DISPERSION IN COMPLEX TERRAIN

        A Report of a Workshop Held

           at Keystone, Colorado

              May 17-20, 1983
  ATMOSPHERIC SCIENCES RESEARCH LABORATORY
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711

-------
       DISPERSION IN COMPLEX TERRAIN

        A Report of a Workshop Held

           at Keystone, Colorado

              May 17-20, 1983
                     by

           Bruce A. Egan, Editor
 Environmental Research & Technology, Inc.
                  for the

      American Meteorological Society
              45 Beacon Street
        Boston, Massachusetts 02108
        Cooperative Agreement 810297
              Project Officer

           Francis A. Schiermeier
    Meteorology and Assessment Division
  Atmospheric Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
  ATMOSPHERIC SCIENCES RESEARCH LABORATORY
     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711

-------
                           NOTICE
The information in this document has been funded by the
United States Environmental Protection Agency under
Cooperative Agreement 810297 to American Meteorology
Society.  It has been subject to the Agency's peer and
administrative review, and it has been approved for
publication as an EPA document.

Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
                             ii

-------
                                    PREFACE

Ong of the most formidable challenges to the meteorological community is
to satisfy the need for a capability to reliably describe, explain, and
predict the behavior of the atmosphere in complex terrain.  The infinite
variations in the size, shape, slope and orientation of ridges and
valleys, coupled with land and water interfaces, cause radical alterations
in the characteristics of the atmosphere's boundary layer over short
distances and time periods.  Natural resources that are important to
economic development and to recreational needs are often concentrated in
such locales.  Consequently, interests of those seeking to preserve
certain amenities of life for themselves and future generations are
often at variance with the interests of those seeking industrial growth.
Equitable resolution of disputes over the use of resources and the
degree of protection of the amenities frequently centers on the capability
to assess the circulation and dispersion characteristics of the near
surface portion of the atmosphere.  The magnitude of the impact on air
quality by human activity is a direct function of the atmosphere's
behavior in complex terrain.

This report contains the thoughts and judgments of 32 atmospheric
scientists who gathered to exchange recently acquired technical information
and research results on atmospheric processes in complex terrain and to
comment on matters relating to adjustments in current air quality
modeling practices.

Since 1979, the American Meteorological Society (AMS) has collaborated
with the Environmental Protection Agency (EPA) through a cooperative
agreement to improve the scientific basis of air quality modeling.  The
organisation of this Workshop on Dispersion in Complex Terrain has been
carried out under this cooperative program.
                                       ill

-------
I am indebted to the following people who participated in this workshop:
S.P.S.  Arya, N'orth Carolina State University;  S.  Barr, Los Alamos Scientific
Laboratory; W.  Blumen,  University of Colorado;  N.  Bowne,  TRC Environmental
Consultants, Inc.; L. Crow, Denver,  CO;  R.  Fisher, EPA, Denver; D. Fox,
                                                                  ğ
USDA Forest Service; S. Hanna, Environmental Research $ Technology,
Inc.; D.  Henderson, U.S. Park Service, Denver;  T.  Lavery, Environmental
Research § Technology,  Inc.; R. McNider, Department of Environmental
Management, State of Alabama; R. Meroney, Colorado State University;
W. Ohmstede, U.S. Army Atmospheric Sciences Laboratory; R. Petersen, NHC
Wind Engineering; R. Pielke, Colorado State University; D. Randerson,
NOAA Weather Service Nuclear Support Office; K. Rao, NOAA Atmospheric
Turbulence & Diffusion Lab.; R. Rowe, University of Calgary; F. Schiermeier,
EPA, Research Triangle Park; H. Slater,  University of North Carolina;
M. Smith, Meteorological Evaluation Services,  Inc.; R. Smith, Yale
University; N.  Snyder,  EPA, Research Triangle Park; G. Start, NOAA Air
Resources Laboratories; R. Sykes, ARAP,  Inc.;  J.  Tikvart, EPA, Durham;
J. Weil,  Martin Marietta Laboratories; F. White,  National Research
Council;  D. Whiteman, Battelle Pacific Northwest Laboratories; G. Wooldridge,
Utah State University;  and J. Wyngaard,  National Center for Atmospheric
Research.

Further,  I would like to thank Lindalu Miller of the USDA Forest Service
and Margit Kohl of Environmental Research $ Technology, Inc. for their
most capable assistance in the production of this report.
                                                  Bruce A. Egan
                                                  Workshop Chairman
                                      IV

-------
                              ABSTRACT







Since 1979, the American Meteorological Society has collaborated with the




Environmental Protection Agency through a cooperative agreement to improve




the scientific basis of air quality modeling.   Under this continuing agree-




ment, the American Meteorological Society conducted a workshop on dispersion




in complex terrain in Keystone, CO, during May 17-20, 1983.   The purpose of




the workshop was to encourage atmospheric scientists working in the area of




complex terrain dispersion modeling to exchange recently acquired informa-




tion on atmospheric processes in mountainous terrain and to  make recommenda-




tions regarding both the present application of air quality  models to




complex terrain settings and the research necessary to meet  future needs.







This report contains the thoughts and judgments of 32 atmospheric scientists




who gathered to exchange such technical information and research results on




atmospheric processes in complex terrain and to comment on matters relating




to adjustments in current air quality modeling practices.

-------

-------
                               TABLE OF CONTENTS
Preface                                                                     iii
Abstract                                                                      v
Executive Summary                                                             1
1.   Introduction                                                             6
2.   State of the Science Review                                              9
     A    Phenomena of Importance                                             9
          (1)  Interaction of Elevated Plumes with Windward Facing            9
                    Terrain Features
          (2)  Lee Side Effects                                              19
          (5)  Valley Situations                                             22
     B    Physical Modeling Capabilities                                     26
          (1)  Comparisons of Field Data with Laboratory Results             27
          (2)  Complex Terrain Applications of Physical Modeling             28
     C    Mathematical Modeling Capabilities                                 29
          (1)  Current EPA Recommended Models                                30
          (2)  Windward Flow about Isolated Hills                            31
          (3)  Mesoscale Flow Models                                         35
          (4)  Land and Sea Breeze Interaction with Complex Topography       36
          (5)  Modeling of Slope and Mountain/Valley Flows                   37
3.   Recommendations                                                         39
     A    Use of the Science                                                 39
          [1)  Comments on Current Research Efforts                          39
          (2)  Need to Quantify Uncertainties in Model Predictions           40
          (3)  Meteorological Input Requirements for Complex Terrain Models  42
          (4)  Suggestions on Improving Present Screening Models             45
          (5)  Suggestions on the Use of Physical Modeling Techniques        47
     B    Future Research and Development Needs                              48
References
Appendix (1)    List of Workshop Attendees                                    64
   "     (2)    Workshop Schedule                                             66
                                      VII

-------
                               EXECUTIVE SUMMARY

The presence of mountainous terrain introduces significant complexities
into the atmospheric transport and diffusion processes affecting ambient
air quality concentrations in a given area.   Terrain acts to distort
otherwise organized flow patterns resulting  in the creation of regions
of converging and diverging flows, enhanced  shear effects, and turbulent
eddies.  These alterations affect both flow  trajectories and ambient
turbulence levels to a large degree.   Terrain also determines the development
of local circulations in mountain-valley settings.  The net effect on
air pollution concentrations depends  critically on the specific geometric
and topographic relationships and on  the characteristics of the flow
fields.  For similar source sizes and release heights, emissions from an
air pollution source located in complex terrain may result in concentrations
on nearby high terrain several times  larger  than the maximum concentration
expected in the absence of the high terrain.

Similarly, stagnation effects in confined valleys can result in the
build-up of concentrations much higher than  those which would be observed
under similar large scale meteorological conditions over flat terrain.
For these reasons, the prediction of  air quality concentrations in
regions of complex terrain has remained a key area of concern for regulatory
agencies and methods for quantifying  atmospheric dispersion processes
have received considerable attention  over the past several years.

This workshop was convened by the American Meteorological Society via a
cooperative agreement with the U.S. Environmental Protection Agency.
Its purpose was to encourage atmospheric scientists working in this area
to exchange recently acquired information on atmospheric processes in
complex terrain and to make recommendations  regarding both the present
application of air quality models to  complex terrain settings and the
research necessarv to meet future needs.
                                       -1-

-------
The following summarises the major conclusions of the workshop about the
present state-of-understanding of dispersion in complex terrain.

(1)  Interaction of Elevated Plume with Windward Facing Terrain Features

     The largest ground-level concentrations associated with elevated
     releases near terrain rising above plume height are often associated
     with stable atmospheric conditions.  The dynamics of the air flow
     depends upon the Froude number of the upstream flow based on hill
     height.  A related parameter, the dividing or critical streamline
     height appears to vertically separate a flow regime which can
     transport a plume up and over a terrain object from a flow regime
     which constrains plumes to stagnate or to pass around the sides of
     a terrain feature.  Verification of the above concepts has emerged
     from physical modeling efforts for a variety of terrain shapes and
     the concepts are supported by field measurement results.  Mathematical
     models are presently being refined for these situations.  Further
     verification is needed, however, to understand the applicability of
     the concept to a broader variety and larger scales of terrain
     geometries.

(2)  Turbulent Dispersion Rates in Regions of Complex Terrain

     Atmospheric turbulent dispersion rates in complex terrain can often
     be expected to exceed those over flat terrain for otherwise similar
     conditions.  This is especially true under stably stratified conditions
     and results from the presence of gravity-driven drainage flows,
     gravity waves, the production of shear from flow, deformation, and
     the creation of eddies from upwind terrain features.  The effect on
     ground-level concentrations depends upon the specific source-
     receptor geometry.

-------
(5)   Lee Side Effects

     The flow on the lee side  of terrain features  can,  under certain
     conditions,  also  cause high ground-level  concentrations.   During
     neutral conditions, flow  "separation"  can occur on leeward slopes,
     causing poor ventilation  of emissions  from lee  side sources within
     the wake cavity region.   Under stably  stratified conditions,
     streamlines  passing over  the crest  may pass closest to  the surface
     on the leeward side.   This  could give  rise to highest ground-level
     concentrations on the  lee side from plumes originating  upwind  of  a
     terrain feature.   No widely accepted models exist  for these situations.

(4)   Valley Situations

     A number of air pollution phenomena are associated with the constraints
     on ventilation that valleys impose  or  with the  gravity-driven  local
     flows  created by  valley side walls.  The  most severe effects  are
     the occurrence of multi-day air pollution episodes within  deep
     valleys during periods when high pressure systems  stagnate over a
     region.  Mathematical  models for air pollution  applications in deep
     valleys are  largely in the  development stage  at the present time.

(5)   Physical Modeling Capabilities

     The use of physical modeling principles with  wind  tunnels  and
     towing tanks has  increased  markedly over  the  past  several  years.
     These  techniques  allow systematic investigations of flow situations
     under  controlled  conditions not practically achievable  with field
     studies.  Properly interpreted,  the results of  these tests add
     substantially to  our understanding  of  phenomena.   Limitations
     remain for the study of two-dimensional terrain features under
     stable conditions and  in  the spectral  range of  turbulence  which can
     be simulated.
                                     -3-

-------
(6)   Mathematical Modeling Capabilities

     As new information is emerging from theoretical  efforts,  physical
     modeling and field experiments,  better and more  refined mathematical
     models are being developed.   Models have been developed or are
     under development to simulate many of the phenomena identified as
     important.  No model exists  which can simulate all of the phenomena
     for a given setting.  The verification of models is difficult due
     to the general lack of extensive data bases.   The trend of lower
     computer costs with time will encourage the development of more
     advanced models capable of using more extensive  input data.

Recommendations emerging from the workshop were in several areas.

(1)   Recommendations on the Use of the Science

     •    The workshop participants supported the approach taken by
          current major research efforts in gathering field and laboratory
          experimental data for ultimate use in developing better mathematical
          models.

     •    A need to quantify uncertainty in model predictions was identified
          together with the recognition that this involved obtaining
          more information about flow conditions than is generally
          needed to estimate mean values from deterministic models.

     •    A viewpoint was taken that the dispersion of pollutants in
          complex topography involved interactions of air flows and
          terrain structure leading to certain natural fluid dynamic
          time and space scales which would not necessarily be important
          for simpler (e.g., flat) terrain problems.   If such time
          scales are larger than the averaging time of application
          interest, estimates of concentrations by deterministic models
                                       -4-

-------
          uould contain larger uncertainties.   This  resulted in the
          suggestion that the  stochastic  nature of the  phenomena be
          recognized and accepted by those  developing or applying models
          for these circumstances.

     •    The meteorological  input  data needs  for  complex terrain models
          is  greater than those required  for level  terrain models.
          Specific  requirements for vertical profiles of temperature  and
          velocity  emerge from the  need to  estimate  Froude numbers  and
          dividing  streamline  height.   On-site turbulence measures  were
          also identified as  being  especially appropriate for complex
          terrain modeling efforts.

(2)   Recommendations on Research and Development Needs

     The  workshop participants identified a number  of specific technical
     areas needing  further research to  further advance  our ability  to
     predict  ground level concentrations  of pollutants  in complex topography.
     Table 1  lists  the  topics  in summary  fashion and identifies the
     status of available observational  data,  physical conceptualizations,
     and  modeling efforts.  The order of  topics in  this  list does not
     signify  the  order  of priorities.
                                      -5-

-------
                      Summation  of  the  Keystone  Workshop
                                   May, 1983
1.   Introduction

The need to estimate reliably the impact of emission sources in regions
of complex terrain for regulatory decision-making purposes remains as a
key challenge to the meteorological community.   It has been identified
as an area for further study arid as a controversial area at all of the
public hearings on the EPA Air Quality Modeling Guidelines, and at
numerous professional society meetings and conferences.  The report of
the AMS/EPA Cooperative Agreement "Air Quality Modeling and the Clean
Air Act" (AMS, 1981) identified dispersion in regions of mountainous
terrain as one of the most important areas for further research and
development activities.

