EPA-600/3-78-021
February 1978
Ecological Research Series
                   THE BIOENVIRONMENTAL IMPACT
                   OF A  COAL-FIRED POWER PLANT
                                Third  Interim  Report
                                    Colstrip,  Montana

                                      Environmental Research Laboratory
                                     Office of Research and Development
                                    U.S. Environmental Protection Agency
                                           Corvallis, Oregon  97330

-------
                RESEARCH REPORTING SERIES

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

      1.   Environmental  Health Effects Research
      2.   Environmental  Protection Technology
      3.   Ecological Research
      4.   Environmental  Monitoring
      5.   Socioeconomic Environmental Studies
      6.   Scientific and Technical  Assessment Reports (STAR)
      7.   Interagency  Energy-Environment Research and Development
      8.   "Special" Reports
      9.   Miscellaneous Reports

This report has been assigned to the  ECOLOGICAL RESEARCH series. This series
describes research on  the effects of pollution on  humans, plant and animal spe-
cies, and  materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                         EPA-600/3-78-021
                                         February  1978
        THE BIOENVIRONMENTAL IMPACT

        OF A COAL-FIRED POWER PLANT

  Third Interim Report, Colstrip, Montana

               December 1977
                 Edited by

    Eric M. Preston and Robert A.  Lewis
   Ecological Effects Research Division
Corvallis Environmental Research Laboratory
            Corvallis, OR 97330
CORVALLIS ENVIRONMENTAL RESEARCH LABORATORY
    OFFICE OF RESEARCH AND DEVELOPMENT
   U.S.  ENVIRONMENTAL PROTECTION AGENCY
           CORVALLIS, OR  97330

-------
                                  DISCLAIMER

This report has been reviewed by the Corvallis Environmental Research Labora-
tory,  U.S.  Environmental  Protection Agency,  and approved  for publication.
Mention of trade names or commercial products does not constitute endorsement
or recommendation for  use.

-------
                                   FOREWORD

     Effective  regulatory  and  enforcement  actions  by  the  Environmental
Protection Agency would be virtually impossible without sound scientific data
on pollutants  and  their impact on environmental stability  and  human  health.
Responsibility for building this data base has been assigned  to EPA's Office of
Research and Development and its 15 major field installations, one of which is
the Corvallis Environmental  Research Laboratory (CERL).

     The primary mission of  the  Corvallis Laboratory is  research on the effects
of environmental pollutants on terrestrial,  freshwater, and marine ecosystems;
the  behavior,  effects  and control  of pollutants  in  lake  systems;  and  the
development  of predictive models  on  the movement of pollutants  in  the bio-
sphere.

     The  Colstrip,  Coal-fired  Power  Plant  Project  is  the  first attempt  to
generate  methods  to  predict  the  bioenvironmental  effects  of air  pollution
before damage  is  sustained.   The methods are to be  framed  within an overall
facility  siting strategy  for  coal-fired power plants in the northern plains.
This will permit planners to assess the ecological  impact of energy conversion
activities on grasslands prior to site selection.  If environmental impacts can
be  predicted,  mitigation  efforts can  be  designed  to minimize  the  adverse
effects of power plant construction and operation.
                                                    A.  F. Bartsch
                                                    Director, CERL

-------
                                 ABSTRACT
     The Environmental  Protection  Agency  (EPA)  has recognized  the  need for a
rational approach  to  the incorporation of ecological  impact information into
power  facility  siting  decisions  in  the northern  great  plains.  The Colstrip
Coal-fired Power Plant  Project  (CFPPP) is  one aspect of the agency's response
to  this  need.   Research  funded  by the CFPPP is a first  attempt to generate
methods  to  predict  the  bioenvironmental   effects  of  air pollution  before
damage  is  sustained.  The  methods  are to  be framed within an overall facility
siting  strategy for  coal-fired power plants  in  the northern plains.  This
document describes  the progress of the investigation.

     Pre-construction  documentation  of the  environmental  characteristics  of
the  grassland   ecosystem  in the  vicinity  of Colstrip,  Montana  began  in  the
summer  of  1974.  This documentation continued  until  Colstrip  generating unit
1  began operation  in  September,  1975.   Since  then, key  characteristics  of
the  ecosystem   have  been  monitored  regularly  to  detect  possible  pollution
impacts upon plant  and  animal  community structure; primary production, inver-
tebrate  animal  consumers,  and  decomposers;   plant  and animal  diseases;  both
beneficial  and  harmful  insects;  indicators  and predictors of pollution (e.g.,
lichens  and honeybees);  physiological  responses  of  plants   and  vertebrate
animals; insect behavior (mainly of honeybees) and production; the behavior,
reproduction  and  development,  population  biology,   health  and  condition  of
vertebrate animals.

     In the  summer  of 1975,  field stressing experiments were begun to provide
the data necessary  to develop  dose-response models for S02 stress on a grass-
land ecosystem.  These  experiments involve  continuous stressing  of one acre
grassland plots with  measured  doses  of S02  during the growing  season (usually
April   through October).   Results  of  the 1975 field season investigations  are
summarized  in this  publication.   The  six-year project will  terminate  in 1980
and a final report will be published after data analyses are complete.-

-------
                                   CONTENTS

                                                                      Page

Foreword	  iii

Abstract	iv

List of Contributors	  vii

Acknowledgements  	  ix

                                 INTRODUCTION

          Assessment of Ecological Impact from the Operation of a
          Coal-Fired Power Plant in the Northcentral Great Plains
               R. A. Lewis, E. M. Preston, N. R.  Glass	2

Sections                       COLSTRIP STUDIES

    1     Effects of S02 and other Coal-Firing Plant Emissions on Pro-
          ducer, Invertebrate Consumer, and Decomposer Structure and
          Function in the Vicinity of Col strip, Montana
               W. K. Lauenroth, J. L.  Dodd, R. K. Heitschmidt,
               J. W. Leetham	13

    2     Plant Community Studies in the Vicinity of Colstrip, Montana
               J. E. Taylor, W. C. Leininger	41

    3     Soil and Epiphytic Lichen Communities of the Colstrip,
          Montana Area
               S. Eversman	50

    4     Investigation of the Impact of Coal-Fired Power Plant Emis-
          sions upon the Disease/Health/Growth Characteristics of
          Ponderosa Pine-Skunkbush Ecosystems and Grassland Ecosystems
          in Southeastern Montana
               C. C. Gordon, P. C. Tourangeau, P. M. Rice	65

    5     Investigation of the Impact of Coal-Fired Power Plant Emis-
          sions upon Insects:   Entomological Studies in the Vicinity
          of Colstrip, Montana
               J. J. Bromenshenk	  140

    6     The Effects of Coal-Fired Power Plant Emissions on Vertebrate
          Animals in Southeastern Montana (A report of progress)
               R. A. Lewis, M. L .  Morton, M. D.  Kern, J.  D.  Chilgren,
               E. M. Preston	  213

    7     Remote Sensing of the Bioenvironmental  Effects of Stack
          Emissions in the Colstrip Vicinity
               J. E. Taylor, W.  C.  Leininger	280

-------
Sections                                                               Page

    8     Integrated Aerosol Characterization Monitoring, Colstrip,
          Montana
               N. L. Abshire, V. E. Derr, G. T. McNice, R. Pueschel,
               C. Van Valin	291

                       FIELD AND LABORATORY EXPERIMENTS

    9     Zonal Air Pollution System:   Design and Performance
               J. J. Lee, R. A.  Lewis	322

   10     First-year Effects of Controlled Sulfur Dioxide Fumigation on
          a Mixed-Grass Prairie Ecosystem
               J. L .  Dodd, W.   K.   Lauenroth, R. K. Heitschmidt,
               J. W. Leetham	345

   11     Monitoring Plant Community Changes Due to S02 Exposure
               J. E. Taylor, W.  C.  Leininger	376

   12     Effects of Low-Level   S02 Stress on Two Lichen Species
               S. Eversman	385

   13     Effects of Low-Level   S02 Exposure on Sulfur Accumulation  and
          Various Plant Life Responses of Some Major Grassland Species
          Under Field Conditions
               C. C. Gordon, P.  M.  Rice, P.  C.  Tourangeau	399

   14     Investigation of the  Impact of Coal-Fired Power Plant Emis-
          sions upon Insects:   Entomological Studies at the Zonal Air
          Pollution System
               J. J. Bromenshenk	473

   15     The Relative Sensitivity of Selected Plant Species to Several
          Pollutants Singly and in Combination
               D. T. Tingey, L.  Bard,  R. W.  Field	508

   16     Progress in Modeling  the Effects of S02 Fumigation on an
          Eastern Montana Grassland
               J.  L. Dodd,  M.  Coughenour, W. K. Lauenroth	514

-------
                             LIST OF CONTRIBUTORS
N.  L.  Abshire
U.S.  Dept. of Commerce
N.O.A.A.
Environmental Research Laboratories
Boulder, Colorado 80302

L.  Bard
U.S.  EPA/CERL
200 S.W. 35th St.
Corvallis, Oregon 97330

J.  J.  Bromenshenk
Environmental Studies Laboratory
University of Montana
Missoula, Montana 59801

J.  D.  Chilgren
U.S.  EPA/CERL
200 S.W. 35th St.
Con/all is, Oregon 97330

M.  Coughenour
Natural Resources Ecology  Laboratory
Colorado State University
Fort Collins, Colorado 80521

V.  E.  Derr
U.S.  Dept. of Commerce
N.O.A.A.
Environmental Research Laboratories
Boulder, Colorado 80302

J.  L.  Dodd
Natural Resources Ecology  Laboratory
Colorado State University
Fort Collins, Colorado 80521

S.  Eversman
Dept.  of Biology
Montana State University
Bozeman, Montana 59715

R.  W.  Field
U.S.  EPA/CERL
200 S.W. 35th St.
Corvallis, Oregon 97330
N.  R.  Glass
U.S.  EPA/CERL
200 S.W.  35th St.
Corvallis, Oregon 97330

C.  C.  Gordon
Environmental Studies Laboratory
University of Montana
Missoula, Montana 59801

R.  K.  Heitschmidt
Natural Resources Ecology Laboratory
Colorado State University
Fort Collins, Colorado 80521

M.  D.  Kern
Dept.  of Biology
The College of Wooster
Wooster,  Ohio 44691

W.  K.  Lauenroth
Natural Resources Ecology Laboratory
Colorado State University
Fort Collins, Colorado 80521

J.  J.  Lee
U.S.  EPA/CERL
200 S.W.  25th St.
Corvallis, Oregon 97330

J.  W.  Leetham
Natural Resources Ecology Laboratory
Colorado State University
Fort Collins, Colorado 80521

W.  C.  Leininger
Dept.  of Animal and Range Sciences
Montana State University
Bozeman,  Montana 59715

R.  A.  Lewis
U.S.  EPA/CERL
200 S.W.  35th St.
Corvallis, Oregon 97330
                                      VII

-------
G.  T.  McNice
U.S.  Dept. of Commerce
N.O.A.A.
Environmental Research Laboratories
Boulder, Colorado 80302

M.  L.  Morton
Dept.  of Biology
Occidental State College
1600 Campus Rd.
Los Angeles, California 90041

E.  M.  Preston
U.S.  EPA/CERL
200 S.W. 35th St.
Corvallis, Oregon 97330

R.  Pueschel
U.S.  Dept of Commerce
N.O.A.A.
Environmental Research Laboratories
Boulder, Colorado 80302

P.  M.  Rice
Environmental Studies Laboratory
University of Montana
Missoula, Montana 59801
J. E. Taylor
Dept. of Animal and Range Sciences
Montana State University
Bozeman, Montana 59715

D. T. Tingey
U.S.  EPA/CERL
200 S.W. 35th St.
Corvallis, Oregon 97330

P. C. Tourangeau
Environmental Studies Laboratory
University of Montana
Missoula, Montana 59801

C. Van Valin
U.S.  Dept. of Commerce
N.O.A.A.
Environmental Research Laboratories
Boulder, Colorado 80302
                                      vm

-------
                               ACKNOWLEDGEMENTS

     Many  individuals  have contributed to the  preparation  of this document.
Editoral assistance was generously provided by Ms. Karen Raldolph.   The help of
Mr.  Thomas Hill  and others is much appreciated.

     Our work could not proceed without the help and support of the people of
southeastern Montana, especially the ranchers on whose land we are working and
the personnel  and persons  residing at and  near Fort  Howes  Ranges Station,
Custer  National  Forest.  The  Kluver's  and the McRae's have been particularly
supportive.                            '
                                      IX

-------
INTRODUCTION

-------
                 ASSESSMENT OF ECOLOGICAL IMPACT FROM THE
                OPERATION OF A COAL-FIRED POWER PLANT IN THE
                         NORTHCENTRAL GREAT PLAINS

                                    by

                 R. A. Lewis, E.  M.  Preston, and N. R. Glass


                                INTRODUCTION

     The principal  short-term constraints on  the utilization of coal reserves
of the northcentral United  States are  the amount of environmental degradation
that  the  American  public is willing to  sustain as the price  for  secure and
abundant energy and the ability of scientists  to forecast the amount and kinds
of environmental impacts that will result from  a  given level  of coal  use.

     Because  of the  rapid  expansion  of  coal  mining  and coal-fired  power
generation in  the  northcentral  great  plains  and because of the very limited
knowledge  and  experience with the responses  of the  range resource  to air
pollution and other impacts  of coal development, it is  important to assess the
potential  responses  of these  ecosystems.  These grasslands exhibit  limited
resilience to  disturbance,  and  there  is  thus  a potential  for irreversible
effects.  It is not now possible to predict the  long-term consequences of even
the  present  level  of  coal  production  in the  west  (Lawton  and McNeil, 1973;
Library of Congress,  1975).

     One of  the primary  concerns of the  Colstrip,  Montana,  Coal-fired Power
Plant  Project  (CFPPP)  is  the  development   of methods  for  the  predictive
assessment  of  impact.   If  this  project  is   successful,  the ability  to
judiciously plan regional  power plant  inventories and  to select  sites for new
power  production  facilities will  be  greatly  enhanced.  Furthermore,  if the
nature and extent of specific impact can be predicted,  steps to minimize or to
ameliorate impacts  may be possible.  If the predictions  are  accepted, it should
be simpler to resolve conflicts among parties concerned  with the siting process
and to arrive at constructive agreements that reduce the overall costs of power
plant siting.

                       PREDICTIVE ASSESSMENT OF  IMPACT

     Federal   legislation  requires  the  prediction of  impact whether  the
capability exists or not.  The National  Environmental Policy Act of 1969 (NEPA)
requires that an environmental  impact  statement (EIS)  be filed  for any project
that requires federal  approval.   The EIS is to identify and describe potential
environmental  impacts  of  such  projects  and   to  evaluate their significance.
NEPA, however,  provides only limited guidelines for the  conduct  and preparation

-------
of  the EIS.   Consequently,  most  EISs  provide  few  data  that are  useful  to
decision makers or that satisfy the intent of the  law.

     The problems of evaluating and forecasting environmental consequences and
developing the technological  methods to reduce or ameliorate impact are complex,
and_CFPPP contributions at this time are  little more than embryonic.  Both strip
mining and coal combustion produce effects on living and non-living components
of  the  environment that  are extensive and  diverse.  While some effects may be
ephemeral (acute toxicity of sulfur dioxide, fluoride, nitrogen oxides, etc.),
others  may  prove  to  be  insidious  and long-lasting  or  irreversible  (acid
rainfall, long-term  soil  fertility decline, decreased energy flow,  retrograde
succession, etc.).

AVAILABLE FORECASTING TOOLS

     Forecasting  tools that  have been more  or less  routinely employed (e.g.,
guessing, extrapolation or rationalization of retrospective information, etc.)
have had  very limited or unknown success, because follow-up on predictions is
infrequent.   Those  tools  that  have  the  greatest  potential  for  predicting
environmental  impacts on terrestrial  ecosystems (e.g., the use of controlled in
situ  field  stressing  with  pollutants  of  concern combined  with  appropriate
laboratory experiments;  simulation  models  such as mass balance models; models of
simple  food webs;  and  diffusion models  that  employ data from the  sites  or
regions of concern) are rarely employed.

     Four  basic  types   of  predictive  tools  can  be  identified.   These  are
qualitative  techniques, retrospective analyses of data and  trend extrapolation,
process models, and empirical methods coupled to predictive  models.  Qualitative
methods are  based  upon judgments and intuition that are  logically arrived at in
the absence  of  any satisfactory theory, direct experience or quantitative data
relating  to  the problem.   Such methods may  lead to conceptual models that may
indicate  what  data are  needed and can generate hypotheses that can be tested.
Qualitative  methods  include  Delphi  methods,  panel  (workshop)   concensus,
intuition  and  visionary  forecast,  and  historical  analysis.   Occasionally,
historical analogy (which is applied widely  in EISs) proves useful for planning
horizons of several years.

     The  use of  scenarios,  while  not a  predictive  technique,  should  also  be
mentioned.   Scenarios  are  hypothetical  narratives  that explore  potential
alternatives,  usually those that seem reasonable or possible (worst case, best
case).   They  are  speculative,  but  can provide some insights into the range of
possible future states of the system and can force consideration of options that
might otherwise be overlooked.

     Methods  that employ  retrospective  data and  trend  analysis  may  be very
effective where relevant  data are  fairly complete for one or more  decades and
when the  functions of interest are noise-free and stable or follow an obvious
trend.   The  basic assumption  underlying these  methods  is that the  observed
patterns will  continue into  the future.  Consequently, such methods  tend to be
better  predictors of  specific short-term  futures  than of  long-term  trends.
Since such methods are based only upon time  series analysis and extrapolation,
the  relationships of  environmental  variables  to  trends  are  unknown.   As  a

-------
result,  predictions are  highly site  and  time specific.   A further  serious
drawback  to the  use  of  retrospective  data  is  that  it typically  includes
information that is not standardized, or is incomplete.  Such data must be used
selectively  and this provides  excellent opportunity  for the introduction  of
social, economic or political  bias.

     Process models can be powerful  predictive  tools.  While costly to develop,
such models  have  a good potential  for  long-term prediction  if  they  are  based
upon  the results  of  appropriate field  and laboratory  experiments  that are
designed  to evaluate  the  effects of  random and  non-random variables on the
output.

     The  environmental residual  technique  has  also been  used in environmental
impact  assessment.   This  technique  could be much improved  by a more thorough
assessment  of  impact  than  is  usual.  Undesirable effects  (residuals) of  power
generation  can be  determined through  a detailed knowledge  of the  power  plant
operations.  One dimension of a two-dimensional matrix defines  each  residual.
The  other dimension lists  management strategies.  By  filling out the cells  or
elements  within this matrix,  it is  possible to optimize  the  management of the
power plant  based  upon any one  or all  of the residuals of interest.  Different
management  methods might  prove to be most effective for different types  of
pollutants.   For  example, it  is possible  that SO  could best  be managed  by
building  a tall stack or by  intermittent  controls, while management of  heavy
metals  might be best accomplished through  the  construction  and  operation of a
baghouse.   The detail  of each  management  strategy  relative to   each  residual
would   in  turn be  dependent  upon  engineering  design and  properties of the
particular  coal which was to  be mined.   The main difficulty is in relating the
residuals to a detailed environmental  impact assessment which can in  turn feed
back  upon the  initial design and  site location of the  power plant  so  as  to
minimize  impacts.

     For  the  environmental  residuals   method  to  be   effectively  applied,  a
detailed  impact assessment with predictive capabilities is required.   Residual
allocation  models  have  been   linked  to  models  that  determine   resource
availability  based on  economics as well  as   geography  and end  uses  of the
particular  energy  activity from which  residuals are  generated  (Federal  Power
Commission,  in press).   Briefly, this  is  accomplished through   modeling  that
depends upon a  knowledge of the  resource being exploited, the technology  being
used, the chemical  and other residuals which  are produced  by the  technology,
and the economics at each stage.   These  factors  can be  used to generate a matrix
of residuals that  can be related to impact on the environment.  This  in turn
determines the  management strategy  for meeting air and water quality  standards
as well  as socioeconomic criteria.

     Controlled field  exposures to selected pollutants,  such as  those  being
conducted  by  the  CFPPP  research  team  (described  subsequently)  can enable
management to determine the exact relationship between the rate  of  exposure  to
a  particular pollutant  and  the   response of  local  biota.   Such  prediction  of
specific environmental  impacts at a  specific site could form the  basis for more
rational siting decisions  (Woodwell,  1970).

-------
                       ECOLOGICAL CONCEPTUAL FRAMEWORK

     A large and expanding body of theory and empirical  information dealing with
stress ecology  serves  as  the basis for  an  ecological  model  of the effects of
coal-fired  power  plant   emissions   on  northern  grassland  ecosystems.   A
conceptualization of this is shown in Figure 1.  The "target population" may be
defined either taxonomically or functionally.  An ecosystem functional model can
be  constructed  by   coupling the  models  of  functionally  important  "target
populations."  The degree of resolution  within  the ecosystem model can be set by
the definitions assigned  to  "target populations."  By examining the response of
the model to changes in air quality over time it is possible to characterize the
"emergent" properties of the system.

     The  quality  and magnitude  of direct effects  of  pollutants  on organismic
function  will  undoubtedly vary  among species and all higher taxonomic groups.
This  differential  selective pressure provides  the  basis for  indirect effects
upon   less   sensitive   biota  and  upon  ecosystem  structure  and  dynamics.
Investigators  in  this  project are interested in the relative  sensitivities of
species  to  both direct and  indirect effects and  the  relative homeostatic and
competitive capabilities  of these  species.  The time lag (which may or may not
be appreciable) between the  occurrence of direct effects and the appearance of
indirect  effects  and the  degree of reversibility of these  effects  are also of
interest.

     A number  of  parameters and functions should be investigated as potential
progenitors  of systems-level indirect  effects  in  northern  plains  grasslands.
Effects on critical resources such as soil moisture,  nutrients, solar radiation,
and trace elements are  likely to be very  important.

     Much of the dynamics of the northern plains grassland ecosystem appears to
be regulated/controlled by the  vagaries of the water  cycle  (Van Dyne, 1973).
Particulate emissions can affect the formation and building  of clouds by acting
as cloud  condensation  nuclei or ice nuclei  upon  which water  vapor condenses.
The hot,  dry air that accompanies these  aerosols as they are  swept into the base
of clouds may also be important.  Under  certain conditions precipitation may be
increased by the  presence of particulate emissions.  Under  other circumstances
precipitation may be reduced  (Library of Congress, 1975).  This could alter the
amount of water  to  biological  components  of the ecosystem and  could  lead to
major shifts in successional trends.   Water stress  may  interact synergistical ly
with  potentially toxic  air emissions,   affecting  the susceptibility  of  some
system components to  direct toxic effects.

     By  influencing  the  degree  and  nature of  cloud  formation,  and  by direct
scattering  and adsorption,  particulate  emissions  may affect  the  quality and
quantity  of  solar radiation impinging on the  ecosystem.  This could influence
primary  productivity  selectively,  by   species,  and  thus  serve  as   a  major
modifying force.

     Certain gaseous emissions may influence the nutrient cycle.  At low levels,
S02 and NO  may serve as  a source of  plant nutrients.  If portions of the system
are nutrient-limited, these  low level emissions could have a stimulatory effect

-------
                                MULTIDIMENSIONAL ARRAY  OF STACK EFFLUENTS
         DIRECT (TOXIC)
            EFFECTS
       FOOD CHAIN
     CONCENTRATION
 DIRECT (TOXIC) EFFECTS
             INDIRECT EFFECTS
                    VIA
             ALTERED PRIMARY
                PRODUCTION
   MULTIDIMENSIONAL
 PREDATOR POPULATIONS
adjustment
    of
pred at ion
              INDIRECT EFFECTS
                    VIA
           ALTERED ENVIRONMENTAL
                  QUALITY
MULTIDIMENSIONAL
TARGET POPULATION
 adjustment
of competitive
   balance
                                                  adjustment of
                                                life table parameters
                                                      T
  MULTIDIMENSIONAL
COMPETITOR POPULATIONS
                                                   EFFECTS  ON
                                              TARGET POPULATION
Figure ].  Autecological perspective of pollutant-ecosystem interaction.

-------
on plant  growth.   At  higher levels, such emissions begin to have direct toxic
effects which may override the stimulatory effects of low  level emissions.

     When  primary critical   resources  are affected,  changes  in physiological
functions or population parameters of "key" species will  quickly follow.  These
may include changes in growth rate, maturation time, reproductive rate, and rate
of phenological  development.  Changes  in populations that  are of sufficient
magnitude will alter  community structure.  Proportions of species may begin to
change,  new  species  may   successfully  invade.   Of  interest   is  how  these
structural  changes affect  community productivity  and  stability,  and  also the
degree of reversibility of these changes.

                     APPLICATION OF PREDICTIVE ASSESSMENT

     There are nearly  as many approaches  to the selection of power plant sites
as  there  are  utilities in this  country.   The  process  is governed  by many
individual circumstances and a  diversity of local and state laws  and procedures.
Further, the business concerns  of  the power company or utilities  involved, value
judgments by various decision makers, legislation, lobbies and the potential for
adversary proceedings can influence all  stages of the decision process.

     In  general,  control   of  the location  of  power  plants  and  attendant
facilities resides with state governments.  Policies differ among states  and in
some, comprehensive siting  laws have been enacted.  Montana, for example, has a
planning mechanism that requires state certification.  Plans  for  specific energy
facilities must be submitted to the state ten years in advance of construction.
The  state evaluates  the  collective prospective  impact  of  all  facilities and
certifies  individual   plants  on  the  basis of  this review.  In  many  states,
however, actual zoning, review, and certification of power plants is delegated
to local  governments;  state review may  not even  be required.   The individual
company  usually  selects a site  and   seeks  whatever zoning  changes   may  be
required.

     The selection of  sites  for new plants appears often to be haphazard  and is
initially dictated by  such  factors as ownership or availability of lease lands
and by  engineering constraints.   From  an often  large  set of available sites,
only a  few will  be subjected to  detailed feasibility  studies.   Perhaps one to
three are  then  evaluated according to  NEPA guidelines to determine whether or
not the proposed facility will have an acceptable level  of  environmental impact.
The selection  of appropriate sites can  be very costly, and in recent years, a
number of proposed facilities have been  denied permits following the review of
environmental impact statements.   Often the utility has no clear  idea until this
time whether or not the proposed facility has an opportunity for approval.

     The objective of  this  project is to  develop a procedure that when applied
will  yield biological  effects information and solutions that will contribute to
the  site  selection  decision.   For  the  present,   efforts  are   restricted  to
consideration of the biological effects  of coal-fired power plant emissions on
terrestrial  ecosystems of   the  northern  great plains  with the  Colstrip CFPP
Project providing  specific  data  and the  first application of certain protocol
elements.   Within  these  constraints  project  investigators  hope to  develop
methods  to  predict a  set of futures  and their range  of  probabilities  and to

-------
utilize this information to influence the course  of  events  so  that  undesirable
impacts of CFPP operation may be ameliorated.

     The prediction of environmental impact,  however effective,  is  of  no  value
to the siting  process unless it is appropriately  wedded  to  the constraints  and
needs  of  the  decision  making  processes.   Representatives  from  government
regulatory  agencies,  producer and  user industries,  environmental   groups  and
citizen groups must be included early  in  the protocol  development  process  in
order to define the policy and user constraints under which the protocol must be
developed.   Continuing consultation  should  be   maintained  with these policy
groups throughout the process of protocol development.

                        ELEMENTS OF A SITING PROTOCOL

     The  protocol  will  consist of a methodology  to  select  suitable sites  for
construction   of  coal-fired  power  plants  given  several  alternatives  and a
continuing  post-siting methodology to ameliorate undesirable biological  effects
should they appear.  Post-siting developments and continuing information inputs
to the protocol should favor flexible or adaptive operational management of  the
working  facilties.  The chronology of information and  action flow is displayed
in Figure 2.

     The  annual patterns of concentrations  of pollutants that would result from
constructing a coal-fired power plant at each of  the alternative sites  could be
predicted by a plume dispersion model or plume tracing studies.

     A set  of  widely acceptable methods for gathering "baseline" information on
the  structure  and dynamics of the terrestrial ecosystems at alternative  sites  is
also needed.   The  pollution predictions and  "baseline"  ecosystems  information
will then serve as  inputs to one or a series of biological response models  whose
outputs predict the biological effects of predicted pollutant concentrations  on
the  ecosystem under consideration.

     This  information  on potential biological  effects on several  alternative
sites must  be  considered in conjunction with outputs  from  a socioeconomic  and
engineering protocol to arrive at a siting  decision.   In specifying outputs  for
the  biological  response  models,  provision  must be made  for  interfacing with a
socioeconomic  segment.  Ecological effects  must  be  translated into  predictive
alternatives   in  man's  relationship  to   the  ecosystem   (e.g.,   ahanges   in
agricultural productivity,  changes  in numbers of esthetic  species,  changes  in
number  of  recreational  wildlife  species,  changes   in  ecosystem  management
practices).

     Once  a  decision  to   site  a plant  has  been  made and  construction   is
completed^the post-siting study  is implemented.   Data gathering is  now  directed
to monitoring pollutant concentrations and biological  effects in the  vicinity of
the power plant.   These  data  are used to  validate and  improve  the  predictive
models employed  in the  site-selection  protocol.  They  will also provide  the
information   base  for  adaptive  management  of  the power  plant.   If pollution
concentrations  or  biological  effects  reach  unacceptable  levels,  government
and/or management will  have the necessary information  to make  rational  modifi-
cations of plant operations.

-------
POLLUTION PREDICTION
   (DIFFUSION MODELS)
    O

SITE CHARACTERIZATION
(SEVERAL ALTERNATIVES)
       O
       o
       o
                 BIOEFFECTS RESPONSE
                        MODELS
                  OUTPUT FROM OTHER
                  PROTOCOL ELEMENTS
                    (SOCIOECONOMIC)
                     SITE SELECTION
                  (BEST ALTERNATIVE)
               o
               I-
               o

               <
                 o
       MONITOR
      POLLUTION
  MONITOR BIOLOGICAL
       EFFECTS
                     OUTPUT  FROM
                 SOCIOECONOMIC POST
                    SITING PROTOCOL
                       ADAPTIVE
                     MANAGEMENT
Figure 2.  Generalized flow diagram of the ecological effects protocol.
                          9

-------
     The final  bioeffects protocol  as presently conceived will consist of five
modules.

1.    A  site  characterization  module  will  include  a discussion  of preferred
methods in collecting  "baseline"  data and a discussion of the constraints and
limitations  of  these  methods.   This  module  should  be amenable  to stepwise
application.

     (a)  Background  Assessments  would  be  conducted  on numerous  sites with
     limited field data supplemented by existing literature.

     (b)  For  a  few  sites with  favorable  characteristics,  a  more rigorous
     Selection Assessment would be conducted.

     (c)  For the site  selected, a  comprehensive  Pre-Construction Assessment
     would be conducted.

     (d)  Finally,  a   long-term  Operations  Assessment  would be  conducted  to
     verify impact predictions, to determine multi-year ecological effects and
     to provide the basis for adaptive management.

2.   An analytical module will define data handling and synthesis requirements
and procedures.

3.   A  predictive module will  detail  and discuss predictive models that may be
used to estimate  future  impacts from CFPP operation at alternative sites.  This
module  should list the site specific parameters that must be measured in order
to  use these models and the  appropriate methods  to be  used  in  gathering the
data.  As far as possible, the accuracy and precision of these models should be
specified here.

4.   A  validation module will  describe the data requirements, data collection
methods, and procedures  for validating and improving predictive models in the
previous module.

5.   An adaptive  management module will discuss alternatives that are available
to  management to  ameliorate impact  once  the CFPP is in  operation.  This might
include the  use of weather forecasts and coal conversion schedules to predict
pollution  levels.  Dose-response (or exposure rate) models from the predictive
module  could  be used  to  define acceptable  pollution levels.   Coal conversion
activities could  then  be regulated  to insure that acceptable pollution levels
are not exceeded.

                                    SUMMARY

     To date, principal  investigators have developed a working knowledge  of the
ecology of grasslands  in the vicinity  of Colstrip,  Montana.   This knowledge,
together with  an  innovative  field  experimental method  for  the  assessment of
ecological  impact of air  pollution, is slowly providing the information required
to develop predictive  capability.  Such a capability will allow managers to plan
and to control  the siting process, to modulate (or ameliorate) impacts through
adaptive management and  to anticipate the possibility and needs for rehabili-

                                      10

-------
tation.  Improved management should be effected not only through more effective
and  efficient site-selection  procedures,  but  also  through evolution  of the
siting protocol  (update  and validation) and  the  adaptive  management of power
facilities through all  phases of their life cycle.

                                  REFERENCES

Federal  Power Commission.  In  press.   The need  for flue  gas desulfurization
     (FDG):  ecological effects  of air pollutants.

Lawton,  J.   H.  and  S.   McNeil.    1973.   Primary  production  and  pollution.
     Biologist 20:3-11.

Library of Congress.   1975.  Effects of chronic exposure to  low  level pollutants
     in the environment.   Congressional Research Service, Serial 0, pp. 402.

Van Dyne,  G.   M.   1973.   Analysis  of  structure,  function and utilization of
     grassland ecosystems.  Natural Resource Ecology Laboratory, Colorado State
     University, Fort Collins.

Woodwell, G.  M.  1970.   Effects of pollution on the structure and physiology of
     ecosystems.  Science, 168:429-433.
                                       11

-------
COLSTRIP STUDIES
     12

-------
                                   SECTION 1

                  EFFECTS OF S02 AND OTHER COAL-FIRING PLANT
                EMISSIONS ON PRODUCER,  INVERTEBRATE CONSUMER,
                  AND DECOMPOSER STRUCTURE AND FUNCTION  IN
                      THE VICINITY OF COLSTRIP, MONTANA

                                     by

      W. K. Lauenroth, J. L. Dodd, R. K. Heitschmidt and J. W.  Leetham


                                INTRODUCTION

     Within the  next  25  years  the northern Great  Plains of the United States
will be subjected to air pollution from energy development.  The air pollution
will come  from  the  conversion  of coal  to electricity through steam generating
power plants.  Most of the coal in the northern Great Plains lies under native
grassland and much of it will be converted to electricity near the open mines.
Sulfur  dioxide  and other  air  pollutants from the conversion  process  will  be
released onto the adjacent native grasslands.

