.3
B.7 0.6 2.5
9.2 0.6 3.3
10. -i 0.* 2.6
. *51
* •
• *
• *
• «
*9
26
12
• *
• »
• •
• •


*6
68
82
88
• •
• •
* »
• •
*9
26
15
» •
32
1*
* »
• •
31
51
67
70
55
6*
16 17
32 284
** 279
6 200
56 <:97
35 273
. 1S6
.106
36 292
2* 237
. 267
37
330 .
73
77
53
62
52
150
100
76
MS
121
111
111
71
b.O 1.0 SPlRAL-lOtOOO' TO 3700
ItJ.b 8.0
19.0 7.5
1S.O *.S
9.0 4.0
1B.O 8.S
14.5 7.0
12.0 -2.0
i







. 21.5 8.5 FIDE EXTINGUISHED DISCHARGED
8*
8*
100
121
ry
71
66
71
76
150
129
iei
61
139
ISO
89
55
150
171
161
55
100
71
76
79
82
8*
300
200
200
50
261
250
79
61
261
326
221
116
139
116
239
1*5
134
229
17.0 7.5 FlrtE EXTINGUISHED DISCHARbED
13.0 1.0 FIRE EXTINGUISHER PROBLEMS
22. b 3.0 POKER SUPPLY PROBLEMS
22. b 3.0 POMER SUPPLY PROBLEMS
16.0 7.0 POHER SUPPLY PROBLEMS
2*.S 9.S
24. b 9.0
20.0 8.5 SO* QUESTIONABLE
20. S 6.S
30. 5 10.5
21.0 9.5
15.5 T.U
16.5 -20.0
21.0 10.5
17.0 9.0
16.0 -2.5
15.5 -20.0
2*.0 10.0
20.0 9.0
16.0 7.5
16.6 -20.0
22.0 10.0
23.0 8.5
20.5 8.5
17.5 7.0
22.0 6.5
37.0 7.0
19.1) 14.0
24. b 11.0
24.0 9.5 0732 SALLATIN PLUME
19.0 -14.0
24.0 10.5
20.0 10.0
19.0 7.0
15.0 2.0 SLIDES 5-8
25.5 13.0
20.0 12.0 SLIDES 9 1 10
15. 0 10.5
25.0 10.0
25.0 9.5 COLBERT PLUME
22.0 8.0
15.5 10.5
26.0 13. 0
21.0 12.0
16.5 11.5












































-------
                                           TREATS AIRCRAFT  FIELD STUDY DATA
DATE  TI^E SHIP  XFRO  YFrtOM XTO YTO  F_LEv STYPE SCLASS S04   N03  NH4   TS  03 BSCAT TEMP  OP   COMMENT
60577 1530
60777 1700
60777 900
60877 635
60877 70S
60877 730
60877 755
60877 1020
60877 1050
60877 1120
60877 1150
60877 1530
60877 1420
60877 545
60877 625
60877 700
60877 730
60877 1015
60877 1050
60877 1120
60877 1155
60877 141B
60877 144E
60877 1550
60877 1615
61377 1115
61377 1145
61377 1215
61377 1250
61377 1305
61377 1230
61377 1305
61877 645
61877 715
61877 1155
61877 1220
61877 1310
61877 1335
61877 625
61877 710
61877 1220
61877 1305
61877 1405
62477 1110
62477 1135
62477 12(15
62477 1350
62477 1415
62477 1440
62477 1520
62477 1550
62477 1655
62477 1725
62477 1610
H
R
H
R
R
R
e
R
R
R
R
R
H
H
h
M
H
H
H
H
H
H
H
H
H
8
R
H
H
f>
H
H
H
R
H
f>
R
H
H
H
H
N
H
a
R
ft
R
U
R
R
n
R
H
9
-IS
-78
-78
30
16
19
16
19
16
19
16
30
14
-78
-30
-78
-30
-7fi
-45
-7?
-*5
-52
-SO
-SI
-ei
-77
-4%
-7M
-36
-77
-78
-*5
-77
-40
-7B
-47
-11
-11
-78
-39
-? H
-119
-77
-29
-55
-4?
16
-5
16
51
74
51
H
11
-123
-156
-156
92
35
114
35
114
35
11*
35
92
62
-156
-162
-15b
-Ib2
-156
-1%9
-16B
-18*
-153
-146
-123
-123
-165
-1H9
-156
-195
-Ifcb
-1S6
-ln76
35 30b
114 610
40 914
92 1676
b? 1372
-156 137«;
-162 91
-156 152-
-162 457
-156 914
-Ib9 1524
-166 1067
-1B9 610
-168 303
91
91
91
-156 l32i
-1B9 106?
-It5 137^
-195 ?2Ho
-IbS 22t«b
. lb«3
-1*9 76^
-15fc 3U3
-191 152
-Ib5 91»
-149 105U
. 129b
. lisa
-165 1296
-191 bl
-Ibfr 3U3
Ike
. 762
-152 437
-109 303
-17U 91*
-lit, 15?*
-Ill 303
-1''9 914
-111 1676
-13U 303
-b4 914
-l£4 13^4
. Ie7e
. 167b
T
T
T
T
T
T
T
T
T
T
T
^
T
T
T
1
T
T
T
T
T
S
S
S
T
T
T
T
S
S
1
T
T
T
c.
S
S
T
T
T
S
c;
T
T
T
T
T
T
T
T
T
T
S
S
0
H
0
K
N
N
N
S
N
N
ft,
S
N
H
H
£
H
r)
H
K
H
H
a
R
H
I
I
I
I
I
I
I
I
I
K,
l\
f»
t\





f*
K
N
N
N
N
N
N
M
N
(^

2.1 0.4 0.6 Ie i*r2
1.1 0.1 0.1 33
1.0 0.3 0.6 49
1.2 0.2 O.b 61
. 0.0 0.0 IB .
1.8 0.2 0.0 . .
0.3 0.3 0.1 36 140
3.1 0.3 1.3 31 135
3.0 0.3 1.1 . 13.
1.6 0.4 1.1 . 107
3.2 0.4 1.1
13.8 0.2 2.6 . .
3.3 0.6 3.9 30
0.3 0.1 0.0 47
3.4 0.6 0.9 BO
1.5 0.3 0.4 73
0.3 0.1 u. 8 44
1.5 0.2 0.5 52
2.0 0.3 O.b 69
2.9 0.6 0.9 Tc
4.8 0.2 l.b 41
8.1 0.3 1.0 332
0.9 0.1 1.2 bl
5.9 0.5 2.2 143
2.7 0.6 1.9 41 190
3.7 0.3 1.4 22 166
3.7 0.2 2.2 . 93
?.4 0.2 1.2 . b7
3.9 0.3 1.7 . 1«0
6.4 0.7 1.5 lib
3.2 0.6 1.7 iOb .
3.4 0.2 0.0 34 160
0.3 0.0 U.I 22 1*3
2.7 0.1 O.b . Ill
6.9 0.1 2.0 . 131
3.4 0.3 1.0 . 134
4.0 0.1 2.3 . 105
6.7 0.3 2.3 261
12.0 0.3 2.b 366
5.H 0.9 1.8 201 .
3.7 0.7 1.0 246
5.5 0.7 1.6 2«1 .
3.5 U.4 U.B 44 169
2.8 0.30.1 2' 141
1.6 0.3 0.0 . B?
?.5 U.4 0.6 13 1*6
6.2 0.4 J.o 1»1
2.1 0.3 0.6 96
2.5 0.4 0.1 lJ
-------
                                               TREATS  AIRCRAFT  FIELD  STUDY  DATA
DATE  TIME SMP XFRO^ YFflOf XTO  YTO  ELEY STYPt SCLASS  S04  N03  NH4  TS
                                                                           03
                                                                                BSCAT TtHP  DP  COMMENT
62*77 1115
62*77 11*5
63*77 1225
62*77 13*5
62*77 1*15
62*77 1*55
62*77 1625
62*77 16*5
62777 1715
62877 710
62877 T35
62877 955
62877 1055
62877 1320
6Z877 1350
62877 1000
62877 1050
62877 1325
62877 1*05
62877 1*30
62877 1620
62877 1650
62877 1725
63077 6*0
63077 715
63077 7*5
63077 1000
63077 13*0
63077 13*5
63077 1*20
63077 1*50
63077 600
63077 625
63077 650
63077 715
63077 950
63077 1020
63077 1050
63077 1115
63077 1320
63077 1*10
70177 11*5
70177 1225
70177 610
70177 650
70577 1020
70577 10*5
70577 1115
70577 1*10
70577 1»*0
70577 1510
70577 1535
70577 1120
70577 1200
H
h
H
h
H
H
h
^
e
e
e
s
e
e
*
H
H
H
H
H
H
H
h
e
e
B
e
B
B
B
B
H
h
h
fc
H
H
H
H
H
H
B
B
H
H
e
B
B
B
B
B
B
H
H
-123
-162
-125
-180
-216
-189
-161
-161
-67
30
100
37
132
30
26
-90
-96
-8*
-123
-116
-8*
-105
-13*
30
162
123
30
127
157
30
6*
-90
-126
-167
-126
-81
-112
-1*7
-111
-86
-1*7
1*
0
-89
-120
-78
-6
-12
36
119
195
119
-8*
-1*1
-189
-165
-152
-211
-170
-198
-179
-179
-13*
92
55
8*
-75
92
104
-137
-t2
-13*
-66
-98
-13*
-11*
-83
52
13
72
52
4*
-73
52
23
-137
-110
-90
-110
-ISO
-10*
-57
-101
-1*4
-57
62
9
-1*4
-56
-156
-130
-cB
66
44
24
44
-152
-105
-162
-123
-179
-216
-180
•
•
-102
30
100
30
132
1*
26
60
-96
-78
-123
-116
-84
-105
-83
-8*
162
123
12
53
157
127
64
30
-126
-167
-126
-87
-112
-147
-111
-83
-1*7
-86
0
17
-120
-89
-6
-12
14
119
195
119
36
-107
-84
-16b
-18S
-210
-170
-211
•
•
-161
92
5b
92
-75
62
104
43
-6<;
-156
-66
-9B
-134
-11*
-134
-134
13
72
au
84
-73
44
23
92
-110
-90
-110
-147
-104
-57
-101
-147
-57
-144
<*
-l*b
-5b
-144
-130
-2B
62
44
2»
44
86
-123
-152
305
914
152
3U5
S14
1067
1067
610
1676
152
610
610
1067
305
S14
457
762
457
914
914
457
457
914
152
610
610
152
457
SI*
610
1128
152
152
610
610
305
305
610
610
762
305
305
457
152
610
610
610
610
305
305
914
914
610
11?8
T
T
T
T
T
S
S
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
I
I
I
I
I
1
I
I
N
0
0
0
0
0
0
I
I
I
I
I
I
I
I
0
0
0
o
0
0
o
0
I
I
I
I
I
1
I
I
I
I
N
N
I
I
ri
B
B
b
B
a
H
B
u
7.1 1.4 0.2 32b .
2.0 1.5 0.6 310
3.0 1.7 0.6 260
2.3 1.7 0.2 195
2.3 1.6 0.1 193
3.4 1.6 0.1 155
2.5 1.4 0.5 . .