The subject has in fact received considerable professional attention
over the past several years as is evident from the large number of
papers presented at AMS and APCA meetings as well as other specialty
meetings and workshops.  In July of 1979, the EPA sponsored a "Workshop
on Atmospheric Dispersion Models in Complex Terrain" (Hovind et al.,
1979) for purposes of developing specific recommendations to EPA with
respect to the design of a multi-year program to address complex terrain
modeling problems.  On the basis of this, the EPA subsequently began a
multi-phased program of field studies, physical modeling experiments and
mathematical model development.  Independently, the Electric Power
Research Institute  (EPRI), as part of their Plume Model Validation and
Development (PMV§D) study  (Bowne et al., 1983), has undertaken a field
program which will include two experiments in complex terrain.  The
Department of Energy's Atmospheric Studies in Complex Terrain  [ASCOT)
program (Dickerson et al., 1980, 1983) has been investigating the problem
from the perspective of energy development needs including geothermal
power and oil shale.  Although none of the above large research efforts
is completed, they are at the stage where considerable field and laboratory
                                      -6-

-------
experience has been gained.  Also, a number of mathematical model development
activities have been pursued and findings from these experiences are
rapidly becoming available.  For these reasons, the Steering Committee
of the AMS/EPA Cooperative Agreement decided that it would be timely to
review the progress made in this field and to encourage a discussion
about what has been learned, what the implications are to present practice
in dispersion modeling, and what these experiences might suggest with
respect to possible augmentations to current research efforts.

To achieve these objectives, this Workshop on Dispersion in Complex
Terrain was organized and conducted.  The agenda for the workshop and a
listing of the attendees are included in the Appendix.  The specific
purpose was to provide a forum for the exchange of technical information
on atmospheric dispersion processes in mountainous terrain and the
relationship of this information to the modeling of air quality concentrations
in terrain settings.  A specific focus for the workshop was to provide
scientific information for use in the activities of EPA.  The workshop's
agenda emphasized measurements and observational data at the beginning.
This focused attention on the new information obtained from field studies
over the past few years and on the importance of understanding meteorological
phenomena for model development in regions of complex terrain.

Following presentations on observed phenomena, the workshop proceeded to
foster discussions focusing on modeling techniques - mathematical and
physical.  These topics were divided into three sections:

(1)   Flow Field Modeling.   Presentation and discussion on theory and
     practice of simulating the wind and temperature fields in complex
     terrain settings.

(2)   Physical Modeling.   Presentation and discussion on the use of and
     results from physical modeling experiments in wind tunnels and
     towing tanks.

-------
(5)  "Regulatory" Modeling.   Presentation and discussion focusing on
     experiences and practice in predicting ambient concentrations for
     use in regulatory decision-making.

The final sessions of the workshop involved dividing the participants
into two groups to address:  (1)  how recent findings should affect current
practice in the application of models;  and (2) future information and
research needs.  These included the development of recommendations or
identification of consensus on modeling approaches, identification of
areas needing further information, and recommendations to the technical
and regulatory community at large regarding how to achieve future needs.

This document presents the conclusions made regarding the state-of-the-
science of dispersion modeling in complex terrain, recommendations made
on the use of the science for the application, and recommendations on
future research needs.
                                      -8-

-------
I.    State  of  the  Science  Review

 A   Phenomena of  I_mpo_rtan_ce

                 •
     In  this chapter,  phenomena  which  are  of special  importance  to  the
     problem of estimating atmospheric dispersion  processes  in or  near
     complex terrain are described.  Where the phenomena are easily
     described by  mathematical algorithms,  these are  presented.

     (1)   Interaction of Elevat_ed Plumes with Windward Facing Terrain Features

          Flow Parameters  Affecting  Flow Trajectories.   If a source of
          emissions  is located near  a  hill that is taller than its  stack
          or release height,  the possibility exists that the highest
          concentrations to  be expected in the area will occur on  the
          facing hillside  when the airflow is from the stack toward the
          hill.  These high  concentrations would be expected either by
          direct plume impaction during strongly stable conditions, or
          by near  misses as  the  streamlines pass close to the hill
          during less  stable, neutral  or unstable  conditions.  This
          section  of the report  addresses  our knowledge of flow  conditions
          on windward facing terrain features.

          The  presence of  mountainous  terrain has  several effects  on the
          flow upwind and  above  the  terrain.   It acts to distort the
          flow field causing accelerations/decelerations and associated
          turning, contractions  and  expansions of  air parcels as they
          pass by  terrain  features.  It also acts  to  alter the structure
          of turbulence within the region  of flow  near the surface.  The
          dynamics of the  ambient flow field upstream of a hill  depend
          critically on the  ambient  density (or temperature)  stratification.
                                     -9-

-------
In very stable conditions,  vertical motion is opposed by
restoring buoyancy forces.   The stratification effect can be
characterized by a Froude number Fr given by
                         Fr=U/Nh                       (1)
where U is a characteristic wind speed for the upstream flow;
N is the Brunt-Vaisala frequency given by

                       - '  ^ ff'1/2                 m

where g is the gravitational acceleration, P is the air density,
and h is the hill height.  The Froude number can be interpreted
as a ratio of inertia! to buoyancy forces in a fluid.  Moderate
to neutral stability  [dominated by inertial effects) includes
the range  1 
-------
This formula is consistent with a simple energy balance argument
for an air parcel as first put forth by Sheppard (1956) .
Sheppard postulated that for a given environmental lapse rate,
one could calculate the value of horizontal velocity far
upwind which would enable air to just surmount a hill by
equating the kinetic energy of a fluid parcel upwind to the
potential energy change associated with lifting the parcel to
the hillcrest.   Snyder et al.   (1985) have extended this
argument to arbitrary velocity and density profiles and to a
wide range of hill shapes with the result
            1/2  oU (H )  ğ  I  pN (h-Z)dz
            A* *•  f*     .Ğ    U*
where H  must in general be determined iteratively.
                          -11-

-------
For two-dimensional ridges or very long finite ridges,  the
fluid can be blocked (i.e.,  effectively become stagnant) ahead
of the obstacle.  For these geometries, a well accepted formula
for the dividing streamline height does not yet exist.   In
addition, there is ambiguity on the upstream extent of the
blocked-fluid region and the variation in depth of the region
with upstream distance.  Thus, while solid evidence exists for
flow field and H  behavior for three-dimensional hills of not
                c
too large an aspect ratio, further experimental laboratory and
field work is required to extend the dividing streamline
concept to more general topography.

In addition to more experimental work, a more complete model
is required for the dividing-streamline height.  The simple
energy argument of Sheppard assumes that the air parcel has a
zero horizontal velocity at hilltop.  However, this assumption
is inconsistent with observed flow fields which show that
fluid speeds up at the top of the hill.  A more complete model
is required to adequately explain the above speedup phenomena
and to ensure reliable extension of the Hc concept to more
complex geometries.  For example, in less strongly stratified
flows  (Froude number greater than, but near unity), the hydrostatic
solution of Smith  (1930) is probably more applicable than
potential flow.  Whereas potential flow suggests symmetrical
streamline patterns upwind and downwind of a  symmetrical
terrain feature, the hydrostatic solution dees not.

In Smith's model,  about halfway between h and H ,  stronger
lateral streamline deflections will begin. . These  deflections
are very sensitive to  initial upstream elevation,  so that a
plume will spread  into a dome shape covering  the top of the
mountain.
                            -12-

-------
Near Hc the flow will be brought to rest and the flow will
split.  The plume will probably go to one side or the other
and may oscillate back and forth, being very sensitive to
upstream flow direction.  Large concentrations are expected
near this level as well as large apparent lateral diffusivity.

Dispersion in Strongly Stratified Flow.  Strongly stratified
flow below Hc has insufficient kinetic energy to pass over the
hillcrest and, neglecting wind shear effects, such flows can
be considered essentially horizontal as they pass around a
hill.  When the plume is flowing along the stagnation streamline,
the plume will "impinge" on the hill resulting in surface
concentrations nearly as large as those in the elevated plume's
center (Snyder and Hunt, 1984) .

Along the stagnation streamline, the flow diverges as it
approaches the hill.  Hunt et al., (1979) show that the large
increase in the crosswind dispersion coefficient, a~y, caused
by diverging streamlines is almost compensated for by the
decrease in wind speed, U, as the stagnation point is approached,
and that at the stagnation point, o~T,U is approximately the
same as it would have been in the absence of the hill.  These
arguments suggest that the concentration at the stagnation
point is approximately equal to that which would occur in the
atmosphere without the hill.  Full doubling of the concentration
values due to surface "reflection" effects does not occur
under these conditions.

Surface concentrations at an assumed point of impingement or
stagnation can be estimated during plume meandering conditions
by integration over the changes in wind direction.  During any
                            -13-

-------
quasi-steady period, the concentration at the stagnation point  is


given by C.(9, 9 ) where 6 is the wind direction and 9


is the stagnation wind direction.  The hourly concentration is  given


by:
     Cmax - '  C.(6,ed)P(e)de                                        (5)
where P(6) is the frequency distribution of wind directions


(Strimaitis et al., 1983).  If the plume spread is due principally


to plume meander, then
     C    = CP(9.)a ,
      max       d  s
where
     C = Q/(/2iro Ua x),
                z  s
     e = e.-Hs /2,
          d— s
and x is the distance from the release to the stagnation point.   If


P(6) is Gaussian, then
               z  8          2erQ
where a  is the standard deviation of the wind direction  about
       Q

the mean wind direction 9  .  P(9) can also be specified


explicitly by the observed distribution of winds during the hour.
                                -14-

-------
Experimental Evidence.  Snyder et al.  (1982, 1985) have shown
agreement of Eqn. 4 with laboratory simulations under stably
stratified conditions.  These tests were made at the EPA Fluid
Modeling Facility and in the stratified wind tunnel at the
Japanese National Institute for Environmental Studies.  The
concept of Hc was examined for a bell-shaped hill, a cone, a
hemisphere, triangular ridges, a sinusoidal ridge, vertical
fences and for a scale model of Cinder Cone Butte  (CCB) near
Boise, Idaho.  Snyder and his co-workers have concluded that
the integral equation for estimating Hc accurately predicted
the separation of the flow regimes for a wide range of hill
shapes and profiles.

The EPA-sponsored Small Hill Impaction Studies, conducted at
Cinder Cone Butte (CCB), Idaho, and Hogback Ridge  (HBR), New
Mexico, (Lavery et al., 1983) have also shown that the integral
formula for HC discriminates between the flow regimes for both
an isolated axisymmetric hill (CCB) and a very long ridge
(HBR).  Photos of oil-fog plumes and ground-level concentration
patterns of two tracers released clearly distinguishes between
the horizontal flow and the flow that goes over the hills.
The HC concept and its ability to predict whether plumes
impinge upon a hill and pass around it or travel up and over a
hill was also found by Ryan and Lamb (1984) and Ryan et al.
(1984) to be valid at Steptoe Butte, a large (335 m), isolated
hill in eastern Washington.

An analysis of the observed tracer gas concentrations and the
meteorological data at CCB showed that the_highest X/Q occurred
when the release height, H, was near or slightly higher than
H .   During this situation the plume was transported directly
toward the hill and produced high ground-level concentrations.
Lower releases tended to be transported around the hill sides
and releases above Hc were transported up and over.  The
highest normalized concentration occurred when H~H .
                          -15-

-------
The HBR experiment also showed that Hc discriminates between
the flow regimes, although the nature of the flow below Hc is
currently under investigation.  A preliminary analysis of 34
tracer-hours showed that the highest VQ occurred when H < Hc.

Further field observations showing horizontal flow about
larger-scale terrain features at low Fr have been reported by
Williams and Cudney (1976).  Field experiments by Rowe et al.,
1982 also tended to validate the dividing-streamline concept
for a ridge with a length about 50 times its height of SO m.

Thus, there appears to be consistent agreement between field
and laboratory observations on the flow structure over a
three-dimensional hill, in particular the horizontal nature  of
the flow below Hc and the dependence of Hc on stratification.
This is true for axisymmetric hills and for hills with small
aspect ratios  (with width/height ratios as large as 16)  (Snyder
et al., 1985) .

Field experiment verification is still needed for the validity
and applicability of the dividing-streamline concept for  large
terrain features  (greater than several hundred meters).

Flow over Terrain during Neutral Conditions.  It is generally
accepted that  the first-order effects of terrain on the  flow
on the windward  face during neutral conditions can be estimated
using modifications to potential flow theory  (Hunt and Mulhearn,
1973; Hunt  et  al., 1979).  For regulatory applications,  a
practical approach involves the superposition of a Gaussian
plume model  (steady state or  "puff" type) onto trajectories
determined  by  potential flow  approximations  (Isaacs et al.,
1979; Hunt  et  al., 1979).  Egan  (1975) demonstrated that  the
"half-height"  terrain correction factor followed from approximations
                           -16-

-------
of the effects of potential flow over a hemisphere but a
"terrain-following" plume path assumption provided first order
estimates for neutral flow over a two-dimensional (ridge-like)
shape.  .Neutral conditions often are associated with high wind
speeds and synoptically persistent meteorological conditions.
Thus neutral conditions can be of importance to the maintenance
of 24-hour average ambient air standards or Prevention of
Significant Deterioration (PSD) increments, especially where
"channeling" effects of terrain features on the wind fields
are important.

Dispersion during Unstable Conditions.  For most regulatory
applications, "worst case" conditions for sources close to
high terrain are expected to occur during stable or neutral
conditions.  For this reason,  phenomena during unstable conditions
have not been studied in depth.  Because of the differential
heating of mountain slopes during daylight hours, convection
effects do result in sustained and significant updrafts and
downdrafts.  Also "fumigation" of pollutant material onto
hilltops, in mid to late mornings has been observed to result
in short durations of high concentrations.  Currently available
models generally use the flow trajectories for neutral conditions
also for unstable conditions.

Turbulence Levels in Regions of Complex Terrain.  In general,
turbulence levels over complex terrain are expected to be
higher over complex terrain than over level terrain for the
same atmospheric stability classification.  These increased
turbulence levels are most likely a result of the following
phenomena.
                            -17-

-------
(a)   Nocturnal,  radiational  cooling  which  produces  surface
     inversions  is  often coupled  with  very low  wind speeds  in
     level  terrain  and is likely  to  result in the generation
     of gravity-driven drainage flows  in complex terrain.
     These  flows result in the mechanical  production of turbulence
     and time-dependent nonstationary  secondary motions which
     periodically sweep the  terrain.

(b)   Topographic alteration  of flow  direction and  speed will
     result in the  production of  shear in  all directions which
     not only contributes to the  production of  turbulence but
     also results in large flow meandering.

(c)   In complex terrain the  presence of flow stratification is
     a key element  in the production of regions of rapid flow
     acceleration and deceleration,  waves, rotors,  and "hydraulic"
     jumps  which tend to produce  shearing  motions,  and associated
     turbulence.

     The effects of the above relative to  flat  terrain seem to
     be largest under stable conditions.   During  neutral and
     unstable atmospheric conditions,  the  effects  of terrain
     on increasing turbulence levels generally  appear to be
     smaller (Start et al.,  1975; Egan, 1975).