     The objectives of this study are to look at the effect of  air pollution on
primary producers, arthropod consumers, and decomposition processes.  The study
has  three  aspects.   At Colstrip,  this  consists  of  baseline monitoring  of
decomposition  processes and  productivity  and  seasonal  biomass  dynamics  of
primary producers and arthropods at four different sites that range from 11  to
18 km east and southeast of the power plant.   Baseline  monitoring began in 1974
and  will   continue  over the  next  several  years  to  determine what,  if  any,
influence air pollution  has on these grasslands.

                          COLSTRIP MONITORING SITES

     The 1975 growing season marked the end of a baseline data collection phase
for  the  Colstrip Power  Plant  Project.  Unit  I  of the  generating  complex  at
Colstrip  began  operation in the  fall  of 1975 and the growing  season  of  1976
represents  the  first  year  of data collection under the influence of the power
plant operation.   An  important task  now to be completed with  respect  to the
baseline  data  is  to compare  and  contrast  the  two  years  of  baseline  data
collection, 1974 and 1975,  with respect to abiotic conditions in an attempt to
relate differences  in  ecosystem response to differences in abiotic conditions
during the two years.   The  objective of this  report is  to present the 1975 data
and make some initial comparisons with the data collected  in 1974.
                                      13

-------
ABIOTIC DATA

     The collection of  abiotic  data at the Colstrip sites was not begun until
the growing season  of  1975.   The U.S.  Department of Commerce weather data for
southeastern Montana can  be  used to make general  comparisons  between the two
years.   Precipitation data for 1974 and  1975 are recorded in Table 1.1.  Data for
1974 are from  the  U.S.  Department of Commerce Weather Station at Colstrip and
1975 data were  collected  at  the four monitoring  sites.   The  Colstrip Station
(USDC) apparently was  not operated in  1975 and data are not available to make
comparisons between 1974 and  1975 to  see  how these  data compare  with data
collected at the monitoring sites.  Total precipitation for the May to September
period was  228 mm in 1974 and 261 mm average for the four sites in 1975.  More
important than the  differences  in total precipitation were the differences in
the May and June precipitation.   In 1974,  121  mm of precipitation were recorded
during May and  June  and in 1975,  221  mm were measured.

     The average monthly  temperatures  in  Table 1.2 are  presented  for general
comparison  of  the  growing seasons during 1974 and 1975 and do not necessarily
represent  average   monthly  temperatures for  the  Colstrip  sites.    The major
differences between the  two  years  included a much  cooler April,  a slightly
warmer May, and a cooler June.

     Soil water data collected at the Colstrip sites in 1975 reflected the high
precipitation during May and June (Table 1.3).  Soil  water to  a depth of  105 cm
remained above 20 cm on all  sites through 30 June.  This was  followed  by a rapid
depletion during the first two weeks of July and a subsequent gradual depletion
throughout  the remainder of  the growing  season.   No major  differences were
evident  among  sites except  for small differences in total soil water observed
throughout  the period.   The Hay  Coulee and Kluver  East sites apparently have the
highest water-holding capacity and the Kluver North site the lowest.   Soil water
at the end  of the growing season was highest at the Kluver East  site  and  lowest
at the Kluver North site and averaged approximately 10 cm to a depth of 105 cm
for al1 sites.

     Soil  surveys  of the  Colstrip sites were made in  1975.  Although physical
and chemical descriptions are incomplete at this time,  profile descriptions for
Hay Coulee, Kluver  West, and Kluver East are included in the Appendix.

PLANT PHENOLOGY

     Phenological  observations  were made  on  a weekly  basis  using  a 14-stage
phenological  classification   (Table 1.4)  during   the  1975  season.   Fourteen
species  were   observed  and  represented  a  range   from  Hood's  phlox  which was
flowering at the time  of the first  observation to fringed sagewort which had
just  begun flowering  at the  time  of  the  last  observation  in mid-September
(Tables 1.5 and 1.6).   Few differences in the timing of phenological events were
among sites with stage  14, drought-induced dormancy, being the most variable of
the stages  and  presumably  reflecting differences in soil water among sites.
                                      14

-------
TABLE 1.1.  GROWING SEASON PRECIPITATION (mm) FOR THE COLSTRIP SITES DURING
            THE GROWING SEASON IN 1974 AND 1975.

Colstrip sites 1975
Month
May
June
July
Aug.
Sept.
Total
Col strip
1974
94
27
50
28
29
228
Hay Coulee
121
94
17
16
5
253
Kluver West Kluver North
121 120
93 116
23 13
21 19
5 5
263 273
Kluver East Average
113 119
104 102
14 17
15 18
6 5
252 261

TABLE  1.2.  AVERAGE MONTHLY TEMPERATURE (°F) FOR SOUTHERN MONTANA DURING 1974
            AND 1975.


Month
April
May
June
July
Aug.
Sept.
Southeastern
1974
46.5
50.8
65.2
73.9
63.7
55.0
Montana average
1975
38.4
52.5
61.3
73.0
67.1
56.2
                                      15

-------
TABLE 1.3.  SEASONAL DYNAMICS OF SOIL WATER (cm),  COLSTRIP STUDY  SITES,  1975.
Depth
Treatment (cm)
Hay
Coulee





TOTAL
Kl uver
West





TOTAL
Kluver
North





TOTAL
Kl uver
East





TOTAL
0-15
15-30
30-45
45-60
60-75
75-90
90-105

0-15
15-30
30-45
45-60
60-75
75-90
90-105

0-15
15-30
30-45
45-60
60-75
75-90
90-105

0-15
15-30
30-45
45-60
60-75
75-90
90-105

3 May
3.73 + .
4.04 + .
4.54 + .
4.79 + .
4.17 + .
3.34 + .
1.73 + .
26.35
3.36 + .
3.34 + .
3.77 + .
3.90 + .
3.31 i .
3.70 + .
3.25 + .
24.63
3.04 + .
3.39 + .
3.83 + .
4.72 + .
3.63 +
2.91 + .
2.42 + .
23.93
3.52 + .
19
37
06
47
31
35
20

11
05
03
16
08
29
33

20
20
20
92
14
34
14

07
4.16 + .42
4.36 + .
4.05 + .
3.99 + .
3.51 + .
2.01 + .
25.61
27
11
19
48
71

14 May
2.95 +
4.00 +
.24
.28
4.39 + .34
4.29 +
3.68 +
3.43 +
3.30 +
26.
2.96 +
3.27 +
3.01 +
2.83 +
3.39 +
4.23 +
4.07 +
23.
2.30 +
2.41 +
3.47 +
3.46 +
3.56 +
3.66 +
3.16 +
22.
2.97 +
3.61 +
4.15 +
4.^7 +
3.99 +
4.35 +
4.11 +
27.
.36
.63
46
.26
04
.29
.03
.56
.78
.18
.20
.25
77
.03
.63
.07
.13
.09
.26
.23
03
.26
.69
.32
.23
.25
.16
.28
24
4 June
1 .95 + .
2.33 + .
3.52 + .
3.97 + .
3.20 + .
3.09 + .
3.16 + .
21.22
1 . 98 + .
2.78 + .
3.12 + .
2.87 + .
3.23 + .
3.02 + .
3.06 + .
20.07
1.93 + .
2.50 + .
3.13 + .
5.33 + 1
2.97 + .
3.50 + .
3.60 + .
22.96
2.21 + .
3.57 + .
3.22 +
3.25 + .
3.14 + .
3.65 + .
3.37 + .
22.42
10
55
34
04
15
50
35

19
23
20
15
52
40
39

06
31
39
.80
53
31
30

13
33
19
17
65
08
28

19 June
4.13 + .
3.54 +
3.36 + .
3.28 + .
3.44 + .
3.45 + .
3.28 + .
24.49
2.03 + .
2.70 - .
2.56 + .
3.46 + .
3.44 + .
2.76 + .
3.54 + .
20.49
2.28 3 .
3.16 + .
3.25 + .
3.44 + .
3.30 + .
3.26 + .
3.43 + .
22.12
2.76 + .
3.46 +
3.15 + .
3.47 + .
3.42 + .
3.37 + .
3.44 + .
23.07

18
49
32
33
15
39
55

07
20
08
61
09
23
63

13
20
19
43
46
38
46

21
40
63
10
27
33
32

30 June
1.99 + .
2.70 + .
3.23 + .
3.19 + .
2.95 + .
3.21 + .
2.93 + .
20.20
2.23 +
2.78 ^ .
3.16 +
3.31 + .
3.14 + .
3.29 + .
3.40 + .
21.31
2.40 + .
2.99 + .
3.15 + .
3.57 + .
3.44 + .
3.46 + .
3.42 + .
22.43
2.26 + .
2.66 + .
2.87 + .
3.02 + .
3.43 + .
3.14 + .
2.96 + .
20.36

21
41
30
33
22
40
34

19
78
44
51
42
36
64

50
47
08
03
18
27
12

16
31
16
07
66
31
09

14 July
0.99 + .
1 .42 + .
1.83 + .
2.03 + .
2.22 + .
2.32 + .
2.53 + .
13.43
0.72 + .
1.21 3 .
1 .50 + .
1 .67 + .
2.64 + .
2.67 + .
2.49 + .
12.91
1.19 + .
1 .61 + .
2.00 + .
2.55 + .
2.54 + .
2.54 + .
2.93 + .
15.36
1.22 + .
1.81 +.
2.14 + .
2.19 + .
2.60 + .
2.64 + .
2.66 + .
15.26
29 July
09
22
18
18
28
40
45

04
08
04
13
82
77
53

19
26
21
40
54
55
42

16
30
09
22
21
26
20

1.33 +
1.55 +
1.92 +
2.44 +
2.67 +
2.88 +
2.23 +
15.
0.96 +
1.51 +
2.17 +
2.53 +
2.56 +
2.40 +
2.60 +
14.
0.94 +
1 .44 +
1.67 +
1.93 +
2.05 +
2.36 +
2.70 +
13.
0.88 +
1.34 +
1.39 +
1.60 +
1.92 +
2.66 +
2.94 +
12.
.09
.18
.12
.16
.49
.66
.12
01
.13
30
.31
.23
.12
.34
.43
73
.08
.20
.14
.27
.27
.16
.26
09
16
.22
17
13
.35
.29
.34
74
9 August
0.80 + .
1.09 + .
1.44 + .
1 .96 + .
2.27 + .
2.72 +
2.02 + .
12.30
0.59 + .
1 . 00 + .
1.07 + .
1.14 + .
1.50 + .
1 . 84 + .
2.13 +
9.27
0.60 + .
0.94 + .
1.06 + .
1.24 + .
1.20 + .
1 .53 + .
2.09 + .
14
25
26
84
36
16
18

04
13
13
07
37
64
18

03
10
11
11
10
19
37
8.66
0.99 + .
1.16 + .
1.19 + .
1.69 + .
2.07 + .
2.29 + .
2.90 + .
12.29
37
12
11
26
09
21
47

25 August
0.59 + .
1.10+ .
1.51 + .
1.56 + .
1 .80 + .
1.85 + .
2.07 + .
10.49
0.73 + .
0.95 + .
0.94 + .
1.10 + .
1.45 + .
2.70 + .
2.38 + .
10.25
0.73 + .
1.13 + .
1.24 + .
1 .67 + .
1.63 + .
1.83 + .
2.21 + .
04
23
25
12
19
21
27

11
09
01
13
18
22
75

09
06
07
31
36
56
52
9 September
0.69
1.21
1.65
1.41
1.84
1.53
2.15
10
0.68
1.28
1.02
1.57
1.74
2.45
1.81
10.
0.63
1.12
1 .30
1.01
1 .32
1.55
1.81
10.43 £
0.78 + .
1.12 + .
1.77 +.
1.28 + .
1.84 + .
2.35 + .
2.47 + .
11.61
11
11
66
06
17
36
51

0.67
1.16
1.35
1.27
1.91
2.28
2.62
11
+ .05
+ .19
+ .46
+ .24
+ .33
+ .14
+ .33
.48
+ .04
+ .37
+ .10
+ .41
+ .54
+ .77
+ .35
56
+ .13
+ .25
+ .21
+ .25
+ .30
+ .47
+ .64
1.73
+ .15
+ .10
+ .12
+ .09
+ .13
+ .06
+ .34
.26

-------
TABLE 1.4.  PHENOLOGICAL STAGES USED DURING THE 1975 GROWING SEASON.





                   Phenological stages



      1..  Winter dormancy



      2.   First visible growth



      3.   First leaves fully expanded



      4.   Middle leaves fully visible



      5.   Middle leaves fully expanded



      6.   Late  leaves fully expanded



      7.   First floral buds



      8.   Mature floral buds



      9.   Floral buds and open flowers



     10.   Floral buds, open flowers, and  ripening fruit



     11.   Buds, flowers, and green and  ripe fruit



     12.   Buds, flowers, green and ripe fruit, and seeds



     13.   Green and  ripe fruit, and dispersing seeds



     14.   Dispersing seeds and drought-induced dormancy
                                 17

-------
       TABLE 1.5.   PHENOLOGY OF MAJOR FORB SPECIES,  COLSTRIP  STUDY  SITES,  1975  (A = Hay Coulee, B = Kluver West, C = Kluver North,  D = Kluver East).
oc
Phlox Tragopoqon Hedeoma Sphaeralcea
hoodji dubius h i s p i da cocci nea
Date
27 May
2 June
11 June
18 June
23 June
30 June
8 July
14 July
22 July
28 July
3 Aug.
10 Aug.
18 Aug.
26 Aug.
1 Sept.
8 Sept.
14 Sept.
ABCD ABCD ABCD A R
11 11 11 11 55554 4
13 13 13 12 7777 4 5
14 14 13 877 4 5
14 9 8 9 9 8 7
14 9 9 11 10 9 8
11 10 11 10 10 10 10 10 10 9
13 13 13 13 12 12 12 10 11
13 13 13 13 13 13 13 12 13 11
13 14 14 14 13 13 13 13 13 11
13 14 14 14 14 13 13
14 13 13
13
14 14




C D
4
5
5

9
10
10
10
12
12
13
13
14




Plantaqo Psora lea
Patagonia araophvlla
ABCD ABC
2 3
2 43
344
7 7
8 7
10 9 9 9 83
10 10 10 9 10 10
10 10 10 10 10 10
13 13 12 12 10 10
13 13 13 13 10 10
13 13 13 13 13 13
14 13 13 13 13 13
14 14 14 13




D
2
3

7
7
8
9
10
10
10
13
13
13

13
14

Artemisia
frig
A
4
4
5
5
5
5
5
5
6
6
6
6
7

9
9
9
B
4
4
4
4
4
4
4
5
6
6
6
6
7

8
8
9
ida
C
4
4
4
4
4
4
4
5
6
6
6
6
7

9
9
9

D
3
4
4
4
4
4
5
5
5
6
6
6
7

9
9
9

-------
TABLE 1.6.  PHENOLOGY OF MAJOR GRASS SPECIES, COLSTRIP STUDY SITES, 1975 (A = Hay Coulee, B = Kluver West, C = i'.luver North, D = Kluver East).

Poa secunda Stipa conata

Date ABCD ABCD
27 May
2 June
11 June 9
19 June 9 889
23 June 10 999
30 June 10 10 10 9 999
8 July 12 12 12 11 10 11 11 10
14 July 13 13 13 12 13 13 13 13
22 July 14 14 14 13 14 13 13 13
28 July 13 13 13 13
3 August 14 14 14 14
10 August
18 August
26 August
1 September
8 September
14 September
Bromus
.iaponicus
ABCD
7777
8888
888
9889
9999
9999
10 9 9 10
10 9 10 12
13 12 13 13
13 13 13 13
14 14 14 14






Koeleria
cristata
ABCD

8
8
9 9
9999
9999
9 9 9 11
11 12 12 12
13 13 13
13 13 13 13
13 13 13 13
13 13 13 13
14 14 14 14




Aristi'cla
loncnseta
ABCD
3
4
4
7
7
8 8
8 8
8 8
10 10
n 10
13 13
13 13
13 14


14

Bouteloua
gracilis
A
4
4
4
4
5
5
7
9
9
10
11
12
13




B
4
4
4
4
4
5
6
6
8
10
11
12
13

13
14

C
4
4
4
4
4
5
6
8
9
10
11
12
13

13
14

D
4
4

5
5
5
7
8
9
10
n
12
13




Aqropyron
smithii
A
4
4
4
6
9
9
10
10
10
11
11
11
12

13
13
13
B
4
4
4
5
6
8
9
9
9
10
n
n
12

13
13
13
C
4
4
4
6
7
9
9
10
10
n
n
12
12

13
13
13
D
4
4

7
7
8
9
10
10
n
n
n
12

13
13
13

-------
ABOVEGROUND BIOMASS DYNAMICS

     Figure 1.1 presents seasonal dynamics of total  aboveground biomass for the
four Colstrip  sites  during  1974 and 1975.  Represented here are current live,
recent  dead  (material  that  died during growing season of measurement) and old
dead (material that  died prior to growing season of measurement) biomass.  The
immediate differences observable on this figure are a larger amount of current
live biomass in 1975  and the slight delay in the recording of the peak in current
live biomass.  Another difference between years, although not an unexpected one,
is the  larger amount of old dead biomass recorded in the spring of 1975 compared
to  1974.   This is a  direct  result  of  the exclusion of  cattle  by  fencing and
points  to a  problem  that must be addressed in analyzing the baseline data.  We
must  be able to sort  out differences  between  the  two years both  in  terms of
differences  in abiotic regimes  and differences resulting  from fencing.  The
Kluver  East  site  which  showed  the  largest  increase in  current  live biomass
between  1974  and  1975 also had the  largest  response  to fencing.   Although we
have not looked at the data in terms of species composition, we expect that the
effects of fencing on this site will  be  larger than on the remaining  sites.

     Figure  1.2 compares the current season production of cool  season grasses,
the most important functional group on the study sites, between 1974 and 1975.
Again,  the  major  differences are the larger absolute amount of current season
production during  1975 and the slight delay in peak production.  In 1974, peak
production was  recorded  on  approximately 15 July 1975.  Although there was an
increase in the production of cool season grasses in 1975,  the relative position
of  each site  remained  constant for both years.  The greatest  amount of cool
season  grasses  was found on the Kluver West  site  in both years and the least
amount  of  cool  season grass production was recorded for the Kluver North site
for both seasons.

     Figures  1.3  and  1.4  present  current  season's  production   for  western
wheatgrass and needle-and-thread grass, the two major contributors to the cool
season  grass  group.  Similar trends of greater production can be seen for both
species  during 1975  than  during 1974.  Western wheat-grass  showed  much less
fluctuation in the measurement of current season's  production  from  date to date
in 1975 than  in 1974.  Needle-and-thread grass productivity was the greatest on
the Kluver West site  and not measurable for the Hay Coulee site.  The Kluver East
site had the  least amount of needle-and-thread grass productivity  in both 1974
and 1975.

     Cool season forb productivity averaged approximately 10 g/m2 for all sites
in 1974  (Figure  1.5).   It  was only slightly higher (approximately 15 g/m2) in
1975.   The Kluver North site had the largest productivity of cool   season forbs
in both years and the Kluver East  site the  least amount  of  cool  season forb
production.

     Half-shrub productivity was also greater in 1975 than in 1974 (Figure  1.6).
The abrupt drop in half-shrub biomass recorded at the beginning of  July in 1974
was not present in 1975.   The Kluver East site had  the  greatest productivity of
half-shrubs in 1975 while the Kluver North site had  the  greatest productivity in
1974.   This  again  is an indication of the  changes that  are  occurring  in the
Kluver East  site, presumably as a result of fencing.

                                      20

-------
  o
  00
      150-
      100-
	Liwt
— R*c*nt d«od
— Old dead
      I5O-
      100-
       50-
        o4r*
          APRIL   MAY
                                            ^&Tt~a asTeM & is 25 s is as
          JUNE   JULY  AUGUST SEPTEMBER  APRIL   MAY   JUNE   JULY  AUGUST SEPTEMBER
Figure  1.1.   Comparison  of aboveground  biomass dynamics  during the 1974  and 1975
              growing seasons at the four  Colstrip sites:   (a)  Hay Coulee,  (b)  Kluver
              West, (c) Kluver North, and  (d)  Kluver East.

-------
     100
      50
 
-------
      100
	Kluver East
       50
 OJ
 c/)
 en
 o

 CD
        0
      100
      50
                — Hay Coulee

                ---Kluver  West

                — Kluver  North
                                                        a
                                                          \
                         15  25  5  15  25
APRIL
MAY
JUNE
                                 JULY
                                                               5  15  25
                                                     AUGUST  SEPTEMBER
Figure 1.3,     Production of western  wheatgrass at the Colstrip sites during

               the 1974 (b) and 1975  (a) growing seasons.
                                    23

-------
     CJ
      GO
      CO
      o
      CD
          100 -  —
           50
  0

100
       	Kluver  West
       	Kluver  North
         — Kluver  East
                                                             a
           50
            0

                             -H—~

                     J	L
               5  15  25  5  15 25  5  15  25  5  15  25  5  15 25
                 MAY       JUNE      JULY     AUGUST  SEPTEMBER
Figure  1.4.     Production of needle-and-thread grass at  the Colstrip  sites
              during the 1974  (b) and 1975  (a) growing  seasons.
                                   24

-------
      100
 Hay Coulee

•Kluver West

 Kluver North

 Kluver East
                                                   a
       50
  CvJ
  co
  CO
  o
  CD
         0
      100
       50
                                           5   15  25   5  15 25   5   15  25
             APRIL
     MAY
JUNE
JULY
AUGUST  SEPTEMBER
Figure 1.5.     Production of cool season forbs at  the  Colstrip sites during
               the  1974  (b) and 1975 (a) growing seasons.
                                     25

-------
        100
         50
    
•	Hay Coulee

	Kluver North

	Kluver East
                                                                  a
    3   100
    s
    o
    GO
          50
           0
              5   15 25   5  15  25  5   15  25  5  15  25   5  15  25

                MAY      JUNE      JULY      AUGUST  SEPTEMBER
Figure  1.6.    Production  of half-shrubs at the  Colstrip sites during  the
              1974 (b)  and 1975 (a) growing seasons.
                                   26

-------
     Warm  season grasses  are  generally an  unimportant  component  in terms of
total productivity of these sites and very small differences among the two years
were recorded  (Figure  1.7).  The  Hay  Coulee  site had the greatest productivity
of  warm season  grasses  in both  years  and  was approximately 15 g/m2.   Slight
differences  were seen  for the remaining sites with an increase  in warm season
grasses for both the  Kluver West and Kluver East sites.

     Aboveground net primary production (ANPP) was  calculated  by  summing the
peak values  of current live plus  recent dead biomass for each functional group.
Figure  1.8 shows ANPP for the four sites for 1974 and 1975 with the contribution
of  each functional  group.   In 1974 ANPP estimates ranged from 106 g/m2/yr for
the Kluver East  site to  123 g/m2/yr for the Kluver  North site.  The ANPP in 1975
was  greater  for  all  sites,  ranging  from 131  g/m2/yr  for  Kluver West  to 165
g/m2/yr  for  Kluver East.   The percentage contribution of the  functional  groups
to ANPP was  similar between years for all sites with the maximum  differences on
the order  of 5% to 7%.  An example  of  this is the data from the Kluver North
site.  Here the percentage contribution  of half-shrubs increased 5% from 1974 to
1975, the same for cool  season forbs decreased 7%,  warm  season grasses increased
5%, and cool season grasses decreased 4%.  The explanation for these differences
is  related  to  the  interaction of exclusion  from   grazing   and  increased
precipitation  in 1975.   At this time  we are  not able to separate the effect of
these influences.

BELOWGROUND PRODUCER BIOMASS

     The  seasonal  dynamics  data  for  belowground biomass require  further
analyses  and  are  not  discussed  in  this  report.   Analyses   of  the  depth
distribution  of  root biomass are complete.  Figure  1.9  presents the vertical
distribution of  root biomass for  six sites  over 2 years.  Although there is some
variability  in  the  depth  distribution  of the root  systems  of  these northern
mixed prairie  grasslands, on the average they are quite  similar.   Averaging over
all sites in both years  approximately 51% of the total  root biomass is found in
the 0-10 cm depth.  This  value decreases  to  4% of the total  root biomass found in
the 50-60 cm depth.  These averages are similar to root distributions reported
by other researchers  (Dodd  et al_. , 1974;  Bartos  and  Sims, 1974).

LITTER DYNAMICS

     Changes in  litter  standing  crop on the Colstrip  sites  for 1974 and 1975
indicate  general similarity  between sites  and years   (Table 1.7).   Although
several dissimilarities  among  sites   are apparent  for  the  1974  season  and the
early portion of the 1975 season, the variation between sites by  the end of 1975
(August and  September) was small.  The early differences are attributed  to lag
effects from differential grazing histories  prior to 1974; the authors conclude
that, with respect  to  litter production and decomposition,  the  sites  are now
comparable and will likely remain so  unless differences are induced by site-to-
site variations in weather  or air pollution.

ARTHROPODS

     The arthropod fauna of the  four EPA field sites near Colstrip, Montana (Hay
Coulee,  Kluver  West,  Kluver North,  and  Kluver East)  were   sampled by  three

                                      27

-------
         100-
         50
     
-------
      150-
                 Season Grasses


                  Season  Grasses


            Cool Season Forbs
                                                   Warm  Season Forbs


                                                   Half Shrubs
                                                                           a
OJ
o
i-
o
=)
a
o
a:
o.
ct:
LU
UJ
      100
       50
o
       50
      100
       50
        0



Figure 1.8.
                                                                 \\\\\\\\\\\N
                Hay  Coulee
                        Kluver West
Kluver North
Kluver  East
               Total aboveground net pv^imary  production for the Colstrip

               sites during 1974 (a) and  1975  (b).
                                     29

-------
       0-10
      10-20
     20-30
E
o
     30-40
LiJ
Q
     40-50
     50-60 -
                       •Colstrip - 1974
                       •Colstrip - 1975
                        Ash Creek - 1974
	Taylor Creek  -1975
                                         VERTICAL  ROOT  DISTRIBUTION
                                Depth
                               0-10
                              10 -20
                             20 -30
                             30 -40
                             40 -50
                             50 -60
        Average
           51
           16
           13
           9
           7
           4
                                _L
                      10
              20
                                           _L
40
50
60
  Figure 1.9.
                             30
                         PERCENT
Distribution of root biomass by depth increments for three northern mixed-grass
prairie sites.

-------
TABLE 1.7.  LITTER STANDING CROP (X ± SE, ash free g/m2) FOR COLSTRIP SITES,
            1974 AND 1975.

Hay Coulee
Date
•974
11 May
11 June
29 June
26 July
16 August
26 September
X
1975
22 April
18 May
20 June
15 July
1 1 August
16 September
X
X

161
172
201
172
128
192
171

162
170
179
154
231
218
186
SE

17
11
13
12
12
15


10
14
11
15
26
18

Kl uver
X

177
175
183
140
139
167
163


159
166
160
207
191
177
West
SE

13
12
6
12
19
13



9
20
10
15
14

Kluver
X

234
157
169
171
198
230
193

193
205
163
218
277
189
207
North
SE

21
22
15
18
10
24


22
19
16
32
33
16

Kl uver
X

158
154
171
161
153
148
157

144

157
152
204
206
144
East
SE

16
12
19
18
18
13


17

19
14
16
39

                                      31

-------
different techniques, each designed for certain components of the total fauna.
For  the  "aboveground  arthropods,"  i.e.,  those  occurring  in the  litter and
standing vegetation, a  quick trap system was  used.   An  open-bottomed cage of
known size was  dropped  over preselected, randomly located plots.  The cage or
quick trap was  dropped  from the end of an 18 ft boom mounted on a two-wheeled
cart.  The insects, vegetation,  and surface litter were removed from the trap by
a vacuum system and the arthropods extracted from the refuse by use of Berlese-
or TulIgren-type extraction.

     The  "soil  macroarthropods,"  i.e.,  those  arthropods occurring  below the
surface  litter and large enough to be sieved from the soil with a 1 mm opening
sieve, were sampled by taking 12.5 cm diameter by 15 cm deep soil cores and wet
sieving  them.   All material  retained  by the  sieve  was  further  separated by
flotation in a saturated solution of magnesium sulfate (MgS04).

     The  "soil  microarthropods,"  i.e.,  those occurring in the litter and soil
and small enough to pass through a 1 mm opening sieve,  were sampled by taking 5.0
cm diameter by 10.0 cm deep soil cores in such a way as to retain,  as nearly as
possible,  the original soil  structure.   The  arthropods  were extracted  by  a
Macfadyen  high  temperature gradient system.   Leetham  (1975)  gives a detailed
discussion of all the above techniques.

      In  the  field at each randomly chosen sample point, one sample of each of
the  above types was taken with five such sample points being selected in each of
two  replicates  of each  treatment, i.e., field site.   All arthropods collected
were  identified  to family  (genus  and species where possible).  Representatives
of each  group or category  were dried at 65°  C  for 25 hr and weighed to obtain dry
weight biomass data.  There were five to six sample dates in each of  the  1974 and
1975 seasons.

     Because of  some recently detected problems in the computer summarizations
of the soil microarthropod data, a discussion of that portion of the arthropod
fauna will be included in a later report.

     Tables 1.8 and 1.9 give  summarizations of the aboveground and soil macro-
arthropods by total counts (Table  1.8) and by the various trophic or functional
groups (Table  1.9).  From  Table 1.8 it is apparent that generally there was a
substantial difference  between  the two seasons with the 1975 season producing
larger numbers of  both aboveground arthropod biomass which shows a substantial
reduction, possibly  due to  a shift in  the dominant  species  between the two
years.  There appears to  be  considerable treatment or field site differences,
however,  the significance of these differences  may be artificial due to  the high
variability of  the data.   Generally,  the Kluver North  (C)  plot  supported the
highest arthropod populations in 1974 but that dominance changed in  1975.  Since
no statistical analyses have been made on the data,  any significant treatment
and/or year effects are unknown.

     Table 1.9 gives a summary of  the arthropod fauna  by trophic group.  Plant
feeding types are all grouped together here rather than splitting them  by their
feeding types, i.e,  chewing,  sucking,  pollen  feeding, etc.  The plant feeding
and omnivorous types appear  to be the dominant feeding types, both by numbers
and biomass for both above and belowground arthropods.  Within the treatments,

                                      32

-------
TABLE 1.8.  SUMMARY, BY TOTAL COUNT, OF THE ARTHROPOD CENSUSING  AT  THE  FOUR
            FIELD SITES NEAR COLSTRIP, MONTANA FOR THE 1974 AND  1975  SEASONS
            (A = Hay Coulee, B = Kluver West,  C = Kluver North,  D = Kluver
            East).

Group
?1/ 21/
Numbers/m Biomass g/m
ABCDABCD
Aboveqround arthropods
        1974            45.2   39.8   96.5   48.3   .143  .269  .345   .200

        1975           136.1   99.3  122.2  109.2   .119  .174  .126   .082

Soil macroarthropods
1974
1975
37.
82.
8
2
49.
148.
5
3
74.7
62.5
42.
69.
7
5
.148
.574
.065
.551
.080
.620
.045
.356

-Time weighted means.
                                      33

-------
TABLE 1.9.   SUMMARIZATION  BY  TROPHIC  OR  FUNCTIONAL  GROUP  OF THE ARTHROPOD
            DATA GIVEN IN  TABLE  1.8  (A = Hay  Coulee,  B =  Kluver West,  C =
            Kluver North,  D = Kluver  East).
2^
Numbers/m

Group
Aboveground arthropods
1974

Soil

1975
macroarthropods
1974
1975
Aboveground arthropods
1974

Soil

1975
macroarthropods
1974
1975
Aboveground arthropods
1974

Soil

1975
macroarthropods
1974
1975
Aboveground arthropods
1974

1975
A
2.9
11.5
23.2
30. 1
28. 1
49.0
5.8
20.9
6.6
17.1
6.2
11.6
0.5
1.0
B
1
15
3
75
C
D
Unknown
.4 2.2 5.8
.7
.9
.2
Plant
21.6
49
8
18
5
8
1
22
0
0
.1
.9
.8
20.
9.
3.
2 18.4
8 6.3
5 19.6
Feeding
60.6 22.6
63.
9.
25.
0 50.4
7 16.4
1 24.8
Predators
.4 8.9 7.6
.9
.4
.5
13.
9.
17.
7 15.3
9
6 6.9
Parasite id
.4 0.8 0.6
.6
1.
4 0.4
A
.001
.002
.091
.072
.132
.085
.023
.332
.005
.008
.023
.064
>.001
.001
21/
Biomass g/m
B
>.001
.001
.016
.013
.249
.150
.027
.112
.005
.005
>.001
.125
>.001
>.001
C
.001
.001
.024
.004
.326
.100
.032
.410
.005
.012
.002
.127
.001
.001
D
.002
.001
.001
.018
.186
.065
.040
.263
.005
.003
--
.043
.001
>.001
                                                                 (continued)
                                     34

-------
TABLE 1.9.  (continued)
                                Numbers/m^
                           Biomass g/m^
        Group
                              B
Soil macroarthropods
        1974

        1975
Aboveground arthropods
        1974             5.0

        1975            21.8

Soil macroarthropods
        1974             3.0

        1975            30.1
Aboveground arthropods
        1974             2.1

        1975            33.8

Soil macroarthropods
        1974

        1975             1.7
Aboveground arthropods
        1974

        1975             1.9

Soil macroarthropods
        1974             1.8

        1975            14.7
2.5
        1.0
   Omnivore

 5.0   20.0    7.2

12.7   16.0   13.6


25.9   23.7    7.0

75.2    3.5   19.6

   Scavenger

 5.9    3.3    4.3

13.5    7.7   11.2


 5.8    2.9    3.0

 6.9    3.8    4.5

  Non-feeding

 0.1    0.1    0.1

 2.9    0.4


 2.4   16.8   11.4

23.2    8.5   10.5
.008   .005
                    .001    .001   .004   .002

                    .014    .009   .010   .011


                    .002    .007   .005   .001

                    .072    .103   .004   .018



                    .003    .013   .005   .004

                    .003    .004   .005   .002


                           .004   .003   .003

                    .006    .019   .020   .023



                         >.001  >.001  >.001

                    ,006    .009   .001


                    ,010    .004   .009   .006

                    ,094    .063   .035   .009
-Time weighted means.
                                     35

-------
Kluver North supports the highest numbers and biomass of plant feeders and the
lowest number  and  biomass of  scavengers.   The switch  in  the relationship of
numbers to biomass for aboveground plant feeders, i.e. ,  lower biomass  for higher
numbers in 1975,  appears to reflect a change in the dominant  species types, as
mentioned earlier.   We  hope  a  more in-depth analysis  of  the data will answer
this question.