1.2 0.5 0.2 . 88.0
1.6 0.5 0.2 33 94.0
1.6 0.1 0.0 35 126.0
2.3 0.6 0.3 16 122.0
1.9 0.5 0.3 . 111.0
4.1 0.9 1.0 16 104.0
*.a o.* ui . isb.o
3.4 3.1 0.3 232
3.4 2.8 0.3 2*2
4.2 3.3 0.6 132
3.6 3.2 0.7 140
3.9 2.7 0.1 133
3.9 2.6 0.4 89
2.6 1.9 0.0 91
3.7 2.5 0.6 98
3.9 0.4 1.1 28 ISb.O
3.0 0.3 0.7 20 lbv.0
2.5 0.2 0.9 20 157.0
3.6 0.0 0.5 . 221.0
3.8 0.3 0.7 16 167.0
2.4 0.2 0.8 19 158.0
0.7 0.6 0.3 17 .
2.3 0.9 2.8 5
2.7 0.4 0.6 80
3.2 1.0 0.0 93 .
2.2 1.1 O.b 106
2.3 1.2 0.8 104
2.2 1.0 O.T 62
3.5 1.5 0.0 70
2.7 1.0 0.7 78
1.5 0.8 0.5 77 .
3.6 1.7 0.9 Sfe
3.6 0.1 1.1 60
1.9 0.2 0.2
4.5 0.2 0.6 . .
3.4 2.1 0.7 53
1.9 1.7 0.2 60
13.2 0.4 2.1 43 2B.3
18.6 0.2 2.8 55 338.9
11.6 0.4 2.0 55 237.0
10.6 0.1 1.0 52 284.0
10.7 0.2 0.9 30 19*. 0
19.1 0.0 0.7 . 176.0
17.4 0.1 1.3 . 19J.O
17. S 0.4 3.1 .
13.0 0.4 4.4 173
66
45
61
45
55
79
45
42
139
171
161
166
171
211
221
97
100
66
79
76
61
79
79
171
134
134
161
139
IbO
64
95
121
13*
*S
100
129
139
150
129
105
89
*
•
121
100
,
•
•
500
400
bUO
600
316
311
26.0 20.0
22.0 15.5
29.0 20.5
29.5 19.0
23. b 17.0
29.0 17.0
27.5 18.5
2b« 0 18.0
17.0 7.0
24.0 19.5
22.5 18.0
23.0 18.0
20.5 17.0
27.0 20.5
22. b 17.0
23.0 19.5
21.5 18.5
2b.O 19.0
21.0 17.0
21.5 17.5
2b.O 18.5
26.0 18.5
22.0 17.0
2b.O 21.0
24.0 19.5
24.5 19.0
26.0 20.5
28. b 19.0
25.0 18.5
21. 5 8.5
17.0 7.5
23. S 19.0
23.5 20.0
24. b 19.0
24.0 19.0
25.0 21.5
25.0 22.0
22.5 20.5
23.0 20. 5
23.5 19. S
28.5 21.0
2b.5 16.5
25.0 18.0
25.0 19.5
22.5 18.5
2B.5 21.0
26.5 20.5
25.0 18.5
28. b 21.0
26. b 21.0
22. b 18.5
21.0 16.5
22. b 17.0
20.5 15.0
                                                                                                UP AND DOWN OF COLBERT
                                                                                                UP AND DOWN OF COLBERT
                                                                                                UP AND DOWN OF CULbERT
                                                                                                UP AND DOWN OF COLBERT
                                                                                                UP AND DOWN OF COLBERT
                                                                                                3500* TO 100>
                                                                                                3500« TO 100«
                                                                                                UP AND DOWN OF COLBERT

                                                                                                PASS THROUGH GALLATIN PLUME
                                                                                                                                           W
                                                                                                CLOUDS AND RAIN  (BNA  PLUME)
                                                                                                CLOUDS AND RAIN  (BNA  PLUME)
                                                                                                K*VILLE 4 POSSIBLY  GALL.  PLUME

-------
                                                        AlkC^AH FIELD STIA'Y UATA




  DATE TIKE SHIP  XFOO  YFRCM  XTO  YTO  tLEV STYPt bCLASS  SU*   NOJ Nn* TS   03 ObCAT   TtMK  CP    COMMENT
70577 1*05
70577 1*30
70577 1*55
70577 1525
70677 7*5
70677 815
70677 8*5
70677 920
70677 1225
70677 1255
70677 1510
70677 1540
70677 1550
70677 555
70677 630
70677 705
70677 735
70677 1010
70677 1050
70677 1*00
70677 1*50
70777 635
70777 710
70777 715
70777 805
70777 1050
70777 11*0
70777 1*50
70777 1525
70777 1550
70777 1015
70777 1100
70777 13*5
70777 1*15
70777 1**5
70777 1510
70877 915
70877 950
70877 1030
121875 832
121875 1126
121P75 1452
121875 1546
121975 925
121975 1027
121975 1228
121975 1316
121975 1*05
121975 145?
H
H
H
M
e
e
B
e
R
B
q
8
H
1-
h
H
H
H
h
h
H
9
H
O
q
f>
e
B
H
6
b
H
h
h
H
H
e
P.
H
P
R
B
e
P
p
P

p
M
-81 -150
-107 -123
-153 -96
-117 -123
36 86
119 4*
*6 7*
128 *6
-23 129
-ei 113
33 flf
122 31
51 66
-86 -1*4
-126 -110
-£b
«b
-UO
-14*
-110
-14*
-117
-IbO
-120
-130
H*
1}
d<»
V2
iCi
1*»0
0
-103
-142
-137
-156
-162
-Ib6
-102
-156
.
.
.
-J*
U2
-1 7V
.
.