Lateral turbulence levels in complex terrain are  generally enhanced to
a greater extent than vertical turbulence.  Plume  meandering and
uplifting as a result of interactions  with upwind  terrain are reasons
cited for an increase in horizontal  dispersion  rates.  The extent to
which such rates are larger at higher  elevations  above terrain is not
well quantified.  When the atmosphere  is stably stratified, generalizations
are difficult.  Air parcels downwind of a  ridge may be rapidly dispersed
upward  (or even upwind) by rotor  zones or  other eddy motions associated
                           -18-

-------
       with  lee  side  phenomena.   Note,  however,  that  terrain  features,
       especially ridge-shaped  features,  can  contribute  to  large
       scale stagnation  of the  air  flow (blocking)  at base  locations
       and can result in very low winds with  little net  transport  of
       pollutant material  into  or out of an area.

(2)     Lee-Side  Effects

       Previous  discussion has  focused  on the phenomena  of  importance
       in determining ambient air quality concentrations on upwind-
       facing slopes. Mathematical  and physical modeling shows  that
       high  concentration can also  be expected,  for certain meteorological
       conditions,  on the leeward slopes of mountains downwind of  a
       source.   Field measurements  are  uncommon  for these situations
       as regulatory  requirements have  generally focused on obtaining
       information on the upwind  facing slopes nearest to a pollution
       source.

       This  section provides a  brief overview of the  current  understanding
       of flow in the lee of hills.

       Neutral Flow.   Simple flow field models,  e.g., potential  flow
       coupled with rapid distortion theory,  work  reasonably  well  for
       predicting surface concentrations on the  upwind faces  of
       hills,  both two-dimensional  and  three-dimensional.   However,
       these simple models are  inadequate for predicting wake effects,
       even  for  hills of moderate slope (i.e., 15  degrees for two-
       dimensional hills and 25 degrees for three-dimensional hills),
       let alone for  steep hills  with separated  wakes.   Flows on the
       lee sides of hills are among phenomena expected to cause  high
       ground-level concentrations  in the vicinity of terrain and  for
       which no  routine  model simulation techniques are  available.
                                  -19-

-------
Snyder (1983)  summarized a variety of idealized neutral-flow
wind tunnel studies of plumes from stacks located both upwind
and downwind of various terrain shapes.   The results show that
ground level concentrations on the lee side of the obstacle
can be many times higher than would be expected if the terrain
were not present.  This increased concentration over the "no-
obstacle" concentration when expressed in a ratio is termed
the "terrain amplification factor" (taf).  It is specifically
defined as the ratio of the maximum concentration occurring in
the presence of the hill to the maximum concentration that
would occur from the same (elevated)  source if it were located
in flat terrain.

Physical modeling experiments using simulated atmospheric
boundary layers show that plumes released downwind of variously
shaped two-dimensional hills all resulted in taf's of 10 to
15, whereas plumes released upwind or on top of the hills
produced taf's of 2-4.  Two lee-side phenomena were observed
to produce high taf's.  One was the reverse flow within a
cavity that recirculated plumes from stacks as high as the
hill.  The other was flow that had not separated but due to
the low transport speeds and high turbulence levels, plumes
dispersed rapidly to the ground farther downwind.  High taf's
were also observed from sources located on the lee-sides of
three-dimensional hills.  The highest taf's occurred when the
source was placed approximately on the separation-reattachmert
streamline.  In the 3-D cases, higher crosswind  aspect ratios
(across wind length of hill divided by hill height) generally
produced higher taf's for downwind sources and smaller taf's
for upwind sources.
                           -20-

-------
Physical modeling results indicate that typical lengths of
reversed flow regions are of order 10 hill heights for two-
dimensional hills and 2 to 10 hill heights for three-dimensional
hills.  In addition, strong trailing vortices downwind of
three-dimensional hills have been observed in laboratory
studies.  The strong downwash caused by these vortices can
result in large surface concentrations.

While no specific mathematical models can be suggested at this
time, it is recommended that sources located 10 hill heights
or closer downwind of a terrain feature be investigated for
possible high impacts due to the obstacle.  While physical
modeling provides a cost-effective means for examining such a
system systematically,  field studies based upon detailed
tracer releases can serve to identify maximum ground level
concentrations.  Snyder (1981) and references in Snyder
(1983) provide information on the proper use of physical
models to account for lee side influences.

Stratified Flow.   Under strongly stratified flows, pollutants
released below the dividing-streamline height on the lee side
of hills have been observed to be recirculated to the hill
surface and to cover a narrow vertical band spread over roughly
a 180 degree sector of the hill surface.  Whereas instantaneous
concentrations are observed to be considerably lower than
those associated with impaction (from upwind sources), long-
term-average concentrations may be larger than impaction
concentrations because the wind meander will significantly
reduce the time-averaged impaction concentrations, but not the
lee-side concentrations.   Even simple models for predicting
lee-surface concentrations from downwind sources are not
available.
                             -21-

-------
(3)     Valley  Situations

       Many air  pollution  sources  such  as  cities,  roads,  industrial
       operations  and  energy production facilities are  located  in
       mountain  valleys.   It has been recognized  for  at least 40
       years (Hewson and Gill,  1944) that  air pollution problems can
       arise from  these sources due  to  the special meteorological
       processes that  occur in  valleys.

       The workshop attendees separated the discussion  of the dynamics
       of individual plumes interacting with high terrain from  discussions
       of the  special  flow conditions associated  with valley settings.
       The processes  identified to be of special  importance  to  the
       valley  situations  include nocturnal drainage flows,  fumigation,
       flow channeling by  valley sidewalls,  and persistent low  wind
       speed stable flows.   For purposes of presentation, it is
       convenient  to  distinguish between relatively shallow  valleys,
       deep, draining  valleys,  and closed valleys.

       Shallow Valleys.  Shallow valleys were  defined by comparison
       of valley sidewall  height to  the effective height of a plume
       from a  source  affecting air quality in  the valley.  A valley
       is "shallow" if the plume is  significantly higher than the
       terrain features.   Under these  conditions, the plume  is  cut
       off from  the valley boundary  layer during  stable conditions
       and reacts  in  a manner analogous to a plume over flat terrain.
       Although  the trajectory of  the  plume may be steered somewhat
       by the  valley  orientation,  the  centerline  of the plume is
       higher than the valley sides  or  ridges  forming the valley.

       Preliminary results reported  at  the workshop and described  by
       Reynolds  et al., [1984)  from  the EPRI Plume Model Validation
       and Development project tracer  studies  at  Bull Run, Tennessee
                                -22-

-------
 (a shallow valley), indicated that terrain influences on
 lifting of the plume paths were not observed as the plumes
 from the facility usually followed trajectories that were
 parallel to the valley axes.  Other analyses  (e.g., Turner and
 Irwin, 1982) indicate that flat terrain models overestimate
 the ground level concentrations on terrain if the plume centerline
 is not deflected upward.

 Deep, Draining Valleys

 Scientific investigations of valley flows have,  so far,  focused
 primarily on improving our understanding of the  physics  of
 valley meteorology, and only a few important research studies
 have focused directly on air pollution investigations (e.g.,
 Start et al,  1975;  Hewson and Gill, 1944).  An improved  understanding
 of valley nocturnal drainage flows is now becoming available
 from the U.S.  Department of Energy's Atmospheric Studies in
 Complex Terrain (ASCOT)  program (Dickerson, 1980,  1983).
 Other work has focused on the breakup of nocturnal valley
 temperature inversions in deep valleys (Whiteman,  1982).  This
 work has led to a thermodynamic model of temperature inversion
 breakup (Whiteman and McKee,  1982) and, more recently,  to an
 initial model  of air pollution concentrations produced  on the
valley floor and sidewalls due to post-sunrise fumigations of
 elevated nocturnal  plumes (Whiteman and Allwine, 1983) .   In
 these studies  the effects of convective boundary layers  which
 grow over heated valley surfaces after sunrise,  the effect of
 upslope flows  produced over the sidewalls, and the effect of
 compensating subsiding motions over the valley floor have been
 simulated but  need  field evaluations.

 In actual valleys,  topographic complications can be expected
 to greatly influence the development of local circulations and
 the dispersion of pollutants  emitted within the  valley.   The
                          -23-

-------
diversity of valley shapes,  orientations,  presence of tributary
valleys, and presence of terrain constrictions along the
valley axes can be expected  to influence the development of
the along-valley circulations, turbulence  levels,  and other
important aspects of valley  meteorology.

Closed Valleys.   Certain valleys with weak or obstructed
outflow have been characterized as trapping valleys in contrast
with draining valleys with vigorous outflow.  The  accumulation
of cool air draining from the sides of the trapping valleys
will build up a deep, stable layer during nighttime hours
that is capped by a stronger inversion at  the interface with
the above-valley air near the top of the cold pool.  Pollution
plumes emitted into this domain are likely to be confined
below the inversion and between the valley sidewalls.

In addition to diurnal trapping regimes, certain synoptic
conditions produce stagnation episodes.  High pressure systems
characterized by low wind speeds, clear or foggy skies, subsidence
inversions and nocturnally-produced ground level inversions,
may exist for 4-5 day periods.  During these episodes, additional
emissions are not compensated by flushing so that air quality
can continually degrade to the point of threatening human
health.  The pollution conditions are not stationary, as there
may be sloshing of air masses within the valley.  The DOE
ASCOT program characterized  diurnal pooling and stagnation
within one California valley in 1979 and 1980.  Murphy et al.,
(1984) studied the accumulation of smoke from space heating in
a trapping valley in Colorado and Marlatt et al.  (1981)
studied the structures of a  trapping valley inversion.

The shape of a closed valley creates unique flow regimes.
Limited field data show  that nighttime radiational cooling of
the surrounding mountain slopes can create a downs lope drainage
                          -24-

-------
flow that appears to reach maximum strength just prior to
local sunrise.  The drainage flow tends to move toward the
lowest point in the valley.  Available data are inadequate to
determine if a gyre usually develops over the valley low point
prior to sunrise.  Some observations from the Nevada test site
(Randerson, 1984) indicate that material released at ground
level within a closed valley at night can be transported out
of the valley.  The implications of this situation for air
quality modeling are clear.  An effluent released into a
stratified environment can be transported out of the valley,
and this is difficult to explain physically.  Mo data are
available to permit a quantitative description of the behavior
of effluents released into a thermally-stratified valley.

Dispersion over Complex Terrain with Superimposed Convective Circulation.
In flat terrain, convective flows frequently lead to the
fumigation of pollutants trapped aloft or the early downwash
of a looping plume resulting in high ground level concentrations.
Convective flows developing over hills, ridges, or more complex
terrain may significantly alter streamline patterns, separation
and stagnation locations, and hill wake turbulence.  It is
possible that non-homogeneous radiative heating caused by
slope orientation could result in different convective scales
than commonly associated with horizontal terrain.  These
perturbed spatial or time scales could result in worst-case
ground-level concentrations from lee impaction, sudden subsidence
or downdrafts.

Orgill (1981)  cites the pioneering studies by Hewson and Gill
(1944) and describes two types of fumigation phenomena within
valleys.  Type I, a simple diurnal fumigation, occurs during
the early morning hours when surface heating develops convection
                         -25-

-------
up to an elevated plume.   Pollution reaches the valley bottom
nearly simultaneously throughout the valley in this case.
Type II, a dynamical fumigation (Tyson,  1968), may also be
related to the morning transition when an up-valley wind
undercuts the well developed, down valley,  mountain wind.  A
marked discontinuity in the wind and turbulence fields is
produced, leading to severe dynamic fumigations.  In this case
the fumigation event propagates up-valley with time.  Whiteman
and McKee (1979)  have identified wind and temperature structures
associated with the conditions for dynamic fumigation.
Associated with the 1982 ASCOT field program, Whiteman et al.,
(1984) performed tracer experiments to further document the
time and space character of the fumigation.

One of the major aims of the DOE ASCOT program is to contribute
to a better understanding of the relationship of the details
of these fumigation phenomena to the structure of the winds,
temperature, turbulence and valley topography.

Physical Modeling Capabilities

The report of the 1979 Workshop on Complex Terrain  (Hovind et
al., 1979) has a fairly detailed discussion on the background
and similarity criteria for physical modeling in mountainous
terrain.  In addition, the workshop document discusses the
relative merits of the wind tunnel and towing tank.  Since the
workshop, the EPA has published a document (Snyder, 1981) that
describes in detail the fundamental principles, practical
applications and the hardware associated with physically
modeling atmospheric dispersion.  EPA also has published a
guideline document  (Huber, 1981) that specifies the necessary
elements of a physical modeling study for assessing the stack
height required so that a plume is not adversely affected by
wakes and eddies of nearby terrain (or buildings).
                          -26-

-------
(1)   Comparisons  of Field Data with Laboratory Results

     Since  the  1979 workshop,  several  comparisons  of physical  model
     simulations  with actual  field observations have been made.
     Weil  et al.,  (1981)  described field and wind  tunnel  experiments
     conducted  for  the Westvaco Pulp Mill situated in the hilly-
     terrain of Western Maryland.   The wind  profiles measured  at
     similar locations under  unstable  stratification in  the  wind
     tunnel and in  the field  showed good agreement as did the
     ground-level concentrations on the windward side of a nearby
     hill.   The observed  concentrations in the wind tunnel were
     consistently lower than  those in  the field on the  leeward side
     of the hill  but only by  a factor  of about 1.6.

     Snyder and Lawson (1981)  reported on a  series of experiments
     attempting to  simulate,  in a towing tank, field results from
     the Cinder Cone Butte, Idaho experiment.   In  particular,  one
     hour  from  the  field  experiment that represented very stable
     atmospheric  conditions was simulated.   The results  of the
     tests  showed that the crosswind concentration distributions in
     the towing tank were exceedingly  narrow and that maximum
     concentrations were  5 to  10 times larger than those  observed
     in the field.   This  was  attributed to the fact that  low frequency
     fluctuations in wind speed and wind direction are not present
     in the towing  tank.   An  ad hoc attempt  was .made to  simulate
     the low frequency wind fluctuations by  superimposing concentration
     patterns from  tows done  at a series of  discrete wind speeds
     and wind directions.  This attempt was  moderately successful
     in that 80% of the model  concentrations were  within  a factor
     of 2.5 of  the  field  concentrations.   Snyder and Lawson  felt
     agreement  would have been better  had wind data at plume altitude
     been  available.
                               -27-

-------
     Other  field/laboratory  comparisons  for complex  terrain  settings
     include studies  by Meroney  (1980)  and Meroney et al.  (1980),
     which  simulated  wind speed  and  direction  measurements at  sites
     in New Zealand and Hawaii.   The measurements  in the laboratory
     correlated  well  with those  in the  field.