     Since  the  data included  here have not been  statistically analyzed, the
homogeneity of the  four field plots is  not clearly defined.  This is due in part
to the high variability of the data,  a  common  characteristic of this kind of
ecological  data.   When  the statistical  analyses are complete, a comparison of
year and date information will  provide a more detailed characterization of the
field sites as  examples  of a northern mixed-grass prairie.

                                 REFERENCES

Bartos, D.  L.  and P. L.  Sims.  1974.  Root dynamics of a shortgrass ecosystem.
     J. Range Manage.  27:33-36.

Dodd, J.  L. , J. K.  Lewis, H.  L. Hutcheson, and C. L. Hanson.  1974.  Abiotic and
     herbage dynamics studies at Cottonwood 1971.   US/IBP Grassland Biome Tech.
     Rep. No. 250.  Colorado State University, Fort Collins.  195 pp.

Leetham, J.  W.  1975.  A summary of field collecting and laboratory processing
     equipment and  procedures  for  sampling  arthropods at  Pawnee Site.  US/IBP
     Grassland Biome Tech. Rep.  No.  284.  Colorado State Univ., Fort Collins.
     49 p.
                                      36

-------
                                   APPENDIX

                              SOIL DESCRIPTIONS-7

     Soil surveys were conducted at the Kluver East, Kluver West,  Kluver North,
and Hay Coulee sites.  The morphological properties were examined to a depth of
152 cm in each of two soil excavation pits  at each site.  Samples were taken from
all horizons  in the profiles. Those samples are now being analyzed for physical
and chemical properties.

     Preliminary analyses  based on morphological properties indicate that the
soils at  the  Kluver  East  site belong  to Kobar and  Lonna  series.  At the Kluver
West and  Kluver North sites  the Yamac series predominates, but much  of the soil
at  the  Kluver West  site  is  very stoney and cobbly  causing description  to be
difficult.   The  Hay  Coulee  site  has  soils of  the Yamac  and  Lonna  series.
Representative profile descriptions from these series are presented  in profiles
1-3.

     Complete  analyses  and  interpretation  of these  soils  await completion of
physical  and  chemical  analyses.   Physical properties  being  determined  are
texture, bulk density,  and water-holding capacity.  Chemical  analyses will yield
information  for  organic  matter (carbon)  pH, total  N,  total  P,  inorganic P,
organic  S,  total  S,  inorganic  S,  cation exchange capacity,  exchangeable bases
(Ca, K,  Mg,  and Na),  and lime.
-Soils were described by R. G. Woodmansee with the  assistance of Lewis
  Daniels,  Soil  Survey  Party Chief, USDA,  SCS,  Forsythe, Montana.
                                      37

-------
PROFILE 1.                       Kobar Clay Loam

                            Kluver East -  North Pit
            Depth
Horizon     (cm)                             Description

  All        0-4          Light brown gray (10  YR  6/2 dry) to dark  brown  (10
                         YR 4/3 moist);  loam;  moderate  fine platy;  hard when
                         dry,   friable   when   moist,   slightly   sticky  and
                         slightly plastic when wet;  no  lime present;  abrupt,
                         smooth boundary; many fine roots.

  A12         4-12        Light alive brown  (2.4 YR 5/2 dry) to dark gray
                         brown  (10  YR  4/2  moist);  clay  loam;  weak  medium
                         platy;  hard  when  dry,  friable  when moist,  sticky
                         and  plastic  when   wet;   no  lime  present;   clear,
                         smooth boundary; many fine roots.

  B2        12-18        Pale brown (10  YR  6/3  dry)  to  dark brown (10 YR 4/3
                         moist);  sandy  clay   loam;  moderate  medium  prisms
                         breaking to  moderate  medium blocks; hard  when dry;
                         friable when moist, sticky  and plastic  when  wet; no
                         lime  present;  clear,  smooth  boundary;  many  fine
                         roots.

  B3ca      18-36        Very  pale  brown (10 YR  7/3 dry) to dark  brown  (10
                         YR 5/4  moist);  sandy  clay  loam;  weak coarse prisms
                         breaking to  moderate  medium and fine  blocks;  hard
                         when  dry,  friable when  moist,  sticky   and  plastic
                         when  wet;  lime present;  gradual,  smooth  boundary;
                         roots common.

  Clca      36-120       Very  pale  brown  (10  YR 7/3 dry)  to yellowish brown
                         (10  YR  5/4 moist); sandy  clay  loam; massive; hard
                         when  dry,  friable when  moist,  sticky   and  plastic
                         when  wet;   lime  present; gradual  smooth  boundary;
                         roots common  to 45 cm and  few roots to 120 cm.

  C2ca     120-152       Very  pale   yellow   (10  YR   7/3   dry)  to  yellowish
                         brown  (10  YR 5/4  moist);  clay  loam; massive; hard
                         when  dry,  friable when  moist,  sticky   and  plastic
                         when wet; lime present; few roots present.
                                     38

-------
PROFILE 2.                         Lonna  Loam

                            Hay Coulee -  North  Pit
            Depth
Horizon     (cm)                              Description


  Al         0-12          Light  brown  gray  (2.5  YR 6/2  dry)  to  dark brown
                          crushed   (10  YR  4/3  moist)  and  dark  gray  brown
                          coated  (10 YR 4/3 moist);  loam;  weak medium platy;
                          slightly  hard when dry,  friable when moist, slightly
                          sticky,  slightly plastic  when wet;  no  lime;  clear
                          smooth boundary; many fine roots.

  B2ca     12-22          Pale  brown   (10  YR 6/3  dry)  to yellowish  brown
                          crushed  (10  YR  5/4  moist) and dark brown coated (10
                          YR  4/3  moist); silty  clay  loam;  weak medium prisms
                          breaking  to moderate  medium  and  fine  blocks;  hard
                          when  dry,  friable   when  moist,  sticky  and plastic
                          when  wet;  lime present;  gradual  smooth boundary;
                          many fine roots.

  Cl        22-60          Very  pale brown (10 YR  7/3 dry)  to yellowish brown
    ca                    (10 YR  5/4 dry);  silty  clay  loam; weak,  very coarse
                          platy  breaking  to moderate fine  blocks (plates not
                          of  pedogenic  origin);  hard  when  dry,  firm  when
                          moist,  sticky and  plastic  when wet;  lime  present;
                          gradual  smooth  boundary; many fine  roots to  30 cm,
                          common fine roots 30-60 cm.

  C2        60-98          Very  pale brown (10 YR  7/3 dry)  to yellowish brown
    ca                    (10  YR  5/4  moist);  silty clay  loam;  massive;  hard
                          when  dry,  friable   when  moist,  sticky  and plastic
                          when  wet;  lime present;  gradual  smooth boundary;
                          common  fine  roots to  80 cm,  few fine roots  to  98
                          cm.

  C3        98-152         Very  pale yellow  (10  YR 7/3  dry) to  light  olive
    ca                    brown  (2.5 YR 5/4  moist);  silty;  massive;  slightly
                          hard  when dry,  friable  when moist;  lime  present;
                          few roots present.
                                      39

-------
PROFILE 3.                       Yamac Silt Loam

                            Kluver West -  West Pit
            Depth
Horizon     (cm)                             Description

  Al         0-10         Brown  (10  YR  5/3  dry)  to dark  gray brown  (10  YR
                         4/2  moist);   fine  sandy  loam;  soft medium  platy;
                         soft when  dry,  very friable when  moist,  nonsticky,
                         slightly plastic  when  wet;  no lime; clear,  smooth
                         boundary; many fine roots.

  B2       10-31         Yellowish   brown  (10 YR  5/4 dry)  to brown to  dark
                         brown  crushed (10  YR  4/3  moist),  dry gray  brown
                         coated (10 YR  4/2  moist);  sandy clay loam; moderate
                         medium  prisms  breaking  to moderate  medium blocks;
                         hard  when   dry,   friable  when  moist,   sticky  and
                         plastic when  wet;  no lime;  clear,  smooth  boundary;
                         fine roots  common.

  B3       31-40         Pale brown (10 YR  6/3  dry) to  dark  gray  brown (2.5
                         YR  4/2  moist);  sandy clay  loam; weak coarse  prisms
                         breaking  to   weak  coarse  and  medium  blocks;  hard
                         dry,  friable  when  moist,   sticky  and  plastic  when
                         wet;  lime  present but  not visible;  gradual  smooth
                         boundary;  roots common.

  Clca     40-118        Very  pale  brown  (10 YR  7/3  dry),   too cobbly and
                         mixed  to   determine  moist  color;  cobbly   (upper  7
                         cm);  sandy  clay  loam;   massive;   hard  when  dry,
                         friable when  moist,  sticky  and plastic when  moist;
                         small  threads  and masses of lime;  gradual  and wavy
                         boundary;   roots  common 40-50  cm,  few  roots  50-118
                         cm.

  C2ca    118-156        Light gray (2.5  YR  7/2  dry), too variable for moist
                         color;  cobbly sandy loam;  massive;   too cobbly for
                         consistence;  lime coatings  on cobbles; few roots.
Lower  horizons  of   this  profile  contain  unweathered  fragments  similar to
nearby  uplands.   Most  Yamac  soils  of EPA  study sites  do  not contain these
cobble  layers.   The east  end  of  Kluver  West  contains  so much  cobble and
gravel that the pit was not described.
                                      40

-------
                                   SECTION 2

          PLANT COMMUNITY STUDIES IN THE VICINITY OF COLSTRIP, MONTANA

                                      by

                       J. E. Taylor and W.  C.  Leininger


                                 INTRODUCTION

     The  plant community monitoring aspects of  the  Colstrip Coal-fired Power
Plant  Project  are generally divided into studies of more-or-less quantitative
plant  population  and  community  attributes  (diversity,  phenology;  canopy
coverage,  productivity)  and photographic  surveillance,  both  from  ground and
aerial vantage points.  This section will  deal  with diversity,  canopy coverage,
and phenology studies.

     The  procedures  and  data  reported here are interrelated with  concurrent
studies of bird and mammal populations (including diversity), plant pathology,
and  lichen ecology.   In  total,  all  of these  studies  are  contributing  to the
development  of a  generalized,  comprehensive approach to air pollution effects
monitoring in biological  systems.

     The objectives of this  aspect of the project  are to:

     1.   Document pre-treatment native plant  communities in areas likely to be
          affected by  the power plants under  investigation and on areas to be
          stressed artifically with pollutants.

     2.   Develop measurement techniques and monitor  changes in plant community
          structure,  diversity,  phenology,   and  speciation  following  air
          pol1ution stress.

     3.   Develop detailed vegetation maps of the study areas.

     4.   Provide  data   for  simulation  models   to  predict  bioenvironmental
          changes resulting from fossil fuel power generation in other areas.

     The  work  was initiated  on July  15,  1974.   Research  from previous field
seasons has been reported by Taylor et a_[.  (1975 and 1976).   The present report
is based on the data collected during the 1975 field season.

     This discussion  is  divided into  two principal parts:  diversity studies,
using standard indices derived from information theory, and phenology studies,
using an index  system of the authors' design.

                                      41

-------
     The underlying hypothesis of  this  work is that,  as  a plant community is
stressed with  the  application  of  any  external  perturbation,  including air
pollution,   and  physiological   changes  in  members  of   the  community  which
result in changed  rates  of vital  functions should be shown by changed species
composition and/or  rates of seasonal  growth processes.   If  species composi-
tion  is  affected,  with  either  loss of  sensitive species, or  re-ordering of
species  dominance  in  the  community,  this  should be  reflected  in the species
diversity observed.  If  rates of  normal  growth and reproductive functions are
influenced, the phenological profiles  of component species should change.  In
the case of the  Zonal  Air Pollution System (ZAPS) plots,  where sulfur dioxide
is applied at controlled rates  to native vegetation, both of these indications
have been observed  (see Section 11).

     Both  responses  to  pollution  could  be  studied  with  biomass/productivity
data,  such as  those  collected  by Dodd and associates (see Sections 1 and 10).
However, the  authors  are  interested  in developing  a non-destructive method-
ology  which might  be  used where biological material was in limited supply, or
where  complementary  observations  could  be  made,  serving to  fortify biomass
data by mutual  corroboration.

                          SPECIES DIVERSITY  STUDIES

      In  1975,  3 sampling  times  were used,  representing  (a)  "peak of green"
(June  10-11),  (b)  summer growth  and maturity and  warm season "peak of green"
(July  10-12),  and (c) late season maturity (September 16-17).

      Plots  within study areas  were  located  by  placing  a cord  with meter-
spaced knots  in a  random meandering pattern through  the  sample area.  The 2
x  5 dm  plot  frames  were  placed  at each  knot  and  canopy  coverage estimated
using  the procedure  of  Daubenmire  (1959).   At  the  same time,  species  were
counted,  the  numerical  data to  go  into diversity indices.   Previous experi-
mentation  had indicated that  2  sets  of  20  frames  per  plot  constituted a
statistically  adequate  sample  (i.e.,  additional  observations  did  not  con-
tribute to increased sensitivity of diversity indices).

     The  diversity index  used  was  the  Shannon-Weaver function  (Shannon and
Weaver 1949):
                                                                    *
                  S
          H'  = -  I      Ni_ ln Nj_        Where H1 = Index of diversity
                 i=l     N     N               S  = number of species
                                              Ni = number of ith species
                                              N  = total number of all species

     The index  was calculated  in  two different  ways.  In using number data
(plant  density),  the  index was  calculated  for  each  of  the  50  individual
frames, and the  mean  of these  values was  tabulated and graphed.  With canopy
coverage data,  the  index was calculated  as an  accumulated figure  over the
entire set  of 40 frames.
                                      42

-------
     The  first approach  allows statistical  comparisons among data  sets;  the
second  was  used  to  evaluate  the  feasibility of  applying  indices  to  canopy
coverage data.

     Further,  the  number-based indices  were  calculated with a modification of
the  Shannon-Weaver function  which  since has  been  discontinued.   Thus,  direct
comparisons   between   indices  are   not   legitimate,  but  relative  tendencies
should be meaningful.

         SPECIES DIVERSITY (CANOPY COVER) FOR THE COLSTRIP STUDY SITES

     Diversity indices were  calculated  using coverage  data  from  all  sites  and
dates of sampling.

     Canopy   cover  diversity   for  the  Colstrip  locations  are  presented  in
Figures  2.1,  2.2,  and 2.3,  showing  data  for 1974, 1975,  and  1976,  respec-
tively.   There are definite  diversity  differences among sites,   and they  are
quite  consistent  over years.   Magnitudes  change,  but relative   diversity  is
reasonably constant.
             3.0
             2.5 -
             2.0 -
              1.5
JULY
AUGUST
SEPTEMBER
                      HAY   KLUVER  KLUVER KLUVER  KNOLL  KNOLL  KNOLL
                     COULEE  EAST   WEST   NORTH    ABC
Figure 2.1.     Shannon-Weaver function (HP) for Colstrip Study sites, 1974.
                                      43

-------
             3.0
             2.5 -
          Q.
          I
2.0
              1.5
                     	1	1	1	1	1	
                       HAY   KLUVER KLUVER KLUVER  KNOLL
                     COULEE  EAST   WEST   NORTH     A
                                                   JUNE
                                                   JULY
                                                   AUGUST
                                           	1	1	
                                            KNOLL   KNOLL
Figure 2.2.     Shannon-Weaver function (HP)  for Colstrip Study sites, 1975.
              3.0 -i
              2.5
              2.0 -
              1.5
                     	1	1	•	—I	1	1	1	1	
                       HAY   KLUVER KLUVER KLUVER  KNOLL  KNOLL  KNOLL
                     COULEE  EAST   WEST   NORTH     ABC
Figure 2.3.     Shannon-Weaver function (HP) for Colstrip  Study site, August,
                1976.
                                         44

-------
     These  site differences  can  be related  to  plant species  dominance and
density.   For  example,  Kluver West has  a low  diversity  due to  the strong
dominance  of needle-and-thread grass (Stipa comata) and  the  paucity of other
species.   Soil  and microclimate  variation, as well  as  grazing  history,  are
the  likely causes  of this  situation.   Thus,  comparisons  among sites probably
are of 1imited  value.

     The  McRae Knolls,  which  are adjacent  sites of  similar  grazing history
and  climate  show   diversity   differences   because  of  edaphic  factors  among
sites.   This   is  reflected primarily  in  species  composition.   When analyzed
for  range condition,  Knolls  A and  C  were in good condition  and  Knoll  B was
excellent.  This  estimate was  based on percent  of climax  vegetation for the
sites.   The  fact that  the diversity indices  show  a  similar  pattern suggests
a  relationship  between  range  condition  and  diversity,  which  is  consistent
with the results of  many  other studies.

     Seasonal  differences  in  diversity  follow   previously   reported  trends
(Taylor  et  al. ,  1976).  As  the season  progresses,  many  of  the  ephemeral
forbs  and  cool-season  grasses disappear  or diminish in  their  contributions
to total canopy coverage.

                               PHENOLOGY STUDIES

     Phenology  data  has been  recorded since  1974 to  establish a baseline
data set.

     The  authors  hypothesize  that phenology will  be a less sensitive approach
than  diversity in  detecting pollution  effects.   This  is  primarily  because of
the  large sample  size   required and the need  to  accumulate observations over
a  number  of years  in order  to discriminate  between  "normal" and "abnormal"
growth  patterns.   Therefore,  the  authors  are  continuing  the  data collection,
and  in  fact  are  increasing  the frequency  of  sampling,  especially  in  the
spring and early summer.

PROCEDURES

     A  phenologic  scorecard designed by the authors was  used (Taylor et al_. ,
1975);  the categories  are shown  in Table 2.1.   This  classification has been
modified  since last season, with  the  insertion of the early  flowering stage.
With  this addition,  the system seems to  fit a number of vegetation types and
has  generated  considerable  interest  from various  fields of ecological  re-
search.   On  each visit  to an  experimental  site, a  reconnaissance  survey is
made.   Each  recognizable  plant   species   is  characterized  as  to  its  modal
phenologic stage.   The  scorecard approach attempts  to  quantify significant
growth stages without resorting to  counts or measurements.

     The data are stored for further comparison and future  analyses.
                                      45

-------
                         TABLE 2.1.  PHENOLOGY CODES.
          Code                               Stages
            1            Cotyledon (newly germinated)

            2            Seedling

            3            Basal  Rosette

            4            Early  greenup, veg. buds swelling

            5            Vegetative growth, twig elongation

            6            Boot stage,  flower buds appearing

            7            Shooting  seed  stalk,  floral  buds opening

            8            Early  flowering

            9            Flowering, anthesis

           10            Late flowering

           11            Fruit  formed

           12            Seed shatter,  dehiscence

           13            Vegetative maturity,  summer  dormancy,  leaf drop

           14            Fall greenup

           15            Winter dormancy

           16            Dead



DISCUSSION

     For purposes of  elucidation, the  index  values  (stages) for  Kluver North
are  displayed graphically  in   Figures  2.4  (grasses)  and  2.5 (forbs).   The
differences among species curves  for grasses  are  clearly related  to longevity
and typical  growth patterns (cool- vs. warm-season species).

     Forbs and shrubs  show  more interspecific variation in  phenology  than do
grasses, but  site differences  are inconsistent.   This  probably is  due to the
unusually  favorable   growing conditions  during  summer,  1975, following  the
later spring.   The  authors  recognize  that  since  the  stages  comprise an un-
sealed,  arbitrary series,  the  resulting  graphs should be viewed  only  as gen-
eralized profiles of plant growth and  not  subjected  to quantitative analysis.

                                      46

-------
                      DEAD


           WINTER DORMANCY


              FALL GREENUP


                   MATURITY


              SEED SHATTER


               FRUIT FORMED


             LATE  FLOWERING


                  FLOWERING

        FLOWER  BUDS  OPENING
        SHOOTING SEED STALK

       FLOWER BUDS APPEARING
                BOOT STAGE

          VEGETATIVE GROWTH


             EARLY GREENUP


             BASAL ROSETTE


                   SEEDLING


                 COTYLEDON
          Agropyron smithii
          Aristida longiseta
          Koe/eria  cristata
          Poo sandbergii
          Stipa coma fa
          ANNUAL  BROMES
                            /NEW SEEDLINGS, \
                            \ANNUAL BROMES/
                                       MAY
                                        14
JUNE
 12
JULY
 13
AUG. SEPT.
 30   10
NOV.
 I
Figure 2.4.  Phenological  profile of selected  species in Kluver  North
               exclosure,  1975.
                                             47

-------
                      DEAD


           WINTER DORMANCY


              FALL GREENUP


                   MATURITY


              SEED SHATTER


               FRUIT  FORMED


             LATE FLOWERING


                  FLOWERING

        FLOWER BUDS OPENING
        SHOOTING SEED  STALK

       FLOWER BUDS  APPEARING
                 BOOT STAGE

          VEGETATIVE  GROWTH


              EARLY GREENUP


              BASAL ROSETTE


                   SEEDLING


                  COTYLEDON
                                  	O
      	Artemisia  frigida
      	 Plantago pafagonica
       • — Sphaeralcea coccinea
      	 Taraxacum  officinale   /NFW GROWTH\
          Tragopogon  dubius
                              \   TRDU   /
                                       MAY
                                        14
JUNE
 12
JULY
 13
AUG. SEPT.
 30   10
NOV.
 I
Figure 2.5.      Phenological profile of  selected  species in  Kluver North
                   exclosure,  1975.
                                              48

-------
     The  development of  analytical  procedures  whereby  quantitative compari-
sons  can  be made  among  species,  locations,  seasons,   and  treatments  is a
continuing part of this  research.

PLANT COLLECTIONS

     An  ongoing  program  of  plant  collection  has  been  under  way  since  the
initiation of  the field phases of the  project.  Voucher  species are deposited
in  the  Herbarium  of  Montana State  University.   In addition, field specimens
are  maintained  for  reference  of crews and  others  for consistency of naming,
correctness  of  identification, elimination  of  duplicate  numbers in Unknowns,
etc.
     Small  field  specimens
dure of Burleson (1975).
are taken in  plastic  covers,  following the proce-
                                  REFERENCES

Burleson,  W.  H.   1975.   A  method  of  mounting plant  specimens  in the field.
     J. Range Manage.   28:240-241.

Daubenmire,  R.  R.  1959.   A  canopy-coverage  method of vegetational analysis.
     Northw. Sci.   33:43-64.
Shannon,  C.  and  W.   Weaver.   1949.   Mathematical
     Univ. Illinois Press, Urbana.   117 p.
                         theory  of  communication.
Taylor,  J.  E.,  W.  C.  Leininger, and
     studies  near  Colstrip,  Proc.
     Sci.  pp.  537-551.
         R.  J.  Fuchs.   1975.  Baseline vegetational
         Ft.  Union  Coal  Field  Symp.,  Mont.  Acad.
Taylor,  J.   E.,  W.  C.  Leininger,  and  R.
     community  changes  due to  emissions
     eastern  Montana.   Section  II  of the
     fired  power  plant,  Second Interim
     Series EPA-600/3-76-013.  pp.  14-40.
               J.  Fuchs.   1976.   Monitoring plant
               from  fossil  fuel  power  plants  in
               bioenvironmental  impact of a coal-
               Report.   USEPA  Ecological  Research
                                      49

-------
                                   SECTION  3

                     SOIL AND EPIPHYTIC LICHEN COMMUNITIES
                         OF THE COLSTRIP, MONTANA AREA

                                     by

                                S. Eversman
                                INTRODUCTION

     Numerous field studies  have  used  lichen distribution patterns to locate
areas of  air  pollution.  The  studies  show that fewer lichen species occur in
polluted  areas  (LeBlanc  and  Rao,  1975),  and that lichen species transplanted
into  polluted areas  lose  vigor  and/or   die  (LeBlanc  and Rao,  1975).   The
Colstrip  project  in southeast Montana provides the rare  opportunity to map
lichen   communities  and  obtain   baseline  physiological   and   anatomical
information  before the  two  350  megawatt  coal-burning  power plants  begin
operation.  This project  will  be able to establish the time  progression of low-
level pollution effects on a previously pristine area.

     The specific objectives  are to:

               1)  Determine  baseline  lichen  community  information  for the
               grassland and  ponderosa pine vegetation types.

               2)   Establish   the  baseline   anatomical   and  physiological
               condition of two native lichen species.

               3) Continually monitor these characteristics to detect changes
               caused by  increasing air pollution.

               4)  Compare  lichen monitoring  results with  measurements taken
               simultaneously  on  associated  vascular  vegetation by  other
               investigators.
                           LOCATIONS AND METHODS

     The  lichen  study sites  in  the Colstrip  area are  shown  in Figure 3.1.
Table 3.1  provides this information for each site.

     Samples of all  observed  lichen and moss  species  on soil  and rocks have
been collected  from sites G1-G7  (grassland sites 8-97  km  from the Colstrip
power plants).   Specimens were taken outside the exclosure fences at the common
CFPP project research  sites  at  Hay Coulee, Kluver North, Kluver West, Kluver
East, (Gl  + G3-G5 on Figure 3.1).

                                      50

-------
                            Forsyth
                             Rosebud County
    Treasure
    County
    Sarpy
    Creek
     PI
P2
 P3       /^P7
9      G4/
          P^  ^G3'
             ^sf \~G5
                        r
                 >P8
     Big Horn County
                          Lame Deer
                             Northern Cheyenne
                                Reservation
                       P9
                                                 Ashland*
                                                                     Custer County
                                                              Pll
                                                            P12|
                                             P16
                               I—JP15
                                  Birneyj
                                                      I
                                                      I
                                                      I
                                                      I
                                                      I
                                                     "h
                                                     	L
                                >P13
                                —-^
                               »P14
                                                                      Custer
                                                                    National Forest
                                                   TC
                                    G7
                                              1 cm-8.0km
Figure  3.1.      Location  of Field  Study Sites  in the Colstrip area.  P1-P16 are
                 ponderosa pine sites on ridges; G1-G7 are  grassland sites.

                                         51

-------
TABLE 3.1.   DESCRIPTION OF LICHEN COLLECTION SITES, COLSTRIP AREA  (Name; air
             distance and direction from Col strip; exposure for ponderosa
             pine sites; location).


                               Grassland Sites

Gl  Hay Coulee; 11.6 km SE; TIN, R42E, Sec. 28
G2  McRae Knolls; 13.8 km SE
G3  Cow Creek (Kluver West); 11.6 ESE; TIN, R42E, Sec. 2
G4  North Pasture (Kluver North); 14.7 km E; TIN, R43E, Sec. 6
G5  School  Section (Kluver East); 18.3 km ESE; TIN, R43E, Sec. 15
G6  Harvey Sage Site, BLM land along Cow Creek Road; 8 km SE
G7  Pasture adjacent Cow Creek Road-Otter Creek Road intersection; 70  km SSE

                        Ponderosa Pine Sites (Ridges)

PI   Sarpy Creek, Charles May Ranch; 44 km W; ENE exp; T2N, R37E,  Sec. 36
P2   Castle Rock; 16 km W; E, NE exp; TIN, R41E, Sec. 36
P3   Kluver NE1; 9.6 km NE; SW exp; T2N, R42E, Sec. 16
P4   Kluver El; 10 km E; W exp; T2N, R42E, Sec.  29
P5   D.McRae hill; 9.6 km SSE; NNW exp; TIN, R42E, Sec. 36
P6   Ridge direction south of Kluver West grassland exclosure (G3);  11.6 km S;
     NW exp; TIN, R42E, Sec. 2
P7   Diamond Ranch Buttes; 28 km ENE; SW exp; T2N, R43E, Sec. 22
P8   Morning Star View picnic ground, North Cheyenne Indian Reservation; 26 km
     SSE; N exp; T2S, R41E, Sec. 12
P9   Stop along Highway 212 between Lame Deer and Ashland; 35 km SE; NE exp.
P10  East Otter Creek Divide Road; 54 km SE; NW exp; T2S, R46E, Sec. 24
Pll  SEAM Site 1; 52 km SE; W exp; T2S, R46E, Sec. 22
P12  SEAM Site 2; 50 km SE; NW exp; T2S, R46E, Sec. 22
P13  Home Creek Butte; 60 km SE; NW exp; T2S, R46E, Sec. 4
P14  Ridge near Three Mile Creek, Lemonade Spring Road; 90 km SE;  NW exp; T2S,
     R46E,  Sec. 4
P15  Fort Howes Ranger Station; 65 km SSE; N exp; T6S, R45E, Sec.  19
P16  Poker Jim Butte; 64 km S; NW exp; T6S, R44E, Sec. 17

TC   Taylor Creek Grassland Fumigation Site; 18  km ESE of Fort Howes Ranger
     Station, Powder River Co.; T7S, R47E, Sec.  9
                                      52

-------
     Ten  fenceposts,   serving  as  permanent observation points,  were placed
inside each of the exclosures where grass  cover was the lowest and  lichen cover
was  the  greatest.  The  point-drop method  (Hanson,  1950)  was  used to record
ground cover  (bare  soil, grass, litter,  moss,  forb,  lichen species) at each
fencepost.  The center of the point-drop frame was placed on the north side of
each fencepost and 200 points per exclosure were  recorded.

     Sixteen ponderosa pine study  sites (P1-P16) are located on ridges 8-97 km
from Colstrip; 13 of  them are exposed ridges that face toward the power plant
stacks.  Ten sites (P1-P5,  P8, P11-P14, P16) have sulfation  plates  placed there
by other  researchers  (Gordon, 1975; Gordon et al_. , 1976).  Site PI0 was used as
a control and source of transplant material; P9 was used as  a transplant source
in  1975  only.   Samples  of  lichens and mosses  from  all  substrates have been
collected from all sites.

     Ten  trees  on each site (excluding P9) are permanently marked for annual
determination  of  percentage of  cover and  frequency  of each  lichen species
present.   Cover classes 1 - 5 (1 = 1 - 5% cover; 2 = 6-  25%; 3 = 26 - 50%; 4 =
51   75%; 5 = 76  - 100%) are recorded for each species encountered on each of
the  four  compass  sides of each  tree  (north = northeast to northwest; east =
northeast  to  southeast; etc.).   Midpoints  of each cover  class are used for
calculations (Tables 3.3 and 3.4).

     Transplants  to  some sites  were made  by wiring  ponderosa pine branches
covered with Usnea hirta onto selected branches of 10 pine  trees facing toward
the power plant.   Site P10 was the source of  all transplant material  in  1975 and
1976.   Transplants  were made  to  sites  P1-P6  to  augment  local  U.  hirta
populations  available  for  collection  and  to  place  sampling specimens  in
locations  more  exposed than those  of  existing  populations.  Transplants were
made  to  sites  P8,   P15,  and  P16 to  provide  comparisons  with   the native
populations  to detect  effects  of  transplanting.   Site  P7  had no epiphytic
1ichens at al 1.

     Parmelia  chlorochroa  or Usnea  hirta  samples  are  collected periodically
from test sites~(except  P9) for manometric determination  of respiration rates,
analysis  of total sulfur (as  S04) content by the Montana State University Soils
Testing Laboratory (Eversman, 1975), and for microscopic examination of algal
color and plasmolysis.

     Plasmolysis  percentage  is  determined by preparing three slides of three
different plants  from each site,  counting  100 algal cells on each slide, and
recording the number of plasmolyzed cells out of  each  100 counted.

                                   RESULTS

     Table  3.2  lists   the  primary lichen  species  on  seven  grassland sites.
Cladonia  spp.  squamules  are  the  dominant  lichen types on the  soil  at all
exclosure  sites;  species  identification  has  not  been   made   because  the
reproductive  structures are  very  rarely  found.   Therefore,  the  other most
abundant species, Parmelia chlorochroa, is used for analytical work.
                                      53

-------
on
     TABLE 3.2.   GROUND COVER OF THE PRIMARY LICHEN SPECIES ON 7 GRASSLAND SITES.   PERCENTAGES DETERMINED BY
                 THE POINT-DROP METHOD.   + =  present,  less than 0.5%  x = present in  undetermined quantities.

Lichen
Cladonia spp.
Parmelia chlorochroa
Collema tenax
Fulgensia fulgens
Buellia epigaea
Dermatocarpon lachneum
Squamarina lentigera
Lecidea decipiens
Acarospora schleicheri
Tonim'a coeruleonigricans
Agrestia hispida

Hay
Coulee
Gl
23.3%
16.7
5.0
+
+
+
+
+
+
+
+
McRae
Knolls
G2
X
X
X
X
X
X
Kluver
West
G3
27.0
16.5
2.5
0.5
+
+
Kluver
North
G4
24.0
9.5
2.0
1.0
+
+
Kluver
East
G5
13.6
10.0
2.7
+
+
+
+
Harvey
G6
X
X
X
X
X
X
X
X
X
Pasture
G7
X
X
X
X

-------
TABLE 3.3.  PONDEROSA PINE SITES:  PERCENTAGE OF TOTAL LICHEN COVER,
            FREQUENCY (Number of Trees Having Lichens), NUMBER OF
            EPIPHYTIC LICHEN SPECIES AT THAT SITE.