-?<:(>
- 104

- 1 y *
610
61 II
121V
1219
1 52
305
610
19H1
305
610
303
610
457
Ib2
457
762
1(167
152
610
305
762
1 52
457
762
457
305
v 14
olO
olO
MO
305
610
152
457
762
1067
152
2347
2652
11 0
5 Ib
762
If>h7
itlV
1 ? 1 v
"105
e 1 0
V14
121"
T
T
1
T
I
r
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
T
S
S
S
T
T
T
S
S
c
T
T
T
T
H
B
O
H





I
I
I
I
0
0
0
0
0
0
0
u
N
N
n.
N
N
N
N
N
N
0
0
0
0
0
0
N
N
N








I
I
4. J
12.0
t».6
2?. 7
lb.5
22.^1
21.3
0.1
20. e
12.3
2u. 7
26. <*
IV. 9
IV. 0
14.2
Ib.b
V.3
IV. 0
13.2
lb.7
lo.7
l4.b
14.0
5.1
18.3
14.3
P.I
17.4
13.1
12.3
4
13.6
IV. 1
17.2
IV. 9
12.3
17.0
1.5
11.3
2.3
3.0
o.v
1.1
1.3



u.v

0.1 0.5
0.4 2.0
U.3 1.6
0.4 2.4
0.7 1 .V
0.2 2.6
U.I 2.6
U.2 0.0
y.j i.b
0.4 2.2
0.3 2.V
0.1 2.7
0.1 2.1
0.5 3.1
0.1 3.1
0.3 2.4
0.3 1.9
o.h *.d
0.7 4.0
1.2 b.b
1.2 3.1
0.4 2.5
0.2 2.2
0.2 0.3
i).l 3.9
0.2 2.3
0.1 1.9
0.4 2.4
0.1 2.8
0.2 2.0
0.0 0.0
0.5 3.2
0.7 3.2
0.5 3.2
0.5 2.9
0.4 2.2
0.2 3.2
U.I 0.3
0.1 2.7
0.0 0.5
0.1 0.7
0.1 0.2
0.1 0.3
0.1 0.4



0.0 U.2

157
184
2u*
205
*0  Ul
34 1*0
. V7
. Ib5
. U7
. Ul
lie
114
65
79
88
87
. 133
04
. Ul
• *
. «
. .
^ .
. ,



• *

J37
too
229
413
*79
350
b50
39
461
bOO
OUO
029
bbO
*OJ
360
434
£74
432
443
£66
1V7
*5U
*69
200
489
321
iiav
bbO
450
411
53*
.
.
.
.
.
*21
bl
350
.
.
.
m
.



.

27.0 17.5
27.0 16.5
21.3 15.0
21.5 16.5
27. U 21.0
26.5 19.5 K'VILLE & POSSIBLY GALL. PLUME
24.0 18.0
19. 0 -5.0
2t).3 £U.5
£3.b 20.0
2V. b 19.5
26. b IB. 5
27. b 18.5
2U.O 19.0
26.3 17.5
2*.U 17.5
22. U 15.0
29. u 20.5
25.5 19.0
30. b 19.5
26. b 18.0
2u.b 20.0
27.0 20.0
24. b ld.0
27.0 20.0
28.0 22.0
24.0 17.0
28.0 20.0
2U.b 19.0
28.5 19.0
28. b 19.0
23.3 17.5
J2.0 20.5
29. b 20.0
26.0 IB. 5
23.5 17.5
2J.3 15.0 500' TO 6300'
7.U -6.5 7700'TO 12900'
20.0 14.0 8700' TO 10UO'
-1*.0
-12.0
-10. 0 .
-B.O . SPIHAL SFC bOO'-3300'
-0.5 . SPIRAL-4000"

o . o .

2.0
I.b
= 265

-------
                 C-l
              APPENDIX C




ALTITUDE CORRECTION, MELOY MODEL SH202




       AND SA285 SULFUR ANALYZER

-------
                                 C-2
                                APPENDIX C
     ALTITUDE CORRECTION,  MELOY MODEL SH202 and SA285  SULFUR ANALYZER
      The flame photometric  detector (FPD)  is  noted for  its  excellent
 sensitivity to low background  levels  of sulfur  and its  linear  logarithmic
 response over several  orders of magnitude.  Unfortunately,  it  is also
 sensitive—in an instrument-specific  manner—to ambient pressure, which
 is  a  function of altitude.  As  a result of  this altitude  sensitivity, a
 study was conducted to develop  altitude corrections  for the Meloy model
 SH202 and SA285 sulfur analyzers used in the  TVA TREATS studies.

      Both sulfur analyzers  were mounted in  an aircraft.   The analyzers
 were  connected in parallel  with a tee connection and  through a 4-way
 valve to three bags.   The Teflon bags were  filled  with  zero air, and low
 and high background concentrations of S02.  The fourth  valve position
 sampled  cabin air.

      An  initial reference response was obtained from  each bag while
 sitting  on the runway  at 152 m  MSL.   Additional responses to the three
 bagged gases  were  obtained  at 762 and 1370  m  MSL.  Results  from the test
 are given in  Table  C.I.
                 TABLE C.I.  MELOY SH202 ALTITUDE TEST
                                 Analyzer response (ppb as S02)
Elevation
(m MSL)
152
762
1370
Model SH202
Zero
4.4
2.3
2.6
Low
20.2
20.4
16.7
a
High
127
116
110
Model 285
Zero
0
-5.1
-0.65
Low
23
16.5
15.8
High
125
120
115
 SH202
response based on March 4,  1976,  calibration (i.e.,  c = 2706A '      )
Modeling Altitude Response

     The change in response (parts per billion as S02) of the flame
photometric detector as pressure drops is characterized by a decrease in
baseline current and a reduction in detector sensitivity.  For the SH202
FPD, both the zero shift and reduction in sensitivity were significant.
These effects were most severe at low concentrations.  For the SA285,
only the zero shift with altitude was significant.

-------
                                 C-3
     The corrected SH202 response is well described by the equation,

          C  =  C1 + 0.46 + 0.000128 (Z - 152)  CT,

where

     C1 =  uncorrected response, ppb,
     C  =  altitude corrected response, ppb, and
     Z  =  elevation, m MSL.

This model explains 98 percent of the altitude variation, and has a
standard error of 1.2 ppb.  This is well within the accuracy of the
instrument and usual calibration procedures.

     The corrected SA285 response is well described by the equation,

          C  =  C1 - 0.6 + 0.00716  (Z -  152),

where

     C1 =  uncorrected  response, ppb,
     C  =  altitude  corrected response,  ppb, and
     Z  =  elevation, m MSL.

This model explains  95  percent  of the  altitude variation,  again well
within the accuracy  of  the  instrument  and usual calibration procedures,

     All total  sulfur readings  were corrected with these models.

-------
               D-l
            APPENDIX D




TABULATION OF TREATS 1976 AND 1977




   HIGH-VOLUME FIELD STUDY DATA

-------
                                 D-2


                 ABBREVIATIONS AND UNITS FOR APPENDIX D


 DATE      Date of sample,  month of year

 STATION   Name of station

 LOCATION  Classification,  I  (inflow),  0 (outflow),  N (neither  I  nor 0)

 X,Y*      Coordinate  location of high-volume sampler,  km

 TSP       Total suspended  particulate  concentration,  |jg m~3

 804       Suspended water-soluble sulfates,  (Jg m"3

 N          Number of samples  averaged for  station

 DURATION  Average duration of sample,  h
*The origin of the coordinate system is Nashville, Tennessee.  The Y-axis
 is rotated 1.8 degrees east of north.