(2)   Complex Terrain  Applications of Physical  Modeling

     Since  the promulgation  of the stack height  regulation in  1979,
     numerous physical  modeling  studies  have been  conducted  to
     evaluate the effect of  terrain  wakes and  eddies on dispersion
     (Greenway et al.,  1981;  Petersen and Cermak,  1979; Petersen,
     1981;  to list a  few).   These assessments  have demonstrated
     that physical modeling  can  be used  in a regulatory environment
     to define  good-engineerir.g-practice stack heights.

     Mountain-valley  and drainage wind  simulations were conducted
     (Petersen et al.,  1980;  Cermak  and  Petersen,  1981) to provide
     qualitative information about flow  in deep  valleys and  quantitative
     information for  use in  refining and calibrating a  numerical
     model.  The physical simulation was also  used to assist in the
     design of a field  experiment.

     Petersen and Twombly (1982)  conducted a wind  tunnel experiment
     to evaluate maximum concentrations  on elevated  terrain  under
     stable stratification.   They compared the EPA Valley model and
     a new  model that included potential flow  theory and the dividing
     streamline  concept with the wind tunnel simulation. The
     average ratio of Valley model prediction  to wind tunnel observation
     was 2.6 whereas  the ratio for the  new model was 1.2.
                              -28-

-------
Snyder  (1983) conducted neutral flow wind tunnel experiments
with a  stack situated upwind, on-top and downwind of three
different shaped hills.  Ground-level concentrations were then
measured with and without the hill present.  An amplification
factor  was defined to be the ratio of maximum concentration
with and without the hill.  Snyder summarized these amplification
factors as a function of hill shape, stack location and stack
height.

These studies by no means represent all the applications of
physical modeling since 1979.  It is apparent, however, that
physical modeling has been used to assess concentrations in
complex terrain for the following situations:  1) plumes
impacting elevated terrain under neutral, unstable and stable
stratification, 2) plumes affected by the wakes and eddies of
upwind  terrain, 3) plumes released in deep valleys under
stable  nighttime conditions.

Mathematical Modeling Capabilities

One of the more important effects of complex terrain on the
plume transport and diffusion problem is the modification of
the ambient flow field.  Flow field assumptions in existing
diffusion models for regulatory applications are quite crude.
For example, diffusion models for computing concentrations on
the windward slope of isolated hills assume that the same
horizontal velocity exists everywhere,  upstream as well as
over the crest of the hill,  in contrast to the streamline
deformation and flow speedup which actually occurs.   Some of
these models attempt to account for the closer passage of
streamlines to elevated terrain and deformation of plumes by
                          -29

-------
using a "terrain correction" factor as discussed earlier,  but
as noted, even these models ignore many details of the corresponding
velocity changes and deformation effects on plume spread.

In this section of the report a brief description of models
currently recommended by EPA as screening techniques is
followed by a discussion of mathematical modeling techniques
for simulating various flows in more detail.

1.   Current EPA Recommended Models
EPA has not yet established a guideline model for use in
complex terrain settings where plume heights are expected to
be below the height of nearby terrain.  EPA does recommend the
Valley, COMPLEX I and sometimes COMPLEX II models for use as
conservative screening techniques for these situations.
Research on the development of a refined model is currently
underway (Lavery et al., 1983).  The following provides a
brief description of present screening models.

The Valley model (Burl:, 1977) uses a Gaussian plume dispersion
equation modified to include 22.5  "sector averaging" of the
crosswind distribution.  Vertical dispersion is calculated
using Pasquill-Gifford dispersion curves.  The model can be
run to calculate 24-hour average concentrations or annual
averages.  Under stable conditions, the model assumes that a
plume travels toward terrain with no vertical deflection until
the plume centerline approaches to within 10 m of a terrain
surface.  Thereafter a minimum stand-off distance of 10 m is
assumed.  Full doubling of concentrations due to reflection
                            -30-

-------
occurs.  At surface elevations above the plume height, concentrations
are assumed to decrease linearly with height, reaching a value
of zero at 400 m above the initial plume height.  Under
neutral or unstable conditions, the plume trajectory assumed
is "terrain-following" - producing concentration values equivalent
to those which would be obtained in the absence of terrain.
For screening analyses, the Valley model is commonly used with
assumed worst-case meteorological inputs of a stack-top wind
speed of 2.5 m/sec. and Pasquill-Gifford stability class F
vertical dispersion coefficients.  A 24-hour average concentration
is obtained by assuming that these conditions could occur for
6 hours within a 24-hour period.

The COMPLEX I (EPA, 1981a) and COMPLEX II (EPA, 1981b) models
are multiple point source models capable of using sequential
meteorological input data.  Under stable conditions, the
algorithms for plume impaction within COMPLEX I are the same
as those within Valley.  COMPLEX II differs from COMPLEX I by
using crosswind dispersion coefficients instead of 22.5
sector averaging.  For neutral and unstable conditions the
models use a 0.5 terrain correction factor to lift the plume
centerline height above terrain height.

2.   Windward Flow about Isolated Hills
The flow models discussed in this section, while highly simplified,
address some of the major effects of hills .on perturbing an
airstream, i.e., streamline convergence and divergence and the
associated fluid acceleration and deceleration.  The models
are considered for two stratification regimes, neutral/unstable
and stable; all are based on inviscid flow.
                          -31-

-------
For neutral/unstable conditions, potential flow can be applied
following the work of Hunt and Mulhearn (1973), Jackson and
Hunt (1975), and Britter et al., (1981).  These authors showed
that for two-dimensional hills and outside a thin "inner
region," the flow behaves like an inviscid shear flow.  Furthermore,
they find that potential flow is a useful concept for flow
speedup, as confirmed with field and wind tunnel measurements.
The Jackson and Hunt theory has been extended to three-dimensional
hills by Mason and Sykes (1979a, b) and found to be in good
agreement with wind observations about a small hill.

Potential flow has been applied in a number of nonregulatory
diffusion models for a variety of terrain shapes:  three-
dimensional hills (Hunt et al., 1979), simple combinations of
hemispherical and cylindrical shapes (Isaacs et al., 1979),
and arbitrarily-shaped three-dimensional hills (Weil et al.,
1981).   The last application employs a potential flow code
developed by the McDonnell-Douglas Company (Hess and Smith,
1962) and is based on the surface source method.  Further work
is required to adapt these methods to actual terrain sites for
use in regulatory applications.  Of particular concern here is
how much terrain detail must be included to get an adequate
prediction of the flow field and the associated concentration
distribution.

For stably stratified flow about three-dimensional hills, we
consider two stratification regimes:  moderately to weakly
stable (Fr > 1) and strongly stable (Fr
-------
mountain for both round hills and also near the end of a. long
finite ridge.  For general finite amplitude hills (not necessarily
low) and Fr -^ 1, the momentum equations can also be linearized
in density stratifications, i.e., by considering a small
perturbation in stratification about the neutral, potential
state.  In contrast to the more commonly used small-height
linearization, it has the great advantage of being applicable
to any terrain slope or height.  The most general outline of
this approach is given in Drazin (1961).   However, its only
application so far has been to two-dimensional hills (Baines
and Grimshaw, 1979).   A central question with regard to the
latter approach, i.e., stratification perturbation to potential
flow, is to how small a Froude number it can be applied.  This
question, as well as how well the above approach predicts flow
patterns about terrain obstacles, must be assessed experimentally.
Physical modeling would be of value in this assessment.

In strongly stable stratification (Fr< 1), the flow about
large three-dimensional hills is confined to horizontal planes
below a certain height H , which depends upon the stratification.
Such horizontal layering has been predicted theoretically by
Drazin (1961) and demonstrated in stratified towing tank
experiments by Riley et al., (1976), Hunt et al., (1978), and
Weil et al., (1981) for various terrain geometries.   It has
also been observed in the atmosphere (Williams and Cudney,
1976; Rowe et al., 1982; and Wooldridge and Furman,  1984).
Drazin suggested that the two-dimensional potential  flow could
serve as a useful ambient wind model for the horizontally
layered regime.  It should be emphasized that potential flow
is applicable only on the windward side of the hill; it cannot
describe flow within the separated wake in the lee of the
                          -00-

-------
hill.  A key problem in applying the model to real terrain
sites is the choice of appropriate lateral boundary conditions,
a problem also discussed below in the context of mesoscale
modeling.

Above the horizontally layered regime, fluid passes mostly
over the hill, as predicted by Drazin and as demonstrated in
the aforementioned laboratory experiments.  Hunt et al.
(1984) suggest a useful ambient flow model for the fluid
passing over the hill.  It is the linear stratified flow
theory (for Fr>>!) applied to a "cut-off" hill, i.e., that
portion of the hill above the horizontally layered regime
defined by H .  Thus, one could use Smith's (1980) theory (for
the limit Frğ 1) and "push" it to FrĞ*l.  We note that in
stable stratification, theory predicts that fluid flowing over
the hill does so asymmetrically, i.e., streamlines pass closer
to the surface on the leeward than on the windward side of the
hill.  Thus, in some instances, for a plume passing over a
hill we should expect higher surface concentrations on the
leeward side as observed in the physical modeling experiments
(Snyder and Hunt, 1984).

Models for stratified conditions, when implemented for regulatory
applications, will require more detailed meteorological data
inputs than are generally available.  Specifically, knowledge
of the temperature and velocity structure as a function of
height from the surface to heights above the terrain features
will be required.
                           -34-

-------
3.   Mesoscale Flow Models

This discussion of mesoscale models considers flow models
which may be applied in complex terrain for distance scales of
the order of 50 kilometers.  Major breakthroughs have been
achieved in recent years in the development of flow models
because of improved high speed computer systems.  Concurrently,
there has been development of improved flow models, thus
making it feasible to use these models for applications in a
variety of complex situations.

Mesoscale flow models may be classified into three major
categories:  1) fluid dynamic models, 2) diagnostic models,
and 3) objective analysis models.  There are many existing
models in each of these three categories.

Improved computer systems, and more efficient numerical schemes
which solve the complex equations, have made practical applications
of the mesoscale models more feasible.  For example, applications
can be simplified by allowing topographic data, models and
graphical output to be integrated with "user friendly," menu-
driven codes (Fox et al., 1983).   These mesoscale flow models
may be used in various ways to assist with air quality analyses.
However, there is a need for additional model evaluation
studies to compare the flow fields generated by the models
with wind flow observations.  Flow models may be used to
calculate wind fields in complex terrain situations where
there are few meteorological data measurements (Pielke et al.,
1983).  The models can be coupled with various advection/diffusion
schemes to simulate dispersion of pollutants.  Insight into
air stagnation in basins and valleys with resultant air pollution
buildup, and flushing of pollutants from basins and valleys
may be analyzed.  The characteristics of mountain and valley
                           -35-

-------
flows can be better understood by simulating cases where local
terrain drives the drainage or the upslope flows.   Diagnostic
or mass consistent models (e.g., Sherman, 1978; Davis, et al.,
1984; King and Bunker, 1984)  have been applied to  produce flow
fields for three-dimensional  complex terrain settings.  These
models to not, however, include thermal effects directly.  The
diagnostic models can be used to obtain a better understanding
of the controlling influences of strong synoptic flow patterns
on the local flow fields.

4.   Land and Sea Breeze Interaction with Complex Topography.
There has been considerable investigation of sea-  and land
breezes over flat terrain;  however, the interaction of these
circulations with complex terrain along and inland from the
coast has been investigated relatively less.  Recent studies
have documented the movement  of pollution offshore in the land
breeze, its fumigation down to the surface layers, and its
subsequent movement back onshore (McRae et al., 1982).  This
recirculation of pollution, which can occur for several days
or more at a time in the Los  Angeles basin, can result in aged
pollutant plumes with high concentrations of secondary pollutants,
In other investigations, numerical studies have demonstrated
that sea and land breezes in  complex terrain are not simply a
superposition of sea and land breezes over flat terrain, with
mountain-valley circulations  in complex terrain.  One reason
for this is that mountain-valley circulations near a coastline
can result in subsidence over the shoreline which will, in
turn, diminish the interaction of the sea breeze.   Thus, such
raesoscale systems must be treated in their entirety.
                         -36-

-------
5.    Modeling of Slope and Mountain/Valley Flows
a.    Existing Models.  Defant's (1951] descriptive model of
slope and mountain/valley flows, while rudimentary, provides
practical guidance for understanding thermally driven circulations
in well defined valleys.  According to Defant, at night shallow-
downs lope flows on the valley sidewalls (directly driven by
temperature deficits near the surface) produce a deep pooling
of cool air in the valley.  If there are variations in the
valley depth or width, an along-the-valley axis flow will
develop.  The along-the-valley axis flow can be quite deep
depending on the depth of the pool of cool air.  Defant's
descriptive model, while addressing the three-dimensionality
of the problems, cannot of course provide details of the flow,
handle interaction with synoptic winds, or address turbulent
structure.

Because the full problem is so complex, recent models have
separated the problem into components.  Rao and Snodgrass
(1981), and Smith and Garrett (1985), and others have concentrated
on the slope drainage flows.  These slope flow models appear
to reproduce quite well the observed temperature and velocity
structure.

Recently, full three-dimensional primitive equation models
have been developed and applied to slope an.d valley flows,
e.g., Egger (1981), McNider (1981), and Yamada (1981).
Despite continuing initialization and boundary problems, these
models offer the promise of including the slope-valley flow
dependence, and of dynamically simulating the evolving thermal
and turbulent structure.

-------
b.   Applications.  While full treatment of the flow with
primitive equation models is possible, actual application of
the models has been limited primarily to idealized terrain,
neglecting synoptic interactions.  Egger (1981), McNider and
Pielke (1981) and Bacler and McKee (1983) addressed a simple
three-dimensional valley opening onto a plain.  The results
from these idealized applications appear to agree qualitatively
with recent observations, (e.g., Whiteman (1982)), and the
descriptive ideas of Defant.

c.   Input Data.  For mass-consistent models the data requirements
can be quite voluminous, especially if the flows are primarily
thermally driven, since thermal effects can only be included
through wind observations.  For dynamic primitive equation
models the input data are fairly simple if the synoptic flow
is assumed spatially uniform and steady.  The basic input
information required is the topography, initial temperature
distribution, and gradient wind speed and direction.