Site
P10, East Otter Creek
PI 2, SEAM Site 2
P14, 3-Mile
Pll, SEAM Site 1
P16, Poker Jim
PI 3, Home Creek Butte
P15, Fort Howes
P5, D. McRae
P3, Kluver NE1
P2, Castle Rock
PI , Sarpy Creek
P4, Kluver El
P7, Diamond Buttes
% cover
39.75
21.00
15.25
13.63
12.88
11.75
10.81
9.13
6.88
6.38
1.75
1.13
0
Frequency
10
10
10
10
10
10
10
8
8
8
4
2
0
Number of
species
1
5
4
6
6
7
7
8
7
9
3
4
0
                                  55

-------
     TABLE 3.4.   LICHEN COVER OF PONDEROSA PINE TREE TRUNKS, COLSTRIP AREA.  (Percentage cover/frequency, 10
                 trees sampled on each site.  Site P7 has 0 for all species.)
en
CTl

Lichen
Usnea
hi rta


Parmel ia
sulcata


Parmel i ops is
ambigua


Alectoria
f uscescens


Hypogymnia
physodes


Parmel ia
infumata



N
S
E
W
N
S
E
W
N
S
E
W
N
S
E
W
N
S
E
W
N
S
E
W
P10
12.0/10
11.1/8
7.4/9
9.3/10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
P12
5.9/10
2.8/7
6.6/9
1.8/7
1.9/5
0.1/1
0.4/3
0
0
0
0
0
0.3/2
0
0.1/1
0
0.9/2
0
0.1/1
0
0
0
0
0
P14
5.9/10
0.8/6
3.3/7
0.8/6
1.8/4
0
0
0.3/2
0.1/1
0
0
0
2.1/7
0
0.1/1
0.1/1
0
0
0
0
0
0
0
0
Pll
3.1/10
3.1/7
4.6/9
0.9/5
0.3/2
0
0.3/2
0
0.3/2
0
0.1/1
0.1/1
0.1/1
0.3/1
0.3/2
0
0
0
0
0
0
0
0
0
P16
3.9/10
0.3/2
1.1/6
1.1/7
2.3/4
0.8/1
0
0
0.8/1
0
0
0
0.4/3
0
0
0.4/3
0
0
0
0.1/1
0.1/1
0
0.8/1
0
P13
2.1/10
0.8/3
3.0/9
0.3/2
0.3/2
0
0.3/2
0
1.0/3
0
1.0/1
0
0.3/2
0
0.1/1
0
0.9/3
0
0
0.1/1
0
0
0
0
P15
3.1/10
0
0.1/1
0.5/4
0.3/2
0
0
0.3/2
3.5/8
0
0.4/3
0.4/3
0.6/6
0
0
0
0.3/2
0
0
0.1/1
0.3/2
0
0
0.1/1
P5
0.9/7
0
0.3/1
0
0.3/2
0
0
0
3.5/8
0
1.6/3
0.1/1
0.6/5
0
0.3/2
0
0.9/2
0
0.1/1
0
0.4/3
0
0.1/1
0
P3
1.6/9
0.3/1
0
0.4/2
0.4/3
0
0
0
3.6/5
0
0.3/1
0.1/1
0.5/4
0
0
0
1.4/6
0
0
0.1/1
0.3/1
0
0.1/1
0
P2
1.0/8
0.1/1
0.1/1
0.3/2
0.3/2
0
0
0
1.4/5
0
0
0
0.1/1
0
0
0
0.5/4
0
0
0.1/1
1.0/4
0
0.1/1
0
PI
0.4/3
0
0
0
0
0
0
0
0.9/2
0
0.1/1
0.1/1
0.1/1
0
0
0
0
0
0
0
0
0
0
0
P4
0.8/1
0
0.1/1
0
0
0
0
0
0.1/1
0
0
0
0
0
0
0
0.1/1
0
0
0
0
0
0
0
                                                                                                   (continued)

-------
TABLE 3.4.   (continued)
Lichen
Lecanora
subfusca


Cetraria
pinastri


Parmel ia
ulophyl lodes


Letharia
vulpina


Total epiphyte

P10
N
S
E
W
N
S
E
W
N
S
E
W
N
S
E
W
cover:
39
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

.75
P12
0
0
0
0
0
0
0
0
0.3/2
0
0
0
0
0
0
0

21.00
P14
0
0
0
0
0
0
0
0
0
0.1/1
0
0
0
0
0
0

15.25
Pll
0
0
0.1/1
0
0
0
0.1/1
0
0
0.1/1
0
0
0
0
0
0

13.63
P16
0
0
0
0
0
0
0
0
0.9/2
0.1/1
0
0
0
0
0
0

12.88
P13
0
0
0
0
0.4/1
0
0
0
1.1/3
0
0
0
0
0
0
0

11.75
P15
0.1/1
0
0.4/3
0.4/3
0
0
0
0
0.1/1
0
0
0
0
0
0
0

10.81
P5
0.9/1
0
0
0
0.4/1
0
0
0
0
0
0
0
0
0
0
0

9.13
P3
0
0
0
0
0
0
0
0
0
0
0.1/1
0
0.1/1
0
0.1/1
0

6.88
P2
2.4/4
0
0
0
0.1/1
0
0
0
0
0
0
0
0.1/1
0
0
0

6.38
PI
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

1.75
P4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

1.35%

-------
     Tables 3.3 and 3.4 give the  epiphytic lichen community data collected and
compiled through  1976.  Only  ten  species  are present consistently  enough to
appear in the  sampling.  The  species  list includes names of  other  collected
epiphytic lichens.

     Respiration rates for  Usnea  hirta samples for  1975  and  1976 are in Table
3.5; those for  Parmelia chlorochroa  are in Table  3.6.   Results  from sulfate
sulfur analyses for 1974 and 1975 are  in Table 3.7; analyses  for 1976 samples
have not yet been completed.

     Determination  of  some  baseline  algal  plasmolysis  counts  have  been
completed  from  the   P10  site.    The  plasmolysis  values (means  ±  standard
deviation) are:   1975, September,  17 ±2;  1976, March, 3 ±3; July,  11 ±2; July, 9
±3.  Counts are continuing for all  other collection.

                                DISCUSSION

     After  the   summer  of  1975,   it  appeared   that   recording  only  the
presence/absence  data for  epiphytic   lichen  communities was not  sensitive
enough to detect and monitor the  low-level chronic  S02 stress, despite reports
using these criteria  from  other  regions (LeBlanc  and Rao, 1975).  However, in
this area any  pollution stress will be  newly-introduced and epiphytic growth is
slight  compared  to  that   in  the  more  humid  climates of  other  studies.
Therefore, the more detailed data were acquired  during 1976.   The marked trees
allow for  annual  repetition of  measurements on the sites.  Three more sites,
closer to the  power plant,  are being added  for analysis beginning in 1977.

     Determination of respiration rates is a method sensitive  enough to detect
adverse  pollution  effects   in U.  hirta (Eversman,  1976).   However,  bacteria
populations increased and/or  changed  in  S02-stressed   Parmelia  chlorochroa
plant bodies.   This probably  caused  the respiration rate of the lichen to be
overestimated.  Photosynthetic rate would be a more specific indicator of the
physiological  condition of  the  lichen plant.   This information combined with
the  plasmolysis  counts gives adequate  information on the vitality  of these
lichens.  Eventually,  the  effects observed  at  the organismic  level  will be
reflected  at  the community level.  Photographs  of  many  communities  are also
being taken regularly.

     Ponderosa pine trees  in the  moister habitats,  with  shrubs Symphoricarpos
albus, Rhus trilobata, and  Ribes spp.,  generally have the  heaviest  lichen cover
on the trunks  and branches  (P10,  Pll,  P12).  Where  the understory  is dominated
by such species as Yucca glauca and Agropyron spicatum (sites PI, P3, P4, P6,
P7), the lichen cover is sparse and restricted to  the lower  30 cm of trunks on
the north and  east sides, if it appears at  all.  None of the sites  in the Custer
National Forest required transplants for monitoring purposes.

     Mosses,   the  only bryophytes  encountered,  have  been collected  from the
grasslands and  from the  soil  and litter  under  the Ponderosa pines.  Verified
identifications of these species, and of mosses  and lichens collected from the
rocks, have not yet been completed.
                                     58

-------
TABLE 3.5.  RESPIRATION RATES OF Usnea Hirta SAMPLES, COLSTRIP AREA, 1975-76.
(Respiration rate is given as micro! iters of
weight/hour, manometrically determined.)

PI,

P2,

P3,

P4,

P5,
P6,

P7,
P8,


P9,

P10





Pll

P12

P13

P14



P15




P16




Site
Sarpy Creek

Castle Rock

Kluver NE1

Kluver El

D McRae
Kluver West transplants
(from P9, 1974)
Diamond Buttes
Morning Star View, transplants
natives
transplants
Road

, East Otter Creek





, SEAM 1

, SEAM 2

, Home Creek Butte

, 3-Mile, top of ridge
bottom of ridge
top of ridge
bottom of ridge
, Fort Howes (transpl from P9)
natives
ii
M
transplants
Poker Jim Butte transplants
' ii
natives
ii
n
Col lection
Date
9-25-75
9-14-76
9-25-75
9-14-76
9-25-75
9-15-76
9-14-75
9-15-76
9-15-75
9-15-76
7-16-76
6-24-76
9-25-75
n
11-20-76
5-01-76
8-11-75
9-25-75
3-23-76
6-23-76
7-15-76
8-09-76
9-15-76
7-16-75
7-15-76
7-16-17
7-15-76
7-16-75
7-15-76
7-16-75
n
8-09-76
II
5-01-75
II
5-13-76
9-15-76
n
9-25-75
9-15-76
7-17-75
7-16-76
9-15-76
X
640*
658
640*
499
640*
633
640*
573
640*
827
669
744*
640*
828
661
563
508
640
709
744
463
668
633
545
453
548
498
462
486
539
528
667
547
829
510
576
470
597
640*
669
630
610
654
02 consumed/g dry
1 s.d.
40
129
40
86
40
98
40
53
40
38
50
45
40
135
60
109
51
40
63
45
57
41
53
35
19
27
48
55
15
64
38
76
40
58
213
113
50
43
40
74
77
31
52
* respiration  rate  of  duplicate  material when  transplanted from P10 site.
                                      59

-------
TABLE 3.6.   RESPIRATION RATES OF Parmelia chlorochroa SAMPLES, COLSTRIP AREA,
1974-6. (Respiration rate is in micro! iters
wt/hour, manometrically determined.)
02 consumed/g dry

Col lection
Site Date
Gl , Hay Coulee 7-20-74
8-30-74
6-26-75
5-14-76
7-16-76
9-14-76
G2, McRae Knolls 9-15-76
G3, Kluver West 7-21-74
7-07-75
7-07-75
7-14-76
9-14-76
G4, Kluver North 7-21-74
7-07-75
7-16-75
9-14-76
G5, Kluver East 7-21-74
6-26-75
7-14-76
9-15-76
9-15-76
G6, Harvey 1 9-15-74
9-15-74
G7, Pasture 5-01-75
5-14-76
X
314
332
215
458
290
274
341
218
349
293
272
301
342
349
242
262
227
316
322
404
239
339
321
337
371
1 s.d.
14
43
26
43
19
22
21
16
21
29
25
15
18
21
4
34
34
40
26
32
15
46
28
132
54
                                    60

-------
  TABLE 3.7.   SULFUR (AS TOTAL SULFATE) CONTENTS OF Usnea hirta AND  Parmelia
              chlorochroa SAMPLES COLLECTED IN 1975.   (Determined by MSU
              Soils Testing Laboratory.)
          Site
Collection
   Date
Usnea hirta
 % sulfur
as sulfate
1  s.d.
     P9, Road                      5-01-75
     P15, Fort Howes transplants   5-01-75
     P15, Fort Howes natives         "
     P9, Road                      8-11-75
     P3, Kluver NE1                9-25-75
     P8, Morning Star View           "
     P13, Home Creek Butte         7-16-75
     P10, East Otter Creek         7-16-75
     P10, East Otter Creek         9-25-75
Parmelia chlorochroa (includes 1974)
     Gl, Hay Coulee
     G3, Kluver West
     G4, Kluver North
     G5, Kluver East
     G6, Harvey
     Taylor Creek
     G7, Pasture
 7-21-74
 9-15-74
 5-01-75
                0.25
                0.25
                0.19
                0.26
                0.03
                0.19
                0.15
                0.08
                0.22
   0.26
   0.22
   0.20
   0.26
   0.23
   0.17
   0.19
             0.04
             0.03
             0.04
             0.04
             0.01
             0.08
             0.16
             0.01
             o.n
0.02
0.02
0.02
0.02
0.01
0.02
0.03
                              ACKNOWLEDGEMENTS

     This  project  is  funded  by  EPA  grant  No.  R803213.   Appreciation  is
expressed  to  Drs.  W.A.  Weber  and J.W.  Thomson  for assistance with  lichen
identification; to Drs. C.C.  Gordon, Jerry Bromenshenk, and Clint Carlson for
field  assistance  and to the  landowners for permitting access to the research
sites.
                                      61

-------
                                 REFERENCES

Bird,  C.D.   Keys   to  the  Lichens  of West-Central  Canada.   The Herbarium,
     Department of Biology, University of Calgary, 1970.

Eversman, Sharon.  Lichens as Predictors and Indicators of Air Pollution from
     Coal-Fired Power Plant Emissions.  The Bioenvironmental Impact of a Coal-
     Fired  Power  Plant,  Second  Interim Report,  Colstrip,  Montana. Corvallis
     Environmental Research  Laboratory,  Office  of  Research and  Development,
     U.S. Environmental  Protection Agency, Corvallis, Oregon, 1975.

Eversman, Sharon.  Effects of Low-Level  S02 Stress on Two Lichen Species.  The
     Bioenvironmental  Impact  of a  Coal-Fired  Power  Plant,  Third   Interim
     Report, Colstrip, Montana.  Corvallis Environmental Research  Laboratory,
     Office of Research and Development, U.S. Environmental  Protection Agency,
     Corvallis, Oregon,  1975.

Gordon, C.C.  Investigations  of the  Impact of Coal-Fired Power Plant Emissions
     upon Plant  Disease  and  upon Plant-fungus  and  Plant-insect  Systems.  The
     Bioenvironmental  Impact  of a  Coal-Fired  Power  Plant,  Second   Interim
     Report, Colstrip, Montana.  Corvallis Environmental Research  Laboratory,
     Office  of  Research  and  Development.    U.S.   Environmental  Protection
     Agency, Corvallis,  Oregon,  1975.

Gordon, C.C., Clinton Carlson, Phillip Tourangeau.  A Cooperative  Evaluation of
     Potential  Air  Pollution  Injury and  Damage  to Coniferous Habitats  on
     National  Forest  Lands  near  Colstrip,  Montana.   Interim  Report  of
     Activities  from June 1,  1975  to May  30,  1976.   USDA,  Forest  Service,
     Northern  Region  and   University   of  Montana,  Environmental   Studies
     Laboratory.   Missoula, Montana, July, 1976.

Hale, Mason. How to Know the  Lichens.  Wm.C. Brown, Dubuque,  Iowa
          1969.

Hanson, Herbert  C.   Ecology  of  the Grassland.  II.   Botanical  Review  16: 283-
     360, 1950.

LeBlanc,  Fabius   and D.  Rao.   Effects   of  Air Pollutants  on  Lichens  and
     Byrophytes.   In:  Mudd, J.B.  andT.T. Kozlowski (eds). Responses of Plants
     to Air  Pollution.  Academic  Press, New York,  1975.

Wetmore, Clifford.   Lichens  of  the  Black  Hills  of  South Dakota and  Wyoming.
     Publications  of the  Museum, Michigan  State  University,  East  Lansing,
     Michigan,  1967.
                                      62

-------
                                  APPENDIX

      LIST OF LICHEN SPECIES COLLECTED FROM THE  COLSTRIP, MONTANA, AREA;
                BIG HORN, ROSEBUD, AND POWDER RIVER COUNTIES.
                  (Collection number(s); site;  substrate.)

 1.   Acarospora schleicheri  (Ach. )  Mass.   565, Gl.   Soil.
 2.   Agrestia  hispida (Mereschk.) Hale &  W.Culb.  569,  G6;  609,  Sarpy  Creek-
          720,  P7; G5.  Soil.
 3.   Alectoria fuscescens  684,  Upper Indian Creek Divide;  690,  P3;  694,  P14;
          556,  Ash Creek.  Ponderosa pine  base.  det.  I.M.  Brodo.
 4.   Caloplaca aurantiaca (Lightf.)  Th.Fr.   547,  P6.   Decorticated  stump.
 5.   Candelariella  aurella (Hoffm.)  Zahlbr.  710, P16;  744,  Pll.  Rock.
 6.   Cetraria  pinastri (Scop.) S. Gray.  548,  Ash Creek;  733,  Pll.   Pine  tree
          trunk.
 7.   Cladonia  pyxidata (L. )  Hoffm.   543,  544 557, G3;  571,  G6.   Soil.
 8.   Cladonia  spp.  squamules   All  sites.   Soil.
 9.   Collema tenax  (Sw.)  Ach.   580,  Ash Creek; 581,  G4;  582, Gl.  Soil.
10.   Cyphelium notarisii   549,  P6.   Decorticated  stump.
11.   Dermatocarpon  lachneum  (Ach.)  A.L.Sm.  584, Gl.   Soil.
12.   Everm'a mesomorpha Nyl.   598,  P15; 660, P10.  Ponderosa pine base.
13.   Fulgensia bracteata (Hoffm.) Ras.  with F. fulgens (Sw.)  Elenk.  716,  P7;
          546,  545, G3; 560,  Gl;  568,  G6.  Soil.
14.   Hypogymnia physodes  (L.)  W.  Wats.  538, P15;  539,  Ash Creek; 688, P3;  698,
          -P16;  703, P8.  Pine tree trunk.
15.   H. bitten' (Lynge) Ahti.   676,  P16.   Ponderosa  pine  trunk.
16.   Lecanora  calcarea (L.)  Somm.   608, P16.   Rock.
17.   L. chrysoleuca  (Sm.) Ach.   669,  P10.   Rock.
18.   L. melanophthalma (Ram.)  Ram.   709,  P16;  669, P10.   Rock.
19.   L. mural is (Schreb.) Rabenh.   665, P8.  Rock.
20.   Lecidea decipiens (Hedw.)  Ach.   566,  G6;  567, Gl.   Soil.
21.   L. marginata Schaer.  682,  Upper  Indian  Creek; 705, P16;  732,  P14.  Rock.
22.   L. rubiformis  (Wahlenb.  ex. Ach.) Wahlenb.  610,  Sarpy Creek;  648,  P16;
          704,  P5; 715, P7; 731,  P16.  Soil in rock crevices.
23.   Letharia vulpina  (L. ) Hue.   654,  P6; 597, P15; 689, P3.  Pine tree trunks
24.   Omphalodiscus virginis  (Schaer.) Schol.   693, P8.   Rock.
25.   Parmelia  chlorochroa Tuck.  G3;  559,  Gl ; 653,  PI;  656,  G5; 658, Sarpy
          Creek.  Soil.
26.   P. infumata Nyl.   554, Ash Creek;  674, P16; 738,  745, Pll; 729,  P16.  Pine
          tree trunk.
27.   P. elegantula (Zahlbr.) Szat.   729, P16.  Pine tree trunk.
28.   P. lineola Berry  666, 671, P10.  Rock.
29.   P. subdecipiens Vain. 562,  P15; 652, PI.  Rock.
30.   P. subolivacea Nyl.  553, 557, Ash  Creek.  Pine tree trunk, branches.
31.   p. sulcata Tayl.  552, Ash Creek;  675,  P16; 696,  P14; 735, Pll.   Pine  tree
          trunk.

                                      63

-------
32.   P.  ulophyl lodes  (Vain.)  Sav.   563,  Ash Creek;  564,  P15;  695, P14;  699,
          P16.  Pine trunk.
33.   Parmeliopsis ambigua  (Wulf.)  Nyl.   605,  P3; 672, P10; 685, Upper  Indian
          Creek;  691,  P3;  734, Pll.  Base of pine  trees.
34.   Physcia caesia (Hoffm.) Hampe  558, P15. Rock.
35.   Peltigera canina  (L.) Willd.  579, P15. Soil.
36.   P.  rufescens (Weis.) Humb.   692, P3.  Soil.
37.   Rinodina confragosa (Ach.)  Koerb.  702, PI.   Rock.
38.   Squamarina lentigera (G.Web.) Poelt.  570, Gl; 718, P7.  Soil.
39.   Toninia caeruleonigricans (Lightf.)  Th.Fr.  714, P7.   Soil.
40.   Usnea hirta  (L.)  Wigg.  603, Whitetail Guard Station; 550,  Ash  Creek;  572,
          P15;  686 Upper Indian Creek.
41.   Umbilicaria  torrefacta (Lightf.) Schrad.   740, Pll.   Rock.
42.   Umbilicaria  vellea (L.) Ach.   608, Sarpy Creek.  Rock.
43.   Verrucaria sp.  706, P16.  Rock.
44.   Xanthoria elegans (Link.) Th.Fr.  609, P10;  707, P16.  Rock.
45.   X.  polycarpa (Ehrh.) Oliv.   673, P15.  Shrub branches.
                                     64

-------
                                  SECTION 4

  INVESTIGATION OF THE IMPACT OF COAL-FIRED  POWER PLANT EMISSIONS UPON THE
 DISEASE/HEALTH/GROWTH CHARACTERISTICS OF  PONDEROSA PINE-SKUNKBUSH ECOSYSTEMS
               AND GRASSLAND ECOSYSTEMS IN SOUTHEASTERN MONTANA

                                      by

                C.  C.  Gordon,  P.  C. Tourangeau, and P. M. Rice


                                 INTRODUCTION

     This  portion  of the  EPA-CERL  (Corvallis Environmental  Research Labora-
tory, Oregon)-sponsored  studies was  initiated to  ascertain  and quantify  the
impact of coal-fired power plant emissions upon the two major  ecosystems of the
Col strip  area of  southeastern  Montana:   (1)  Ponderosa pine-skunkbush (Pinus
ponderosa-Rhus trilobata) and (2) cool season-short grass.   Before the two,  350
MW capacity coal-fired power  plants were  constructed and began  operations  at
Col strip,  air monitoring studies by  federal  and  state  agencies,  the utility
companies, and private research  institutes established that ambient air in the
Col strip area was as pristine and free of phytotoxic gases as any area in  the
contiguous United States thus far studied (Northern Cheyenne  Research Project,
1976).   Because  of  the  pristine  nature of  this  area of  Montana,  it  was
hypothesized  that  if  the baseline characteristics of the chemical,  physical,
and  growth/health/disease  parameters  of  indigenous  flora and  fauna species
could be established and quantified before the power plants began operations,
many future direct and indirect  impacts of the atmospheric emissions upon these
species  could be measured and  quantified.  To test  this  hypothesis,  several
study objectives were  carried  out.  These encompassed  the establishment of:

     (1)  Baseline  levels  of  sulfur and  fluoride concentrations in  the domi-
          nant indigenous  plant species of both the ponderosa pine-skunkbush
          and cool  season-short  grass  ecosystems;

     (2)  baseline growth/health/disease parameters of ponderosa pine foliage;

     (3)  baseline  insect  population  and  insect  damage parameters for ponde-
          rosa pine of the  area;

     (4)  baseline  fungal  population  and  damage  to selected indigenous plant
          species; and

     (5)  baseline physical (pH) and chemical parameters of precipitation.

     These baseline parameters and characteristics were used to study species
of  flora and fauna collected  from  permanently  established  ponderosa  pine-


                                     65

-------
skunkbush  sites  located  at  varying distances  and directions  from Colstrip
during the 1974 and 1975 growing seasons.

     After the 1974 and 1975 baseline studies on the ponderosa pine-skunkbush
sites were completed  (see EPA-CERL  1976 Second Interim Report, EPA-600/3-76-
013). some of the earlier study objectives were deleted (i.e., fungal popula-
tions and host-parasite relationships on indigenous plant species) and others
were modified (i.e. , the number of ponderosa pine-skunkbush sites and precipi-
tation  chemistry  studies).   The modification  or deletion  of those original
objectives were  initiated  to allow concentration on those objectives which more
adequately tested the  original hypothesis.

     During 1975 and 1976, air monitoring by the Montana Department of Health
and Environmental Sciences (DHES) Air Quality Bureau at their monitoring site 4
km southeast of Colstrip disclosed that S02,  N02,  03,  and F  concentrations in
the ambient  air  were  all  substantially below both  federal  and Montana state
standards for these pollutants (Maughan, 1977).  Wind  speed  and direction data
collected by personnel  of  the DHES Air Quality Bureau at their Colstrip site
disclose that the prevailing winds during the  growing season  (May to September)
are to  the  southeast  and  east and  to the  west  and  northwest,  and average
between two to  six miles per hour.

     Since September  1975, when Colstrip Unit 1 went  on line, and July  1976,
when Unit  2  started up, these two units have operated at an average megawatt
capacity of  1/3  (33%)  to  2/5 (40%) of their  total rated capacity of 700+ MW.
The estimated atmospheric emissions of the Colstrip units when operating at 1/3
and 2/5 capacity  are presented in Table 4.1  below.

        TABLE 4.1.  ESTIMATED ATMOSPHERIC EMISSIONS OF COLSTRIP UNITS.
                          Unit 1  & 2
                        at 1/3 capacity
  Unit 1  & 2
at 2/5 capacity
   Unit 1  & 2
at full capacity
S02 (tons/yr)
NO (tons/yr)
Particulate (tons/yr)
Fluoride (Ibs/yr)
6,046
6,920
536
4,666
7,256
8,304
644
5,600
18,142
20,760
1,612
14,000

Source:  Montana Department of Health and Environmental Sciences, 1975

                            MATERIALS AND METHODS

COLLECTION OF VEGETATION

     Five ponderosa pine-skunkbush sites located east to south of Colstrip at
distances  of  5 to  80  km were  utilized during both the  1975  and 1976 study
periods.   The  locations of each of these  five sites are listed in Table 4.2.
                                      66

-------
         TABLE 4.2.   LOCATIONS OF  FIVE  PONDEROSA  PINE-SKUNKBUSH SITES.
Ponderosa Pine-
Skunkbush Sites
Plot
Plot
Plot
Plot
Plot
S-3
S-5
SE-2
SE-4
E-l
T2S,
T6S,
T1S,
T47E
T2N,
Location
R41E,
R44E,
R42E,
, R47E
R42E,
Sec
Sec
Sec
, Sec
Sec
Distance
from
NE 12
SE 8
SW 1
NW 19
SW 29
26
66
16
56
5
and Direction
Col strip
km
km
km
km
km
SSE
SSE
SSE
ESE
E

Ponderosa  pine  foliage was collected  during  the  1975  field season from these
plots  for  fluoride and sulfur  analysis,  evaluation  of seven selected foliar
pathologies,  determination of  percent needle  retention,  and measurement of
needle length and fascicular cross-sectional area.

     Each  of  the five plots consisted of  ten  permanently  marked  trees.  Based
on  tree  height  and  diameter  at breast height  (DBH),  five  of the trees were
classified  as "older"  and five were  classified  as  "younger."   From each of
these ten trees,  four branches  were  removed from  the upper third of the crown
and four branches  from the lower third of  the crown from the side of the trees
facing the  Colstrip steam  generating complex.  The branches from each tree were
placed  in  plastic  sacks  with  appropriate identification  of  tree number and
crown position and transported  to  the  field laboratory or to the  Environmental
Studies Laboratory at the  University of Montana for subsequent preparation for
chemical analyses, pathological  evaluations, and measurements.

     The percent  needle retention  for  1972, 1973, and  1974 year's foliage was
determined  for each internode from  each branch for the respective  categories of
crown position and tree age by counting all fascicular scars on each internode,
then removing all  retained fascicles,  counting them,  and computing the percent
needle retention as shown:

           	fascicles present	-, x -|00
          ^-fascicles  present  +  number  of fascicular scars;

The mean percent  needle retention  for  each year's internode was computed, and
subsequent data analyses used  these basic values.

     One hundred  fascicles were randomly  selected from the group of retained
fascicles from  each  internode,  the  fascicular  sheaths  were  removed,  and one
needle was  randomly  selected from each fascicle  for pathological evaluation.
Ten fascicles were also selected for measurement of fascicular cross-sectional
area.   After evaluation, 25 needles  were randomly selected for measurement of
needle length.
                                      67

-------
     Each needle  was  evaluated  for  the presence  of  seven pathologies:   (1)
percent basal necrosis,  (2)  percent  basal  scale,  (3)  percent defoliator, (4)
percent tip burn,  (5)  percent total  necrosis,  (6) percent mottled needles, and
(7) percent  healthy needles.  Percent  healthy needles is  the  percentage of
total   needles  without  any  visible  pathology whatsoever  and,  as  such,  are
completely green.   Percent total  necrosis  is an estimate of the total necrotic
surface area of the 100 needles.   This category includes  chlorosis or yellow-
ing.  The  percent occurrence of  each pathology was  computed,  and  all  data
analyses used these  derived values.

     The method of sampling results  in the partitioning of nine  categories of
tree  age  and crown position within  each  plot for  each year's  foliage  is as
follows:
                      Upper Crown - Younger Trees
                      Upper Crown - Older Trees
                      Upper Crown - All  Tree Ages
                      Lower Crown - Younger Trees
                      Lower Crown - Older Trees
                      Lower Crown - All  Tree Ages
               All Crown Positions - Younger Trees
               All Crown Positions - Older Trees
               All Crown Positions - All Tree Ages

These  categories  permit a detailed  analysis  of within-plot and between-plot
variability as a function of duration of foliage exposure.   Studies have shown
that  sulfur  and  fluoride  are  partitioned in  greater concentrations  in the
taller  dominant vegetation  at  a  site which is impacted by the emissions of a
coal-fired power plant and which  includes  vegetation species similar to those
discussed  here  (Gordon  ert  al. ,  1977); other  studies  have demonstrated that
fluoride  is partitioned in the upper crown of conifers.  Any investigation of
the effects  of  coal-fired power  plant emissions upon  ponderosa  pine sites in
southeastern  Montana  requires a background  preoperational analysis  of the
partitioning  of  fluoride  and sulfur, pathologies  associated  with air pollu-
tion,  needle length, and fascicular cross-sectional  area in the upper and  lower
crowns of ponderosa  pine.

     Understory species of grass, shrubs,  and forbs were  collected at the five
ponderosa pine-skunkbush sites during both the 1975 and 1976 summer collection
periods.  A list of the  understory species collected is presented in Table 4.3.
At  each site, at least eight to ten  different  species  of  understory were
collected where  possible.   Collection  of  leaf foliage  from  each understory
species was obtained from a minimum of three plants and these were lumped and
tagged with the collection number, placed in plastic sacks, and brought back to
the laboratory for chemical analysis.

     Needles  manifesting  the  various  needle  pathologies  (i.e.,  mottling,
weevil, fungal damage)  were  selected for  histological studies.   Histological
preparation and staining procedures have been described by Jones  £t al_. (1965).
Photomicrographs were  taken with  the  use of a Reichert  Zetopan  phase~microscope
adapted with Leitz cameras.
                                      68

-------
                        TABLE 4.3.  UNDERSTORY SPECIES.
                Scientific
        Common
          Artemisia cana
          Gutierrezia sarothrae
          Prunus virginiana
          Rhus trilobata
          Festuca idahoensis
          Agropyron spicatum
          Artemisia 1udoviciana
          Balsamorhiza sagittate
          Andropogon scoparius
          Lupinus sp.
          Psoralea sp.
          Yucca glauca
          Aristida longiseta
          Stipa comata
          Vicia sp.
          Artemisia frigida
          Chrysothamnus nauseosus
          Juniperus scopulorum
          Artemisia tridentata
Silver Sage
Broom Snakeweed
Chokecherry
Skunkbush
Idaho Fescue
Bluebunch Wheatgrass
Prairie Sage
Arrow!eaf Balsamroot
Little Bluestem
Lupine
Scurf-pea
Yucca
Red Three Awn
Needle-and-thread
Vetch
Fringed Sage
Common Rabbit-brush
Rocky Mountain Juniper
Big Sage
CHEMICAL ANALYSIS OF VEGETATION
     The  methods  of  chemical  analyses  for  sulfur  and  fluoride have  been
reported  in  detail  elsewhere  (Kay et a_L ,  1975; EPA, 1976).  Therefore,  these
procedures are summarized here.
     Vegetation was dried under forced draft at 90°F, ground in a Wiley mill to
pass 30 mesh, and  stored in capped plastic vials.   For  fluoride analysis, 0.50
g of dried ground plant material was slurried with  CaO,  charred under infrared,
and ashed overnight at  600°C  in a muffle furnace.  The ash was  dissolved in 30%
HC10   made  to 100 ml with 50% TISAB, and the fluoride  activity was determined
within  Orion specific  ion electrode.   The  calculations  for  determining the
fluoride concentration in the  original plant sample were computerized.
     Sulfur  determinations  were carried out  using a  Leco induction furnace.
Before the data analysis, each derived variate (percents basal necrosis, basal
                                     69

-------
scale, needle retention, tip burn, mottled needles,  total  necrosis, and healthy
needles) was coded by the arcsine transformation for percentages where:
          coded variate = arcsine /variate x 57.29578
                                     100

All coded variates and sample statistics were decoded for the presentation of
results.