-------
                                    D-3
                TREATS 1976 C 1977 HIGH-VOLUME  FIELD STUDY DATA

DATE    STATION            LOCATION    X        V     TSP     S04    N

21076   LBL
21076   JOHNSONVILLE
21076   CUMBERLAND
21076   PARADISE
21076   GALLATIN
21076   WIDOWS CREEK
21976   LBL
21976   JOHNSONVILLE
21976   COLBERT
21976   CUMBERLAND
21976   GILES
21976   GALLATIN
21976   WIDOWS CREEK
22076   LBL
22076   JOHNSONVILLE
22076   COLBERT
22076   CUMBERLAND
22076   PARADISE
22076   GILES
22076   GAUATIN
22076   WIDOWS CREEK
22476   LBL
22476   JOHNSONVILLE
22476   COLBERT
22476   CUMBERLAND
22476   PARADISE
22476   GILES
22476   GALLATIN
22476   WIDOWS CREEK
31176   LBL
31176   JOHNSONVILLE
31176   COLBERT
31176   CUMBERLAND
31176   PARADISE
31176   GILES
31176   GALLATIN
31176   WIDOWS CREEK
31876   LBL
31876   JOHNSONVILLE
31876   COLBERT
31876   CUMBERLAND
31876   PARADISE
31876   GILES
31876   GALLATIN
31B76   WIDOWS  CREEK
31976   LBL
31976   JOHNSONVILLE
31976   COLBERT
31976   CUMBERLAND
31976   PARADISE
31976   GILES
31976   GALLATIN
31976   WIDOWS  CREEK
32376   LBL
DURATION
0
N
N
0
N
0
0
N
I
N
N
0
N
0
N
I
N
0
N
N
I
N
N
I
N
0
N
N
N
N
N
I
N
0
N
N
0
N
N
I
N
0
N
N
D
0
N
I
N
D
N
N
I
D
-118
-105
-83
-30
27
90
-118
-105
-99
-83
-24
27
90
-118
-105
-99
-83
-30
-24
27
90
-118
-105
-99
-83
-30
-24
27
90
-118
-105
-99
-83
-30
-24
27
90
-US
-105
-99
-B3
-30
-24
27
90
-118
-105
-99
-63
-30
-24
27
90
-118
71
-18
24
114
24
-144
71
-18
-159
24
-108
24
-144
71
-18
-159
24
114
-108
24
-144
71
-18
-159
24
114
-108
24
-144
71
-IS
-159
24
114
-108
24
-144
71
-18
-159
24
114
-108
24
-144
7i
-18
-159
24
114
-108
24
-144
71
18.0
33.5
36.5
29,5
39.0
30.5
41.0
76,0
37,5
61.5
54,0
53.0
50.5
50.0
39,0
44,5
45.0
77.5
34.0
49.0
38.0
23.0
24.0
19,0
30.0
87.0
32,0
34.0
29.5
14,0
27,5
28.0
38.5
63.5
12.0
47.0
189.5
33.0
48.0
35.5
43.0
70.0
37, 0
72.0
43.0
24.0
68,5
24.5
67.0
60.5
47.0
55.5
11.0
30,0
9.0
7,5
7.5
7,5
8.5
10,5
5,0
15,0
3.0
3.0
4.0
2,5
3.0
8.0
5.0
7.0
6.0
6.5
13,0
6.0
9,0
2.0
3.0
2.5
2,5
4.0
4,0
4.0
3.5
5.0
8.5
5.5
9.0
13.0
4,0
12.0
12,5
3.0
5.5
5.5
5,0
5,5
6.0
5.0
6,5
6.0
9.5
5.5
9.0
9.5
11,0
11.0
5.0
10,0
1
2
2
2
2
2
1
1
2
2
1
2
2
1
i
2
2
2
1
1
2
1
2
2
2
2
1
2
2
1
2
2
2
2
1
2
2
1
2
2
2
2
1
1
2
1
2
2
1
2
1
2
1
1
9.0
9,0
9,0
9,0
9,0
9.0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9.0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
9,0
4,5
9.0
9,0
9,0
9,0
4,5
9,0
9,0

-------
                                     D-4
                 TREATS  1976  C  1977  HIGH-VOLUME  FJCLO  STUDY DATA
 DATE     STATION

 32376    JOHNSONVILLE
 32376    COLBERT
 32376    CUMBERLAND
 32376    PARADISE
 32376    GILES
 32376    GALUTIN
 32376    WIDOWS CREEK
 32*76    LBL
 32*76    JOHNSONVILLE
 32476    COLBERT
 32*76    CUMBERLAND
 32*76    PARADISE
 32*76    GILES
 32*76    GALLATIN
 32*76    WIDOWS CREEK
 60277    SHAWNEE
 60277    JOHNSONVILLE
 60277    COLBERT
 60277    CUMBERLD.
 60277    HENDERSON
 60277    OWENSBORO
 60277    BOWLING GREEN
 60277    WIDOWS CREEK
 60277    KINGSTON
 60377    SHAWNEE
 60377    JDHNSONVILLE
 60377    COLBERT
 60377    CUMBERLD.
 60377    HENDERSON
 60377    OWENSBORO
 60377    PARADISE
 60377    GILES COUNTY
 60377    BOWLING GREEN
 60377    WIDOWS CREEK
 60377    KINGSTON
 60*77    SHAWNEE
 60*77    JOHNSONVILLE
 60477    COLBERT
 60477    CUMBERLO,
 60477    PARADISE
 60477    GILES COUNTY
 60477    WIDOWS CREEK
 60477    KINGSTON
 60577    SHAWNEt
 60577    JOHNSONVILLE
 60577    COLBERT
60577   CUMBERLO.
60577   PARADISE
60577   CUES COUNTY
60577   WIDOWS CREEK
60577   KINGSTON
60877   SHAWNEE
60877   JOHNSONVILLE
60877   COLBERT
LOCATION

   N
   I
   N
   0
   M
   N
   I
   0
   N
   I
   N
   0
   N
   N
   1
   I
   N
   0
   N
   0
   0
   D
   N
   0
   N
   I
   I
   I
   0
   I
   0
   N
   N
   N
   0
   N
   I
   N
   0
   N
   I
   N
   N
   N
   N
   0
   0
   0
   I
   I
   B
                                                       TSP
SO*   N   DURATION
-105
-99
-83
-30
-24
27
90
-118
-105
-99
-83
-30
-24
27
90
-176
-105
-99
-83
-72
-36
29
90
203
-176
-105
-99
-83
-72
-36
-30
-24
29
90
203
-176
-105
-99
-83
-30
-2*
90
203
-176
-105
-99
-83
-30
-24
90
203
-176
-105
-99
-18
-159
24
11*
-108
24
-1*4
71
-18
-159
24
114
-108
24
-1*4
101
-18
-159
24
166
176
92
-1*4
-26
101
-18
-159
2*
186
176
11*
-108
92
-144
-26
101
-18
-159
24
114
-108
-144
-26
101
-18
-159
24
114
-108
-144
-26
101
-18
-159
29,5
74.0
31.0
50.0
29,0
50.0
67,0
35.0
42.0
43.5
41.5
66.0
50.0
53.0
37.0
77,5
58,5
57,0
27,0
58,0
118.0
46,0
53.5
40.0
81.0
61.5
75.5
60.0
0.0
117,0
69.0
49.0
45,0
109.0
82.5
114.5
77.0
75.0
64.0
97.0
18.0
96.0
76,5
94,0
71.5
103.5
77,0
98.0
63,0
205.5
65,0
165.5
54,5
50.5
4.5
7,0
4.0
4.0
7,0
6.5
9,0
4.0
6,0
7,0
5.0
6.0
5,0
5,5
5,3
9.0
10.0
12,0
5.5
1,0
9.0
6,0
6.0
8.5
10.5
8.0
17.0
9.0
0.0
11.0
3.0
13.0
1.0
11.0
11,0
18.5
19,5
12,0
12,0
14,5
4,0
14,0
10,5
11,5
12.5
17.0
8,5
9.5
18,0
23.0
14,0
7.5
6.0
9.0
2
2
2
2
1
2
2
1
2
2
2
1
1
2
2
2
2
2
2
1
I
1
2
2
2
2
2
2
I
1
2
1
1
2
2
2
2
2
2
2
1
2
2
2
2
2
2
2
I
2
I
2
2
2
9,0
9,0
9,0
9,0
9,0
9,0
9,0
24.0
24,0
24,0
24,0
24,0
24,0
24,0
2't,0
9.0
9,0
9,0
9,0
«,0
8.8
8,0
9,0
9,0
9,0
9,0
4,0
9,0
0,0
9,3
6,7
9,0
0,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9.0
9,0
9,0