As mentioned, definite problems for dynamic models still exist
in the areas of flow initialization and boundary conditions.
Computation speed is also a major drawback, with computer
memory and speed requirements outside the reach of most applications,

d.   Treatment of Dispersion Processes.  Because of spatial
variations in both wind direction and turbulent structure, it
is unlikely that simple Gaussian-dispersion models can be
utilized with such numerical flow models.  .Small-grid, finite-
difference, advection-diffusion calculations might be applicable.
Lagrangian particle models for point source transport and
dispersion are also promising.
                          -38-

-------
5.         Recommendations

          A number of recommendations evolved from the workshop.   These
          included statements regarding current research efforts  and the
          use of current technology,  suggestions of specific research
          needs focused on near-term needs (and which presumably  could
          be accomplished in the near-term)  and suggestions regarding
          the longer term research needs.   As will be seen, some  of
          these latter suggestions may equally well relate to other
          applications of geophysical fluid dynamics.  The recommendations
          are organized into two categories relating to recommendations
          on the Use of the Science and recommendations on Future Research
          Needs.

 A         Use of the Science

     (1)   Comments on Current Research Efforts.   First of all,  it should
          be noted that there seemed  to be a consensus among the  workshop
          participants that the approach taken over the past several
          years with a focus on gathering more field data on transport
          and diffusion processes  in  complex terrain settings was appropriate,
          The need for large efforts  to gather field data was strongly
          recommended at previous  workshops  on this topic (e.g.,  EPA
          Workshop on Atmospheric  Dispersion Models in Complex Terrain
          (Hovind et al.,  1979),  DOE  Workshop on Research Needs for
          Atmospheric Transport and Diffusion in Complex Terrain  (Barr
          et al.,  1977)),  and the  research programs of EPA, DOE and EPRI
          have been very responsive to these needs.  The increased use
          of wind tunnels  and water channels in applying physical modeling
          techniques to study flow phenomena was also supported by the
          participants.   The notion of performing scale model tests in
          parallel with full scale field experiments performed  in complex
                                    -59-

-------
     terrain was  viewed as  providing an excellent  methodology to
     understand  the  effects of physical scale  on the  phenomena of
     concern and  to  investigate systematically the effects  of
     changes of dispersion  patterns  with flow  conditions.   Such
     systematic  investigations are difficult and expensive  to
     accomplish  in field experiments.   Emerging mathematical  modeling
     techniques  for  predicting concentration values in complex
     terrain account for the physical  phenomena of importance
     better than  earlier models.  A  number of  these developments
     have  been made  hand-in-hand with  the increased data bases
     emerging from the  field and physical modeling programs.

(2)   Need  to Quantify Uncertainties  in Model Predictions.   One of
     the recurring issues surfacing  at this workshop  was the  need
     for modelers to incorporate uncertainty of predictions into
     the process  of  using models.  The group recognized the regulatory
     application  needs  of having decisions made on the basis  of
     certainty and on the basis of computations made  for both short
     and long averaging times and for  infrequent events. An admonition
     to include  uncertainty into the modeling  evolves from  the fact
     that  the flow over complex terrain, like  any  other atmospheric
     flow, is turbulent in  most regions at most times.  This
     turbulence  thus has both temporal and spatial scales.   These
     scales are  not  determined simply by the universal flow physics;
     they  are also strongly influenced by the  structure of  the
     topography,  by  the interactions of this topography with the
     ambient, larger-scale  atmospheric circulation, and, in dispersion
     applications, by the nature and geometry  of the  pollution
     sources.

     The stochastic  nature  of the atmosphere coupled with its
     nonstationarity and spatial complexity leads  to significant
     uncertainty  in  any prediction of its behavior.  It is  virtually
                               -40-

-------
impossible to measure initial conditions in a manner which
represents this complex flow.  The measurement problem is
magnified in complex topography where small scale boundary
forcing is related to small scale variability of the surface.
In idealized conditions (flat terrain) a recent AMS/EPA workshop
(Fox, 1984)  recognized that inherent uncertainty would remain
even if the measurement problem could be removed due simply to
the difference between individual realizations of seemingly
identical flows.  In practice, physical models offer an opportunity
to investigate components of this source of variance.  It is
possible that the physical constraints of topography might
tend to reduce this variance because of the more frequent
occurrence of locally organized secondary flows (slope flows,
etc.).   To the extent this may occur, such inherent variability
may be less over complex terrain.  This notion, however, is
hypothetical at this time - it has not been demonstrated.

The recommendation to include uncertainty into the modeling
process has several implications.  Firstly, it will not be a
simple task.  Deterministic models have evolved from development
techniques attempting to achieve "best fits" of predictions
with observations without (until rather recently)  attempting
to match uncertainties of predictions with uncertainties in
the observations.  Further, the amount of information required
to quantify uncertainty is greater than that needed to estimate
"mean" values since uncertainty of all the various input
values and parameters needs to be assessed.  For example,
theoretical  and observational work on uncertainty indicates a
strong dependence of concentration fluctuations on the source
exit conditions even at large downwind distances (Sawford,
                            -41-

-------
     1983;  Fackrell  and  Robins,  1982).   Secondly,  there  follows  a
     suggestion  that the effects  of  uncertainty  in input data  and
     parameters  should be studied with  physical  models where a
     degree of control of variables  with time  can  be  maintained.
     Thirdly,  the  recommendation  is  supportive of  pursuits  of
     higher-order  closure modeling which includes  consideration  of
     higher statistical  moments  of distributions and  of  large-eddy
     simulation  modeling wherein  uncertainty can be quantified by
     interpretation  of output  results of multiple  simulations.
     Finally,  the  recommendation  is  consistent with the  charge to
     rely on field data  (with  appropriate consideration  of  stochastic
     variability)  to evaluate  the statistical  performance of models.

(3)   Meteorological  Input Requirements  for Complex-Terrain  Models.
     An underlying concept in  the field of dispersion modeling is
     that one  cannot expect a  dispersion model to  perform well
     without appropriate input data.  Generally, "appropriate" is
     interpreted as  being of good quality, being representative  of
     the flow  field  of interest  and  providing  reliable  information
     on the flow parameters of importance. Some specific recommendations
     appropriate for dispersion  models  as applied  to  regulatory
     applications  are provided below.

     The set of  meteorological data  collected  must be representative
     for the site  of the source  and  the probable paths  of pollution
     diffusion,  including those  paths which are  highly  repeatable
     and those that  are  expected to  produce the  highest  levels of
     ground level  impact for the time periods  that have  regulated
     pollution standards.  The data  sets must  be adequate to be
     used for the  modeling of  the annual, 24-hour, and  3-hour
     averaging times of  pollution impacts.

-------
All sites in complex terrain will have unique characteristics
and will deserve careful examination of the surrounding topography
The initial plan for collection of meteorological data should
be discussed with the meteorological staffs of applicable
regulatory bodies before data collection begins, so that any
special modeling input needs can be met.

For an elevated source, an adequate data set would probably
include, but would not be limited to, the following:

     •    At least one instrumented tower with the data feeding
          to a data logger and strip chart backup for wind and
          temperature.  The instruments should include wind
          direction and speed at two or.more levels,  turbulence
          measurements at the top of tower (as near to plume
          height as practical), temperature measurements and
          delta T measurements.

     •    Solar radiation measurements near the base  of tower.

     •    Collection of additional wind data on a continuous
          basis at other locations at or near the suspected
          points of high impact on air quality.

     •    Radar or theodolite-tracked tetroons carrying minisondes
          They will describe three-dimensional trajectories at
          specific sites over ranges of a few to some tens of
          kilometers.  The temperature sensor will provide
          information on the energetics of flow along slopes
          and in mountain valleys.
                           -43-

-------
     •    To apply the integral form of calculating H ,
          vertical profiles of wind U(Z) and temperature T(Z)
          measured upwind of the terrain to heights at least
          as high as the terrain height are desired.  It is
          impractical to construct towers to heights of 200 or
          300 m or higher, so remote sensing devices are
          needed.  Doppler acoustic sounders can provide
          information on wind profiles.  Alternatively,  tethersondes
          or minisondes could be used with a shorter tower to
          measure temperature profiles.

Possibilities for special studies to augment the basic data
described above might include tracer studies over limited time
periods during the season(s) of primary concern.  Such a study
would involve release of a tracer gas from the effluent point
of interest along with samplings at sites in the region of
anticipated impact.  By identifying "worst-case" atmospheric
conditions, tracer releases would be necessary on only a
limited number of days.  Wind tunnel studies allow systematic
investigation of cause and effect and provide a means of
extrapolation from field information (see Section 2B).

While the above suggestions relate rather specifically to a
single category of sources, there appeared to be a consensus
among workshop participants that detailed meteorological data
including direct measures of turbulence intensity should be
required for use in complex terrain dispersion models.  The
need for such data is greater for mountainous settings than
for flat terrain settings because of the importance of site
specific phenomena on turbulent dispersion processes.
                           -44-

-------
     .Another recommendation relating to the use of meteorological
     input data in models evolved from the recognition that more
     advanced models will probably require more detailed inputs.
     Current regulatory practice requires  that one or more years of
     hour-by-hour values of wind speed and direction, stability
     class, and mixing depth be  input sequentially,  so that all
     sets of three-hour and daily averages are computed.  The
     purpose of this procedure is to allow evaluation of the highest
     and "highest,  second-highest" concentration averages.  Concern
     is expressed that the associated volume of computations may be
     excessive unless simplifications are  made in the simulation
     algorithms to reduce total  computation costs.  An alternative
     approach which should be considered would be to be selective
     in identifying worst-case meteorological scenarios to be used
     and thereby encourage the use of more computation time-intensive
     algorithms if they were deemed to be  superior.

(4)   Suggestions on Improving Present Screening Models

(a)   For impingement on high terrain, the  group recognized the
     importance of screening models, as applied by regulatory
     agencies, to eliminate the  need for detailed assessments for
     facilities having minor ambient air impacts.  A screening
     model was defined in this context as  a conservative model
     which can be used without any meteorological data, e.g.,
     worst-case meteorological assumptions are made.  Worst-case
     Valley is the present EPA screening model for terrain higher
     than a source height and assumes a specific wind speed, a
     specific duration of wind direction,  specified  temperature
     profiles and dispersion rates,  and arbitrary "miss" distances
     for plume paths.  The group felt that because of concerns over
     the arbitrariness and elementary nature of the  assumptions
                               -45-

-------
     already in  Valley,  that  it  would  not  provide  a  good  basis  for
     improvement.   Adding  more complexity  would  not  necessarily
     make  this model  any more reliable,  because  further arbitrary
     meteorological assumptions  would  be required.   Use of  the
     critical streamline height  concept, which was raised specifically
     in discussion  (Rowe,  1982),  requires  assumptions  about the
     vertical variations of temperature  and  velocity.  Members  of
     the group thought  it  unwise  to  add  complexity requiring more
     detailed meteorological  data inputs to  a model  which would be
     used  in the absence of such  data.

     It was  felt that better  screening models could  be developed by
     a two-fold  effort:

     1.   Examine existing data  available  from ambient air  quality
          or tracer measurements  for a variety of  sources in terrain
          settings  to see  if  uniformly-conservative  screening
          algorithms  could be developed  on the basis of comparing
          peak observed  and predicted  concentrations.

     2.   Develop screening techniques which were  consistently
          conservative.  That is, develop  methods  which would not
          result in biases in the decision-making  depending upon
          differences of source-topographic  situations (i.e.,
          biased against sources  located close to  terrain vs.
          further from  terrain,  etc.).   The  degree of  overprediction
          expected  from such  models  should be quantified.

(b)   At present, no screening model  is available or  can be  recommended
     for predicting maximum air  pollution  concentrations  due to
     diurnal fumigations or light wind high-stability  dispersion
                                -46-

-------
     conditions in deep valleys.   Further research and*evaluation
     of actual concentration data are necessary before such models
     can be developed and used with confidence.

(5)   Suggestions on the Use of Physical  Modeling Techniques

     The advances made in developing laboratory techniques  and in
     the understanding of the limits of  applicability of physical
     modeling techniques suggest  that increased reliance be placed
     on their use.   Application of these techniques should  be
     considered for those situations where they are appropriate.

(a)   Consideration for Screening  Models.   Physical modeling can
     provide a source of useful information regarding the expected
     maximum concentrations at a  proposed site.  In particular,
     fluid modeling could be used to estimate the maximum short
     term average (e.g., 1-hour or less)  concentrations  under plume
     impaction conditions as a function  of hill Froude number,
     stack height,  and stack location.   The optimum stack location
     and height could then be selected so that concentrations are
     mimimized.   In addition, fluid modeling could be used  to
     estimate the short term concentrations due to terrain  wake or
     cavity effects.   For screening purposes,  it would have to be
     assumed that the conditions  modeled would actually  occur.  If
     on-site or representative off-site  data are available, more
     realistic estimates can be made.  For mountain-valley  or
     drainage-type  flows, the wind tunnel may not be able to simulate
     the worst-case condition which appears to occur during transition.
     Another technique would have to be  employed for these  situations.
                               -47-

-------
(b)   Detailed  Evaluations.   If the  screening evaluation suggests
     that plume  impact  or terrain wake  or cavity effects will  cause
     standards to  be  exceeded,  a more refined modeling analysis
     could be  undertaken.

     Physical  modeling  could be used during this evaluation to help
     refine and  validate a site-specific model (if the length
     scales are  not  too large).   The basic components  of the model,
     such as plume trajectory and dispersion parameters, could be
     evaluated as  a  function  of stability, wind direction and wind
     speed. The model's ground-level concentration estimates  could
     then also be  compared with the wind tunnel  results.  Once the
     validated model  (for hourly concentration estimates)  is fully
     tested, it  can  be  applied using on-site meteorological data  to
     estimate  concentrations for the prescribed  averaging times  (3-
     hour, 24-hour,  annual).  Post-concentration monitoring is
     suggested at  the location predicted to give the highest concen-
     trations.  This  information could  be used to assess the performance
     of the technique.   Physical models can also be used to assist
     in the design of field programs for complex terrain settings.