                                   RESULTS

DATA ANALYSIS OF VEGETATION

     The data for both 1975 and 1976 were interrogated by Analysis of Variance
(ANOVA).  The analyses were  performed  by the ANOVA computer program of Sokal
and Rohlf (1969).*

     A  four-level  nested  ANOVA  was used because the 1975 data required tests
for each variable for:  (1) Differences between plots;  (2) differences between
upper and lower crown positions; (3) differences between the younger and older
trees,  and  (4)  differences  between the years  of .foliage (1974,  1973, 1972).
The design for the four-level ANOVA is  schematically shown in  Table 4.4 for any
individual plot.   In  the  actual  analysis of all variables, the data for each
plot  were  arrayed  exactly  as in  Table  4.4  and  all  plots  were  tested
simultaneously.    In Table 4.4  the  highest level, Level  4, is  the  individual
plot.  At Level 3 are the two  crown positions  (upper and  lower) within each
plot.  At Level  2  are the tree ages (younger  and older), and at Level 1 are the
three years of foliage (1974, 1973, 1972).  The lowest level, Level  0, is the
individual  variates,  denoted as 1, 2,  3,  4, 5 to  indicate  the  five  samples
measured for fluoride content.  Since each plot consists of  five older and five
younger  trees, the  most  fundamental data block in any plot would be the five
variates  for  a particular  year's  foliage for  younger  or older trees,  from
either  crown  position.   As  can  be  seen from this design, tests are performed
between  the years of foliage, between the tree ages, between the crown posi-
tions, and between the plots.
*The  Program,  as  published  in  this  edition,  contains  errors.   Interested
 readers should contact the authors of the text for  current listings.
                                     70

-------
 TABLE 4.4.   THE STRUCTURE OF THE FOUR-LEVEL NESTED ANOVA FOR EACH INDIVIDUAL
             PLOT, 1975 COLLECTION, BOTH CROWN POSITIONS.
Level  4 - Plots

Level  3 - Crown Positions

Level  2 - Tree Ages

Level  1 - Age of Foliage

Level  0 - Five Individual
  Observations for Each
  Variable within Each
  Age of Foliage
              Plot
Upper
Lower
Younger
J_
1
2
3
4
5
2
1
2
3
4
5
3
1
2
3
4
5
Older
1
1
2
3
4
5
2
1
2
3
4
5
3
1
2
3
4
5
Younger
I
1
2
3
4
5
2
1
2
3
4
5
3
1
2
3
4
5
Older
1
1
2
3
4
5
2
1
2
3
4
5
3
1
2
3
4
5
                                     71

-------
     TABLE 4.5.   FOUR-LEVEL ANALYSIS OF VARIANCE TABLE FOR FLUORIDE,  1975
                 COLLECTION.
          Source
            of
         Variation
 Sums
  of
Squares
Degrees
  of
Freedom
 Mean
Squares
  F
Ratio
Between the Plots       29.004
  (Level  4)

Between Crown Positions
  within the Plots       2.412
  (Level  3)

Between the Tree Ages
  within the Crown       4.876
  Positions (Level 2)
                 10
                                   7.251
                                    .482
                    .487
                               15.031*
                                 .989
             1.032
Between the Years of
Fol iage within the
Tree Ages (Level 1)
Within the Years of
Foliage (Level 0)
18.889 40
103.868 240
.472 1.091
.432

*F ratio significant @ p ^ .05
 TABLE 4.6.   THE STRUCTURE OF THE THREE-LEVEL NESTED ANOVA FOR EACH INDIVIDUAL
           PLOT, 1976 COLLECTION*

Level 3 Plots
Level 2 - Tree Ages within the Plot
Level 1 - Age of Foliage within the Tree Ages
Level 0 - Five Individual Observations for
Each Variable within the Ages of Foliage




Plots
Younger
!_
1
2
3
4
5
2
1
2
3
4
5
3
1
2
3
4
5
Older
1
1
2
3
4
5
2
1
2
3
4
5
3
1
2
3
4
5

*This design is precisely the same as that used for individual upper or lower
 crown positions in the 1975 collection.
                                     72

-------
     The results of the four-level ANOVA for fluoride for the 1975 collections
are shown in Table 4.5, wherein the levels as detailed in Table 4.4 are listed
under Source of Variation.The  only  level  in Table 4.5 which showed a signifi-
cant effect was  Level  4,  Between the Plots.  The conclusion to be drawn from
Table 4.5 is that the means for the five plots  are sufficiently different, that
they  represent  different  populations  as  regards   fluoride  concentrations.
Also, it may be concluded that within the plots mean values for the respective
crown positions, tree ages, and individual years of foliage would represent the
same population.

     In  1976,  collections  were  obtained  from  upper crown positions  only—
therefore the level supporting upper and lower crown for the ANOVA for 1975 is
not  needed.   The design for the  1976  collection is shown  in Table 4.6  as a
three-level nested  ANOVA*,  where Level  3  is the individual plot.   At  Level 2
are  the  tree  ages within  the  plots,  at Level  1 are the individual years  of
foliage within the ages, and at Level 0,  the individual variates.

     The results of ANOVA  for fluoride for 1976 are shown in  Table 4.7.  None of
the F ratios are significant,  and it may be concluded that fluoride means from
plots, tree ages, and years of foliage from the upper crown represent  samples
from the same population.

     TABLE 4.7.  ANALYSIS OF VARIANCE TABLE FOR FLUORIDE,  1976 COLLECTION.

Source
of
Variation
Between the Plots
(Level 3)
Sums
of
Squares
3.912

Degrees
of
Freedom
4


Mean
Squares
.978


F
Ratio
2.025

Between the Tree Ages
  within the Plots       2.415
  (Level 2)

Between the Years of
  Foliage within the     6.303
  Tree Ages (Level 1)

Within the Years of     46.176
  Foliage (Level 0)
 20


120
.483



.315


.384
                                1.532
.819
     The 1975  collections  from the upper crown and the 1976 collections were
compared by a four-level nested ANOVA shown as Table 4.8,  where, for brevity.
only two plots are shown.  At  Level 4 are the two collections, at Level  3 are
the plots within the collections, at Level 2 are the two tree ages within
*In 1975  collections,  ANOVA was performed
 separately and used this design.
   on the upper  and  lower  crown  data
                                     73

-------
     TABLE 4.8.   THE STRUCTURE OF THE FOUR-LEVEL NESTED  ANALYSIS  OF  VARIANCE  FOR BOTH  1975 AND  1976
                 COLLECTIONS, UPPER CROWNS.


Level 4 - Collections                1975 Collection                          1976 Collection
Level  3 - Between the
  Plots within Collections      Plot 1               Plot  2               Plot  1              Plot 2
Level 2 - Between Tree
  Age within the Plots    Younger    Older    Younger    Older     Younger    Older    Younger    Older

Level 1 - Between the
  Ages of Foliage within
  the Tree Ages           Z4. ?3 72  747372  747372   747372   757473   757473  757473  757473

Level 0 - Five Individual   1  1  1
 Observations for Each
 Variable within the
 Ages of Foliage
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
^
0
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5

-------
the plots within the collections, at Level 2 are the two tree ages within the
plots, at Level 1  are the years of foliage, and  at Level 0 are the  five variates
for each year of foliage.
     The results of the 1975 and 1976 fluoride  analyses are shown  in Table 4.9.
The F ratios  for Level 4 (Between Collections) and Level 3  (Between the Plots
within the  Collections) are  significant  at p ^ .05.  It may be concluded that
mean  values  for  fluoride  from the  two  collections  would  be  sufficiently
different such that they would  represent samples from different populations;
similarly  fluoride  values  from the plots  represent samples  from  different
populations.   Any  level  which  showed  effects significant  at  p ^  .05 were
further interrogated by t-tests.

HISTOLOGICAL  STUDIES

     Six  plates  of photomicrographs  depicting  three  of  the needle  tissue
pathologies  being studied  on  ponderosa pine  foliage were  prepared  for this
report.  Other pine foliage pathologies caused by insects are presented by Dr.
Bromenshenk in his portion of this report.

Needle Mottling

     Air pollution  studies have  demonstrated that the macroscopic symptoms of
conifer needle mottling are a typical host manifestation to phytotoxic gases.
However,  mottling of  ponderosa  pine  needles  can and does  occur in  pristine
areas where no or very low concentrations of phytotoxic gases are  present.  The
tissue pathology of mottling is  first manifested in the mesophyll cell beneath
the rows of stomatal opening in localized areas of the needles.  As illustrated
in  Plate  1,  which shows both cross and  longitudinal sections of  mottled pine
needles, the  cell  walls of the  mesophyll cells beneath the stomatal  openings
are  destroyed  in  some  cases   and  the  surrounding cells  then lose  their
chlorophyll  (photo  D,  Plate 1).  In other cases, the mesophyll  cells beneath
the stomatal  openings lose their chlorophyll but retain intact cell walls which
lose none of their form (photo B, Plate  1).

     Histological  studies of  mottled  needles  from the ponderosa  sites in the
Colstrip area have  not established any association thus  far  with  either fungal
or  insect  infestation.   Consequently,   it appears  that the  needle  mottling
recorded in  this  study is being caused  primarily  by abiotic factors such as
heat (i.e.,  concentration  of the sun's  rays by water droplets along  stomatal
rows).

Basal Necrosis and Scale

     During the  1974  and  1975 study periods,  there  was  a strong possibility
that needle  tissue  pathologies  occurring beneath or at the apical  opening of
the fascicular sheath were caused by both biotic (insects) and abiotic (acidic
solutions)  causal  agents  because  there  were  two  major  macroscopic  symptoms
involved with these pathologies.  These  two  macrosymptoms were classified as
basal  needle  necrosis and  basal  needle  scale.   While  both   symptoms  are
associated with necrotic  tissues,  basal  scale is associated only with tissue
swelling (hypertrophy) and,  in many cases, with the splitting of  the epidermal
tissues  on  the  swollen  area of the  needle.  Although  the authors  are  not

                                     75

-------
  TABLE 4.9.  ANALYSIS OF VARIANCE TABLE FOR FLUORIDE IN  BOTH  1975  AND  1976 COLLECTIONS, UPPER CROWNS.
    Source of Variation
Sums of Squares      Degrees  of  Freedom     Mean Squares     F Ratio
Between Collections
  (Level 4)

Between the Plots
  within the Collections
  (Level 3)

Between the Tree Ages
  within the Plots
  (Level 2)

Between the Years of Foliage
  within the Tree Ages
  (Level 1)

Within the Years of Foliage
  (Level 0)
    23.914
    25.314
     5.340
    12.335
    94.360
 10
 40
240
                  23.913
                   3.164
.534
.308
.393
              7.557*
              5.925*
1.731
 .784
*F ratios are significant @ p S .05

-------

-------
completely convinced that some insects may not cause tissue damage similar to
that classified  as basal  needle  scale, histological  studies  of both tissue
necroses during 1976 and early 1977 strongly indicate that the two pathologies
are caused by abiotic agents.

     Photomicrographs were prepared to depict the damage  associated with basal
needle necrosis  (Plates  2  to 4).   On Plate 2 are three photomicrographs of a
cross section  of  the  needle base and sheath of ponderosa pine collected from
the McRae pine site (SE-2).  In photo B of this plate, one notes  that there is
no  endodermal  tissue (see circled area  on  photo) on  the  dorsal  side  of the
needle  where   tissue  necrosis occurred.   This  means  that the  causal  agent
induced this damage before the parenchymatous cells in this  area differentiated
into  endodermal   cells   during  early  morphogenesis  of  the  needle.   Needle
morphogenesis   of  ponderosa pine  takes  approximately  three  months  after the
spring bud  break-candle  stage period, and thus it is during this time that a
needle is damaged by the causal agent.

     Both saprophytic fungi  and small insects (see Bromenshenk's 1977 report on
insects)  are  common residents  of the  fascicular sheath  area.   In  all  four
photographs on Plate 2 one notes the saprophytic fungus Phylctaena sp.  on the
outer cataphylls  of the  fascicular sheath.  This saprophyte, which produces a
pycnidial fruiting  body (see arrow in  photo D), is the most common fungus found
in  these histological  studies of  needle  basal  tissues.  Since  its  hyphal
filaments  (vegetative  structure)  are  not  found  in  either the  necrotic  or
healthy  tissues  of the  pine needle,  it  is  suspected that this saprophyte is
restricted to  the dead sclerenchymatous cells of the fascicular sheath.

     On  Plate  3  are photomicrographs which depict  cross  sections  of healthy
(photos A and  B)  and necrotic (photos  C and D) basal needles collected from the
Kluver pine site  (E-l) during 1976.  The structure in the middle of the three
needle  set  is the  dwarf shoot  bud which usually  remains  dormant  during the
entire  life of the  pine  needles.  However, depending on its positioning along
the  internode, the dwarf shoot bud can  initiate  growth  and form the male or
female cones.   Also, dwarf shoot buds  can form lateral branches  if the terminal
bud of the branch  is destroyed.

     Photos A  and B of Plate 3 depict the  healthy tissue of a needle lacking any
basal tissue or cell necrosis.  Photos C and D depict an interfacial (between
the needle) necrosis of the epidermal and hypodermal tissue area of all three
needles  of  this needle bundle (see arrow in photo  C).  This necrosis occurred
during early needle morphogenesis, since the epidermal and hypodermal cells had
not differentiated  from  the thin-walled parenchymatous cells.  Therefore, the
causal agent(s)  of the  damage entered this  area at an early stage of needle
tissue   development.    To   facilitate   understanding  of  this   interfacial
epidermal-hypodermal  tissue   necrosis,   Plate  4  was  prepared  showing  a
peridermal  longitudinal  section  of  a  ponderosa  pine  needle  (Site  E-l)
demonstrating  basal tissue necrosis of the interfacial area of the needle.  As
depicted  (see  arrows)  in photos  A, B,  and  C,  the necrosis occurs in all the
hypodermal  cells  from the  dwarf  shoot bud to the  outer periphery  where one
notes the epidermal  one-cell  layer.   When necrosis occurs in the hypodermal-
epidermal  tissues  of  the  needle,  easy  entry  to  the  mesophyll   tissue  is
available to both parasitic or saprophytic fungi.  However, this  rarely occurs

                                      78

-------
B

-------

PLATE 3

-------
and there is an extreme dearth of either type of fungi in needles manifesting
basal needle necrosis.
                 /
     One notes  in  photos  A, C, and  D  of Plate 4 that the dwarf shoot bud is
composed entirely of thin-walled parenchymatous cells and that no cell necrosis
is occurring in the  apical portion of  this bud (photo  D  shows  a different
section of  the  same bud).   Furthermore,  one notes  that  cellular  necrosis is
occurring at the base of the dwarf shoot bud, which  strongly suggests that the
causal agent  in this case  is  lodged only  at the base of the  bud  during the
period of penetration  for these cells.  The photographs  on  both  Plate 3 and
Plate 4 strongly support our theory that abiotic causal agents are responsible
for basal  needle necrosis.

     On Plate  5 are photomicrographs of cross sections of basal  needle areas
manifesting  various  stages  of basal  scale.  In photo A, the swelling beneath
the  necrotic  epidermal  and hypodermal  tissues is caused by the production of
excessive parenchymatous  cells (hyperplasia);  if  these  cells  had developed
normally during early needle morphogenesis they would have differentiated into
mesophyl1 cells containing  chloroplasts.   In photo  B, the needle cross section
depicts  no  external tissue necrosis but  is manifesting the  hypertrophy of
mesophyll cells and the symptoms of basal scale.  One notes  that the mesophyll
cells adjacent  to  the hypodermis (see arrow) are undergoing cell  enlargement
(hypertrophy)   but  not  hyperplasia.   In  both  photos  C  and  D,  excessive
hypertrophy  of  mesophyll cells has caused the splitting of the hypodermal and
epidermal layers in the  basal needle area.

     The last plate (Plate 6) depicts cross sections of a  ponderosa pine needle
manifesting  basal  necrosis  (photos  A and  B)  and a  longitudinal  section of a
ponderosa pine  needle manifesting basal  scale (photos C and D).  As can easily
be  discerned from  these  two tissue pathologies,  the necrotic cells  of the
"basal  necrosis"  symptom  do not undergo hypertrophy prior to dying (photos A
and  B), while hypertrophy of mesophyll  cells does occur prior to death in the
"basal  scale"  symptom.   Photos A  and  B of this plate  also demonstrate (as
photos A and B, Plate 2) that the causal  agent caused the damage to the basal
needle tissues  before formation of the endodermal cells in the area of damage
(see arrow in photo B).

1975 DATA ANALYSIS

     As previously mentioned in the introduction section of this reportK each
of   the  12  pine  foliage  characteristics  being   measured  is  expected  or
hypothesized  to change if  and when the atmospheric  emissions  of  coal-fired
power plants impact the permanent ponderosa pine-skunkbush sites.  The results
of the  1975 ponderosa pine-skunkbush studies are presented first so that the
reader realizes where the similarities and significant differences  within the
12 characteristics occurred before any coal-fired power plants went on-line at
Colstrip.  After the 1975 data, the results of the 1976 data  are presented, and
the  variability between the 1976 data and 1975 data are then compared utilizing
four-level Analysis of Variance (ANOVA).

     The data for the 12 growth/health/disease characteristics  from both crown
positions and  the  upper and lower crown  positions  separately  from 1975 were

                                     81

-------
  ; "v*"' :   Si
         .'&
i&w^^jwiiTO" wii



   ^^MP^-^:"
•
                PLATE 4
                 82

-------
PLATE 5




 83

-------
PLATE 6



  84

-------
interrogated  by nested ANOVA.  These  ANOVA  designs  were previously outlined
under Materials and Methods.

     The F ratios and the percent variance components for all 12 characteris-
tics  from  the  four-level  nested ANOVA,  employing  both crown positions, are
shown  on Table 4.10.   The F  ratios  and  percent variance components from the
ANOVA  for  upper crown positions are shown on Table 4.11, and those for  lower
crown positions are arrayed  in Table 4.12.  The data on these three  tables are
summarized in  Table 4.13, and significant treatment effects (p ^ .05) of any of
the 12 characteristics are indicated by asterisks (*).

     Inspection of Table 4.10 shows that  all growth/health/disease  character-
istics  have  significant  differences  between  the  plots  except for  cross-
sectional  area.   The cross-sectional  area,  however,  was the only  one which
showed  a  significant treatment effect between  the crown positions within the
plots  (Table  4.10,  Level  3).  Sulfur levels,  cross-sectional  area, percent
total  necrosis,  and percent  mottled needles  showed  significant  treatment
effects due to tree age within the two crown positions  (Table 4.10,  Level 2).
Percent needle retention, both basal needle pathologies, percent tip burn, and
percent healthy needles showed significant treatment effects  between the three
different years of foliage within given tree ages (Table  4.10,  Level 1).

     As  mentioned,  Table  4.11  shows  the  F  ratios  and   percent  variance
components  for 1975  data from  the  upper crown.  Seven  of  the  12 measured
characteristics (i.e., fluoride, cross-sectional area,  needle length, percent
basal necrosis, percent defoliator, percent tip burn,  percent mottled needles)
for the  upper crown position show significant  treatment  effects between the
five plots (Table  4.11).

     Treatment  effects  due  to tree age were  found  for  sulfur,  percent total
necrosis, and  percent mottled needles in  the upper crown foliage as  was found
in the  ANOVA  for  both crown positions (Table 4.10) but not  for needle cross-
sectional  area (Table 4.10,  Level 2).  Significant treatment effects  due to
year of foliage of the upper crown are present for percent basal  scale, percent
basal  necrosis,  percent  tip  burn,  and percent  healthy needles,  but  not for
percent needle retention as was found in  the ANOVA using both crown  positions
(Table 4.10,  Level  1).

     Table 4.12 contains data obtained from three-level  nested ANOVA of the 12
growth/health/disease  characteristics  of foliage for the 1975  data from the
lower  crown  position.  Five  of  the seven  foliage  characteristics  which had
significant treatment  effects in the upper crown between plots (Table 4.11,
Level  3)  showed  significant  treatment   effects  in  the  lower  crown;  cross-
sectional area  and  percent basal necrosis were not significant at this level
(Level 3).  A  treatment effect due to  tree  age (Table 4.12, Level  2) was present
only for needle cross-sectional area.  Percent needle  retention, the two basal
needle  pathologies,  percent  tip burn,   and  percent  healthy  needles  showed
significant treatment effects due to years of foliage (Table 4.12, level  1) as
was found in  the ANOVA using  both crown positions (Table 4.10,  Level 1).  The
data on T-ables  4.10,  4.11,  and 4.12 at Level 0 demonstrate  that more than 50
percent of  the total variance was found within  the  individual year's foliage.
                                      85

-------
CO
       TABLE  4.10.   SUMMARY  OF  F  RATIOS  AND  % VARIANCE  COMPONENTS  FOR THE  INDICATED LEVELS FOR ALL VARIABLES
                    FROM THE 1975 COLLECTION,  BOTH  CROWN  POSITIONS.

Level 4
Between
Plots
Variable
Fluoride (ppm)
Sulfur (ppm)
Area (mm2)
Needle length (mm)
% Needle
Retention
% Basal Necrosis
% Basal Scale
% Defoliator
Percent
F Variance
Ratio Component
15.0313*
15.6357*
.3812
8.1780*
6.0267*
6.4512*
25.1439*
26.5435*
% Tip Burn 559.6725*
% Total Necrosis
% Mottled Needles
% Healthy Needles
6.2189*
14.0596*
6.3088*
20.3
23.7
-12.7
28.1
5.5
1.0
7.0
37.2
35.9
10.1
30.9
-5.5
Level 3
Between
Crown Positions
Within Plots
Percent
F Variance
Ratio Component
.9893
.4764
6.4182*
2.7015
.7143
2.2886
.2011
.7619
.0407
.2298
1 . 2688
.2325
.17
-5.6
19.2
5.3
1.7
9.3
2.5
-.77
-1.0
-7.0
-1.7
7.3
Level 2
Between Tree Ages
Within
Crown Position
F
Ratio
1.0325
4.0962*
3.0008*
1.2374
.7599
.1746
.5748
1.5119
.8610
2.3471*
4.5441*
.2533
Percent
Variance
Component
.18
8.2
5.4
1.1
-2.4
-12.8
-6.0
2.5
-1.0
9.0
5.3
-13.7
Level 1
Between Years
Fol iage Within
Tree Ages
Percent
F Variance
Ratio Component
1.0912
.4883
.4092
1.1608
1.8939*
3.3341*
3.1461*
1.2710
2.0059*
1.1983
.2996
3.8790*
1.4
-8.3
-11.7
2.0
14.4
32.5
28.9
3.1
11.0
3.3
10.6
40.9
Level 0
Residual
Percent
Variance
Component
77.9
82
99.8
63.3
80.7
69.8
67.4
57.9
55.0
84.4
76.0
71.0

    *F ratios significant at p ^  .05

-------
  TABLE 4.11.   SUMMARY OF F RATIOS AND % VARIANCE COMPONENTS FOR THE INDICATED  LEVELS  FOR ALL VARIABLES
               FROM THE 1975 COLLECTIONS, UPPER CROWN POSITIONS.



Level
3
Between
Plots
Variable
Fluoride (ppm)
Sulfur (ppm)
Area (mm2)
Needle Length (mm)
% Needle Retention
% Basal Necrosis
% Basal Scale
% Defoliator
% Tip Burn
% Total Necrosis
% Mottled Needles
% Healthy Needles
F
Ratio
9.
3.
5.
10.
1.
17.
3.
8.
7.

6.
1.
.1449*
,8730
,6601*
.4802*
,4999
,2936*
.6641
.9789*
8807*
6175
5830*
5612


Level 2
Level 1
Between
Tree Ages
Within Plots
Percent
Variance
Component
27
12
8
32
3
18
11
33
34
-6
26
7
.5
.9
.6
.9
.5
.8
.1
.4
.8
.9
.5
.4
F
Ratio
1.9399
7.6885*
1 . 3488
1.2172
1.5102
.0862
.2469
1.9726
1.4351
3.6924*
6.8488*
.2442
Percent
Variance
Component
3.
10.
1.
1.
4.
-14
-10
4.
2.
16.
9.
-10.
4
2
1
2
1


0
9
0
1
8
Between
Fol iage
Tree
F
Ratio
.7511
.2550
.5007
1 . 3740
1.4177
3.7867*
2.9838*
1.0508
1.9609*
.9812
.3136
2.8365*
Years
Within
Ages
Percent
Variance
Component
-3.
-13.
-10
4.
7.
34
28.
•
10
-3.
-10.
27.
6
4

5
1

2
6

4
2
7
Level 0
Within
Years
Foliage
Percent
Variance
Component
72.6
90.2
100
61.2
85.2
61.6
71.2
61.7
52.1
91.2
74.5
75.6

*F ratios significant at p g .05

-------
  TABLE 4.12.  SUMMARY OF F RATIOS AND % VARIANCE COMPONENTS FOR THE INDICATED LEVELS FOR ALL VARIABLES
               FROM THE 1975 COLLECTIONS, LOWER CROWN POSITIONS.

Variable
Fluoride (ppm)
Sulfur (ppm)
Area (mm2)
oo Needle Length (mm)
% Needle Retention
% Basal Necrosis
% Basal Scale
% Defoliator
% Tip Burn
% Total Necrosis
% Mottled Needles
% Healthy Needles


F
Ratio
6.
4.
3.
13.
4.
4.
2.
11.
20.
1.
29.
•
4123
2540
1453
7680
6608
4609
3684
9581
1396
0794
0267
4061
Level 3
Between
Plots


Level 2
Level 1
Between
Tree Ages
Within Plots
Percent
Variance
Component
* 14
16
9
* 32
11
12
13
* 36
* 35

* 27
9

.3
.2
.8
.3
.8
.2
.9
.0
.54
.1
.0
F
Ratio
.6068
2.3751
6.9123
1.2750
.2831
.3060
.9208
1.1603
.4319
1.7427
1.3147
.2599
Percent
Variance
Component
-2.
6.
* 10.
1.
-8.
-9.
-1.

-4.
5.
•
-14.
9
3
5
1
7
2
0
9
4
7
49
8
Between
Fol iage
Tree
F
Ratio
1 . 3845
.8698
.2856
.9044
2.4079*
2.8310*
3.3777*
1.5131
2.0409*
1.3306
.2819
5.2829*
Years
Withi
Ages
n
Percent
Variance
Component
6.
-2
-13.
-1.
21.
25.
27.
5.
11.
5.
-12.
48.
3

3
2
4
8
9
7
9
8
1
8
Level 0
Within
Years
Foliage
Percent
Variance
Component
82.5
79.3
93.6
67.3
76.0
70.5
59.8
56.3
57.4
87.8
84.4
56.9

*F ratios significant at p g .05

-------
                         TABLE 4.13.   SUMMARY OF ANALYSES OF VARIANCE FOR 1975 COLLECTION.
                                                    Between the Tree Ages
                      Between the Plots	         Within the Plots	   Between the Years of Foliage
                 (Level 4) (Level 3) (Level 3)  (Level 3) (Level 2) (Level  2)  (Level  2) (Level  1)  (Level  1)
                   Both      Upper     Lower     Bothtt   Upperttt  Lowerttf     Both       Upper     Lower
                  Crown      Crown     Crown     Crown    Crown     Crown       Crown       Crown     Crown
      Variable   Positions Position  Position   Positions Position  Position   Positions Position  Position
    Fluoride          *        *         *
      (ppm)

    Sulfur (ppm)      *                              *        *

    Area (mm2)t                *                     *                  *
    Needle            *        *         *
      Length (mm)

    % Needle          *                                                             *                   *
      Retention
"•  % Basal
      Necrosis

    % Basal
      Scale
    % Defoliator      *        *         *

    % Tip Burn        *        *         *

    % Total           *
      Necrosi s

    % Mottled         *        *         *
      Needles

    % Healthy         *
      Needles
  t This variable showed significant differences between the Crown Positions
 tt This level is actually Between the Tree Ages Within the Crown Positions
ttt This level is Between the Tree Ages Within the Plots
    Significant treatment effect at p ^ .05
      *

-------
Fluoride

     The only significant effect for fluoride in the three-or four-level nested
ANOVA was between the plots for all  crown positions as shown in Table 4.13.  No
significant treatment effects for fluoride were  detected between tree ages or
between years  of foliage.   T-tests  were performed between  the  categories of
tree age and  crown  positions for fluoride data  for.foliage from all  plots to
confirm this,  and the  t-statistics  for plot SE-2 are presented  in Figure 4.1
(plot SE-2 was representative of all five plots  for fluoride).   Figure 4.2 was
prepared to  show the mean  values for fluoride in  1974 foliage  for all  upper
crown samples from all  five plots, and this figure illustrates the difference
in mean values between the  plots.  The mean fluoride values  shown in Figure 4.2
between plots SE-4 and  those  of E-l, SE-2, and S-5 were significantly different
at p ^  .05.

Sulfur

     Significant treatment  effects between the plots and between the tree ages
for sulfur levels were detected (Table 4.10, Levels 2 and 4) by ANOVA employing
both crown positions and by ANOVA employing  the  upper  crown  position (Table
4.11,  Level  2),  which  suggests greater variation between upper  crowns of the
two tree ages than between the lower crown positions.

     T-tests  of  sulfur levels  between  the individual years  of foliage from
upper crowns of both older trees and younger trees, tested separately, revealed
no  significant  differences   (p  S  .05).   Thus,   the  mean  values of  sulfur
concentrations for upper crown younger trees and upper crown older trees were
recomputed by  pooling  the measurements for the individual years.  These means
for  upper  crown  younger trees and upper crown older trees  were  then compared
utilizing  the  t-test.   Plot  SE-4 was the only plot  which  showed  significant
differences  for  sulfur  between  the  tree  ages  at  p ^  .05,  and the  basic
statistics are presented in Table 4.14.  The means and  95% confidence intervals
for  sulfur levels  between  upper crown, younger  and  older  trees are shown in
Figure 4.3 for plot  S-5.
         i

Cross-Sectional Area

     For needle  fascicular  cross-sectional  area, ANOVA indicated significant
treatment  effects (Table 4.13) between the plots for the  upper crown position.
Also,  significant effects  were found between the tree ages for the lo^er and
for  the combined  crown positions  (see  Table  4.13)  and  between  the  crown
positions  for  both crown analyses.  It should be  expected that the variability
of cross-sectional area of fascicles from lower crown foliage  within the plots
will be greater  than between the plots.  Figure 4.4 illustrates the mean values
and  95% confidence  intervals  for fascicular cross-sectional areas  from the
upper  and  lower  crown  positions  for plot SE-4.  As is depicted  in Figure 4.4,
the  difference  between the crown positions was most apparent for the younger
trees and  fascicular cross-sectional area was generally smaller for the lower
crown position.
                                      90

-------
               1974
                               UO
LY
LO
UY
DO
LY
.416


.162
.461

1.068
1.473
.913
XY
UZ
ZO LZ
.625


1.163
               1973
                               UO
LY
LO
UY
UO
LY
.035


.069
.063

.382
.654
.698
ZY
UZ
ZO LZ
.388


.328
                               UO
LY
LO
               1972
UY
UO
LY
.100


1.343
1.921

.149
.331
1.572
ZY
UZ
ZO LZ
1.023


1.279
 *t.05 = 2.306 for UY, UO, LY, LO Comparisons
  t.05 = 2.101 for ZY, ZO, UZ, LZ Comparisons

**UY = Upper Crown, Younger Trees            ZY
  UO = Upper Crown, Older Trees              ZO
  LY = Lower Crown, Younger Trees            UZ
  LO = Lower Crown, Older Trees              LZ
            All  Crowns, Younger Trees
            All  Crowns, Older Trees
            Upper Crowns, All Tree Ages
            Lower Crown, All  Tree Ages
 Figure 4.1.     Results  of t-tests* between the categories** of tree age  and
                crown  position  for 1974,  1973,  and 1972 years of foliage  from
                plot SE-2.
                                     91

-------
                                                                      PLOT SE-4
                                              -|  PLOT S-5
                                           PLOT E-l
ro
                                     ^ PLOT SE-2
                                                   ^ PLOT S-3
                                                 PPM FLUORIDE

    Figure 4.2.     Mean  values and 95%  confidence  interval  widths for fluoride in 1974 foliage,  upper crown,
                   al1 tree ages.

-------
                                                             1974
                                                   -| 1973    OLDER TREES
                                                                                                1  1974
                                                                                                  1973
                                                                                              1972
                                                                                         YOUNG TREES
I
1
200
1 ft I 1972

I " ' • 	 1 YOUNG TREES*
g ^^ OLDER TREES'1
*ALL YEARS FOLIAGE COMBINED
1 1 ' 1 1 1 ' 1 ' 1 ' 1 ' 1 '
300 400 500 600 700 800 900
                                            PPM SULFUR
Figure 4.3.     Variation  in total sulfur concentration in upper crown younger and older trees for plot S-5.

-------
                                                            | — + - 1  1974
                          YOUNGER TREES
                                                                             1973     UPPER CROWN
                                                                             1972
                                                1974
                                                 1973
                           LOWER CROWN
                                           — | 1972
to
                          OLDER TREES
                                                          t-H
                                       UPPER CROWN
                                                             LOWER CROWN
                    .5
 I            I            I           I           I
1.0        1.5         2.0         2.5         3.0
      FASCICULAR  CROSS  SECTIONAL AREA, nm2
  I
3.5
    Figure 4.4.     Mean  values and  95% confidence intervals for fascicular cross-sectional area, upper and
                   lower crown positions, younger and older trees,  plot SE-4.

-------
TABLE 4.14.  BASIC STATISTICS FOR SULFUR CONCENTRATIONS IN UPPER CROWN YOUNGER
             TREES AND UPPER CROWN OLDER TREES, ALL YEARS FOLIAGE COMBINED,
             PLOT SE-4*


       Statistic            Upper Crown, Younger     Upper Crown, Older
N
Maximum Value
Minimum Value
Mean
Variance
Standard Deviation
Standard Error
95% Confidence Limit
Median
Standard Error
Coefficient of Variation
Standard Error
Sampling Precision**
15
550
300
420
5285.714
72.03
18.772
40.266
450
23.5
17.31
3.25
9.587
15
500
250
348
6291.666
79.320
20.480
43.930
350
25.666
22.771
4.36
12.612

 *t-statistic @ p £ .05 = 2.048
  t-statistic between the mean values = 2.580
**Defined as
                        100
Needle Length

     Significant treatment  effects  were  found for needle length between plots
but  not within plots  (Table 4.13).  The mean values for needle length for each
year's foliage for upper and lower crown, younger and older trees for each plot
were compared  by  t-tests and the results are summarized in Table 4.15.   As  is
depicted in this table,  needle length across the plots was generally shorter on
the  lower crowns than on the upper crowns of younger trees, while needle length
of older trees of both upper and lower  crowns was generally the same.  The upper
crowns of younger trees  tended to have longer needles than the upper crowns  of
older trees.

Needle Retention

     ANOVA showed  significant treatment effects  for  needle  retention between
the plots (Table 4.13) and between the years of foliage for both crown positions
and  the  lower  crown  position but not  for  the upper crown position.  The mean
values and 95% confidence intervals for percent needle retention for upper and
lower crown positions for plot E-l are shown in Figure 4.5  and these values are
generally representative of the other plots.  The obvious differences between
1974 percent needle retention and older foliage from the lower crown as depicted
in  Figure  4.6 is representative  of  the  patterns at the other  plots  with the
exception of plot S-3 where the  percent needle retention on  the lower crowns was


                                      95

-------
  TABLE 4.15.  RESULTS OF T-TESTS BETWEEN MEAN VALUES FOR NEEDLE LENGTH FOR
               TREE AGES AND CROWN POSITIONS FOR ALL PLOTS.