-------
                                    D-5
                TREATS 1976 t 1977 HIGH-VOLUME FIELD STUDY DATA

OATE    STATION            LOCATION      X      Y     TSP    SD4   N

60877   CUM6ERLO.
60877   HENDERSON
60877   OWENS80RD
60877   PARADISE
60877   GILES COUNTY
60877   BOWLING GREEN
60877   WIDOWS CREEK
60877   KINGSTON
61377   HENDERSON
61377   DWENSBQRO
61377   BOWLING GREEN
61777   SHAWNEE
61777   JOHNSONVILLE
61777   CUMBERLO.
61777   HENDERSON
61777   aWENSBORO
61777   PARADISE
61777   GILES COUNTY
61777   BOWLING GREEN
61777   WIDOWS CREEK
61777   KINGSTON
61877   SHAWNEE
61877   LBL
61877   JOHNSONVILLE
61877   COLBERT
61877   CUMBERLAND
61877   OACD
61877   PARADISE
61877   GILES
61877   GALLATIN
61877   HYTQP
61877   WIDOWS CREEK
61877   KINGSTON
62477   SHAWNEE
62477   LBL
62477   JOHNSONVILLE
62477   COLBERT
62477   CUMBERLAND
62477   OACD
62477   HENDERSON
62477   OWENSBORO
62477   PARADISE
62477   GILES
62477   GALLATIN
62477   BOWLING  GREEN
62477   HYTOP
62477   WIDOWS  CREEK
62477   KINGSTON
62877   SHAWNEE
62877   JOHNSONVILLE
62877   COLBERT
62877   CUMBERLO.
62877   HENDERSON
62877   OWENSBORO
DURATION
I
N
N
N
N
N
B
0
I
I
1
N
N
I
0
0
D
0
0
N
0
I
I
I
I
I
N
0
N
0
N
D
0
I
I
I
I
I
N
0
a
D
N
a
D
N
0
0
N
N
0
N
I
I
-83
-72
-36
-30
-24
29
90
203
-72
-36
29
-176
-105
-83
-72
-36
-30
-24
29
90
203
-176
-118
-105
-99
-83
-81
-30
-24
27
63
90
203
-176
-118
-105
-99
-83
-81
-72
-36
-30
-24
27
29
63
90
203
-176
-105
-99
-83
-72
-36
24
186
176
11*
-108
92
-144
-26
186
176
92
131
-18
24
166
176
114
-10B
92
-144
-26
101
71
-18
-159
24
-156
114
-108
24
-138
-144
-26
101
7i
-18
-159
24
-156
186
176
114
-108
24
92
-138
-144
-26
101
-18
-159
24
166
176
85.0
97,0
163,0
115,0
62.0
110. 0
130.5
38.5
69,0
118.0
408.0
31.5
20.0
22.0
78,0
140.0
39,0
29.0
66.0
32.0
51.5
65.3
39,0
30,2
33,0
45.3
36,0
63.3
35,0
32,3
35,0
60,2
55.5
50,5
40,0
44.2
63,4
64,0
96,0
41,0
163,0
42,0
82.0
63.8
71.0
89.0
100.7
36,0
120. 0
126,0
127.0
144,0
125.0
259,0
7,5
5.0
2,0
8.5
11.0
4.0
14,5
7.0
10,0
31,0
5,0
6.5
7,5
6.5
2.0
3.0
11.5
7.0
6,0
9.5
14.0
8.6
7,9
8,7
8.4
10,6
9,3
13.3
10,6
9.8
10,2
14,1
17,1
6.2
4,5
7.1
4.5
6.0
5.8
7,0
4.0
6,8
6,8
8.7
12,0
4.6
9.4
9,8
8,5
6.0
20,0
8,5
12,0
6.0
2
1
1
2
1
1
2
2
1
I
1
2
2
2
1
1
2
1
I
2
2
4
1
!>
4
6
2
6
1
4
1
6
2
4
1
5
5
7
2
1
1
5
i
4
1
1
6
4
2
2
I
2
1
1
9,0
8,0
8,7
9,0
9,0
8.0
9,0
9,0
24,0
9,3
8,0
9,3
9,0
9,0
B,2
9,0
B,9
9.0
8,0
8,8
9,0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
24, 0
24.0
24,0
24, 0
24,0
24, 0
24,0
24,0
24,0
6,5
9,1
24,0
24, 0
24,0
6,0
24,0
24,0
24,0
9,0
9,0
9,0
9,0
8,8
8,3

-------
                                    D-6
                TREATS  1976  C  1977 HIGH-VOLUME FIELD STUDY DATA
DATE
         STATION
                           LOCATION
                                                     TSP
SU4   N
DURATION
62877
62877
62877
62877
62877
63077
63077
63077
63077
63077
63077
63077
63077
63077
63077
63077
63077
63077
63077
63077
70577
70577
70577
70577
70577
70577
70577
70577
70677
70677
70677
70677
70677
70677
70677
70677
70677
70677
70677
70777
70777
70777
70777
70777
70777
70777
70777
PARADISE
GILES COUNTY
EfOwLING GREEN
WIUOWS CREEK
KINGSTON
SHAWNEE
LBL
JOHNSONVILLE
COLBERT
CUMBERLAND
OACD
HENDERSON
QWENSBORCJ
PARADISE
GILES
GALLATIN
BOWLING GREEN
HYTOP
WIDOWS CREEK
KINGSTON
SHAWNEE
JOHNSONVILLE
COLBERT
CUHBERLD.
PARADISE
GILES COUNTY
WIL'OWS CREEK
KINGSTON
SHAWNEE
JOHNSONVILLE
COLBERT
CUMBERLAND
QACO
PARADISE
GILES
GALLATIN
HYrop
rtlUQWS CREEK
KINGSTON
SHAWNEE
JOHNSONVILLE
COLBERT
CUMBERLD.
PARADISE
GILES COUNTY
WIDOWS CREEK
KINGSTON
I
N
I
N
N
I
I
I
I
I
N
G
0
0
N
0
0
N
0
0
0
N
I
N
0
I
I
N
I
N
0
N
D
I
0
N
0
0
N
I
N
0
N
N
N
0
0
-30
-24
29
90
Z03
-176
-118
-105
-99
-83
-81
-72
-36
-30
-24
27
29
63
90
203
-175
-105
-99
-83
-30
-a*
90
203
-176
-105
-99
-83
-81
-30
-2*
27
63
90
203
-176
-105
-99
-83
-30
-24
90
203
11*
-108
92
-144
-26
10;
71
-IB
-139
24
-156
186
176
114
-108
24
92
-138
-144
-26
101
-18
-159
24
11*
-108
-144
-26
101
-18
-159
24
-156
11*
-108
24
-138
-144
-26
101
-18
-159
24
11*
-108
-144
-26
151.5
136.0
69,0
120,5
71.0
59,5
46.0
45.5
63.6
39.4
64.5
110.0
200.0
65.0
44.0
58.5
76.0
39.0
63,8
60,8
92,0
90.0
116,5
91.3
132.3
97.0
123.0
88.5
101.5
131.7
101.8
123.4
108,0
129.8
88.0
99,3
66.0
97.5
103,3
80,7
90,0
134,5
116,5
141.0
146,0
207,0
91,5
7,5
15,0
2,0
7.0
6,0
7.1
3.7
7,6
5.3
6.9
4.7
13,0
4.0
10,2
B.4
10,0
2,0
5,0
9.4
12,7
19.0
30.0
25.0
27,5
29,5
29.0
24,0
25.0
25,9
38,2
27,6
32.6
29.8
25.2
27,6
29,0
17,9
27.8
26,8
13,3
28,0
25,5
29,5
25.0
42.0
33.5
29.5
2
1
1
2
2
4
1
6
5
7
2
1
1
5
1
4
1
1
5
4
2
2
2
2
2
1
2
2
4
6
5
5
1
5
1
3
1
6
4
3
2
2
2
2
1
2
2
9,0
9,0
a,o
9(0
9,0
24,0
24,0
24,0
24,0
24,0
24,0
8.2
9,0
24,0
24,0
24,0
8,0
24,0
24,0
24,0
9,0
9,0
9,0
9,0
6,9
9,0
9,0
9,0
24,0
24,0
24.0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
24,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
9,0
N«209

-------
                   E-l
                 APPENDIX E




INVERSION HEIGHTS AND AVERAGE WIND VELOCITIES




      FOR LAGRANGIAN AND EULERIAN DAYS

-------
                                E-2
                ABBREVIATIONS AND UNITS FOR APPENDIX E


Sampling class:  I = inflow, 0 = outflow

Inversion type:  R = radiation; S = subsidence

Station        :  BNA = Nashville, TN (NWS)
                 JSV = Johnsonville Steam Plant, New Johnsonville, TN
                 KIN = Kingston Steam Plant, Kingston, TN
                 COL = Colbert Steam Plant, Pride, AL
                 PAR = Paradise Steam Plant, Drakesboro, KY
                 GAL = Gallatin Steam Plant, Gallatin, TN
                 WID = Widows Creek Steam Plant, Bridgeport, AL

Wind velocity  :  Angle defines average direction from which wind is blowing.
                 Directions in degrees from true north, speed in meters per
                 second.  Numbers in parentheses are the resultant wind
                 through the transport layer.