     Future Research and Development Meeds

     The workshop  attendees elected to  identify  research needs,  and
     establish some  priority associated with them, for the problems
     of predicting ground-level concentration of air pollutants  in
     complex topography.  These problems are focused on concerns
     with the  highest concentration values which can result, and  of
     course with understanding the  physics that'contribute to  these
     high values.
                                -48-

-------
     After considerable deliberations a set of problems were established,
     which in the opinion of participants,  bracketed the physical
     situations which can lead to maximal  ground-level  concentrations
     in complex terrain.   Of the  seven problems finally selected,
     four involved stable stratification,  two involved  the unstable
     planetary boundary layer, and three involved detailed flow
     field considerations.   The problems were further recognized to
     be composed of subcomponents of a more fundamental nature.
     They were qualified by, in addition to their subcomponents:
     (a)  whether some observational data had been collected on the
     problems; (b)  whether a physical conceptualization of the
     problem had been developed;  (c)  whether physical or numerical
     modeling efforts had been called for  or already conducted;  and
     (d)  whether a further major  field effort was deemed necessary.
     Table 1 lists the topics and summarizes these considerations.
     The  ordering of tasks  does not necessarily imply priority.   A
     brief synopsis of the issues for each  of the tasks is provided
     below.

(1)   Stable  Plume Impaction.  Physical modeling and field studies
     have established that in stable flows  a critical elevation
     exists  which differentiates  the flow  which goes around a hill
     from the flow which goes up  and over  a hill.  In combination
     with information on plume height and  dimensions, better estimates
     of the  location and magnitude of ground-level concentrations
     can  be  made.  Field studies  should now be directed toward
     establishing the validity of this concept in larger scale and
     more complex (less idealized) terrain  situations.   It is
     likely  that mathematical modeling of  this phenomenon would
     yield additional insights not otherwise obtainable.
                                -49-

-------
(2)   Lee-Side  Effects   Laboratory modeling of both neutral  and
     stable  flows  over 2-D  and 3-D topography has  shown that  elevated
     concentrations  may be  associated with a source located downwind
     of the  terrain.   A sounder,  physically based  understanding of
                     •
     these phenomena is needed.   A combination of  theoretical and
     laboratory modeling is called for in an attempt to provide
     such an understanding.  Some field research will be important,
     especially as the theoretical understanding develops.  Numerical
     models  capable  of predicting separation and reattachment
     points  for neutral conditions and two-dimensional hills  are
     presently feasible arid should be extended and evaluated  with
     data.

(3)   Fumigation.   The  upset of a  stable layer in a valley has been
     identified for  many years as contributing to  high ground level
     concentrations.   Although the physical concepts of this  process
     are reasonably  well understood,  it was suggested that laboratory
     modeling  might  help to refine the understanding.  Equally
     important is  the  conduct of  a field experiment focused on
     these events.  Numerical models  and fluid modeling (e.g., Bell
     and Thompson, 1980) can be used  to understand the diffusion
     processes near  the valley top where strong wind shears occur.
     More generally,  mathematical models need to be developed to
     describe  dispersion processes in the transition region just
     above and below hill crests.

(4)   Trapping  Valley.   There exists a class of valleys which  because
     of their  orientation cause drainage winds to  pool at their
     base forming  a  thick stable  air  mass.  When circumstances are
     such that diurnal heating and momentum transport from above
     are insufficient  to break this inversion, very high concentrations
                                -50-

-------
     may occur.   A combination of laboratory and field experiments
     was recommended to quantify the existing understanding of this
     situation.   Studies of how to develop adequate meteorological
     inputs for  these conditions are also needed.   Further guidance
     on the development of parameterizations of physical  processes
     could be provided by numerical simulation models.  The
     interactions between the evolving circulations which develop
     on different length scales are thought to be  one of  the key
     scientific  problems in dealing with valley meteorology.

     Another important need is data to document the depth of the
     stable boundary layer over the valley.  The relationship
     between the depth of this stratification and  the strength and
     persistence of drainage winds should be determined.

     Future studies should address both diurnal and synoptic stagnation
     conditions  in trapping valleys -- especially  those with populations
     and pollutant sources.   Field and modeling studies should
     address the critical topographic and meteorological  conditions
     for occurrence, as well as the resulting air  quality degradation.

(5)   Convective  Boundary Layer over Complex Terrain.   As  a result of
     intensive theoretical,  numerical and physical modeling studies,
     as well as  definitive field experiments,  we have achieved a
     fairly complete understanding of the evolution of the convective
     planetary boundary layer (PEL)  over homogeneous  terrain.
     Similar systematic studies of the convective  PBL over complex
     topography  are generally lacking.   There  are  also no studies
     of the interaction between the topographically induced motions
     (such as in the separated cavity and wake regions) and the
     buoyancy-driven convective motions and their  combined effects
     on the dispersion of pollutants in complex terrain.
                                -51-

-------
     In  order  to  understand  the  above  problem,  the  following  types
     of  studies should  be  conducted  in a  sequential  manner:

     a.    Construct  simple slab  models for  the  convective  PBL over
          a  uniformly but.  significantly sloping terrain  and over
          2-D  hills  or  ridges.

     b.    Conduct physical modeling  studies  of  boundary  layers over
          isolated 2-D  and 3-D hills of different slopes and  shapes
          in both neutral  and unstable conditions.

     c.    Conduct numerical  modeling (large  eddy simulation)  studies
          of the  convective  PBL  over 2-D  and 3-D isolated  hills.

     The studies  proposed  here should  be  very useful in  improving
     the understanding  of  the spatially inhomogeneous convective
     PBL in  complex  terrain  and  its  parameterization in  mesoscale
     numerical models of flow and  dispersion.   These will  also
     indicate  to  what extent and under what  types of topographical
     situations our  current  knowledge  of  the homogeneous flat-
     terrain convective PBL  can  be transferred  or applied  for
     modeling  purposes.

(6)   Mesoscale and Longer  Range  Transport and Diffusion.   Although
     not contributing to the highest ground-level concentrations,
     the transport of pollutants over  scales upward of 15  km  to as
     far as  200 km is important  because of  Class I  PSD increments.
     The research necessary  here is  twofold. First, a more fundamental
     understanding of the  physics  is probably necessary  before much
     significant  progress  can be made. Second, although a great
     many models  exist,  they are all unverified. It is  recognized
     that chemical transformation  and  removal processes  become
                               -52-

-------
          important aspects of models applied to longer distance transport
          issues.  The problem of longer range transport and diffusion
          in complex terrain requires the combination of a wind field
          model with an appropriate diffusion scheme.  Evaluation of
          such a combined approach is not easily accomplished.  Both the
          wind field and the diffusion components require attention.  As
          was discussed earlier, various procedures are available to
          simulate winds, all of which involve some compromise.  The
          best that can be expected is that ranges of appropriate performance
          can be defined for each class of model.  These can be developed
          by combinations of theoretical and field studies which are
          recommended.  Evaluation of the diffusion component can rely
          on work done near the source but will ultimately require a
          tracer study conducted over complex terrain.  The CAPTEX
          experiment (CAPTEX, 1984) may provide this requirement for
          eastern rolling terrain; however, there remains a need for
          western complex terrain tracer studies.

     (7)  Sea/Complex Terrain Interface.  Analysis of dispersion in
          coastal terrain involves further meteorological complications
          associated with the coupling of land-sea breezes with mountain-
          valley flows, discontinuities of surface temperature, etc.  It
          was felt that although it is common for pollution sources to
          be located in such regions, relatively little research has
          been done in this area.

A number of more general concerns  were expressed by the group during
discussions of the above research  needs.  There was concern that we
ought to better recognize that dispersion of pollutants in complex
topography is a special subset of the geophysical fluid dynamics of
stratified flows.  As such, under  stable stratification, influences of
topography can be felt well above, up and downwind of the topography.

-------
As a result a caution was raised regarding the extrapolation of conceptu-
alizations which are valid and useful for simple (flat)  terrain problems
to problems in complex topography.   Interactions between the physics of
stratified flow and the topographic structure lead to certain natural
time and space scales in the flow.   In cases where such time scales are
larger than the averaging time of application interest,  estimates of
concentration by deterministic models will contain larger uncertainties.
There were repeated references to the difficulty of developing any
deterministic and generalized solution.  Hence it was suggested that the
basic stochastic nature of the phenomena should be accepted, and that
statistical approaches be used to characterize the uncertainty.

Consideration of the more fundamental aspects suggested that the seven
problems in Table 1 depended upon an understanding of the following,
more basic items:

          a.   The space and time scales of the flow as a function of
               topography and other types of forcing or boundary conditions

          b.   The depth of the planetary boundary layer in complex
               terrain.

          c.   The interaction of the planetary boundary layer with the
               free air above it.

          d.   The space and time variation of turbulence as a function
               of stability and topography.

          e.   The generalization of the dividing-streamline concept to
               arbitrary shapes and scales of hills.

          f.   The effects of radiation on the flow field in complex
               terrain.
                                    -54-

-------
          g.   Effects of wakes in the flow over complex topography.

          h.   The identification and measurement of meteorological and
               other data which are representative of the flows.

This list represents a very comprehensive set of basic issues stemming
from the fundamental complexity of stratified fluid dynamics, when
coupled with nongeneral specific topographic environments (real topography)
The associated uncertainty in the ability of simple models to simulate
these processes needs to be a key concept kept very much in the forefront
of any applications in complex topography.
                                    -55-

-------
<£ _
       ^c

oo.      CD

cu!    oo
                                       O -3
                                       —4  CD
                                       CD -3
                                       >  CD
                                       CD  CU
                                       Q Z
                                    o
                                   en
              o
              e
              o"
              o
              o
ft CD
O CD
                                                                                 "3
                                                                                  O
                                                                                 O
                                                                                 U
                                                                                 •J
CD
S
O
                                                                              O
                                                                              o
                                                                              u
                                                                              CD
                                                                              E=
                                                                              O
                                                                                                       o
                                                                                                      -3
                                                                                                       CD
                                                                                                       O
                                                     CD
                                                    •3
                                                     O
-J
c oo
c a
              U  00

              CO  E^
                         -3
                         CD
           i-
           O
           u
           o

           CD
           S
           O
           co
                        T3
                         CD
                                       C.
                                       e
                                       o
                                       CJ
                                       CJ
                                       CD

                                       O
                                       oo
                                                 -3
                                                  CD
                                                 -3
                                                  O
                                                  CD
CO
C
CD

O
             "3
              CD
             •3
              O
              CD
                                              •3
                                              O
                                              0)
                                              f-.
                                              O
                                                                                  DO
                                                                                  c
"O
0
2

X
£_t
o
4-1
OS
fl
o
r— 1
P ,
X
tu









-3
0)
-3
CD
CD
Z

CD
.—4
f^
"OS
CJ
•r-4
r-H
CU
cu
•e^

4-J
O
Z
•3
CD
•3
CD
CD
Z

CO
c
•H
,—)
CD
-^
,0

CO
U
U
oo
tu

Q
3
DATA GATHERING

si co
O E-
Z
cn tu
LJ Z
00 O
tu cu
T* ^
H O
O U
a. CQ
>- 3
r: oo



















STATUS/ NEEDS
mline Data Available or
03

a
tu
E—
D
!_i
4-J
co
oo
tU DO
8-

tu
09

o
£
•H
•a
• H
^
•i-l
E- a







CD
e

rH
MM

CD
C_} r— 4
i-i -Q
— OS
0 4-1
H 00
of Scheduled to be
Gathered
tn
CD X
i— 1 4-1
03 •<-!
CJ —1
00 "H

.• OS
4-J O
r~ .^
00 — (
•H CU
_CD CU











C
O
•H
4J
CJ
03

£

-3
CD
-3
CD
CD
Z
03
4-1
oS
a
"3
r— 4
CD
• r-4
co
c
-3
• H
>
• H
Q

• n
J^
tn
03
2
C
2
O
Q
CD
•3

cn

CD
CD


c
0

to
4-1
CJ
CD
4-1
UH
tu
JZ
co
•H
^CD


CD
C
• r-4
r-H
S
C3
CD
^
*J
co












to
CD
O
f-i
3
O
00
03
•3
r— 1
CD
a,
CD
ft
_O
4-1
a.
^
FH
CD
X
03


X

03
-3

3
O
02






C
• H

f-
O
• H
4J
OS
CO

S

tu
-3
CD
-3
O
CD
CD
CJ
03
f-i
oS
JZ
U

CD
CJ
C
CD
,-1 cn
3 0
Xl 'H
f-l 4-1
3 '-n
H -H




C
•H
03
f-i

CD
H

X
CD
,— 4
eu
(3
O
U
4-1
CD
X
4->
O
C
03
4-1
03
r— t
CD
• rH
U,
X
•3
3
O
ca
14-4
o

c
0
•-H
4->
o
03
f^
CD
4-J
C
1—4
to
CD
cn
tn
03
^*

f-i
•H
<

4->
£
03
£
SO

4_>
00
-3
CD
CD
4-J
CD
i^
4-J C
< 03 m
o
CD CO 4J
CD C 03
ft "t c^
o, x: ^

JZ 0 0

— i CO 4-1
2 03
• * r— 1
f-l 





to
X

1— t
-H
03
>

C
•H
x:
4_)
• r4
2;
CD
X
O
03
4-1
03
Q
-3
r-H
•r-t
X
03
C
3
0
33
U-4
0

c
o

4_>
CJ
03
i-i
CD
4Ğ1
C
h-4
C
• H

m
2
O
r— 4
CU

CD
>
• H
4-J
O
CD
>
£^
O
u
CU
-3
CD
Z
m
CD
0*
T-^
0}

1^*,
4-J
•H
5

*%^
CD
X
03





^
• H
03

Ui
CD
E-

X
CD
r— 4
CU
E
O
CJ
•3
CD
-3
CD
CD
Z
oS
4->
03
Q
CD
|
in
r— 4
CD
•O
O
2

•3
r— 4
CD
.H
U,

2
O

u-
^
f-i
O
CU
tn
C
03

H

CD
^-4
03
•J
in
O

S +-*
0 C
U M
^^^
—i C
OS •!-(
4-J 03
tn f-i
03 ft
O CD
u e-
                                                        -56-

-------
References:

American Meteorological Society,  1981:   Air Quality Modeling and the
     Clean Air Act.  A Report Prepared under Cooperative Agreement with
     the Environmental Protection Agency.   Amer.  Meteor. Soc.,  Boston,  MA,
     288 pp.

Bader, D.C., and T.B. McKee, 1983:   Dynamical Model Simulation  of the
     Morning Boundary Layer Development in Deep Mountain Valleys.
     J. Clim. 5 Appl. Meteor. 22:341-551.

Baines, P.G., and R.H.J.  Grimshaw,  1979: Stratified Flow over Finite
     Obstacles with Weak  Stratification.  Geophys.  Astrophys. Fluid Dyn.
     13:317-334.

Barr, S., R.E. Luna, W.E.  Clements,  and H.W. Church, 1977:  Workshop on
     Research Needs for Atmospheric  Transport and Diffusion in  Complex
     Terrain, NTIS CONF-7609160,  U.S.  Energy Research and Development
     Administration, Washington,  D.C.,  23  pp.