Plot
SE-4
S-5
E-l
SE-2
S-3
Year of
Foliage Origin
1974
1973
1972
1974
1973
1972
1974
1973
1972
1974
1973
1972
1974
1973
1972
Mean
UY* vs
110.7
127.6
117.3
128.3
150.2
146.1
151.0
160.8
160.2
134.1
145.7
130.5
130.1
138.0
120.3
Values for
UO*
131. 2t
124.6
1 11 . 5t
115. 5t
141.5t
136. 7f
152.9
153. 3t
145.91
119. 9f
137. 5t
135. 2t
119.01
123. 5t
1 1 1 . 4t
Needle
LY*
126.6
118.1
111.6
120.0
127.5
122.5
152.7
144.1
145.1
127.3
125.3
121.8
127.8
126.8
116.9
Length
vs. LO*
135. 9f
125. Of
111.7
122.5
125.4
110.3f
146. 5f
151.lt
140.4
122. 3f
138.lt
128.2f
113. 7f
114. 6f
110. 4t

Plot
SE-4
S-5
E-l
SE-2
S-3
Year of
Fol iage Origin
1974
1973
1972
1974
1973
1972
1974
1973
1972
1974
1973
1972
1974
1973
1972
Mean
UY* vs
110.7
127.6
117.3
128.3
150.2
146.1
151.0
160.8
160.2
134.1
145.6
130.5
130.1
138.0
120.3
Values for
LY*
126. 6t
118.lt
111.6t
120. Of
127. 5t
122. 5t
152.7
144.lt
145. If
127. 3f
125. 3f
121. 8f
127. 8f
126. 8f
116.9
Needle
UO*
131.2
124.6
111.5
115.5
141.5
136.7
152.9
153.3
145.9
119.9
137.5
135.2
119.0
123.5
111.5
Length
vs. LO*
135.9
125.0
111.7
122. 5f
125. 4f
110. 3f
146.5
151.0
140.4
122.3
>38.1
128.2f
113. 7t
114. 6f
110.4

*UY = Upper Crown, Younger Trees   LY = Lower Crown, Younger Trees
 UO = Upper Crown, Older Trees     LO = Lower Crown, Older Trees

fMean values are significantly different @ p ^ .05
                                     96

-------
                                                                               •\  1974
                                         |  1973
                                                                         1972
                                                                                           UPPER  CROWN
UD
•-J
                                                                    -\  1973
                                                                                  1974
                                                         LOWER  CROWN
                                                                   1972
                       I

                      50
 I
60
70
 I
80
 i
90
  I
100
                                                 PERCENT NEEDLE RETENTION
    Figure 4.5.     Mean values  and  95% confidence intervals for percent  needle  retention  for  upper  and  lower-
                   crown positions,  plot  E-l.

-------
                         PLOT
                         SE-4
                                                                  1974
   1973

   1 1972
                      PLOT
                      S-5
^  1973
                                                        -\  1974
                                                         1972
            PLOT
            E-l
                                                1972
                                                            ^  1974
         1973
               PLOT
               SE-2
                                                  1972
                                                                1974
           1973
                            PLOT
                            S-3
         •4-
         1973

         1972
                                                                  1974
                   I           I           I           i

                  20          40          60          80
                          PERCENT NEEDLE RETENTION
                   100
Figure 4.6.     Mean  values  and 95%  confidence  intervals  for percent needle
               retention for upper crown positions for all plots.
                                     98

-------
90 percent for all internodes.  At all  plots,  mean percent needle retention de-
creases with internode age for upper crown positions, but the differences were
usually significant  only  between the youngest internodes (i.e., 1974 foliage)
and the oldest  internodes  (i.e., 1972 foliage)  being examined.

Basal Necrosis

     As depicted in Table  4.13, significant treatment  effects  for basal necrosis
were found  between the plots and between  the  years  of foliage for both crown
positions and for  the upper crown position, while  significant treatment effects
in the  lower crown position were  found only between the years of foliage.  The
mean values for percent basal needle necrosis and the 95% confidence intervals
for upper and lower crown positions for plots E-l  and SE-2 are shown in Figure
4.7.   As is depicted in this figure, the mean  values  for percent basal necrosis
increased with internode age in general, but differences between the mean values
were not always significant  across the plots.

Basal Scale

     The  ANOVA of basal  scale showed  that  there were  significant  treatment
effects in this pathology for both crown positions between the plots and between
the  years  of  foliage.   For the  upper and  lower  crown positions,  significant
treatment  effects  were indicated between the different years  of  foliage (see
Table 4.13).  No significant treatment effects were detected  for basal scale due
to tree age.   Percent  basal scale  is quite variable,  and the  mean values are
imprecisely  known.   On  Table  4.16 are the basic  statistics  for  percent basal
scale  for  the  three different-aged foliage from  plot  S-3,  and these data are
illustrative of the variability of the data from the other plots.

Defoliator

     For percent  defoliator,  the ANOVA detected significant treatment effects
between  the plots for  both  crown  positions  and for lower and  upper crown
positions, but not between the tree  ages  nor between the years of foliage, which
indicates that the variability for  this pathology was greater between the plots
than within  the plots  (see  Table  4.13).  As was found for basal scale, percent
defoliator  was a  very  variable needle pathology.  The  mean  values  and 95%
confidence intervals for all plots,  upper and lower crown positions, are arrayed
in Table 4.17.

Tip Burn

     In Table 4.13,  significant treatment effects between the plots and between
the different years  of foliage were  detected for percent tip  burn by each ANOVA.
No significant effects were found between the younger and  older  trees or between
crown  positions.   The  basic  statistics  for  percent  tip  burn for  the three
different-aged foliage from plots S-3 and SE-2 are  presented in Table 4.18.
                                       99

-------
1974
1973
1972
1974
1973
1972
1974
1973
1972


1974
, UPPER CROWN
' 1
I t t
1 * 1
H
1, 	 1 LOWER CROWN
1 ' 1
1 t 1
1 * 1
J UPPER CROWN
iJI
^ PLOT
SE-2
H
1973 JL 1
     1972
                                                      PLOT
                                                       E-l
r -
0
i
5
I
10
l
15
i
20
i
25
                            PERCENT BASAL NECROSIS
Figure 4.7.     Percent basal  necrosis  for upper  and  lower crown  positions,
               plots E-l  and SE-2, illustrating differences between  internodes
               and between plots.

                                     100

-------
TABLE 4.16.
BASIC STATISTICS FOR PERCENT BASAL SCALE FOR 1974,  1973,  AND 1972
YEARS FOLIAGE FROM PLOT S-3.
                                     1974
                                      1973
1972
N
Maximum Value
Minimum Value
Mean
Standard Deviation
95% Confidence Intervals

Median
Standard Error
Coefficient of Variation
Standard Error
20
12
0

6.
+ 1.


1.
125.
40.



69
00
05
57
251
682
1
25
20
37
0
8.
10.
+5.
-4.
8.
3.
65.
14.



23
94
55
22
49
06
66
17
20
54
0
12.
13.
+8.
-6.
10
3.
66.
14.



.52
.72
29
42

84
22
34

 TABLE 4.17.  MEAN VALUES AND 95% CONFIDENCE INTERVALS FOR PERCENT DEFOLIATOR
              FOR UPPER AND LOWER CROWN POSITIONS,  ALL PLOTS.

1974
Upper Crown
1973
1972
Plot
SE-4
S-5
E-l
SE-2
S-3
0
.44
2.93
2.89
3.41

1.3
3.67
3.03
5.44

.44
2.28
1.97
2.93

1.
4.
2.
5.
11
44
41
59
99
.43
2.0 1.
3.70 2.
3.23 1.
4.14 3.
11
15
62
59
10
.02
1.20
7.51
8.41
8.18
.19
1.62
11.12
4.19
10.19


6.
3.
6.
,01
.94
.34
.40
53
Lower Crown


1974



1973


1972


Plot
SE-4
S-5
E-l
SE-2
S-3
.07
1.89
1.98
2.06
6.22
.42
2.16
4.78
2.79
6.38
.05
1.35
1.94
1.63
4.24
.04
2.74
5.96
6.37
7.85
.21
2.64
7.98
4.77
5.43
.03
1.77
4.75
3.51
4.10
.11
4.10
9.09
8.68
9.08
.63
3.45
4.58
4.38
6.74
.08
2.43
3.72
3.56
5.01
                                     101

-------
TABLE 4.18.   BASIC STATISTICS FOR % TIP BURN FOR THE INDICATED YEARS FOLIAGE
             FOR PLOTS SE-2 AND PLOTS S-3.

Statistic
N
Maximum Value
Minimum Value
Mean
Variance
Standard Deviation
SE-2*
1974 1973
20 20
12 52
0 0
2.01 10.5
50.9 200
7.13 14.1
95% Confidence Interval +1.95 +8.0
-1.30 -5.98
Median
1.45 5.0
95% Confidence Interval +2.26 +8.0
-1.22 -4.35
Coefficient of Variation 87.59 74.6
Standard Error
22.0 17.1
Sampling Precision** 81.2 66.6
S-3*
1972 1974 1973
20 20 20
73 4 6
000
23.2 .08 .03
335 12.7 11.21
18.3 3.56 3.35
+13.64 +.25 +.017
-11.26 -.08 -.020
25.9 ' 0 0
+17.7 +.13 +.11
-14.4 -.13 -.11
63.5 212.8 337.0
13.5 106.7 259.6
53.6 199 315.3

1972
20
2
0
.07
7.87
2.8
+ .17
-.07
0
+ .08
-.08
180.4
78.2
168.8

*The data arrayed
position combinec
**Defined as (t
X
herein are for all categories of tree age and crown
5Sx-* 100



                                    102

-------
Total Needle Necrosis

     ANOVA showed significant treatment effects between the plots and between
the tree ages for both crown positions (Table 4.13).   Significant effects were
detected for the different-aged trees for upper crown but not lower  crown.  No
significant effects  of  percent  total needle necrosis were  found  between the
crown positions or between the different years of foliage.   The mean  values and
the 95% confidence intervals for percent needle necrosis  for SE-2 are presented
in Figure 4.8.

Needle Mottling

     As shown in Table 4.13, the ANOVA detected significant treatment effects
in percent mottled needles between the plots for the three categories of crown
positions,  and  in the two  tree  ages for both crown  positions  and  the upper
crown position.  No differences  were detected due to years of foliage origin.
The mean values and 95% confidence  intervals for the foliage of upper crowns of
younger and older  trees from plots SE-2 and S-3 are depicted in Figure 4.9.

Heal thy Needles

     Significant  treatment  effects  were  detected between plots for the ANOVA
employing  both  crown  positions  and  for  all  three  combinations  of  crown
positions  between the years  of  foliage  (see  Table 4.13).   Mean  values for
percent  healthy needles  and the  95% confidence intervals for the upper crown
positions for all  five ponderosa sites are depicted  in Figure 4.10.

1975-1976 DATA ANALYSES  COMPARISON

     The data  from  the  1976 collections were treated by a three-level  nested
ANOVA, and the results are summarized as Table 4.19.  In  addition, the data for
foliage from the  upper crown positions between collections for both years were
treated  by  four-level nested ANOVA, and the results are shown in Table 4.20.
Tables 4.11,  4.19,  and  4.20 are summarized in Table 4.21.  In this  table the
four-level  nested ANOVA  between  the 1975 and  1976  collections  is  designated
ANOVA K,  the  three-level  nested ANOVA for  upper  crown  foliage  from the 1975
collections is  designated ANOVA  B, and the three-level  nested  ANOVA  for the
1976  collection  is  designated  ANOVA J.   For each variable  in  Table  4.21,  a
significant treatment effect (p ^ .05) for a particular level, for each ANOVA
(K, B, or J, as above), is  denoted by an asterisk (*).  Then,  for ppm fluoride
(refer to Table 4.21) significant treatment effects were  detected:  (1) Between
the two years' collections (Level  4,  ANOVA K), and (2) between the plots (Level
3, ANOVAs K and B).

     Significant  treatment  effects  between  the  1975  and  1976  collections
(Table 4.21,  ANOVA  K) were detected only for:  (1) ppm fluoride,  (2)  percent
total necrosis, and  (3) percent healthy needles.   Significant treatment effects
only between the  plots were indicated for area (mm2), needle  length  (mm), and
percent  defoliator  by   each  ANOVA.   For percent  basal   scale,  significant
                                     103

-------
 1974



 1973




 1972
     1974
o


     1973



     1972
                                                                        YOUNGER
                                                                    OLDER
                                           8        10        12


                                             PERCENT TOTAL NECROSIS
                                                                            14
 I

16
18
20
Figure 4.8.     Mean  values and 95% confidence intervals for percent total necrosis illustrating differences

               in mean values for upper crown, younger and older trees at plot SE-2.

-------
   1974
   1973
   1972
o
Ul
   1974
   1973»-
   1972
1974
1973
1972
1974 4.
1973>|
1972*-
                                                       YOUNGER TREES
                                                                          PLOT
                                                                          SE-2
                                OLDER TREES
                                                                          YOUNGER TREES
                                                                             PLOT
                                                                             S-3
                                          OLDER  TREES
                                                                  10
                                                                            I
                                                                           12
1
 I
16
li
    Figure 4.9.
                                          PERCENT MOTTLED NEEDLES
                Mean  values  and  95%  confidence   intervals  for  percent  mottled  needles  illustrating
                differences in mean values for upper crown, younger and older trees for plots SE-2 and S~3.

-------
                        PLOT
                        SE-4
                                                  I
                                       1974
                      1   1973
                           1972
                I
                        PLOT
                         S-5  I	•	1  1973
                                                          1974
                                                     -\  1972
 PLOT
  E-l
    »
                                                1972
                                                            ^ 1974
                                                         1973
PLOT
SE-2    h
                                                 1972
                                                          »    1974
                                                          1973
                           PLOT
                           S-3
                             •*•
                                                         1973
                                                         1972
                                                                  1974
                  20
    I
   40
60
                          PERCENT HEALTHY NEEDLES
80
 I
100
Figure 4.10.    Percent  healthy  needles  for all  plots, upper crown positions,
               illustrating differences between  internodes.
                                    1Q6

-------
  TABLE 4.19.   SUMMARY OF  F  RATIOS  AND % VARIANCE  COMPONENTS  FOR THE  INDICATED  LEVELS  FOR ALL VARIABLES
               FROM THE 1976 COLLECTIONS,  UPPER  CROWN  POSITIONS.




Level
3
Between
Plots

Fl
Variable
uoride (ppm)
Sulfur (ppm)
Area (mm2)
Needle Length (mm)
%
%


-------
      TABLE 4.20.  SUMMARY OF F RATIOS AND % VARIANCE COMPONENTS FOR THE INDICATED LEVELS FOR ALL VARIABLES

                   FROM THE  1975 AND  1976 COLLECTIONS, UPPER CROWN POSITIONS.
o
CO
Level 4
Between
Col lections
Variable
Fluoride (ppm)
Sulfur (ppm)
Area (mm2)
Needle Length (mm)
% Needle
Retention
% Basal Necrosis
% Basal Scale
% Defoliator
% Tip Burn
% Total Necrosis
% Mottled Needles
% Healthy Needles 1
F
Ratio
7.5576*
2.2318
1.2522
.0479
.8059
2.7653
2.7453
.6829
.1085
5. 3700*
4.1453
1.2915*
Percent
Variance
Component
21.
6.
-.
-11.
-1.
3.
-1.
-5.
-10.
6.
11.
10.
45
06
80
99
28
57
65
85
38
97
19
21
5
7
6
10
1
15
1
8
7
1
19
6
Level 3
Between
Plots Within
Col lections
F
Ratio
.9252*
.2439*
.2892*
.3693*
.5374
.1392*
.5825
.1998*
.7407*
.4337
.5443*
.4417*
Percent
Variance
Component
13.65
27.74
14.20
35.68
5.55
24.34
13.01
31.03
32.04
5.68
24.19
15.52
Level 2
Between
Withi
F
Ratio
1.7318
5.5503*
1.2935
1.4814.
.8097
.2987
.3352
1.9918
.6565
1.0276
.4014
.1244
Tree Ages
n Plots
Percent
Variance
Component
2.49
7.99
1.30
2.49
-2.09
-6.87
-13.78
4.41
-4.21
.30
-3.63
-15.19
Level 1
Between Years
Foliage Within
Tree Ages
F
Ratio
.7843
.3939
.7410
1.0661
2.0411*
2.3314*
6.1854*
.9365
3.2129*
2.4593*
1.4591*
5.5676*
Percent
Variance
Component
-2.81
-8.12
-4.66
.96
16.85
16.58
52.14
-.90
25.32
19.66
5.73
42.70
Level 0
Within
Years
Foliage
Percent
Variance
Component
65.20
66.32
89.95
72.85
80.96
62.28
50.28
71.31
57.22
67.36
62.50
46.75

    ^Indicates F ratio significant at p ^ .05

-------
 TABLE 4.21.   SUMMARY OF  RESULTS  OF  THREE-LEVEL  NESTED ANALYSIS OF VARIANCE FOR UPPER CROWN FOLIAGE FROM
              THE 1975 COLLECTIONS  (ANOVA  B),  THE  1976 COLLECTIONS (ANOVA J), AND THE FOUR-LEVEL NESTED
              ANALYSIS OF VARIANCE  BETWEEN THE 1975 AND  1976 COLLECTIONS (ANOVA K).


Variable
Fluoride (ppm)
Sulfur (ppm)
Area (mm2)
MaaHIo I annth fmm^
Level 4

Between
Collections
ANOVA
K
*



Level 3

Between the Plots
ANOVA
K B J
* *
* *
* * *
* * *
Level 2
Between the
Tree Ages Within
the Plots
ANOVA
K B J

* * *


Level 1
Between the Years of
Foliage Within the
Tree Ages
ANOVA
K B J




% Needle Retention

% Basal Necrosis

% Basal Scale

% Defoliator

% Tip Burn

% Total Necrosis

% Mottled Needles

% Healthy Needles
* Indicates F ratio significant at p ^ .05

-------
treatment effects  only between  the  years of  foliage  were  detected  by each
ANOVA, while for percent  tip  burn,  significant treatment effects between the
plots and between the years of  foliage were noted by each ANOVA (Table 4.21).

     Since no significant  treatment  effects  for tree ages  or age of foliage
origin  were  detected, values  for all   tree  ages and  years of  foliage were
combined,  and  the  resultant  mean  values and  95%  confidence  intervals  for
fluoride from upper crown positions from all plots, both collections, are shown
in  Figure  4.11.   Whereas   in  1975 the  plots  were significantly  different in
fluoride content,  in  general,  in 1976,  with  the exception  of plot SE-2, the
plots were not significantly different.

     For  ppm sulfur,  significant treatment  effects between the  plots were
detected by ANOVAs K and J, but not for 1975 data alone (ANOVA B).   ANOVAs B, K,
and J detected significant differences for sulfur between the tree ages.  These
results indicate that changes  in sulfur  concentrations  in 1976 were sufficient
to  cause  the plots themselves to be different, but  the changes were not such
that  the  two year's collections would  be different.  The  only statistically
significant difference between older and younger trees in 1976 was at plot S~5.
The  basic  statistics  for  sulfur concentrations in upper  crown,  younger and
older trees for plot S-5 are arrayed  in Table 4.22.

     The  mean  sulfur  values  for younger and  older trees  for  all  plots are
arrayed  in matrix  form  as  Table  4.23.   For younger  trees, plot SE-4  was
significantly lower than all other plots, while S-5  had significantly greater
concentrations than  all other plots.  Although plot E-l  had more sulfur than
plot  SE-4, it was  lower than plots SE-2, S-3, and S-5.   There were no signif-
icant  differences  between  sulfur concentrations between plots  S-3 and SE-2.
For older trees (Table 4.23),  plot SE-4 was again significantly lower in sulfur
than all other plots.  In  addition to plot SE-4,  plot S-5 had higher levels in
foliage than plot  E-l.  No other significant differences could  be shown for
older trees between plots.

     The  mean values  and  95%  confidence intervals  for  sulfur in upper crown,
trees for 1975 and  1976 collections for each plot are shown in Figure 4.12.  The
increases shown for plots  S-5, SE-2, and S-3  in 1976 over 1975 are significant
at p ^ .05.

     For percent needle retention, the  ANOVA  for 1975-1976 detected a signifi-
cant  treatment effect  between the years of  foliage, as did the ANOVA £or the
1976 collection (Table 4.21).  No significant effect for year of foliage'origin
was detected for the 1975  collection (Table 4.21, ANOVA B).  The mean percent
needle retention and 95% confidence intervals for plot  E-l, both collections,
are shown in  Figure  4.13.   The  mean values for 1975 and 1974 years'  foliage from
plot E-l are essentially the same, while that for 1973 year's foliage is nearly
12 percent lower.   Whereas for the 1975  data, significant differences occurred
usually only between the 1974  and 1972 internodes, the 1976  data generally show
significant differences between  1973 and both  1974 and 1975 internodes.

     For  percent  basal necrosis  (Table 4.21)  significant  treatment effects
were shown for  1975-1976  and 1975 between the years of foliage, but not  for the
1976  collection.   The mean  values  and 95%  confidence intervals  for basal

                                     110

-------
                                                    PLOT
                                                    SE-4
                                                    PLOT
                                                    SE-5
                                                    PLOT
                                                    E-l
                                                      PLOT
                                                      SE-2
                                                PLOT
                                                S-3
                                              2            3
                                              PPM  FLUORIDE
Figuv
ure
kll.    Mean values and  95% confidence
        combined, for  all plots.  1975 (-
intervals for fluoride
    -) and 1976 ( — 4—)
in upper crowns
col lections.
all years foliage

-------
ro
                                           1975
PLOT
SE-4
                                                  1976
                                                         1975
                      PLOT
                       S-5
                                                                          1976

200
| 	 • 	 1 1975 PLOT
E-l
1 	 	 * 	 | 1071; PLOT
SE-2
I.., 	 A 	 	 I 1 07C
| 	 • 	 1 1975 PLOT
S-3
1^ 1 1 r\-7 r
» I i y / u
1 1 1 1 1 1
300 400 500 600 700 800
PPM SULFUR

1
900
    Figure 4.12.    Mean values and 95% confidence intervals  for sulfur in all plots.  Upper crown,  older trees,
                   both collection sets.

-------
                                 1975 COLLECTION
                                                                   1972
                                                                         1975
                                 1976 COLLECTION

                                             h-
                                                                           1974
                                                    1973
                              i
                            60
                           i
                         70
80
 i
90
100
                                         PERCENT NEEDLE RETENTION
Figure 4.13.
Mean  values  and  95%  confidence  intervals  for percent  needle  retention,  1975 and  1976
collection,  upper  crown, for plot E-l.

-------
TABLE 4.22.   BASIC STATISTICS FOR SULFUR CONCENTRATIONS  IN  UPPER  CROWN
             YOUNGER TREES AND UPPER CROWN OLDER TREES,  ALL YEARS FOLIAGE
             COMBINED,  PLOT S-5*
Statistic
N
Maximum Value
Minimum Value
Mean
Variance
Standard Deviation
Standard Error
Upper Crown, Younger
15
1050
550
746
23023
151.7
39.1
95% Confidence Interval 84.0
Median
Standard Error
700
49. 1
Coefficient of Variation 20.3
Standard Error
3.8
Sampling Precision** 11.2
Upper Crown, Older
15
700
450
613
4809
69.3
17.9
38.4
650
22.4
11.3
2.0
6.2

*Di fference between mean values is significant @
between the individual years foliage were not si
pooled.
**Defined as (t „
X
5Sx') 100

p ^ .05. Since differences
gnificant, these values were

                                  114

-------
 TABLE 4.23.   MEAN VALUES FOR TOTAL SULFUR FROM UPPER CROWN YOUNGER AND OLDER
              TREES, ALL PLOTS, 1976 COLLECTION.
 Plot & Mean
   Sulfur
Concentration
Plot &
   Mean Sulfur
    Concentration
                                 Younger Trees
S-3
SE-2       E-
S-5
               SE-4
                                         640
                                    410
                                    620
                               410
                         528
                    410
                         746
                                                                         *.,
                    410
                                         640
                                    620
                         528
               S-5
                                    746
                               746
                    746
                                         640
                                    620
               E-l
                                    528
                               528
                                         640
               SE-2
                                    620
                                  Older Trees
 Plot & Mean
   Sulfur
Concentration
Plot &
   Mean Sulfur
    Concentration
S-3
SE-2       E-l
S-5
               SE-4
                                         590
                                    388
                                    603
                               388
                         533
                    388
                         613
                    388
                                                    603
               S-5
                                               533
                                    613
                                          613
                                                    603
               E-l
                                    533
                                         590
               SE-2
                                    603
 'Indicates mean values shown are significantly different @ p ^ .05
                                     115

-------
necrosis for upper crown positions for the 1975 and 1976 collections are shown
in Figure 4.14.  Although the  mean  values for  1974 year's  foliage  increased
slightly in  1976 over 1975, the values  for  1973 foliage have not  (Figure 4.14).

     For percent mottled needles,  the  ANOVA for 1975 showed significant treat-
ment effects between the  tree ages,  while the  ANOVAs  for 1975-1976 and 1976
indicated significant treatment effects between the years  of foliage,  but the
ANOVA for 1975  did not.

     The mean values and 95% confidence intervals for percent mottled  needles
for the 1976 collections from plot S-3 are shown in Figure 4.15,  which illus-
trates the lack of effect between the tree  ages (compare Figure 4.9).

     For  percent  total  needle necrosis,  the  ANOVA  for  1975-1976  detected
significant treatment effects between  the two  different years' collections and
between the years of foliage.  The ANOVA for 1975 detected significant effects
only between the  tree  ages.   The  ANOVA for 1976 did not indicate differences
between the mean  values  for tree ages but did detect differences between the
years  of  foliage.   For the  individual  years  of foliage,  the  95% confidence
intervals enclosed the mean values for the tree ages at plot  E-l (Figure 4.16)
for the 1975 collection as well as  the  1976 collection.   The mean values and 95%
confidence intervals for older and younger trees combined  are shown  in Figure
4.16  for both  collections   from  plot  E-l.   The  1976  data  in   Figure  4.16
illustrate the gradual  increase in percent total necrosis fo the other plots in
1976.

     For percent healthy needles,  the  ANOVA for  1975-1976 detected significant
treatment effects between the collections.  The 1975-1976  ANOVA did  not indi-
cate differences between the plots,  nor did it  for the 1975 collections.  The
1976  ANOVA  detected  significant  treatment effects  between the  plots.   All
ANOVAs  indicated  significant differences  between the years of foliage (Table
4.21).   The mean  values  and  95% confidence  intervals  for percent  healthy
needles for all plots from the 1976 collection  are shown in  Figure  4.17.

     The results of the analysis of the data from the 1975 collections demon-
strated  that  samples collected  from  the  upper or lower  crown  positions of
individual trees  to  evaluate  the pre-operational baseline conditions  for the 12
characteristics measured  herein  may  be considered samples  which are repre-
sentative of the individual  tree as  a  whole,  with the exception of fascicular
cross-sectional area.  However, if such samples were collected from geograph-
ically separate plots, from different  tree ages within a plot,  or  from differ-
ent ages of foliage, they cannot a priori be considered to be samples from the
same population.

     Furthermore, the results of the data analysis for the 1976 collection and
the  comparison of  both years'  collections show that, with  the  exception of
fluoride, percent total needle necrosis, and percent healthy needles,  the two
collections represent  samples from  the same  population,  even  though  differ-
ences  were  indicated for  certain variables within  the  two collections.  For
example,  treatment  effects  between  the plots  for  sulfur  concentrations were
noted  in  1976  but  not in  1975,  and  treatment effects between  the years of
foliage were noted in 1976 but not in 1975.  These effects,  however, were  not

                                      116

-------
       1974
                           1973
                                           1975 COLLECTION
                                                                    -\  1972

(
j — •» 	 1 iy/o
1 , -A, 	 i Q7/i i O7f, rni i rrTTnN
• „„,.,...„ 1 O7T
	 — — j iy/j
i i 1
3 5 10 15

1
20
                                            PERCENT  BASAL  NECROSIS
Figure 4.14.
Mean values and 95% confidence  intervals  for percent basal necrosis,  upper crown positions.
1975 and 1976 collections, plot  E-l.

-------
oo

1 t
I f
1 *
1 *
1 t
1 '
1 -•
1 *
III III
0 5 10 15 20 25 30
YOUNGER TREES
1 Q73
| 1Q71 OLDER TREES


1 1 1 1 i
35 40 45 50 55
                                            PERCENT  MOTTLED NEEDLES
    Figure 4.15.
Mean  values  and  95%  confidence intervals  for  percent mottled  needles  for upper  crown,
younger and older trees, plot S-3.

-------
              I   »    I  1974


            I          >                | 1973
            1                           '1975 COLLECTION

              |          •               |  1972
                                ^  1975
                                PERCENT TOTAL NEEDLE NECROSIS
                                                                          .j 1974           1976
                                                                                        COLLECTION
                                                                                     •I 1973
                                                    I            i           i            i
                                                    8           10           12           14
Figure 4.16.    Mean values  for  percent total  needle  necrosis  for upper crown,  older  and  younger trees
               combined,  1975 and 1976 collections.

-------
               PLOT
               SE-4
                                                                 1975
                                           1974
          PLOT
          S-5
          PLOT
           E-l
                                             1973
                                                     1975
                  ^  1974
                          ^  1973
                                                               1975
                           -j  1974
                          1973
                   PLOT
                   SE-2
                                                              1975
                                     1974
                                1973
                   PLOT
                    S-3
                                                        1975
                                  1974
                                ^  1973
                   20
 40           60           80

PERCENT HEALTHY NEEDLES
                                                     I
                                                    100
Figure 4.17.
Mean values  and  95% confidence intervals for percent healthy
needles,  1976 collection.
                                    120

-------
sufficient  to permit  the  conclusion  that  the  two years'  collections  were
samples from different populations.

CHEMICAL ANALYSIS OF UNDERSTORY SPECIES

     Table 4.24 depicts the results of fluoride and sulfur analyses and basic
statistics for the  foliage of nine understory  species  collected at the five
permanent sites  during both 1975 and 1976  collection  periods.   Also  on this
table are the results  of fluoride and sulfur analyses of ten other understory
species  which were  not  abundant at all  the ponderosa pine-skunkbush sites;
these species were collected during either  1975 or 1976 at one or more of the
five sites.   The basic statistics for sulfur and fluoride content of these 19
total species is presented  in Table 4.25.  During 1977,  eight separate samples
of  each  understory  species from  each  site will  be collected  for chemical
analysis.

     The mean fluoride concentration found for all  understory  species was below
5 ppm.   Little bluestem  and bluebunch wheatgrass, the  two most  common grass
species  at the ponderosa sites,  usually had  the lowest fluoride  levels (i.e. , <
2 ppm) of any understory species collected.  Sulfur levels found  in foliage of
silver sage, broom snakeweed, and arrowleaf balsamroot usually were higher than
other  common understory species  and  two to three times  higher  than  the two
commonly found grass species.

     In  a study of fluoride- and sulfur-polluted areas in  which ponderosa pine
foliage  and  understory species of skunkbush,  broom  snakeweed,  and bluebunch
wheatgrass were utilized, Gordon et aj_. (1977) demonstrated that the foliage of
the  dominant species  (ponderosa  pine)  accumulates  excessive  levels  of both
sulfur  and  fluoride  at  a  faster rate than  the understory species  during an
exposure time of 15  months.  The researchers do not anticipate any significant
increase of  fluoride or sulfur accumulation  in understory species at the five
ponderosa pine-skunkbush sites until the ponderosa pine foliage reaches two to
three times the baseline  for fluoride and 1.5 times  the baseline for sulfur.

                            DISCUSSION AND  SUMMARY

SULFUR

     Several  past  studies  of air pollution  impacts on conifer species (i.e.,
Linzon,  1973a; Dreisinger,  1965;  Ellertsen  et aj_. , 1972;  Berry et aJL , 1964;
Miller  et aj_. ,  1963;  Carlson,  1972)  have encompassed  one  to  four  of the
growth/health/disease  characteristics  (i.e.,   tip  burn,  needle  retention,
needle  length,  and  percent necrosis)  which have  been used  in  the Colstrip
ponderosa pine sites during the last two years to quantify the variability of
these characteristics  prior to  the incidence of air pollution damage.  Also,
there have  been  several  air pollution studies  of  conifer trees  to determine
sulfur and/or fluoride concentrations in damaged and undamaged foliage (i.e.,
Katz et al_. ,  1952; Carlson, 1974;  Compton et al_. , 1961 ;  Ellertsen  et al_. , 1972;
Knabe,  1968;  Linzon,  1972,  1973b).