-------
                          E-3
TABLE E.I INVERSION HEIGHTS AND AVERAGE WIND VELOCITIES
           FOR LAGRANGIAN AND EULERIAN DAYS
Sampling
Date class Time (L)
2/10/76


2/19/76

I :



3/11/76




3/18/76





3/23/76






I
0
0
I
I
I
I
I
0
I
I
I
I
0
I
I
I
I
0
0
I
I
I
I
I
0
0
0515
1345
1715
0810
0930
0810
0930
0930
1643 (avg)
0815
0830
1015
0923 (avg)
1530 (avg)
0655
0815
0655 (avg)
0655 (avg)
1340
1700
0815
0810
0815
0810
0810
1705
1415
Inversion
Upper air height (type)
station (m AGL)
BNA
JSV
BNA
COL
JSV
COL
JSV
JSV
BNA & WID
COL
COL
JSV
COL & JSV
JSV & BNA
WID & COL
COL
WID & COL
WID & COL
JSV
BNA
COL

COL

COL
BNA
JSV
626 (S)
915 (S)

244 (R)

1829 (S)


1900 (S)
366 (R)

1829 (S)

1250 (S)
519 (R)
1524 (S)


1372 (S)

305 (R)

1219 (S)


1767 (S)

Wind
velocity (m s"1
(219/14.0)

(215/13.2)

237/4.7

278/13.8
(276/12.4)
(269/6.9)

232/2.0
235/2.7
(235/2.6)
(183/5.2)
230/8.4

249/10.3
(243/9.5)

(212/12.2)

086/2.1

021/1.3
(043/1.3)

(213/6.6)
Layer
) (m)
SFC-626
SFC-915
SFC-915

SFC-244

245-1829
SFC-1829
SFC-1900

SFC-366
367-1829
SFC-1829
SFC-1250
SFC-519

520-1524
SFC-1524

SFC-1372

SFC-305

306-1219
SFC-1219

SFC-1767

-------
                                        E-4
TABLE E.I (continued)
Sampling
Date class
3/24/76 I
I
I
I
0
0
6/03/77 I
I
I
I
I
I
I
0
0
0
0
0
0
0
0
0
6/04/77 I
I
I
I
I
I
I
I

I
Time (L)
0655 (avg)
0825 (avg)
0655 (avg)
0655 (avg)
1332
1315 (avg)
0640 (avg)
0615
0630 (avg)
0950
0805
1600 (avg)
1605
0845
1110
0845
1110
1110
0950
1110
1400
1510
0650
0630
0650
0630
0630
1015
1200
1435 (avg)

1325
Upper air
station
COL & WID
COL & WID
COL & WID
COL & WID
JSV
JSV & KIN
PAR & SNA
PAR
PAR & BNA
JSV
JSV
JSV & BNA
JSV
COL
COL
COL
COL
COL
JSV
COL
JSV
COL
GAL
GAL
GAL
GAL
GAL
JSV
JSV
JSV, COL,
BNA
JSV
Inversion
height (type)
(m AGL)
564 (R)

1752 (S)

1829 (S)

305 (R)
1646 (S)

1646 (S)

1524 (S)

366 (R)

1645 (S)


1646 (S)

1524 (S)

457 (R)

1219 (S)


1128 (S)

& 1433 (S)


Wind
velocity (m s"1)

192/6.9
197/9.9
(196/8.9)

(201/10.6)
061/6.0
348/6.3
(358/5.6)

(064/3.9)

(046/2.2)

068/2.7

009/5.2
(016/4.4)

(016/4.4)

(009/2.8)

125/2.9

145/1.0
(132/1.7)

(020/1.4)


(349/2.3)
Layer
On)

SFC-564
565-1752
SFC-1752

SFC-1829
SFC-305
306-1676
SFC-1676

SFC-1646

SFC-1524

SFC-366

SFC-1645
SFC-1645

SFC-1646

SFC-1524

SFC-457

458-1219
SFC-1219

SFC-1128


SFC-1433

-------
TABLE E.I (continued)
                                        E-5
Sampling
Date class
6/04/77
(Cont . )





6/28/77










6/30/77
















0
0
0
0
0
0
0
I
I
I
I
0
0
0
0
0
0

I
I
I
I
I
I
I
I
I
0
0

0
0

0

Time (L)
0845
0725
06lO(avg)
0725
0725
1235
1510
0940 (avg)
1110
1400
1510
0700
0620
0700
0700
1040 (avg)
1140 (avg)

0430
0710
0430
0710
0710
1220
1140 (avg)
1650 (avg)
1445 (avg)
0650
0655 (avg)

0650
0655 (avg)

0655 (avg)

Inversion
Upper air height (type) Wind
station (m AGL) velocity (m s~l]
COL
COL
COL (2)
COL
COL
COL
COL
COL & JSV
COL
JSV
JSV
GAL
JSV
GAL

KIN & JSV
KIN, WID &
JSV
COL
COL
COL
COL

COL
COL & JSV
COL
JSV & COL
GAL
GAL, PAR, &
KIN
GAL
GAL, PAR, &
KIN
GAL, PAR, &
KIN
366 (R)

1280 (S)


1524 (S)

1524 (S)

1372 (S)

365 (R)
1524 (S)


1372 (S)


366 (R)

1220 (S)


1524 (S)

1524 (S)

548 (R)


1220 (S)





(124/2.6)

315/2.7
(322/1.2)

(005/2.9)

(239/9.0)

(232/8.1)
244/10.6

266/11.8
(262/10.8)

(237/9.0)


234/7.9

240/9.8
(238/9.2)

(234/9.8)

(233/10.0)

223/8.5


240/9.3

(233/8.8)

Layer
) (m)

SFC-366

367-1280
SFC-1200

SFC-1524

SFC-1372

SFC-1372
SFC-365

366-1524
SFC-1524

SFC-1372


SFC-366

367-1220
SFC-1220

SFC-1524

SFC-1524

SFC-548


549-1220

SFC-1220


-------
                                        E-6
TABLE E.I (continued)
Date
6/30/77



7/06/77















Sampling
class
-0
0
0
0
I
I
I
I
I
I
I
0
0
0
0
0
0
0
0
0
Time (L)
1145
1155 (avg)
1510
1420
0810
0800
OSOO(avg)
0750
0755(avg)
1005
1415
0820
0715
0820
0715
0715
1230
1110
1700(avg)
1510
Upper air
station
KIN
KIN & JSV
KIN
JSV
GAL
GAL
PAR & GAL
PAR
PAR & GAL
JSV
JSV
COL
COL
COL
COL
COL
COL
COL
COL (2)
COL
Inversion
height (type)
(m AGL)
1220 (S)

1829 (S)

365 (R)

990 (S)


1067 (S)

366 (R)

1372 (S)


1372 (S)

1036 (S)

Wind
velocity (m s"1)

(234/8.5)

(230/12.0)

057/0.6

014/0.9
(026/0.7)

(333/1.9)

082/2.9

012/4.6
(024/3.8)

(020/2.2)

(355/3.2)
Layer
(m)

SFC-1220

SFC-1829

SFC-365

366-990
SFC-990

SFC-1067

SFC-366

367-1372
SFC-1372

SFC-1372

SFC-1036

-------
            F-l
           APPENDIX F




SYNOPTIC METEOROLOGICAL SUMMARIES




FOR LAGRANGIAN AND EULERIAN DAYS

-------
                                F-2


                               APPENDIX F


   SYNOPTIC METEOROLOGICAL SUMMARIES FOR LAGRANGIAN AND EULERIAN DAYS
     Meteorology for the two field studies is analyzed for the Lagrangian
 and Eulerian measurement days shown in Table 6.  Thus, six days were
 analyzed for the 1976 spring study and eight for the 1977 summer study.
 The following analyses describe by study and date the most significant
 meteorological parameters.


 1976 STUDY

     In the Tennessee Valley region, February and especially March are
 typically characterized by spring type weather; that is, frequent frontal
 passages are the rule, with good ventilation (wind speed times mixing
 height) and rapidly changing airmass characteristics.  Typical inversion
 heights and wind speeds are about 600 m and 5.5 m s"1 in the morning and
 1800 m and 7 m s"1 in the afternoon.  Typical daily maximum temperatures
 average 15°C, whereas daily minimum temperatures average around 3°C.
 Surface wind speeds average 5 m s"1; this represents the strongest average
 flow for any period during the year.  Precipitation typically occurs on
 one out of every three days, and mean cloud cover averages 65 percent.
 Analysis of weather types (see the Meteorological Characterization subsec-
 tion of Section 3) indicates that the spring season has similar charac-
 teristics to the annual average weather.  High-pressure systems occur
 most frequently (27 percent), with the centers of these systems typically
 located northeast and east of Nashville.  Maritime tropical and modified
 polar airmasses occur equally during this time of the year and account
 for 80 percent of all airmass types.