Bell, R.C.,  and Thompson,  R.O.R.Y.,  1980:  Valley Ventilation by Cross Winds,
     J. Fluid Mech., 96:757-767.

Bowne, N.E., R.J. Londergan, D.R. Murray,  and H.S.  Borenstein,  1983:
     Overview, Results, and Conclusions for the EPRI Plume Model Validation
     and Development Project:  Plains  Site.  EPRI EA-3074,  Electric
     Power Research Institute,  Palo  Alto,  CA, 234 pp.

Brighton, P.W.M., 1978:  Strongly Stratified Flow Past Three-Dimensional
     Obstacles.  Quarterly Journal  of  the  Royal Meteorological  Society
     104:289-307.

Britter, R.E., J.C.R. Hunt, and K.J. Richards, 1981:  Analysis  and Wind
     Tunnel  Studies of Speedup, Roughness  Effects and Turbulence over a
     Two-Dimensional Hill.  Quart. J.R. Met. Soc.,  88.

Burt, E.W.,  1977: Valley  Model  User's  Guide.  EPA-450/2-77-018,  U.S.
     Environmental Protection Agency,  Research Triangle Park, NC,  112 pp.

CAPTEX, 1984: Acid Rain:  A Search for  the  Source. NOAA, winter  '84, 9-11.

Cermak, J.E., and R.L. Petersen,  1981:  Physical Modeling of Downslope
     Mountain Wind and Atmospheric Dispersion, Proceedings  of the  Fourth
     U.S. National Conference on Wind  Engineering Research,  Seattle, WA.

Davis, C.G., S.S. Bunker,  and J.P. Mutschlecner,  1984: Atmospheric Transport
     Models  for Complex Terrain,  J.  Climate and Appl. Meteorol., 23: 235-238.
                                       -57-

-------
Defant, F.,  1951:   Local Wind.   In "Compendium of Meteorology" ed.
     T.F.  Malone.   Amer. Meteor.  Soc.,  Boston, NLA.   pp.  655-675.

Dickerson, M.H.  and P.H. Gudiksen, 1980:   The Department of Energy's
     Atmospheric Studies in Complex Terrain (ASCOT)  Program.   Conference
     Papers, Second Joint Conference on Applications of Air Pollution
     Meteorology,  American Meteorological Society,  Boston,  MA 467-473.

Dickerson, M.H.  and P.H. Gudiksen, 1983:   Atmospheric Studies in  Complex
     Terrain; Technical Progress  Report FY-1979 through FY-1983.   UCID-19851,
     Lawrence Livermore National  Laboratory,  Livermore,  CA  367 pp.

Drazin, P.G., 1961:  On the Steady Flow of a  Fluid of Variable Density
     Past on Obstacles.  Tellus.  13:239-251.

Egan, B.A.,  1975:   Turbulent Diffusion  in Complex Terrain.   Workshop
     Proceedings,  Lectures on Air Pollution and Environmental Impact
     Analyses.  .American Meteorological Society, Boston, MA,  244-267.

Egger, J., 1981: Thermally forced Circulations in a Valley, Geophys.
     Astrophys.  Dynamics, 17: 255-279.

EPA 1981a: Source Code for COMPLEX I -  Version 81052.  U.S. Environmental
     Protection Agency, Research  Triangle Park, NC,  37 pp.

EPA 1981b: Source Code for COMPLEX II - Version 81051.  U.S.  Environmental
     Protection Agency, Research  Triangle Park, NC,  37 pp.

Fackrell, J.E. and A.G. Robins, 1982:  Concentration Fluctuations and
     Fluxes in Plumes from Point  Sources in a Turbulent Boundary  Layer.
     J. Fluid Mech., 117:1-26.

Fox, D.G., 1984:  Uncertainty in  Air Quality Modeling.  Bull. .Amer.
     Meteor. Soc., 65:27-36.

Fox, D.G., D.G.  Ross, D.L. Dietrich, and W.E. Marlatt, 1983:  Om Modeling
     Dispersion in Mountainous Terrain  - an Application of TAPAS/CITPUFF/WINDS.
     Proceedings Fourth Int. Clean Air  Congress, May, 1983, Paris.

Greenway, A.R.,  J.E. Cermak, R.L. Petersen and H.C.  McCullough, 1981:
     Physical Modeling Studies for GEP  Stack Height Determinations,  74th
     Annual Meeting of the APCA.   Paper No. 81-20.5.

Hess, J.L.,  and Smith, A.M.O.,  1962:  Calculation of Nonlifting Potential
     Flow about Arbitrary Three-Dimensional Bodies.   Douglas Aircraft Co.,
     Report No.  ES40622.

Hewson, E.W. and G.C. Gill, 1944:  Meteorological Investigations  in
     Columbia River Valley near Trail,  British Columbia - Part II.
     Report submitted to the Trail Smelter Arbitral Tribunal, Bulletin 453,
     U.S. Bureau of Mines, Washington,  D.C.,  228 pp.
                                        -58-

-------
Hovind, E.L., M.W. Edelstein, and V.C. Sutherland, 1979:  Workshop on
     Atmospheric Dispersion Models in Complex Terrain.  EPA-600/9-79-041,
     U.S. Environmental Protection Agency, Research Triangle Park, NC,
     215 pp.

Huber, A.H., 1981: Guideline for Use of Fluid Modeling to Determine Good
     Engineering Practice Stack Height.  EPA-450/4-81-003, U.S.  EPA/OAQPS,
     Research Triangle Park, NC, 47 p.

Hunt, J.C.R. and P.J. Mulhearn, 1973:  Turbulent Dispersion from Sources
     Near Two-Dimensional Obstacles.  Journal of Fluid Mechanics 61:245-274.

Hunt, J.C.R., J.S. Puttock, and W.H. Snyder, 1979:  Turbulent Diffusion
     From a Point Source in Stratified and Neutral Flows Around  a Three-
     Dimensional Hill; Part I - Diffusion Equation Analysis. Atmospheric
     Environment 13:1227-1239.

Hunt, J.C.R., W.H. Snyder,  and R.E. Lawson, Jr., 1978:  Flow Structure
     and Turbulent Diffusion Around a Three-Dimensional Hill; Fluid
     Modeling Study on Effects of Stratification; Part I - Flow  Structure.
     EPA-600/4-78-041, U.S. Environmental Protection Agency, Research
     Triangle Park, NC.

Hunt, J.C.R., K.J. Richards and P.W.M. Brighton, 1984: Stratified Shear
     Flow over Low Hills: II. Stratification Effects in the Outer Flow
     Region.  To be submitted to Quart. J. Roy.  Meteorol. Soc.

Isaacs, R.G., A. Bass, and  B.A. Egan, 1979:  Application of Potential
     Flow Theory to a Gaussian Point Source Diffusion Model in Complex
     Terrain.  Extended Abstracts, Fourth Symposium on Turbulence,
     Diffusion, and Air Pollution, .American Meteorological Society,
     Boston, MA 189-196.

Jackson P.S. and J.C.R. Hunt, 1975:  Turbulent Wind Flow over a  Low Hill.
     Quart.  J.  R.  Met. Soc., 101:929-955.

King, D.S.  and S.S. Bunker, 1984:  Application of Atmospheric Transport
     Models  for Complex Terrain.  J. Climate and Appl. Meteor.,  23:239-246.

Lavery, T.F., B.R. Greene,  B.A. Egan, and F.A.  Schiermeier, 1983:  The
     EPA Complex Terrain Model  Development Program.  Extended Abstracts,
     Sixth Symposium on Turbulence and Diffusion, American Meteorological
     Society, Boston, MA  126-130.
                                        -59-

-------
McNider, R.T.  and R.A.  Pielke,  1981:  Numerical Simulation of Slope and
     Mountain Flows.   Second Conference on Mountain Meteorology,  AMS
     Steamboat Springs, CO.

McNider, R.T., 1981:  Impact of Topographic Circulations on the Transport
     and Diffusion of Air Pollutants,  Ph.D.  Dissertation, University of
     Virginia, (NTIS-PB82-102435).

McRae, G.J., F.H. Shair, and J.H.  Seinfeld,  1982:  Convective Downmixing
     of Plumes in a Coastal Environment.  J.  Appl.  Met.,  20:1312-1324.

Marlatt, W.E., J.D. Beach and D.G.  Fox, 1981: The  Shape of Air Temperature
     Inversion Surfaces of Mountain Valleys.   Second Conf. on Mountain
     Meteor.,  Nov. 9-12, 1981,  Steamboat Springs,  CO.   Amer. Meteor.
     Society,  Boston, MA, p. 173-178.

Mason, P.J., and R.I. Sykes, 1979a:  Three-Dimensional Numerical  Integrations
     of the Navier-Stokes Equations for Flow over  a Surface-Mounted
     Obstacle, J. Fluid Mech.,  91:433-450.

Mason, P.J., and R.I. Sykes, 1979b:  Flow over an  Isolated Hill of
     Moderate Slope.   Quart. J.R.  Met.  Soc.,  105:383-395.

Meroney, R.N., 1980:  Wind-Tunnel Simulation of the Flow Over Hills and
     Complex Terrain, J. of Industrial  Aerodynamics, 5:297-321.

Meroney, R.N., H.C. Chien, and V.A. Sandborn, 1980: Sites for Wind-Power
     Installations: Physical Modeling of the Wind  Field over Kahuku Point,
     Oahu, Hawaii, Proceedings of the Third International Symposium on
     Wind Energy Systems, August 26-29, 1980, Copenhagen, Denmark, pp. 75-90.

Murphy, D.J., R.M. Buchan and D.G.  Fox, 1984: Ambient Total Suspended
     Particulate Matter and Benzo(a)  Pyrene Concentrations from Residential
     Wood Combustion in a Mountain Resort Community.  Amer. Indust.
     Hygiene Assoc. Journal, 45(7):431-435.

Orgill, M., 1981:  A Planning Guide for Future Studies 1981; Report No.
     PNL-36-56, ASCOT/80/4.

Petersen, R.L. and J.E. Cermak, 1979: Evaluation of "Good Engineering
     Practice" Stack Height at the ASARCO Smelter, Hayden, Arizona --A
     Physical Model Study, Technical Report prepared for North American
     Weather Consultants by Colorado State University, CER79-80RLP-JEC2.

Petersen, R.L., J.E.  Cermak and R.A.  Nelson, 1980: Simulation of Pollutant
     Dispersion in the Mountain Wind, Second Joint Conference on Applications
     of Air Pollution Meteorology,  pp.  540-545.
                                       -60-

-------
Petersen, R.L., 1981: Determination of GEP Stack Height for Bunker Hill
     Company Smelter and Zinc Plant Stacks, Technical Report prepared
     for Brown, Peacock, Keane and Boyd, NHC Wind Engineering Report.

Petersen, R.L. and E.E. Twombly, 1982: Site-Specific Concentration
     Estimates for Stable Plume Impact on Elevated Terrain.  Presented
     at the 75th Annual Meeting of the Air Pollution Control Association,
     June 20-25, 1982.

Pielke, R.A., R.T. McNider, M. Segal,  and Y. Mahrer, 1983:  The Use of a
     Mesoscale Numerical Model for Evaluations of Pollutant Transport
     and Diffusion in Coastal Regions  and over Irregular Terrain.
     Bull. AMS, 64:243-249.

Randerson, D. 1984: Personal Communication.

Rao, K.S. and H.F. Snodgrass, 1981:  The Nocturnal Boundary Layer over a
     Sloping Surface.  Fifth Symposium on Turbulence, Diffusion and Air
     Pollution.  AMS, Boston, MA.

Reynolds, S.D., T.C.  Myers, J.E. Longstaff, M.K.  Liu, G.E.  Moore and
     R.E. Morris, 1984: Operational Validation of Gaussian  and First-
     Order Closure Plume Models at a Moderately Complex Terrain Site.
     EPRI EA - Electric Power Research Institute, Palo Alto, CA  210 pp.

Riley, J.J., H.T. Liu, E.W. Geller, 1976:  A Numerical and  Experimental
     Study of Stably Stratified Flow Around Complex Terrain.  EPA-600/4-76-021,
     U.S. Environmental Protection Agency, Research Triangle Park, NC
     41 pp.

Rowe, R.D., 1982: Stack Design for Stable Air Flow in Complex Terrain,
     265-278, Symposium/Workshop Proceedings: Acid Forming  Emissions in
     Alberta and their Ecological  Effects, Alberta Environment, Edmonton,
     Canada.

Rowe, R.D., S.F. Benjamin,  K.P. Chung, J.J. Havlena, and C.Z.  Lee, 1982:
     Field Studies of Stable Air Flow  over and Around a Ridge, Atmos. Envir.,
     16:643-653.

Ryan, W., and B. Lamb, 1984:  Determination of Dividing Streamline Heights
     and Froude Numbers for Predicting Plume Transport in Complex Terrain.
     J. Air Pol. Control Assoc., 34:152-155.

Ryan, W., B. Lamb, and E.  Robinson, 1984:  An Atmospheric Tracer Investigation
     of Transport and Diffusion Around a Large Isolated Hill.   Atmospheric
     Environment, 18:2003-2021.

Sawford, B.L., 1985:   The Effect of Gaussian Particle-Pair  Distribution
     Functions in the Statistical  Theory of Concentration Fluctuations
     In Homogeneous Turbulence.  Quart.  J. Roy Met.  Soc., 109:359-354.

Sheppard, P.A., 1956:  Airflow Over Mountains.  Quarterly Journal  of the
     Royal Meteorological  Society,   82:528-529.
                                    -61-

-------
Sherman, C.A., 1978:  A Mass Consistent Model  for Wind Fields over Complex
     Terrain.  J.  Appl.  Meteor.,  17:312-318.

Smith, F.G., and A.J.  Garrett,  1983:  Two-Dimensional  Simulation of
     Drainage Flow and Tracer Gas in  the Geysers Area.   Sixth Symposium
     on Turbulence and Diffusion, Amer.  Meteor.  Soc., Boston.

Smith, R.B., 1980:  Linear Theory of  Stratified  Hydrostatic Flow Past an
     Isolated Mountain.   Tellus,   32:348-364.

Snyder, W.H., 1981:  Guideline for Fluid Modeling of  Atmospheric Diffusion.
     EPA-600/8-81-009, U.S. Environmental Protection  Agency, Research
     Triangle Park, NC  185 pp.

Snyder, W.H., 1983:  Fluid Modeling of Terrain Aerodynamics and Plume
     Dispersion - A Perspective View.  Extended  Abstracts,  Sixth Symposium
     on Turbulence and Diffusion, American Meteorological Society,
     Boston, MA, 317-320.