     Linzon's sulfur studies in the Sudbury  area of Ontario revealed that 15-
month-old jack pine foliage (Pinus banksiana),  growing 27  km  northeast and 160

                                     121

-------
TABLE 4.24.  FLUORIDE AND SULFUR CONCENTRATIONS IN UNDERSTORY SPECIES FROM THE 1975 AND 1976 COLLECTIONS.
SE-4
Species
Silver Sage

Broom Snakeweed

Chokecherry

Skunkbush

Idaho Fescue

Bluebunch Wheatgrass

Prairie Sage

Arrowleaf Balsamroot

Little Bluestem

Lupine

ppm F
1975
1976
2.3
2.3
—
1.7
1.8
2.4
6.4
0.9
0.8
1.4
1.1
1.7
2.2
3.1
—
4.0
—
—
—
2.1
ppm S
1975
1976
1300
1950
	
950
650
800
950
900
700
950
800
950
900
1000
	
2400
	
	
	
1400
s-
ppm F
1975
1976
3.
3.
1.
--
--
--
1.
1.
--
2.
2.
0.
5.
4.
4.
4.
--
--
2.
2.
0
0
5
-
-
-
9
7
-
0
8
9
7
0
9
1
-
-
1
0
5
ppm S
1975
1976
1300
1550
1675
	
	
	
950
950
	
800
1100
750
1150
1000
1150
750
	
	
1100
950
E-l
ppm F ppm S
1975 1975
1976 1976
1.3 1300
1.8 1400
— 	
— 	
— 	
— 	
1.8 1800
700
— 	
— 	
1.1 500
	 	 _
0.9 1000
— _ —
— 	
— 	
2.5 300
3.3 250
— 	
5.5 1150
SE-2
ppm F
1975
1976
2.0
2.7
2.2
4.9
2.5
	
4.1
4.9
—
—
2.1
2.1
4.2
_ ~ _
—
—
2.3
1.2
—
2.1
ppm S
1975
1976
1400
1250
1800
2500
600
	
800
1000
	
	
600
600
1100
_ _ —
	
	
400
300
	
950
s-
ppm F
1975
1976
--
2.
--
2.
0.
4.
3.
2.
--
2.
--
2.
4.
— ~
--
—
1.
1.
4.
3.
-
5
-
3
7
7
6
0
-
2
-
5
0
~
-
-
2
0
9
4
3
ppm S
1975
1976
	
850
	
1200
300
500
600
600
	
550
	
700
650
"• — — ~
	
	
300
400
750
800
                                                                                            (continued)

-------
TABLE 4.24.   (continued)
                              SE-4             S-5             E-1             SE-2            S-3

                         ppm F   ppm  S    ppm  F  ppm S   ppm F   ppm S   ppm F   ppm S   ppm F   ppm S
      Species             1975    1975     1975    1975    1975    1975    1975    1975    1975    1975
                          1976    1976     1976    1976    1976    1976    1976    1976    1976    1976
Scurf Pea                  ---     	      ---    	     2.8     950     1.9     900     —    	
                                  	      ---    	     4.7    1350     ---    	     2.9     900

Yucca                      ---     	      2.8     800     ---    	     ---    	     ---    	
                                  	      1.9     700     5.6     750     —    	     1.6     650

Red Three-Awn              	     	      ---    	     2.9     500     ---    	     	    	
                                  	      ---    	     1.7     650     ---    	     ---    	

Needle and Thread          —     	      —    	     2.7     700     2.2    1100     —    	
                                  	      ---    	     ---    	     4.4    1200     ---    	

Vetch                      ---     	      ---    	     2.4    1200     —    	     —    	


Fringed Sage               —     	      —    	     —    	     —    	     —    	
                           2.6     1400      2.4    1250     5.3    1250     1.8    1100     1.9    1200

Green Rabbit Brush         —     —       —    	     —    	     —    	     —    	
                                  	      2.0    1550     ---    	     ---    	     ---    	

Rocky Mountain Juniper     —     	      —    	     —    	     —    	     —    	
                                  	      —    	     ---     600     1.9     800     —    	

Big Sagebrush              —     	      —    	     —    	     —    	     —    	
                                  	      ---    	     —-    	     2.4    1300     ---    	

-------
TABLE 4.25.   UNDERSTORY SPECIES, PONDEROSA PINE - SKUNKBUSH SITES.

Species and Year
of Collection
Silver Sage

Broom
Snakeweed
Chokecherry

Skunkbush

Bluebunch
Wheatgrass
Prairie Sage

Little
Bluestem
Lupine

Scurf Pea

Idaho
Fescue
Arrowleaf
Balsamroot
Yucca

Red Three-Awn

Needle-and-
Thread
Vetch
Fringed Sage
Common
Rabbi tbrush
Rocky Mountain
Juniper
Big Sage
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976
1975
1976

1976

1975
1976
Mean
F"
2.15
1.85
1.85
2.97
1.67
3.55
3.56
2.38
1.78
1.8
3.4
3.55
2.0
1.83
3.5
3.0
2.35
3.8
.8
1.87
4.9
4.05
2.8
3.03
2.9
1.7
2.45
4.4
2.4
2.8

2.0

1.9
2.4
S
1230
1400
1737
1550
516
650
1020
830
750
750
960
1000
333
316
925
1050
925
1125
700
766
1150
1575
800
700
500
650
900
1200
1200
780

1550

700
1300
Standard
Deviation
F~
.70
.49
.49
1.7
.91
1.63
1.88
1.75
.83
.68
1.87
.64
.7
1.27
1.97
1.5
.63
1.27
	
.42
	
.07
	
2.23
	
	
.35
	
	
1.44

	

	
	
S
216
403
88.3
832.1
189
212
459
171.7
264
147.2
198. 1
0
57.7
76.3
247
231.1
35.3
318
	
202
	
1166
	
50
—
—
282
	
	
713.6

	

141.4
	
Standard
Error
F"
.35
.35
.35
.98
.52
1.15
.84
.87
.42
.34
.84
.45
.40
.73
1.4
.67
.45
.9
	
.24
	
.05
	
1.29
	
	
.25
	
	
.64

	

	
	
S
96.7
180.2
62.5
480.4
109
150
205.3
76.8
132.2
73.6
88.6
0
33.3
44.0
175
103.6
25
225
	
116.6
—
825
—
28.8
—
—
200
—
—
318. 1

* —

100
—
                               124

-------
km west of the non-ferrous smelters, contained sulfur averaging 2,200 to 1,000
ppm, respectively.  Jack  pine  foliage  of the same age collected at two inter-
mediate-distance  sampling sites, 45  and 67 km  northeast,  had average sulfur
concentrations  of  1,700  and  1,300  ppm, respectively,  while three-month-old
foliage from these same two intermediate sites had average sulfur concentrations
of 1,500 to 1,400  ppm respectively.

     Katz ejt al.  (1952) worked with both field-grown and  potted ponderosa pines
and fumigated with S02 concentrations of 22 to 500 pphm  for durations of 35 to
1,656  hours (1.4  to  69  possible  day  periods),  respectively.   These  studies
revealed  that  the most  elevated  sulfur  levels  (5,600 ppm)  were found  in
one-year-old  foliage  exposed  to  the  lowest concentration  (22 pphm)  for  the
longest period of  time (1,656 hours).

     Katz's fumigation  studies also revealed that one-,  two-, and three-year-
old ponderosa pine foliage fumigated for  96 hours with S02 concentrations of 500
pphm contained 120, 40, and 300 ppm of sulfur, respectively.   In these studies,
there  appeared  to be  a relationship between excessive sulfur accumulation and
the  incidence of  the  S02-caused needle necrosis that Katz  was  attempting to
determine.  For instance, ponderosa pine  trees fumigated with 500 pphm of sulfur
for 96 hours developed no foliage necrosis but  had 530 ppm of sulfur  in the one-
year-old needles  and 520 ppm in the two-year-old  needles  after the fumigation.
In another S02 fumigation study utilizing 500 pphm for 40  hours, Katz found that
100  percent  necrosis  occurred,  and  in  the  one-,  two-,  and  three-year-old
foliage, the sulfur levels were found to be elevated above 750 ppm.

     In all of Katz's, as well  as Linzon's (1958 to 1973)  reported S02 field and
fumigation  studies, no  details  were  presented  on  the  variation  of  sulfur
concentrations  in healthy foliage  of  pines within  and  between  the different
years'   growth.   Thus,  it  is  difficult  to  compare  the  variability  in  sulfur
concentrations of ponderosa  pine from the Col strip  study  area with either of
their studies.

     Another field study  in the area of the Trail, B.C.,  smelter (besides  that
of Katz) was carried out by U.S. Forest Service personnel (Sheffer and Hedgcock
1955) between 1928 and 1936.  One of the forest species collected and studied by
these  two  investigators  was ponderosa pine.   Sulfur analysis of damaged  and
undamaged ponderosa pine  foliage revealed that foliage with concentrations of
2,700 to 2,300 ppm collected at sites 18 and 37 km,  respectively,  south of the
emission source were damaged but that foliage with 2,000  to 1,600 ppm collected
42 to 70 km, respectively, south of the source were  undamaged.  Baseline sulfur
content of healthy ponderosa pine foliage collected by Scheffer and Hedgcock 79
to 97  km  south of the source  was  found  to  range from 800 to  1,000 ppm.   Over
25,000 ponderosa  pine  trees  were utilized in this eight-year field  study,  and
the  various  types of  damage   include  decreased  cone production and  seedling
reproduction, loss of annual  increment, premature needle  casting, and increased
sulfur content of  the foliage.   This study was one of the first and  only  field
studies in which the investigators set up their studies and carried them out to
compare and understand the variabilities of pathologies at control sites versus
the impact sites.
                                     125

-------
     Some field  investigations  during the  last 15 years,  such  as  Carlson's
(1974) study on damage to ponderosa pine  from a small  sulfur dioxide source in
the Missoula Valley  of  Montana,  demonstrated that elevated sulfur content of
pine (Pinus  ponderosa) is related to the  severity of the foliage damage; other
studies, such  as those  by  Ellertsen  (1972)  and Berry and  Hepting  (1964) on
sulfur and  fluoride  levels  in damaged white pine  (Pinus  strobus) foliage in
Tennessee,  did  not demonstrate this.

     To date,  after  an  intensive search  through the sulfur dioxide pollution
literature  involving  both  field  and controlled fumigation chamber investiga-
tions, the investigators have not found any studies dealing with S02 damage to
plants that quantify the variability of baseline sulfur content within foliage
of any conifer  species between different years'  foliage or  between  foliage from
different  control   sites.   However,  numerous  field  investigators  and  some
control  fumigation   investigators  have  utilized the  sulfur content  of  S02-
damaged foliage as one of the major indices of their pollution  studies.  If the
variability  of the   sulfur  content of foliage  from controlled  areas  is not
known, then  sulfur  levels  in damaged foliage  is much  less useful for under-
standing if  and when  airborne sulfur emissions are affecting ecosystems.

     Every  field  investigator   mentioned  above used  at  least   one  foliage
pathology or characteristic  to  demonstrate  air pollution impact  on the study
sites.  Also,  each was  studying  an area  which  was  being severely impacted by
the  emissions  from  one  or more  large  stationary  sources.   Investigations in
areas  where acute fumigation damage  is  occurring may not  require  an under-
standing of the baseline quantities  of needle pathologies or the variability of
those  pathologies  between  different  years'  foliage,  different-sized trees,
different  crown  positions,  and  different   sites.  However,  these  baseline
quantities are absolutely essential for understanding  if and where chronic air
pollution damage is  occurring within ecosystems.

     The baseline studies on the ponderosa pine  sites clearly demonstrated that
abiotic and biotic causal agents normally occurring in a pristine  area cause a
measurable amount of damage to pine foliage; this damage has to be quantified
prior to understanding if and where chronic air  pollution damage is occurring.

FLUORIDE

     Fluoride  studies  from both field and  fumigation  chamber investigations
have  revealed  that  this element is the most phytotoxic of the normal  gaseous
emissions from large stationary sources of air pollution  (Lione et al_.  , 1962).
More controlled S02  fumigation studies have been completed and published than
fluoride studies,  but  there  are more published  field  studies  dealing  with
fluoride pollution than with any  other phytotoxic gas because:   (1) Gaseous and
particulate fluorides are the major emissions of large stationary  sources such
as aluminum  and  steel  smelters,  and phosphate fertilizer, animal supplement,
and P205 plants; (2) in most cases, baseline fluoride  levels in vegetation are
below  10 ppm,  and fluorine has  not proven  to  be an  essential ion  for plant
metabolism  or  structure, and  (3)  fluoride  pollution problems,  in  many  past
instances,   have  involved domestic  animals  or  agricultural  crops,  provoking
intensive scientific investigation.
                                     126

-------
     Fluoride is not a major pollutant within the emissions  of coal-fired power
plants or non-ferrous smelters, and few studies around either pollution source
have included this element or compound.   However, because of the phytotoxicity
of even  low  concentrations  (.5 ppb) of  fluoride in  the ambient air to plant
pollen (Facteau et al.,  1973, 1977), its low translocation rate from the leaf
receptor to  other parts of  the  plant  (thus remaining  in  the  leaf),  and its
rapid build-up  in  the  food chain, the  investigators  have  spent considerable
time determining  the baseline  levels of this  element  in  the  ponderosa pine-
skunkbush ecosystems of  the Colstrip area.

     The fluoride  studies  in upper crown foliage at the ponderosa sites were
set up  to  best utilize  the evidence gained by Knabe (1968) in forest canopy
fluoride studies which demonstrated that the most elevated  fluorides in conif-
erous foliage were in the highest and most exposed upper crown  foliage.  Field
studies  in the  vicinity of aluminum plants by various  investigators (Shaw et
al_. ,  1951;  Adams et al_. ,  1952;  Compton et al_. ,  1961;  Facteau  et  al_. ,  1976;
Carlson, 1972;  Gordon,  1974)  in the United States have demonstated that pon-
derosa  pine  is  very  susceptible to fluoride pollution.  Current and one-year-
old foliage of ponderosa pine with fluoride levels less than 20 ppm have been
reported to  manifest needle  tip necrosis  by  Compton et  al_.  (1968),  Carlson
(1972), Hindawi (1973),  and Gordon et ah (1976).

     Partitioned  ponderosa  pine  needles damaged by fluoride  pollution have
been  reported  by  Gordon  et  al_.   (1976) to contain up to three  times  more
fluoride in  the 3 cm tip portion  than  whole  needles and over six times more
fluoride in  the  needle  tips than  middle  and basal portions.  Both Jacobson et
aj_. (1966) and Compton et al_. (1960) have demonstrated with fluoride-fumigated
gladiolus plants that after fluoride enters the  leaves,  it is  translocated to
the leaf tips where tissue necrosis then occurs.  Jacobson1s  (1966) fluoride
fumigation  studies  with  other   species  of  plants  (cotton  and  tomato)  to
determine fluoride translocation  and partitioning in plant tissues led him to
suggest  that  the  ability of  a  plant  to  readily translocate  and partition
fluoride may influence its  susceptibility or resistance to fluoride  damage.

     The 1976 data from this study demonstrated an increase in fluoride con-
centrations  in ponderosa pine foliage from Kluver (E~l),  McRae (SE-2), and Fort
Howes,  Custer National  Forest (S-5) sites over that occurring in foliage of the
same age during  1975.  This was not accompanied by any significant  increase in
sulfur  levels in the needles at  these  three  sites  between the 1975 and 1976
study periods.   Because  sulfur  taken in from the ambient air  is translocated
out of  the  pine needles, no significant increase is expected  at any of these
sites until  after the  emissions  of the Colstrip coal-fired power plant units
increase from the  estimated  average 1976  level  of .8 tons per  hour  to the
higher potential levels  of +2 tons per hour.

NEEDLE PATHOLOGIES

     While there was a significant increase in both the  amount  of total needle
necrosis and  loss of healthy needles at, all  the ponderosa pine-skunkbush sites
during  1976 over that of 1975, at this time the  increase is not attributed to
phytotoxic emissions of Colstrip  units.   Houston (1974), fumigating white pine
with a  combination of S02 (.025 ppm) and 03 (.05 ppm) for six hours, reported

                                     127

-------
that  all  sensitive  clones  manifested tip  necrosis on  75  to  100  percent of
current year  needles.   However,  these are hypersensitive clones of white pine
and are much more sensitive to both of these gases than ponderosa pine (Miller
et al_. , 1975).  Constantindou et al_. (1976) reported a significant decrease in
the chlorophyll content of red pine (Pinus resinosa)  cotyledons  when fumigating
with  0.5  ppm  S02  for both 15-and 30-minute periods.  Kress and Skelly (1977),
when  fumigating loblolly pine (Pinus taeda) with S02  (0.14 ppm) + 03  (.05 ppm) +
N02  (0.10 ppm) for  six hours per  day for 28  consecutive  days reported  a 26
percent reduction in growth.   The studies of both Constantindou et 
-------
                                  REFERENCES

Adams, D.  F. ,  D.  J. Mayhew, R. M. Gnagy,  E.  P.  Richey,  R.  K.  Koppe,  and  I.  W.
     Allan.  1952.   Atmospheric pollution  in  the  ponderosa pine blight  area,
     Spokane County, Washington.  Ind. Eng. Chem.  44:1356-1365.

Berry,  C.  R.  and  G.  H.  Hepting.    1964.   Injury to  eastern  white pine  by
     unidentified atmospheric constituents.  Forest Sci.  10:1-13.

Carlson, C. E.   1972.   Monitoring Fluoride  Pollution in Flathead National Forest
     and Glacier  National  Park.   U.S.  Forest  Service,  Division of  State  and
     Private Forestry, Missoula, Montana,   pp.  1-25.

Carlson, C. E.   1974.   Sulfur Damage to Douglas-Fir  Near a Pulp and Paper Mill  in
     Western  Montana.   U.S.  Forest  Service,  Division  of  State  and  Private
     Forestry, Missoula, Montana.   Report No. 74-13.  41 pp.

Compton, 0. C.  and L. F. Remmert.  1960.  Effects of air-borne fluorine on injury
     and fluorine content  of  gladiolus leaves.   Proc.  Amer.  Soc.  Hort.  Sci.
     75:663-675.

Compton, 0.  C.,  L.  F.  Remmert,  and  J.  A.  Rudinsky.  1961.  Needle  Scorch  and
     Condition of Ponderosa Pine Trees in The Dalles Area.  Ore. Agr.  Exp.  Sta. ,
     Oregon State University,  Corvallis.  Misc. Paper 120.

Compton, 0. C., F. W. Adams, W.  M.  Mellenthin,  S.  Elliott, N. Chestnut, and  D.  W.
     Bonney.    1968.    Fluorine  Levels in  Crops  of The  Dalles Area,  1965-67:
     Cherry, Peach,  and  Pine  Trees and the  Ambient Air.   Ore. Agr.  Exp.  Sta.,
     Oregon State University,  Corvallis.  Spec. Rpt. 261.  40 pp.

Constantindou,  H. ,  T.  T.  Kozlowski,  and  K.  Jensen.  1976.  Effects  of sulfur
     dioxide  on  Pinus resinosa  seedlings  in  the cotyledon  stage.   J.  Env.
     Quality.  5:141-144.

Dreisinger,  B.  R.   1965.  Sulfur  Dioxide Levels  and  Effects  of  the Gas  on
     Vegetation Near Sudbury, Ontario.  Presented at 58th Annual  Meeting of  the
     Air Pollution Control Association.  Paper No. 65-121.

Ellertsen,  B.  W. ,  C.  J.   Powell, and  C.  L. Massey.  1972.   Report on study  of
     diseased white pine  in Tennessee.  Mitt.  Forstl.  Bundesversuchanst.,  Wien.
     97:195-206.

Environmental Protection  Agency.  1976.   The Bioenvironmental Impact of a  Coal-
     Fired Power Plant, Second  Interim Report, Colstrip,  Montana—June 1975  (R.
     A.  Lewis,  N.  R.  Glass, and A.  S.  Lefohn, eds.).  Environmental  Research
     Laboratory, Office of  Research and Development, Corvallis,  Oregon.  EPA-
     600/3-76-013.

Facteau, T.  J. , S.  Y. Wang,   and  K.  E.  Rowe.   1973.  The effect of  hydrogen
     fluoride on  pollen  germination and pollen tube growth  in Prunus  avium  L.
     cv.  "Royal Ann."   J.  Amer.  Soc.  Hort.  Sci.  98(3): 234-236.
                                      129

-------
Facteau,  T.  J.  and  W.  M.  Mellenthin.   1976.   Fluoride Investigations  in  The
     Dalles  Area  1968-1974.   Ore.  Agr.  Exp.  Sta.,  Oregon State  University,
     Corvallis.  Tech.  Bull.  132.  56pp.

Facteau,  T. J. and R. E.  Rowe.  1977.   Effect of hydrogen fluoride and hydrogen
     chloride  on pollen tube growth  and  sodium fluoride on pollen germination in
     "Tilton"  apricot.   J. Amer. Soc. Hort. Sci.   102(1):95-96.

Gordon,  C. C.  1974.   Environmental  Effects of Fluoride:  Glacier National Park
     and Vicinity.   U.S.  Environmental  Protection  Agency,  Region VIII, Denver,
     Colorado.  EPA-908-1-74-001.  150pp.

Gordon,  C. C. , P.  C.  Tourangeau, and C.  E.  Carlson.   1976.  Fluoride Emissions of
     Coal-Fired  Power Plants  and  their  Accumulation in and/or Impact Upon Plant
     and  Animal   Species.    Paper  presented  at   American  Chemical   Society
     Symposium, San Francisco, California.   10 pp.

Gordon,  C.  C., P. C. Tourangeau, J.  J.  Bromenshenk, C.  E.  Carlson,  and P.  M.
     Rice.  1977.  Pre- and Post-Operational Investigations  into the Impacts of
     Coal-Fired  Power  Plant Emissions  in the  Northern Great  Plains.   Paper
     presented to the  National Academy of Sciences/National  Research Council
     Meeting,  Washington, D.C.  85 pp.

Hindawi, I.  J.  1973.   Fluoride in Glacier  National  Park:   A  Field Investiga-
     tion.    U.S.   Environmental   Protection  Agency,   Region   VIII,   Denver,
     Colorado.  EPA-908-1-73-001.  69pp.

Houston, D.   B.    1974.    Response  of  selected  Pinus   strobus  L  clones  to
     fumigations with sulfur dioxide and ozone.  Can. J.  For. Res.  4:65-68.

Jacobson, J.  S. ,  L.  H.  Weinstein, D. C.  McCune,  and A. E. Hitchcock.  1966.  The
     accumulation  of fluoride by plants.   J. Air  Poll.  Contr.  Assoc.   16:412-
     417.

Jones,  B.  L.  and  C.   C.  Gordon.  1965.   Embryology and  development  of  the
     endosperm haustorium of Arceuthobium Douglasii.  Amer.  J. Bot.  52(2): 127-
     132.

Katz, M.  and A. W. McCallum.  1952.   The effects of sulfur dioxide an conifers.
     Proc.  U.S.   Technical  Conference  on Air  Pollution  (New York,  Toronto,
     London),   pp.  84-96.

Kay, C.   E.,  P. C. Tourangeau,  and C.  C.  Gordon.   1975.   Fluoride levels  in
     indigenous  animals and  plants collected from  uncontaminated  ecosystems.
     Fluoride.  8(3):125-131.

Knabe,  W.   1968.   Experimentelle Prufung  der  Fluoranreicherung in Nadeln  und
     Slattern von Pflanzen in Abhangigkeit von deren Expositionshoh  uber Grund.
     Miedzynarodowej  Konf.,  Wplyw.  Zanieczyszczen  Powietrza  N  Lasy, 6th,
     Katowice,  pp. 101-116.
                                      13Q

-------
Kress, L. W. and J.  M.  Skelly.  1977.  The Interaction of 03, S02 and N02  and  its
     Effects on  the Growth of Two  Forest Tree Species.   Cotterell Centennial
     Symposium—Air Pollution and Its  Impact on Agriculture,  pp.  128-152.

Leone, I. A., E.  Brennan, and R.  H. Daines.  1962.  Effects of Air Pollution on
     Vegetation.   Presented  at  the  Air  Pollution Control  Association, Mid-
     Atlantic States Section  Meeting,  Wilmington,  Delaware.  6 pp.

Linzon, S. N.  1958.  The Influence of Smelter  Fumes on the Growth of White Pine
     in the Sudbury Region.  Joint Pub., Ont.  Dept. Land  Forests,  Ontario Dept.
     Mines, Toronto, Ontario.  45 pp.

Linzon,  S.  N.  1972.   Pre-Pol1ution Background Studies in  Ontario.  Proc. 27th
     Ann. Meeting of the  Soil Conservation Society of America,  pp. 1-7.

Linzon, S. N.  1973a.  The Effects of  Air  Pollution of Forests.  Paper  presented
     at the 4th Jt.  Chem.  Eng. Conf. pp.  1-18.

Linzon,  S.  N.   1973b.   Sulfur  Dioxide Air Quality Standards for  Vegetation.
     Paper  presented  at  the  66th Annual  Meeting  of the  Air  Pollution  Control
     Association.  Paper No.  73-107.   19 pp.

Materna,  J.   1973.    Criteria for  Characterization of  Pollution  Damages  in
     Forests.  Proc.  Int.  Clean  Air Congr. 3rd, Dusseldorf,  West  Germany,   pp.
     A121-A123.

Maughan,  D.   1977.   Personal Communication and monthly  computer readouts  for
     S02,  03,   N02,  HF,  and  particulate  for  the Burlington  Northern   air
     monitoring site at Colstrip,  Montana.

Miller,  P.  R. ,  J.  R.  Parmeter,  Jr.,  0.  C. Taylor, and  E. A.  Cardiff.  1963.
     Ozone  injury to the foliage of ponderosa  pine.  Phytopathology.  53:1072-
     1076.

Miller,  P.  R. and J.  R.  McBride.   1975.   Effects  of air pollutants  on forests.
     Chapter 10,  pp.  195-235.   j.n Responses of Plants to  Air Pollution  (J. B.
     Mudd and T.  T.  Kozlowski, eds.).  Academic  Press, New York.

Montana  State Department of  Health and  Environmental Sciences.  1975.  Findings
     of  Facts and Conclusions of Law.   Colstrip Units 3 and 4 Proceedings Before
     the  Board of  Natural Resources  and  Conservation, and  the  Board  of  Health
     and Environmental Sciences.   3 pp.

Northern  Cheyenne  Research Project.   1976.   The  Northern  Cheyenne  Air  Quality
     Redesignation   Report  and   Request.   Prepared   for   the   Environmental
     Protection Agency, Region VIII, Denver, Colorado,   pp.  5-6.

Scheffer, T. C.  and G. G.  Hedgcock.   1955.  Injury to Northwestern Forest Trees
     by  Sulfur Dioxide from  Smelters.   U.S. Forest Service Tech. Bull.  1117:1-
     49.
                                       131

-------
Shaw, C. G. ,  G.  W.  Fischer, D. R. Adams, M.  F.  Adams,  and D.  W.  Lynch.   1951.
     Fluoride injury to ponderosa pine:  A summary.   Northwest Sci.  15:156.

Sokal,  R.  S. and F.  J. Rohlf.   1969.  Biometery, the Principles and Practices of
     Statistics  in   Biological  Research.   W.  H.   Freeman  and  Company,  San
     Francisco.  775 pp.
                                     132

-------
                                                   APPENDIX A

                LESS THAN ONE-HOUR EXPOSURE OF SO? ALONE AND IN COMBINATION WITH OTHER POLLUTANTS
       TIME
CONCENTRATION  RESPONSE OF VEGETATION
PLANT SPECIES
   AUTHOR
    1.   15 min.   .5 ppm S02
CO
CO
               Significant decrease in chlorophyll content of
               cotyledons.   Significant decrease in chloro-
               phyll content of cotyledons at 1, 3, and 4 ppm.
               Highly significant reductions in chlorophyll
               in primary needles at .5, 1, 3,  and 4 ppm.

               Highly significant decrease in increment of
               total dry weight (growth)of cotyledons at .5,
               1, 3, and 4 ppm.  Highly significant decrease
               in increment of total dry weight (growth) of
               primary needles at .5, 1, and 3  ppm; signifi-
               cant decrease reduction at 4 ppm.

               No necrosis of tips of cotyledons and primary
               needles at 1 day after treatment of .5, 1,  3,
               or 4 ppm.
  Red pine
  seedlings

Pinus resinosa
Constant!ndou
  et al.,  1976
    2.  30 min.  .5 ppm S02
               Significant decrease in chlorophyll content of
               cotyledons.  Highly significant decrease in
               chlorophyll content of cotyledons at 1,  3, and
               4 ppm.

               Highly significant decrease in chlorophyll
               content of primary needles at .5, 1, 3,  and
               4 ppm.

               High-ly significant decrease in increment of
               total dry weight (growth) of cotyledons at
               .5, 1, 3, and 4 ppm.  Highly significant
               decrease in increment of total dry weight
               (growth) of primary needles at .5 and 1  ppm;
               significant at 3 and 4 ppm.
  Red pine
  seedlings

Pinus resinosa
Constantindou
  et al., 1976

-------
   APPENDIX  (continued)
       TIME
CONCENTRATION  RESPONSE OF VEGETATION
PLANT SPECIES
AUTHOR
                               37% of  seedlings  had  necrosis  of tips of
                               cotyledons  and  primary  needles  1 day after
                               exposure  to 4 ppm.
3. 1 hour .5 ppm S02 Significant decrease in chlorophyll content
of
in
3,
cotyledons.
chlorophyl 1
and 4 ppm.
Highly
content

significant decrease
of cotyledons at 1 ,

Red pine Constant! ndou
seedlings et al_. , 1976

Pinus resinosa
00
-Pi
               Highly significant decreases in chlorophyll
               of primary needles at .5 and 1  ppm;  signi-
               ficant decreases at 3 and 4 ppm.

               Highly significant decrease in  increment of
               total  dry weight (growth) of cotyledons at
               3 and  4 ppm.   Exposures .5 and  1  ppm were
               significant.   Highly significant decreases
               in increment of total dry weight (growth) of
               primary needles at .5 and 1 ppm;  significant
               at 3 and 4 ppm.

               15% of seedlings had necrosis of tips of
               cotyledons and primary needles  1  day after
               exposure to 3 ppm; 25% had necrosis  after
               exposure to 4 ppm.  "Continuous fumigation
               with S02 at much lower dosages  than  used
               may have inhibitory effects on  seedling
               development and on regeneration of plant
               communities."

-------
     APPENDIX (continued)
        TIME
        CONCENTRATION  RESPONSE OF VEGETATION
PLANT SPECIES
AUTHOR
CO
en
15 min.    .5 ppm S02   Highly significant decrease in chlorophyll
to 2                   content of cotyledons for .5,  1,  3,  and 4
hours.                  ppm.

                       Highly significant decreases in chlorophyll
                       content in primary needles.

                       Highly significant decrease in increment of
                       total  dry weight at 3 and 4 ppm.   Exposures
                       to .5  and 1  ppm were significant.

                       Highly significant decrease in increment of
                       total  dry weight (growth) of primary needles
                       at .5  and 1  ppm; significant at 3 and 4 ppm.

                       15% of seedlings had necrosis  of tips of
                       cotyledons and primary needles 1-day after
                       exposures to 3 ppm; 35% after  exposure to
                       4 ppm.
                                                                                 Red pine
                                                                                 seedlings


                                                                                Pinus resinosa
               Constant!ndou
               et al.,  1976
5.
6.
7.
2
2
2
hours
hours
hours
0.
0.
0.
25
25
05
ppm
ppm
ppm
S02
S02
S02
6.5% of exposed parts i
lesions.
4.5% of exposed parts i
lesions.
Caused tip necrosis on
njured. Needle
njured. Needle
new needles.
Eastern white
pine
Pinus strobus

Jack pine
Pinus banskiana

East. White pine
Pinus strobus

Berry,
Berry,
Costoni
1973
1971
1971
s,

-------
    APPENDIX  (continued)
        TIME
CONCENTRATION  RESPONSE OF VEGETATION
                                               PLANT SPECIES
AUTHOR
         2  hours
 Alone

 0.04 ppm
 Followed 3 min
 later by 0.05
 ppm S02 for 2
 hrs.  and then
 fumigated 24
 hrs.  later
 with a mix-
 ture at same
 concentration.
[Needle necrosis  (author  stated  necrosis
   was  most severe  at  this  exposure).]
co
8. 4 hours


9. 6 hours
per day
for 28
days
0.05 ppm S02
+ 0.05 ppm
ozone

0.14 ppm S02
+ 0.05 ppm
ozone +
0.10 ppm N02
Severe needle necrosis (2-3 cm leaf tip
injured) on new needles, expressed most
vividly 72 hours after exposure.
OH TM PHMDTMATTnM \JTTU MTUCD DHI 1 IITAMTC
01/2 IN LUMB1NA 1 1UN Wll n UmhK PULLU 1 AN 1 o
Significant growth reduction (measured as
height) compared to either ozone and sulfur
dioxide combined, or ozone alone. Needles
were significantly narrower than for any
other exposure. This study is an example
of growth reduction with slight foliar
symptoms. Foliar symptoms response most
sensitive in early July.
East. White pine
Pinus strobus


Loblolly pine
Pinus taeda
(2 wks old)
Sycamore
(1 wk old)
Seedl ings
Costonis ,
1973


Kress and
Skelly,
1977

-------
   APPENDIX (continued)
      TIME
CONCENTRATION  RESPONSE OF VEGETATION
 PLANT SPECIES
  AUTHOR
    10.  .6 hours  0.025 ppm
                 S02 + 0.05
                 ppm ozone
               20-28 day-old white pine needles on ramets,
               exposed 9 am - 3 pm.  All sensitive clones
               adversely affected.  Response judged in terms
               of needle elongation (growth) and foliar
               lesion or tip necrosis.  Author calls
               response synergistic.  0.025 ppm S02 or 0.05
               ppm = threshold; 0.10 ppm 03 = 20% necrosis.
East.  White pine

 Pinus strobus
  Houston,
    1974
11. 22 hours. 100 to 500
ppm S02
aqueous
sol.
Swelling of thylakoid disks in chloroplasts
disintegration of intro-chloroplast membranes
in mature needles. Absence of mitochondria.
Author state 100 - 500 ppm aqueous S02 is
equivalent to 1000-fold lower concentration
in air or 0.1 - 0.5 ppm.
Lodgepole pine
Pinus contorta



Malhotra,
1976
CO
                               Older  needles  more  sensitive.   Photosynthesis
                               reduced/measured  by evolution  of 03.
    12.   Annual    0.06
         Average   ppm
         of  24-
         hour
         measurements
       - 0.009 20 ± 5% loss in radial growth.  Premature
               drop of some needles.  Reduction of above-
               ground plant organs that can  lead to impor-
               tant losses of water in catchment areas (at
               Hamr air monitoring station,  Erzgebirge,
               Germany).  At higher S02 levels (0.026 - 0.037
               ppm, annual average) rapid death of individual
               trees and whole groups of trees (at Brandov
               air monitoring station, Erzgebirge, Germany).
  Fir forests
   Abies sp.
Materna,
  1973

-------
    APPENDIX (continued)
GO
00
TIME CONCENTRATION
13. Average 0.045 ppm
30 min.
S02 cone.
for period
of 1954-
1963
14. 4-8 hr/ 0.1 ppm
day, 5 S02 + 0. 1
days/wk ppm ozone
4-8 weeks
RESPONSE OF VEGETATION
Foliar injury, premature defoliation,
reduced photosynthesis, decreased radial
and volume growth - 10 cubic feet per acre
per year ($1,171,000 in 1963), premature
death, rough bark, purple bark, reduced
blister rust, small leaf growth, more severe
symptoms with high pollution levels. 750
square miles in Sudbury, Ontario.
16% needle necrosis (chlorotic, yellow
spots, current year needles thin and twisted)
shedding of older needles far exceeded
damage responses from single exposures.
.1 ppm S02 injured 4% of needle area; .1 ppm
03 injured 3% of area.
PLANT SPECIES AUTHOR
East. White pine Linzon,
1971
Pinus strobus

East. White pine Dochinger
et al . ,
Pinus strobus 1970


-------
                             REFERENCES - APPENDIX

Berry,  Charles  R.   1971,   Relative  Sensitivity  of  Red,  Jack  and White  Pine
     Seedlings to Ozone and Sulfur Dioxide.   Phytopathology.  Vol. 61.  pp.  231-
     232.