 February 10, 1976

     As noted previously, February (and March) weather is typified by
 strong ventilation and frequent frontal passages.   This is exemplified by
 the weather pattern of February 10, 1976.  On the morning of February 10,
 1976,  a developing low-pressure center and associated cold front were
 located in the Great Plains area of Iowa and Kansas, while at the same
 time,  a large modified continental polar airmass was centered over central
Florida.   The combined effect of these two systems produced an influx of
maritime tropical air, with accompanying clouds.  Early-morning inflow
 ceilings (near Muscle Shoals, Alabama) were around 1000 m AGL, but gradually
 rose above 1500 m AGL at the afternoon outflow boundary (near Bowling
 Green,  Kentucky).  No significant rainfall occurred during the sampling
 day, and the maximum average surface temperature was 20°C (68°F).  Tempera-
 ture soundings showed that warm air advection produced a morning mixing
 layer (~600 m AGL)  nearly equal to the climatological average, whereas
 the afternoon mixing height (~900 m AGL) was significantly below the
 afternoon climatological average.   As might be expected, winds under this
pressure configuration were southwesterly and strong; Valley-wide mixing-
 layer wind velocity averaged 215 degrees at 13 m s~l throughout the sampling

-------
                                F-3


period.  The average of the surface and 850-mb 24-h back-trajectory for
February 10, 1976, showed that the airmass originated along the Louisiana-
Texas border area (Figure F.I).


February 19, 1976

     The weather pattern on this date was characterized by a weak surface
frontal passage through the study area during the daylight hours, which
caused little change in the prevailing modified maritime polar airmass
weather characteristics.  Generally, fair skies with southwesterly surface
winds early in the morning and westerly winds by midday characterized the
prevailing weather conditions.  Areawide temperatures were warm, with
highs generally around 20°C (68°F).  Winds aloft were westerly from near
surface to 1800 m AGL, the top of the afternoon mixed layer.  The morning
radiation inversion was about 300 m AGL.  The 24-h back-trajectory for
February 19, 1976, indicated the source region to be eastern Kansas
(Figure F.2).


March 11, 1976

     The synoptic weather pattern for March 11, 1976, was predominantly
influenced by an east-west-oriented stationary front located through
central Kentucky and a high-pressure center located along the Georgia-
South Carolina border.  This combination produced high cloud ceilings
over the field study area and an influx of maritime tropical air.  Winds,
both surface and aloft, were southwesterly during the morning hours,
whereas a gradual backing produced a more southerly component by after-
noon.  The morning radiation inversion height was about 400 m AGL, and
the morning subsidence inversion height was about 1800 m AGL.  Due to
warm air advection, the afternoon inversion height decreased by about 500
m.  No precipitation occurred during the measurement period, and high
temperatures were again around  20°C.  Also, the 24-h back-trajectory
indicated that the airmass originated in central Mississippi  (Figure F.3).


March  18, 1976

     Somewhat similar  to  the weather of March  11,  1976,  a  high-pressure
center was  located along  the Georgia-South Carolina border, with a warm
front through northern Kentucky.  Again,  southwesterly  flow of  maritime
tropical air occurred  across the  study  area.   Partly  cloudy skies  prevailed,
with temperatures again  reaching  20°C.  The morning  radiation and  subsi-
dence  inversions  were  about 500 and 1500  m AGL respectively.

     By  afternoon,  the subsidence inversion  had decreased to  about 1400  m
AGL  due, again,  to warm  air advection.   The  24-h trajectory analysis
indicated that the  source region included southern Louisiana  and Mississippi
 (Figure  F.4).

-------
Figure F.I.  The 24-h boundary-layer back-trajectories for 1800 h CST, February 10, 1976.

-------
Figure F.2.  The 24-h boundary-layer back-trajectories for 0600 h CST,  February 19,  1976.

-------
Figure F.3.   The 24-h boundary-layer back-trajectories for 1800 h CST, March. 11, 1976.

-------
Figure F.4.  The 24-h boundary-layer back-trajectories for 1800 h CST, March 18,  1976.

-------
                                 F-8
 March 23, 1976

      A meteorological pattern different from the previous two occurred
 on this date.  A modified continental high-pressure airraass was centered
 over the Chesapeake Bay area, with a very weak pressure gradient over
 the study area.  Boundary-layer winds were light and northeasterly early
 in the morning, but became stronger and southwesterly by afternoon.   Skies
 were generally clear throughout the sampling period, with the area tempera-
 ture again peaking around 20°C.   Morning radiation and subsidence inversion
 heights averaged 300 and 1200 m AGL, respectively, whereas the afternoon
 subsidence inversion height rose to about 1800 m AGL.   The 24-h back-
 trajectory shows a complex trajectory path resulting from the changing
 nature of the high-pressure center's location (Figure F.5).


 March 24,  1976

      The high-pressure center of the previous day was well off the eastern
 seaboard by  the morning of March 24, 1976.   An approaching cold front
 with a pressure-frontal orientation similar to that described on February
 10,  1976,  caused an influx of warm,  moist tropical air.   Middle and high
 cloud  cover  increased throughout the day;  however, rainfall did not occur
 in the study area until after dark.   Morning radiation and subsidence
 inversion  heights were about  600 and 1750 m AGL respectively  with the
 afternoon  subsidence inversion height rising slightly to about 1800 m
 AGL.  The  24-h trajectory shows  that once again the source region was
 southern Louisiana and Mississippi  (Figure F.6).
 1977  STUDY

      Summertime  (June  and July) weather  in  the Tennessee Valley region,
 as  in most  of  the  eastern United  States,  is  typified by infrequent frontal
 passages  and slow-moving high-pressure systems.  Due to increased solar
 insolation, vertical mixing  is usually good, with cliraatological morning
 and afternoon  averages being about 450 and  1900 m AGL respectively.  How-
 ever, total ventilation is significantly  less, on the average, than the
 spring season  due  to reduced wind speeds.  Transport layer speeds average
 only  3.5  m  s"1 in  the morning and 5ms1"1 in the afternoon.  The average
 daily maximum  temperature is 31°C (twice  the daily springtime average
 maximum), whereas  the daily  minimum averages 20°C.  Typical surface wind
 speeds average 3 m s l, the  slowest average wind speed of any season.
 Precipitation occurs on an average of one third of the days, and mean
 cloud cover averages about 55 percent.  Analysis of weather typing indi-
 cates that stationary fronts and high-pressure centers without associated
 frontal systems predominate.   Most pressure centers fall within a 300-
 to 399-km radius of Nashville or are greater than 500 km and favor the
 45- through 134-degree sector (northeast through southeast of Nashville).
Maritime tropical airmasses cover the area 75 percent of the time and
 either one or more of the four Tennessee weather stations record
measurable precipitation 60 percent of the time.

-------
Figure F.5.  The 24-h boundary-layer back-trajectories for 1800 h CST, March 23,  1976.

-------
Figure F.6.  The 24-h boundary-layer back-trajectories for 1800 h CST, March 24,  1976.

-------
                                F-ll


June 3, 1977

     Early in the morning of June 3, 1977, a weak continental polar front
lay east-west across northern Alabama and Georgia.  Behind it, to the
north, was a modified continental airmass centered over central Michigan.
This high-pressure system produced northerly winds, no precipitation,
and morning radiation and subsidence inversion heights that averaged
about 350 and 1650 m AGL respectively.  The average field study after-
noon inversion level decreased to about 1500 m due to subsidence of
superior air.  The average maximum temperature was 30°C, with mostly
sunny skies prevailing.  The 24-h trajectory data showed that the source
region was in southern Indiana and Illinois (Figure F.7).


June 4, 1977

     The high-pressure center of the previous day moved southeastward
and was centered over West Virginia on the morning of June 4, 1977.  Again,
airflow was light and from a northerly direction.  Average morning radiation
and subsidence inversion heights were 400 and 1250 m AGL respectively.
The afternoon inversion height rose to around 1400 m AGL.  Again, no signi-
ficant cloud cover was present over the study area, and the average maximum
temperature was 31°C.  The trajectory data indicated that the airmass
originated along the Ohio River Valley area (Figure F.8).