Snyder, W.H., and R.E. Lawson,  Jr., 1981: "Laboratory Simulation of
     Stable Plume Dispersion over Cinder Cone Butte:   Comparison with
     Field Data" Appendix to U.S. EPA Doc. No. EPA-600/3-82-036, Research
     Triangle Park, NC.

Snyder, W.H., R.S. Thompson, R.E. Eskridge, R.E. Lawson, Jr., I.P. Castro,
     J.T. Lee, J.C.R.  Hunt, and Y. Ogawa, 1982:  The Structure of Strongly
     Stratified Flow over Hills,  Dividing Streamline  Concept.  Appendix
     A to EPA-600/3-83-015, U.S.  Environmental Protection Agency, Research
     Triangle Park, NC  320-375.

Snyder, W.H. and J.C.R.  Hunt, 1984: Turbulent Diffusion from a Point
     Source in Stratified and Neutral Flows around a Three-Dimensional
     Hill; Part II: Laboratory Measurements of Surface Concentrations,
     Atmos. Envir. 18:1969-2002.

Snyder, W.H., R.S. Thompson, R.E. Eskridge, R.E. Lawson, Jr., I.P. Castro,
     J.T. Lee, J.C.R.  Hunt, and Y. Ogawa, 1985:  The Structure of Strongly
     Stratified Flow over Hills:  Dividing-Streamline Concept, J. Fluid Mech.,
     52:249-288.

Start, G.E., C.R. Dickson, and L.L. Wendell,  1975: Diffusion in a Canyon
     within Rough Mountainous Terrain.  Journal  of Applied Meteorology,
     14:333-346.

Strimaitis, D.G., A.  Venkatram, B. Greene, S. Hanna,  S. Heisler, T. Lavery,
     A. Bass, and B.  Egan, 1983:  EPA  Complex Terrain Model Development.
     Second Milestone Report, 1982.  EPA Doc. No. EPA-600/3-83-015.
     Research Triangle Park, NC  April 1983.
                                      -62-

-------
Turner, D.B., and J.S. Irwin, 1982: Extreme Value Statistics Related to
     Performance of a Standard Air Quality Simulation Model using Data
     at Seven Power Plant Sites.  Atmos. Envir. 16:1909-1918.

Tyson, P.O., 1968: Velocity Fluctuations in the Mountain Wind.   J.  Atmos. Sci,
     25:381-384.                                           .

Weil, J.C., J.E. Cermak and R.L. Petersen, 1981:  Plume Dispersion about
     the Windward Side of a Hill at Short Range:  Wind Tunnel Versus
     Field Measurements,  Proceedings of the AMS Fifth Symposium on
     Turbulence, Diffusion and Air Pollution, AMS,  Boston,  MA.

Weil, J.C., S.C. Traugott, and O.K. Wong, 1981: Stack Plume Interaction
     and Flow Characteristics for a Notched Ridge.   Report  No.  PPRP-61.
     Martin Marietta Corp., Baltimore, MD.

Whiteman, C.D., 1982: Breakup of Temperature Inversions in  Deep Mountain
     Valleys: Part I. Observations.  J. of Appl.  Meteorol., 21:270-289.

Whiteman, C.D.  and K.J. Allwine, 1983:  VALMET -  A Valley Air Pollution
     Model.  PNL-4728, Pacific Northwest Laboratory,  Richland,  WA,  163 pp.

Whiteman, C.D., A.H.  Huber, R.W. Fisher, and B.D.  Zak, 1984:  Atmospheric
     Tracer Experiments in a Deep Narrow Valley.   Fourth Joint  Conference
     on Applications  of Air Pollution Meteorology,  American Meteorological
     Society, Boston, MA.

Whiteman, C.D.  and T.B. McKee, 1979:  Air Pollution Implications of
     Inversion Descent in Mountain Valleys.  Atmospheric Environment
     12:2151-2158.

Whiteman, C.D.  and T.B. McKee, 1982:  Breakup of  Temperature Inversions
     in Deep Mountain Valleys:  Part II - Thermodynamic Model,  Journal
     of Applied Meteorology, 21:290-302.

Williams, M.D., and R. Cudney, 1976: Predictions  and  Measurements of
     Power Plant Plume Visibility Reductions and  Terrain Interactions.
     Third Symposium on Atmospheric Turbulence, Diffusion and Air Quality.
     Oct. 19-22, 1976.  Amer. Meteor. Society,  Boston, MA,  p. 415-420.

Wooldridge, G.L. and  R.W. Furman, 1984:  The Use  of Froude  Numbers  to
     Represent Flow Patterns Around an Isolated Mountain.  Eighteenth
     International Conference on Alpine Meteorology,  Opatija, Yugoslavia,
     5 pp.

Yamada, T., 1981:   A  Numerical Simulation of Nocturnal Drainage Flow.
     J. Meteor. Soc.  Japan, 59:108-122.
                                     -63-

-------
                                 APPENDIX  1
                   WORKSHOP ON DISPERSION IN COMPLEX TERRAIN
                              Keystone,  Colorado

                                May 17-20,  1983

                               List of Attendees
Dr. S.P.S. Arya
Professor of Meteorology
Department of Marine, Earth $
     Atmospheric Science
North Carolina State University
Raleigh, NC  27650
(919) 737-2210

Dr. Sumner Barr
Group H-8
Los Alamos Scientific Laboratory
Box 1663
Los Alamos, NM  87544
(505) 667-2636

Dr. William Blumen
Department of Astrogeophysics
Box 391
University of Colorado
Boulder, CO  80309
(305) 492-8770

Mr. Norm Bowne
TRC Environmental Consultants, Inc.
800 Connecticut Blvd.
East Hartford, CT  06108
(203) 289-8631

Mr. Loren W. Crow
Consulting Meteorologist
2422 South Downing Street
Denver, CO  80210
(305) 722-8665
Dr. Bruce A. Egan
Environmental Research
696 Virginia Road
Concord, MA  01742
(617) 369-8910
Mr. Richard W. Fisher
Region VIII, Air Programs
Environmental Protection Agency
1860 Lincoln Street
Denver, CO  80295
(303) 837-3763
Technology, Inc.
Dr. Douglas G. Fox
USDA Forest Service
Rocky Mountain Forest fğ Range
  Experiment Station
240 West Prospect Street
Fort Collins, CO  80526
(303) 221-4390
                Dr. Steven R. Hanna
                Environmental Research
                696 Virginia Road
                Concord, MA 01742
                (617) 369-8910

                Mr. Don Henderson
                U.S. Park Service  --  Air
                P.O. Box 25287
                Denver, CO  80225
                (303) 234-6620
                         Technology, Inc.
                                        §  Technology,  Inc.
Mr . Thomas F . La very
Environmental Research
696 Virginia Road
Concord, MA 01742
(617) 369-8910
Dr. Robert N. Meroney
Engineering Research Center
Colorado State University
Ft. Collins, CO  80523
(303) 491-8572

Dr. Richard T. McNider
Air Program/ Planning § Development  Section
Department of Environmental Management
State Capitol
Montgomery, AL   36130
(205) 834-6-570

Mr. William Ohmstede
U.S. Army Atmospheric Laboratory
White Sands Missile Range, NM   88002

Dr. Ronald Peterson
NHC Wind Engineering
22477 72nd Ave.  S
Kent, WA  98032
(206) 872-0218
                                          64

-------
Dr.  R.  Pielke
Department of Atmospheric Science
Colorado State University
Fort Collins, CO  80523
(303) 491-8293

Dr.  Darryl Randerson
NOAA Weather Service
  Nuclear Support Office
P.O. Box 14985
Las Vegas, NV  89114
(702) 295-1260

Dr.  K.  Shankar Rao
NOAA Atmospheric Turbulence $ Diffusion
  Laboratory
P.O. Box E
Oak Ridge, TN  37830
(615) 576-1238

Dr.  Richard D. Rowe
Department of Mechanical Engineering
University of Calgary
2500 University Drive, NW
Calgary, Alberta, CANADA T2N 1N4
(403) 284-7185

Mr.  Francis A. Schiermeier
Chief,  Terrain Effects Branch
Meteorology & Assessment Division
Environmental Protection Agency
Research Triangle Park, NC  27711
(919) 541-4551

Mr.  Herschel H. Slater
1310 Willow Drive
Chapel Hill, NC  27514
(919) 929-5889

Mr.  Maynard E. Smith
Meteorological Evaluation Services, Inc.
134 Broadway
Amityville, NY  11701
(516) 691-3395
Dr. Ronald Smith
Department of Geology
Yale University
210 Whitney Avenue
New Haven, CT  06520
(203)  436-1077
                        Geophysics
Dr. William H. Snyder
Chief, Fluid Modeling Section
Meteorology § Assessment Division
Environmental Protection Agency
Research Triangle Park, NC   27711
(919) 541-2811

Mr. Gene Start
NOAA Environmental Research  Laboratories
550 Second Street
Idaho Falls, ID  83401
(208) 526-2329

Dr. Ian Sykes
ARAP, Inc.
P.O. Box 2229
Princeton, NJ  08540
(609) 452-2950

Mr. Joseph A. Tikvart
Chief, Source Receptor Analysis  Branch
(MD-14)
Monitoring § Data Analysis Division  (OAQPS)
Environmental Protection Agency
Research Triangle Park, NC   27711
(919) 541-5561

Dr. Jeffrey C. Weil
flartin Marietta Laboratories
1450 South Rolling Road
Baltimore, MD  21227
(301) 247-0700

Dr. Fred D. White
3631 North Harrison Street
Arlington, VA  22207
(202) 334-3515

Dr. David Whiteman
Pacific Northwest Laboratories
Richland,  WA  99352
(509) 375-3875

Dr. Gene Wooldridge
Department of Soil Science S,
  Biometeorology
Utah State University (UMC 48)
Logan, UT  84321
(801) 750-2178, -2179

Dr. John Wyngaard
National Center for Atmospheric  Research
P.O. Box 3000
Boulder, CO  80307
(303) 497-0632
                                         65

-------
                                  APPENDIX  2


                   WORKSHOP ON DISPERSION IN  COMPLEX TERRAIN


                              Keystone,  Colorado

                                May 17-20,  1983


                             Preliminary Schedule


Tuesday, May 17, 1985        "Review of Phenomena"

     8:00           Introduction,  Objectives  of Workshop    B. Egan

     8:30-10:00     EPA-CTMD Study Experiences    T. Lavery (1 hour Presentation,
                                                             h hour Discussion)

    10:30-12:00     DOE-ASCOT Program Field       S. Barr   (1 hour Presentation,
                         Experience                          h hour Discussion)

     1:30-3:00      EPRI Bull Run Study           N. Bowne  (1 hour Presentation,
                         Field Experience                    ^ hour Discussion)

     3:15-5:30      Group Discussion on           M.E. Smith,  Moderator
                         Phenomena of Importance

                    Participants will be encouraged to present short Contributions
                         on this Topic to be followed by more general Discussion
Wednesday, May 18, 1983    "Review of Modeling"   D. Randerson, Moderator

     8:30-10:00     Flow Field Modeling           R. Pielke  (1 hour Presentation,
                                                             4 hour Discussion)

    10:30-12:00     Physical Modeling             W. Snyder  (1 hour Presentation,
                                                             h hour Discussion)

     1:30-3:00      Regulatory Modeling           S.R. Hanna  (1 hour Presentation,
                                                              % hour Discussion)

     3:15-5:30      Group Discussion on Modeling  D. Randerson,.Moderator
                         Approaches

                    Participants will be encouraged to present short Contributions
                         on this Topic to be followed by more general Discussion

     5:30           Subgroup Assignments
                                        66

-------
Thursday,  May 1983     "Develop and Document Findings"

     8:30-12:00     Meet as two working Subgroups to define Areas of Agreement
                         and Disagreement

                    Group 1:  Suggestions for Model  Applications, B. Egan,  Moderator

                    Group 2:  Suggestions for Future Information Needs,
                                                                  D. Fox,  Moderator

     1:30-2:30      Reconvene entire Workshop to review Progress

     3:00-5:30      Return to separate working Groups, Report Writing


Friday, May 20, 1983          "Wrap up Discussion"

     8:30-10:00     Continue with Report Writing

    10:30-12:00     Discussion of Workshop Findings  in Draft Form

    12:00           Departure
                                        67

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
                              2.
                                                            3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
  DISPERSION IN COMPLEX TERRAIN
  A  Report of a Workshop Held  at
  Keystone,  Colorado, May 17-20,  1983
                                                            5. REPORT DATE
                6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

 Bruce  Egan
                                                            8. PERFORMING ORGANIZATION REPORT NC.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 American  Meteorological Society
 45  Beacon Street
 Boston, Massachusetts 02108
                10. PROGRAM ELEMENT NO.
                  CDWA1A/02/0279 (FY-85)
                11. CONTRACT/GRANT NO.
                                                              CR 810297
12. SPONSORING AGENCY NAME AND ADDRESS
 Atmospheric  Sciences Research  Laboratory
 Office  of Research and Development
 U.S.  Environmental Protection  Agency
 Research  Triangle Park, NC  27711
                13. TYPE OF REPORT AND PERIOD COVERED
- RTF, NC
FY 83-85
                14. SPONSORING AGENCY CODE
                  EPA/600/09
15. SUPPLEMENTARY NOTES
15. ABSTRAC1
     A workshop was convened  by  the American Meteorological  Society under  a  cooperative
agreement  with the U.S. Environmental  Protection Agency.   The purpose of the workshop
was to encourage atmospheric  scientists working in  this  area to exchange recently
acquired  information on atmospheric processes in complex terrain and to make recom-
mendations regarding both the present application of  air quality models to complex
terrain settings and the research  necessary to meet future needs.

     This  report contains the thoughts and judgments  of  32 atmospheric scientists who
gathered  to exchange such technical information and research results on atmospheric
processes  in complex terrain  and to comment on matters relating to adjustments in
current air quality modeling  practices.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                               b.lDENTIFIERS/OPEN ENDED TERMS
                              c.  COSATI Field/Group
18. DISTRIBUTION STATEMENT

     RELEASE TO PUBLIC
   19. SECURITY CLASS (Tins Report)
   UNCLASSIFIED
                                                                          21. NO. OF PAGES
                                               20. SECURITY CLASS (This page)
                                               UNCLASSIFIED
                                                                          22. PRICE
EPA Form 2220-1 (RĞv. 4-77)    PREVIOUS EDITION is OBSOLETE

-------