June 5, 1977

     By the morning of June 5, 1977, the high-pressure center of the previ-
ous two days dissipated and began merging with the Bermuda high.  Winds
were west-northwesterly throughout the boundary layer.  Inversions averaged
350 and 1400 m AGL during the morning hours and 1750 m AGL during the
afternoon.  Again, no significant cloud cover was present and the average
high temperature reached 33°C.  The 24-h trajectory data indicated that
the airmass originated along the Mississippi River west of the study region
(Figure F.9).
June 8, 1977

     A modified continental high-pressure airmass was centered over northern
Alabama on June 8, 1977.  Morning radiation and subsidence inversion heights
averaged 300 and 1600 m AGL respectively.  By afternoon, the subsidence
inversion rose to about 1800 m AGL.  Boundary-layer winds were light and
variable over northern Alabama and west-northwesterly over the northern
half of the field study area.  Trajectory analysis shows that the 24-h
airmass source region was in central Indiana and Illinois (Figure F.10).
Temperatures rose to around 27°C under fair skies.

-------
Figure F.7.  The 24-h boundary-layer back-trajectories for 0600 h CST, June 3, 1977.

-------
Figure F.8.  The 24-h boundary-layer back-trajectories  for 0600 h GST,  June  4,  1977.

-------
                                                                                                        I
                                                                                                       I—I
                                                                                                       *-
Figure F.9.  The 24-h boundary-layer back-trajectories for 1800 h CST, June 5, 1977.

-------
Figure F.10.  The 24-h boundary-layer back-trajectories for 0600 h CST,  June 8,  1977.

-------
                                F-16
June 28, 1977

     A significant departure from the earlier June northerly wind flow
regime was evident during the last one third of the month.   On June 28,
1977, an approaching cold front located in the Great Plains combined with
the Bermuda high to produce steady southwesterly flow of maritime tropical
air.  Boundary-layer winds throughout the study area were southwesterly
and relatively strong (7 to 10 m s'1).  Morning radiation and subsidence
inversion heights averaged about 400 and 1500 m AGL, respectively, whereas
afternoon inversion heights were around 1400 m AGL.  Despite the approaching
frontal system, only mid-level small cumulus and altocumulus clouds were
evident.  Average high temperatures reached 32°C.  Significant rainfall
did not materialize until after sunset, and most of that was confined  to
the western part of the  field study area.  The 24-h trajectory data indi-
cated  that airmass source regions included the Gulf Coast  area of  eastern
Texas  and southern Louisiana  (Figure F.ll).


June  30,  1977

      By the  morning  of  June 30,  1977,  the  southward moving cold  front of
June 28,  1977,  was  located  over  Kentucky and had begun moving northward
 as a warm front in response to the  approach of a second maritime polar
 airmass.   This new frontal  system was situated through central Iowa,  then
 southwestward to southern Colorado.  Again, warm moist tropical air accom-
 panied relatively strong (8 to 11 m s"*1) southwesterly winds.  Morning
 radiation and subsidence inversion heights averaged 450 and 1200 m AGL,
 respectively, with the afternoon subsidence inversion rising to about
 1650 m AGL.   Hazy but fair skies were prevalent in the morning; however,
 by midafternoon isolated cumulus congestus and cumulonimbus clouds pro-
 duced shower activity near the northern study area boundary.  Maximum
 temperatures were around 35°C.  The 24-h trajectories indicated that  the
 airmass originated in the  northern Louisiana area  (Figure F.12).


 July  6,  1977

       Weather  conditions within  the Tennessee Valley  area  and, in  fact,
 across most of the  eastern United  States were dominated by a  large,
 sprawling high-pressure system  on  July  6,  1977.   The surface  center  of
 this  maritime tropical airmass  was located over southern  Louisiana,
 whereas  the upper-level core was located over the bootheel area of
 Missouri.   Boundary-layer  winds were light (<4  m s"1) and northerly.
 Morning radiation and subsidence inversions were relatively low,  with
 values averaging 350 and 1100 m AGL, respectively, whereas afternoon
  subsidence  inversion levels remained relatively stable.   Maximum tem-
  peratures peaked around 35°C while subsiding air effectively eliminated
  cloud cover.  Trajectory data indicated that the little air movement
  present came from the northwest (Figure F.13).

-------
Figure F.ll.  The 24-h boundary-layer back-trajectories for 0600 h CST,  June 28,  1977.

-------
                                                                                                         i
                                                                                                        <—i
                                                                                                        Co
Figure F.12.  The 24-h boundary-layer back-trajectories for 1800 h CST, June 30,  1977.

-------
                                                                                                       I
                                                                                                       l->
                                                                                                       vo
Figure F.13.  The 24-h boundary-layer back-trajectories  for 0600 h CST,  July 6,  1977.

-------
                                F-20
July 7, 1977

     Similar to July 6, 1977, this day's weather was characterized by a
continuation of the Bermuda high-pressure system over much of the eastern
United States.  The upper-level high-pressure cell was centered over the
study area.  Maritime tropical air continued to pervade the area.  Boundary-
layer winds were again light (<5 m s"1) and generally from the WNW.  Morn-
ing radiation and subsidence inversion heights averaged 200 and 1700 m AGL.
Skies were generally fair, and average maximum surface temperatures once
again reached 35°C.  The 24-h trajectories indicated that the airmass
affecting the study area once again originated near the Mississippi River
(Figure F.14).

-------
Figure F.14.  The 24-h boundary-layer back-trajectories for 0600 h CST,  July 7,  1977.

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 1. REPORT NO.
     EPA-600/7-80-126
2.
                             3. RECIPIENT'S ACCESSION NO.
 4. TITLE AND SUBTITLE
     TRANSPORT AND TRANSFORMATION OF SULFUR OXIDES
     THROUGH THE TENNESSEE  VALLEY REGION
                             5. REPORT DATE
                                 June 1980
                                                           6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
     T. L. Crawford and L. M.  Reisinger
                                                           8. PERFORMING ORGANIZATION REPORT NO.
                                                               N581
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
     Office of Natural Resources
     Tennessee Valley Authority
     Muscle Shoals, AL  35660
                             10. PROGRAM ELEMENT NO.

                                  1NE-832
                             11. CONTRACT/GRANT NO.
                                                                81  BDL
 12. SPONSORING AGENCY NAME AND ADDRESS
     U.S. Environmental Protection Agency
     Office cf Research & Development
     Office of Energy, Minerals  &  Industry
     Washington, D.C.  20460
                             13. TYPE OF REPORT AND PERIOD COVERED
                                 Milestone
                             14. SPONSORING AGENCY CODE
                                  EPA-ORD
 15. SUPPLEMENTARY NOTES
     This project is part of  the EPA-planned and coordinated Federal  Interagency
     Energy/Environment R&D Program.
 16. ABSTRACT
        This report is directed  to scientists interested in the  long-range atmospheric
   transport and transformation  of sulfur compounds.
        Statistical and climatological analyses of historical data  and the results of
   two long-range transport  studies are presented.  The^two long-range atmospheric
   transport field studies were  conducted over a 300-km  area of the  southern United
   States centered on the Tennessee Valley region   The first study was conducted dur-
   ing the spring of 1976, and the second was conducted during the  summer of 1977.  The
   field study region contains seven large coal-fired power plants  and one large city.
        Results indicate that the  predominant flow and mass transport direction is from
   the southwest to the northeast.   Also, aerometric measurements obtained by aircraft
   and ground sampling compared  favorably with results obtained  with  an analytical
   transport-transformation  model  developed for this study.  Results  indicate that,
   during prevailing southwesterly airflow, large sulfur influxes are present.  These
   influxes, which are at the same order of magnitude as the Tennessee Valley regional
   emission fluxes, can only partly be explained by upwind anthropogenic sources.
   Natural source emissions  are  hypothesized to account for about half of this sulfur
   influx.
 7.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                                           c.  COSATI Field/Group
               Earth Atmosphere
                  Transport Processes

                  Charac..Meas.  & Monit.
 6F    8A    8F

 8H   10A   10B

 7B    7C   13B
 8. DISTRIBUTION STATEMENT
                19. SECURITY CLASS (ThisReport>
                      Unclassified
21. NO. OF PAGES
      153
                Release to public
                20. SECURITY CLASS (Thispage)
                      Unclassified
                                                                         22. PRICE
EPA Form 2220-1 (9-73)

-------