Berry,  Charles  R.   1971.   Age  of Pine  Seedlings  with Primary Needles  Affects
     Sensitivity to Ozone  and  Sulfur Dioxide.  Phytopathology.   Vol.  64.   pp.
     207-209.

Constantindou,  H.,T.  T.  Kozlowski,  and  K.  Jensen.   1976.   Effects of  Sulfur
     Dioxide  on  Pinus resinosa Seedlings  in the  Cotyledon Stage.  Journal  of
     Environmental Quality.  Vol.  5.  pp. 141-144.

Costonis,  Arth     .   1971.  Effects  of Ambient  Sulfur Dioxide  and Ozone  on
     Eastern  Wl K>-  Pine  in a Rural  Environment.  Phytopathology.   Vol.  6.   pp.
     717-720.

Costonis,  Arthur  C    1973,   Injury to Eastern White  Pine by Sulfur  Dioxide  and
     Ozone Alone  ^ .;  in  Mixtures.  European  Journal  of  Forest  Pathology.  Vol.
     3.  pp.  50-55.

Dochinger, L. S. F  w   Rpnder,  F.  L. Fox, and W.  W. Heck.  1970.   Chlorotic Dwarf
     on  Eastern White s'ine caused by an Ozone  and Sulfur Dioxide  Interaction.
     Nature.   Vol.  225.  .January 31.   p.  476.

Houston,  Daniel  B.    19/4.   Response  of Selected  Pinus strobus  L.  Clones  to
     Fumigations  with Sulfur  Dioxide  and Ozone.  Canadian  Journal of  Forest
     Research.  Vol. 4.  pp. 65-68.

Kress,  Lance W.  and  John  M  Skelly.   January 13,  14,  1977.    The Interaction of
     03, S02, and N02 and  Its  Effect on the  Growth of Two Forest  Tree Species.
     Cottrell   Centennial   Symposium,  Air   Pollution  and   Its  Impact   on
     Agriculture,   pp. 81-86.

Linzon,  Samuel  N.  1971.   Economic Effects of Sulfur  Dioxide on Forest  Growth.
     Journal  of the Air Pollution  Control Association.  Vol. 21:  No. 2.   pp.  81-
     86.

Malhotra,  S.   S.   1976.   Effects  of Sulfur Dioxide on Biochemical Activity  and
     Ultrastructural   Organization   of   Pine   Needle   Chloroplasts.     New
     Phytologist.   Vol. 76.  pp.  239-245 plus plates.

Malhotra,  S.   S. and D. Hocking.   1976.  Biochemical  and Cytological  Effects  of
     Sulfur  Dioxide on  Plant Metabolism.   New Phytologist,  Vol.  76.  pp. 227-
     237.

Materna,  Jan.   1973.   Criteria   for  Characterization of  Pollution Damage  on
     Forests.   Proc.  Int.  Clean  Air Conqr.  3rd;  Dusseldorf,  W.  Germany,   pp.
     A121-A123.


                                      139

-------
                                  SECTION 5

               INVESTIGATION OF THE IMPACT OF COAL-FIRED POWER
                PLANT EMISSIONS UPON INSECTS:  ENTOMOLOGICAL
                STUDIES IN THE VICINITY OF COLSTRIP, MONTANA

                                     by

                              J.  J. Bromenshenk


                                INTRODUCTION

     Although this report  concentrates  on  the entomological  investigations in
the vicinity of Colstrip,  Montana,  from June 1,  1975,  to March  1, 1977, it also
includes  a  synopsis  of earlier  reports and,  therefore,  serves as  the  major
report  of the  baseline  phase (pre-operational)  and  monitoring phase  (post-
operational)  of  the  study.   The  rationale,  methodologies,   and  preliminary
results were published in  the  Second Interim Report (EPA,  1976).  Insect studies
at the sulfur dioxide treatment plots (ZAPS) are presented in section  14 of this
publication,  while   related   fungal   and   vegetational   investigations  are
elaborated in sections 4 and 13.

     The unifying hypothesis of the EPA-funded Coal-Fired Power  Plant  Project is
that methods can be developed to predict the bioenvironmental  impacts of coal-
fired power plants before any damage occurs.  The basic assumption  is that air
pollutants are major contributors to damage, defined as negative changes  in the
components  and  processes  of  an  ecosystem.   Our  ancillary hypothesis  is that
methods can  be  developed  to serve this purpose,  based on the  use of indicator
species of plants, fungi,  and  insects as early warning  systems and as continuous
monitors of atmospheric pollutants.

     The  southeastern Montana  area  was  essentially  free  of  anthropogenic
pollutants in the  ambient air prior to the construction of the Colstrip power
plants (U. S.  Environmental  Protection Agency (EPA),  1976; Montana Department of
Health and  Environmental  Sciences  (DHES),  1972;  Montana Department of Natural
Resources  and Conservation (DNRC),  1974;  Westinghouse  Environmental  Systems
(WES), 1973).  Therefore,  it was hypothesized that if we measured,  quantified,
and  mathematically  described   pollutant-sensitive  characteristics  of  key
indigenous species of the  flora and fauna of the  major ecosystems (cool season-
short grass and  ponderosa  pine-skunkbush) prior to the  operation of the Colstrip
generating  units,   we   could   monitor  any  observable  changes   in  these
characteristics  as the ecosystems were subjected to pollutant challenge.  Also,
by the  appropriate  analysis   of  temporal   and   spatial  factors,  we  hoped to
correlate these  changes  with the  level  of air pollution and relate them to other
ecosystem components.

                                      140

-------
     The  uniqueness  of this  approach lies in the  opportunity  it  presents to
participate  in  a multi-disciplinary, multi-agency  study,  to  conduct baseline
studies in a basically uncontaminated area, and to pursue these studies during
initial and subsequent impact of emissions from the Colstrip power  plants.  The
use of  biological  indicators  or estimators is not  new,  and the rationale for
their  use  was stated  well  by  Thomas e_t  al.   in  a 1974   review of biological
indicators of environmental  quality:

          A continuous monitor  of  any environmental variable is superior
     to  periodic  sampling  because  the  concentration  of  pollutants  or
     intensity of  stress  varies with time.   Because organisms integrate
     their responses through time and  because they  react to all synergistic
     and antagonistic effects of combined pollutants or stresses, they do
     provide convenient full-time  monitors.   These  biological  indicators
     give  the  actual responses  of individual organisms  or populations,
     rather than predict  biological  responses from physical measurements
     obtained through instrumentation.

     However, it  is essential  that   these indicators be  sensitive to pertur-
bations, practical  to monitor and measure, and  reliable predictors of ecological
and economic  impacts.   To  satisfy these criteria and to  test  the previously
stated  hypothesis,  we proposed  and   carried out,  in progressive  order,  these
study objectives:

     (1) Selection and establishment  of permanent  study sites  (ponderosa pine-
skunkbush, grassland, and apiary) distributed to maximize gradient exposures to
the Colstrip emissions;

     (2)  compilation of  inventories   of  insect and  fungal  populations,  infes-
tations, and damage to indigenous plant species at the  study sites  to provide a
basis  for  the   selection  of  more   specific  groups for   use  as  biological
estimators;

     (3)  analysis   and  selection  of insect  and fungal  populations with  a
diversified but understandable interrelationship with indigenous plant species
in order to examine effects of air contaminants  on  symbiotic associations;

     (4) selection and testing of disease- and injury-causing fungal and insect
species to be utilized for more intensive  study;

     (5) selection and testing  of  beneficial  fungal and  insect  species to be
utilized for more intensive study;

     (6) chemical analysis of the selected indigenous plants, insects, and fungi
to  establish  baseline  levels   and  to discern  any  accumulation of  materials
(fluorides and sulfurs) released by the conversion  of  coal to electricity, and

     (7)  determinations   of physical  (pH)  and  chemical   characteristics  of
precipitation.

     After the  inventory, selection,  and testing phase,  baseline  data during
1974 and 1975 was gathered to determine:


                                      141

-------
     (1)  Growth,   health,   and  pathologies  of  ponderosa  pines,   including
associated fungal and insect populations and damage parameters;

     (2) concentrations  of  fluorides  and  sulfurs in the  dominant  species of
indigenous plants of both the ponderosa pine and the grassland ecosystems;

     (3) concentrations of these substances in an  insect pollinator (honeybees)
of grasslands and croplands, and

     (4) precipitation  chemistry.

     In 1976 the precipitation studies were abandoned, the number of pine study
plots were  reduced and studies were intensified  at  the  S02  fumigation (ZAPS)
plots.   The objective was to increase the depth,  understanding, and precision of
hypotheses  regarding  those  systems, processes,  and  species  which  preliminary
studies indicated were  the components most sensitive or vulnerable to measurable
pol1utant stress.

                                    INSECTS

     At  least 1,400  species  of insects inhabit  the grassland ecosystems of
Montana (personal communication, N.  Reese,  USDA  Rangeland Research Laboratory,
Bozeman,  Montana).   From among these,  we elected  to  study:   (1)  Pest  and
beneficial  insects associated with ponderosa pine, the dominant tree species in
the  Fort  Union  Basin,  and a species susceptible to  damage by phytotoxic gases
and  acid  rains (EPA,  1973);  (2) pollinators,  especially honeybees,  and  (3)
ground-dwelling  beetles,  particularly  members  of   the  families  Carabidae,
Silphidae,  and Scarabaeidae (see section on ZAPS).  These  three sets were chosen
because of  probable  sensitivity to  air pollution and ecological  and economic
importance.

     A  brief review of the economic considerations follows.  Timber resources,
composed entirely of one commercial  species, ponderosa pine,  are harvested  from
the  Custer National  Forest  and the  Northern Cheyenne  Reservation.   Based on
allowable  annual cut  and timber sale data, these pine stands have a wholesale
value of $1,584,000 ($180 per 1,000  board feet) (Gordon,  1975).  Harmful insects
attack  healthy  or weakened  trees,  impairing vitality  and causing mortality.
Bark  beetles, historically,  have  destroyed  more standing  timber  in western
forests than all other insects combined, and defoliators have ranked second as
destructive agents  (Keen,  1952).

     Agriculture is Montana's leading industry,  according to annual reports of
the  Montana Department of Agriculture  Statistical Service.  In  1973 the  cash
receipts for the seven-county area of southeastern Montana totaled $110,042,600
or 10 percent of Montana's marketable produce.  Pollinating insects  are vital to
various crops and indigenous plant species.  The honeybee is  important not  only
to Montana's agriculture, but also to that of California and  other states where
25 percent of Montana's bees are taken in the winter for pollination purposes.

     Montana is  an  important contributor to the nation's honey industry, ranking
first in  the nation  for  honey yield  per colony and fifth for  total yield.
Montana honey is premium grade white "clear" honey and is used to upgrade honey

                                      142

-------
from  other  regions.   Approximately  eight  percent  (6,600)  of  the  honeybee
colonies of Montana are located in our study area.  These colonies  produce about
739,200  pounds  of honey  annually  (based  on  state average  yield), with  a
wholesale  value  of  $332,640  and  a  retail  value  of  $724,416  (personal
communication,  W.  Kissinger, State Apiarist).  But  estimates  of  the economic
value of  honeybees  as pollinators range  from  20  to 100 times their wholesale
value as  honey  and wax  makers  (Winski,  1974;  E.  McGregor,  1976; F.  Moeller,
personal communication).   Using these figures, it appears that the honeybees of
southeastern Montana  are worth from $6.6 million to $33 million annually just as
pollinators, excluding their value to California growers.

                              LITERATURE  REVIEW

     Insects and other invertebrate animals seldom are studied in  air pollution
investigations, yet the available  literature suggests significant interactions
of  insects  and pollutants.  Sulfur  oxides,  nitrogen  oxides,  carbon oxides,
particulates,  hydrocarbons, fly ash, acid mists,  and numerous trace elements
impact  entomological   systems  (bibliography   in  Appendix  A).   For  example,
Hillmann and Benton (1972)  conducted investigations  near a 615 MW coal-burning
generator  in  central  Pennsylvania and  observed several consequences to insect
populations which were linked to  sulfur dioxide air pollution,  while a European
study associated honeybee mortalities with arsenic  emitted  from  a coal-fired
power plant, 6 km distant (Ferencik, 1961).

     Air  contaminants may  accumulate in  the  tissues of insects by ingestion,
respiration,  or penetration  through  the  integument (Debackere,  1972).   Toxic
substances  may  be  passed  along  the food  chain  and accumulate  in  predatory
insects,  resulting  in host-parasite or host-predator imbalances  (Dewey,  1972;
Hillmann,  1972;  Hay,  1975).  In addition, pollinating  insects, especially the
social  insects, appear  to  be  very  susceptible  to toxicosis from atmospheric
pollutants  (Dewey, 1972; Knowlton  e_t a]., 1950; Bromenshenk and Carlson, 1975).
Social  insects such  as  domestic  and wild  bees  forage  over large distances,
transport  materials back to the colony  and  store  them as food  supplies.  These
insects are morphologically  and behaviorally specialized for  the collection and
transport  of  small  particles (i.e., pollen) which may  lead to the inadvertent
accumulation of small, particulate contaminants.

     Studies  of pesticides  and of anesthesia  with  carbon  dioxide and nitrous
oxides  indicate that  the  effects of toxic  substances  on  insect physiology,
biochemistry,  and  behavior  include sublethal  effects,  such as disorientation,
memory  loss,  and  permanent changes in behavior  as  well  as  the  more obvious
lethal  effects  (Schricker  and  Stephens,  1970;  Ribbands, 1950).   Gerdes et al_.
(1971a  and  1971b)  found that   1.3  to  2.9  ppm of  hydrogen  fluoride reduced
hatchability and fecundity  of fruit flies, Drosophila melanogaster, probably as
a result of genetic damage,  and Ramel (1967) induced chromosome disjunction by
feeding  D.  melanogaster  0.25  mg mercury.    Recently,  European  investigators
(Debackere,  1972;  Toshkov  et  aj. , 1974) concluded  that some insect species,
particularly  honeybees,  have a very high potential  as  sensitive  and efficient
indicators of toxic substances and acidic solutions.

     Numerous reports indicate that "trees weakened or injured by pollutants are
more likely to  be attacked  by insects that normally  require weakened trees for

                                      143

-------
successful  reproduction"  (Haegle,  1973).   The number of arthropod  pests  that
have  been  associated  with  pollution-predisposed  trees  is  small.   However,
certain sucking insects or  those  that feed in concealed places,  such as under
bark, are more  likely  to  proliferate, while foliage-feeding  insects  are  more
subject to poisoning because of their greater exposure both to airborne toxins
and through feeding (Hay,  1975).  Two questions repeatedly are  raised in reviews
of the subject (Haegle, 1973; Hay, 1975; Ciesla, 1975):   Are trees predisposed
to  arthropod  attack  and  damage   because  of  air  pollution  stress?  Do  air
pollutants affect the predators and parasites of a  pest  arthropod and, as such,
affect important  regulators of pest  population  numbers?   These  questions  are
particularly important in  view of the conclusion of  Mattson  and Addy (1975) that
phytophagous  insects  act  as  regulators  of primary  production  and  nutrient
cycling  of  forests  and  thus  have   long-term  interactions  with  fundamental
ecosystem processes.

                            MATERIALS AND METHODS

     The techniques  and materials  employed  for the  entomological  investiga-
tions, including the monitoring and analysis for the accumulation of sulfur and
fluoride in insects, are lengthy and have been  published previously (EPA, 1976;
Gordon et al_. ,  1976; Kay et al_. , 1975;  Carlson et a_L ,  1974).  Summaries of these
procedures,  modifications,  deletions, and  additions  are  indicated  under  the
reports on specific objectives.

                                OBJECTIVE #1

             SELECTION AND ESTABLISHMENT OF PERMANENT STUDY SITES
         DISTRIBUTED SO THAT GRADIENT EXPOSURES AND RESPONSES TO AIR
                POLLUTANTS FROM THE COLSTRIP POWER PLANTS OCCUR

     In  September of  1976, participants  from  federal  agencies  and  research
institutions convened  to  examine  the role of applied ecology in environmental
problem  solving  (Johnson,   1976).   They  concluded   that   post-action  impact
monitoring programs are potentially valuable aids to the  improvement of modeling
and  predictive  capabilities and provide valuable opportunities to measure the
results of  controlled  experimental  perturbations of diverse  ecosystems.  They
stressed the need  for an experimental design that could identify unforeseen or
underestimated  impacts,  provide a  base against which  to  compare post-action
impacts, improve future predictions, and be amenable to information transfer.

     We believe that plot  selection  is a critical element of  pre-  and post-
action impact evaluations.   First, any controlled study of perturbations in this
type of field  investigation depends on plot distribution which in turn should
consider known  or suspected wind  channels from  the  stationary source and the
degree of  exposure of  the  plot  to  contaminants emitted  by  the source.  For
example,  because  fluorides  are transported in  the  atmosphere,  vegetation on
ridges or  elevations downwind  from  the emission source is more  likely to be
subjected  to fluoride  stress than  vegetation upwind  from the source  or in
valleys (Carlson and Dewey,  1971j.  Second, air pollution impacts  may  be short-
term and acute  or long-term and chronic.  The  latter  may  be discernible only
after a period of several years.  Therefore, for maximum  usefulness, study plots
                                      144

-------
should be  relatively  permanent  as  regards  location and  accessibility, which is
why plots were located on public domain whenever possible.

     For  the  Colstrip studies, permanent  study  plots  were established in the
ponderosa  pine-skunkbush and  the cool season-short grass ecosystems.  Ponderosa
pine-skunkbush  sites  were  located at varying  distances from Colstrip in eight
directions.   General  locations  were  determined by plotting  intersects of eight
compass  radii  (45.0  degree  sectors)   on  a topographic  map,  with concentric
circles representing  a two-fold geometric  progression of distance (8 to  120 km)
from the power plants. Final  site selection was made  in the field based on these
criteria:  (1) Each site should be on public domain whenever possible; (2) each
should be  on  the most elevated and  exposed terrain  in  the  immediate area, and
(3) each should be composed of ponderosa pines  of different ages  associated with
a diversity of understory species.

     Prior to the initiation  of this project,  we conducted baseline studies for
the state  of  Montana  at  35 pine sites  in  the  Colstrip area  (DNRC,  1974).  From
these  sites,   14  were  chosen  for  use  in the  pre-operational  phase  of the
entomological  studies  for EPA and three  sites  were  added  (3 to 5  km  from
Colstrip)  where  extensive  air  monitoring  was  being  conducted  by the state of
Montana  and  by Battelle Northwest  Laboratories.  The  majority  of these sites
were  located  in  the  path   of the  prevailing  winds,  downwind  from the  stacks
(Figure 5.1A,  locations published by  EPA, 1976).

     Twenty-six commercial  apiaries  (Figure 5.IB, locations in Appendix  B) were
established at  distances of  6.7 to  130 km from Colstrip as permanent plots in
the  short grass  ecosystems  and  bordering  agricultural   lands.   Apiaries  of
southeastern  Montana  are  situated along  river  and  creek floodplains  and,  as
such, are  in  the valleys.   Most of  the beeyards  within 20 km of Colstrip are
along Rosebud Creek;  only  three are  located elsewhere.  The choice of bee sites
was  limited  to  existing  apiaries, which tend  to  be  permanent  because  of
Montana's  legal  restrictions  on location.  Four men own and manage all  of the
6,600 commercial bee  colonies in the study area.  They all  are interested in the
results  of the  study and  have given permission to  obtain  samples from their
hives at any time  and  from any location.  Thus,  we believe that we  shall continue
to have access to  these sites  and resources.

     To date,  we  have one of the  most  extensive field plot systems in south-
eastern Montana.  In  addition to the EPA  sites, similar study sites have been
established for  use  in  related studies sponsored by  other state and  federal
agencies  (USFS,   ERDA,   State  of  Montana).   There   is  baseline data from  50
ponderosa  pine-skunkbush  sites,  26   apiary  locations,  and  six  grassland
enclosures, and  entomological  studies  are being conducted at  17  of  the  pine
sites, all of  the apiaries, and  on the two ZAPS  grassland exclosures.

     The initial results in the post-operational investigations at these sites
confirm the belief  that we not only have  established  an extensive network of
monitoring points  but that  these  points  are  being  subjected  to differential
levels of air  pollution (see Objective #5 and section  by Gordon et al_. ).
                                      145

-------
Figure 5.1.  Colstrip study area;  locations of (A) ponderosa pine-skunkbush and
            (B) apiary sampling sites.
                                      146

-------
                                OBJECTIVE #2

          COMPILING INVENTORIES OF INSECT POPULATIONS, INFESTATIONS,
             AND DAMAGE TO INDIGENOUS PLANT SPECIES AT THE STUDY
             SITES TO PROVIDE A BASIS FOR THE SELECTION OF MORE
                SPECIFIC GROUPS FOR USE AS INDICATOR SPECIES

     These  surveys  wtfi  Inltittid  at thi  ponderost  plne-skunkbush  sites  to
dettrmlne thu Insect ftuna pristnt,  to ditirmlnt thi stitus of Insist pests  of
pondirosa pint and thilr potential for outbreak, ind to dittrmlnt ths itatus  of
binsfldil  Insicts,  such 11  Initeti  prtdidous on  folligt fisdtri.  It wis
hypothiilnd that this  1nformit1on  would provide a bisls for thi  itlietlon  of
men  spidfle  groups  of Instcti  as  1nd1e§tor iptclti,  ts  well is  provldi
supplement!! 1nformit1en, such is hibltat Information ind distribution records.
Wt did not Inttnd to expend largs amounts of study resources amassing  extensive
species  lists  or collecting  quantitative data such ts  estlmatss of absolute
population size or diversity Indices,   It was felt that the natural variability
of many Insect populations precluded or greatly reduced the probability of using
this type of  data for meaningful post-action evaluations of pollution-induced
perturbations or of  separating these perturbations from the interaction of other
physical and biotic  variables.

     Investigators  from  the  Colorado  Natural  Resources  Ecology  Laboratory
(CNREL) attempted to document changes in total  arthropod  population numbers and
biomass  on   grassland  sites  using  techniques  that  had  successfully  shown
responses of arthropods to artificially-induced environmental  stresses  such  as
moisture and grazing pressure.  They encountered problems  applying this approach
at  the  EPA  S02  fumigation  plots because of  low population  numbers and high
natural variability.  However,  their extensive insect  sampling provided  species
inventories   for the  grassland  ecosystems;  these  inventories  allowed  us  to
concentrate  our attention on the pine ecosystems.

     Several techniques  were used to  conduct  insect  inventories:  (1)  Exami-
nation  of state  and  federal entomological  collections  (State Department  of
Forestry  and USDA  Forest Service, Region 1)  for  identified  species  collected
from southeastern Montana timber  stands; (2) searches of records and collections
pertaining  to  insects  and  insect  outbreaks  in  ponderosa  pines of  eastern
Montana;  (3)  on-site  observations  and  sampling  procedures  (Appendix  C)
conducted in August and October, 1974, .and once every two or three weeks from
May 1   through  October 15,  1975;  (4)  examination of insects  in or on  foliage
samples obtained from all of the  ponderosa pine  sites,  both  from the EPA studies
and those from the Forest Service project, and  (5) utilization of  sticky traps
(after Williams,  1973) which  consisted   of  35 x 45  cm  rectangles   of  1.5  ml
aluminum coated with Tack-Trap®and fastened 3.5 meters above the ground to the
trunks of each of five trees at  five  locations.  The sticky traps  were  changed
every  six weeks from  May  5  through  November  20,  1976.   The purpose  of this
trapping, begun  in  1976, was to  trace  the  phenology  of some of  the major  or
dominant insects on  pines and, hopefully,  to obtain specimens of ephemeral adult
forms, which were necessary for the  identification of a few  species which we had
seen only in immature  stages.  Results  from the sticky traps  were not complete at
the time (March 1, 1977) of preparation of this  report.
                                      147

-------
     All insect  specimens  were  labeled according to site, date of collection,
and  indexed  by  number in a log containing  comments.   The specimens were pre-
served in 80  percent ethanol in glass vials until they could be identified.

     The emphasis for this part of the entomological investigations  has been to
provide supportive data for the ponderosa growth-health studies (see section by
Gordon  et aJL for study plan and results).  The growth-health studies quantify
pathologies occurring on pine foliage obtained from ten  permanently marked trees
at each  of  the  plots.  Foliage  samples were  taken  from upper and lower crown
positions  on the  side of the  trees  facing  Colstrip.   The foliage  was  then
subjected to analysis of  sulfur and fluoride content, and the percent occurrence
of ten needle pathologies (e.g., insect, fungal,  pollution damage), the percent
needle  retention,  needle  length  and cross-sectional  area,   and  chlorophyll
content.  Pathology determinations  were made for each group of  100 needles from
the preceding four years of growth.  The lengthy procedures have been published
(Gordon et aJL ,   1976); procedures and results are summarized in the section by
Gordon et aJL of this report.   These procedures,  including sample size and data
handling, were evaluated by PEDCO Environmental Specialists, Cincinnati,  Ohio,
and their recommendations  regarding the study plan have  been incorporated.

     An  insect  checklist  is  included in Appendix D.  On-site measurements and
evaluations  dealing with  tree  physiology and insect-damaged portions of trees
other than foliage are summarized in Appendix E.   The most frequent and serious
insect damage syndromes are illustrated in Figures 5.2 and 5.3.

     From  our pre-operational  studies,  the following  conclusions  were  drawn
regarding the status of insect species present at the ponderosa pine-skunkbush
plots:

     (1)  Compared to  most  stands  of ponderosa pine in Montana,  trees on the
study  plots  in  southeastern Montana were exceptionally healthy and generally
demonstrated  little or almost no insect, disease, or pollution damage, except in
a few localized  areas.

     (2)  The  primary  causes  of  tree  mortality  appeared  to be  porcupine
girdling,  storm breakage  and  blow-down,  and  fire.   Insects and  disease as
primary causes  of  death  were  rare, except  on  the Cheyenne Indian Reservation
which  has  had a history  of  occasional insect outbreaks  in  timber  stands and
continues to have some localized insect problems.

     (3)  Insect populations were generally  low and insect damage was minimal.

     (4)  The most prevalent destructive insects and damage included:  (a) Pine
cone boring by the larvae of beetles and moths (Conophthorus, Laspeyresia, and
Dioryctria  spp.),   which  occurs  on  all  plots  and  occasionally  destroys
significant numbers of cones and seeds; (b)  scale  insects (Matsucoccus secretus
Morrison)  found  on  most trees  under the  fascicular  sheaths or  between the
needles in two year  or older needles;  (c)  foliage  feeding by several weevils
such as  Scythropus  elegans (Couper)  observed on  all   plots  and  Magdalis sp.
observed on new  growth at several  plots; (d) pine  loopers (Phaeoura mexicanaria
(Grote))  seen at  six plots  but  in  very  low  numbers;  (e)  several  sawflies
(Neodiprion  spp.)  seen  feeding  on  foliage  at  several plots  but only on

                                      148

-------
Figure 5.2.  A.  Ponderosa pine killed by bark beetles  (Dendroctonui fionderosae).    B.  Young P°"derosa
             pine girdled by a porcupine, which will  cause  the tree  to die.   C.  Immature pine cone of
             ponderosa pine attacked by Conophthorus  pondgrosae Hopk.  The cone is dwarfed by the attack
             of the insect.   D.   A pine seed moth la^T^oWbly Lasjjorresia fiiprana (Kearf.). working
             through the pith into seeds.  The larval stage of three species of pine seed moths are
             indistinguishable.
                                                  149

-------
Figure 5.3.  A.   Basal  portion of a needle  of Pinus  ponderosa  showing  the necrotic  symptom of damage
             found under the fascicular sheath.    B.  An  infestation of  pine  needle scales (Phenacaspis
             pinifoliae) on ponderosa  pine.    C.   Notches chewed  in the  needles of  Pinus  ponderosa  by
             weevils such as those of  the genus  Scythropus.    D.  A cluster of larvae  of  a species  of
             Neodiprion (sawfly)  feeding on needles  of  Pinus ponderosa.   E.  Close-up of needles of
             Pinus ponderosa damaged by a pine needle miner.   A larva  is inside the needle.   F.  Short-
             ened needles with bulbous bases characteristic of attack  by several pine  gall midges,  family
             Cecidomyiidae.  Many of the western species  have  not yet  been named.
                                                   150

-------
occasional  branches;  (f)  infrequent,  localized infestations  of  bark beetles
(Dendroctonus  ponderosae Hopk. and  Dendroctonus  brevicomi  LeC.)  and engraver
beetles  (Ips  cal1igraphus  (Germar)  and !_.  pini  (Say)),  and  (g)  ubiquitous
populations of aphids.

     (5)  The  most  prevalent  insects during  the summer  months  were lady bird
beetles  (particularly  Cleis sp.  and Neomysia sp.) and a  leaf-footed plant bug
(Leptoglossus  sp.),  which  was  common  in the autumn months.   The beetles are
predacious  and are reported  to  feed on a variety of  foliage  feeders  such as
mites,, scale insects,  and aphids.

     (6)  Many  other  insect predators,  parasites, and parasitoids prey on the
harmful  insect  species.    Particularly  conspicuous   beneficial  insects  were
lacewings  of  the  family  Chrysopidae   and  Mantispidae,  several  Icheumonids
(parasitic   Hymenoptera),  and   beetles   of  the   families  Cleridae   and
Coccinellidae.  In  addition,  a profusion of  dead aphids  exhibiting large exit
holes in the abdominal region  indicated  the interaction of  an unidentified but
prevalent parasite.

     (7)  The  potential  for an immediate epidemic of  destructive insects was
considered  low except in a  few  localized  areas.    The  pine  looper,  Phaeoura
mexicanaria, occurred in  epidemic numbers on  the Northern Cheyenne Reservation
and  the  Custer National   Forest in  1969  and  1970  (Dewey, 1970).  About 20,000
acres were  heavily defoliated and  another 43,000 were  moderately  to lightly
defoliated.  The defoliation appeared to  predispose many trees to attack by bark
beetles  (Ips cal 1 igraphus,  1^.  pini,  and  Dendroctonus valens LeC.).  A collapse
of  the   pine  looper  population  in  1970 was  attributed  to  the  spread  of  a
pathogenic bacterium.

     M.  McGregor, leader  of the Bark Beetle Evaluation and Control Group, USFS,
Region 1, inspected several Col strip pine sites during August 3-8,  1975, and his
report  (in  Gordon  e_t a!. , 1976)  closely  agreed  with our  own observations.
McGregor conc]uded that a weakening  of stands might  lead  to another population
build-up  of  the   pine   looper (£.  mexicanaria)  but  discerned  no  immediate
potential for  outbreak.   He noted infestations of the mountain pine beetle (D.
ponderosae) and the pine  engraver beetle (!_. pini) in a blow-down at  one plot in
the  Custer  National  Forest and in  areas of  thinning of  second-year growth of
ponderosa  pine in  the  Three-Mile   Creek  area,  which  had  a potential  for  a
localized build-up by  1976.

     Besides providing  species lists  and  information  on the pre-operational
status of insects associated with ponderosa pines, these  insect surveys may be
extremely useful for post-impact monitoring.  Rather than attempting to directly
quantify  insect  population  dynamics,  we  are  monitoring  the insect  damage
incurred by  foliage  by assessing the percent needles  attacked  or consumed by
insects.   In his  book  on ecological methods as applied to insect  populations,
Southwood  (1975)  considered  estimates   of  the amount of  a  plant  consumed,
expressed either as a quantity of material  consumed  or as some index of damage,
to  be  the  most  meaningful biological   measure  of  the  effect  of an  insect
population-on  a plant stand,  although these estimates are only an approximate
index of the size of the  insect population.   Damage per se is of more intrinsic
interest to us  than absolute insect numbers because  insect  damage affects growth

                                      151

-------
rate and is reflected by growth rings (Mott et al_. ,  1957) and because increased
insect  damage  to conifers  has  often been  associated  with  pollution-stressed
trees  (review  by  Haegle,   1973).   However,  in  every  case,  reports of  tree
mortality  or decreased  production due to insect attack on  pollution-stressed
trees relied on  studies  conducted after the  stationary  source(s)  had been in
existence  for  some  period of  time.  Generally,  there was  little, if  any,
documentation of insect conditions prior to the operation of the source.   Thus,
if an  insect outbreak occurred,  the argument often was made that the outbreak
was a natural part of the life cycle of the pest and  was  a random or coincidental
occurrence rather than the result of increased levels of a toxic pollutant.

     In the Col strip area, we quantified foliage damage  caused by several  agents
including insects (see section by Gordon et aK ) prior to the  operation  of the
power plants, determined the current status of insects known to be destructive
to  ponderosa  pine,  and obtained  records  concerning  the  history of  insect
problems in the  area on the nearby  commercially usable stands  of timber, the
Custer National  Forest and the Indian reservations.  We consider the area to be
free  of  serious  insect  problems,  as  evidenced both by  records  and  by
observations of the tree stands at  each site.  However,  some  insect species were
identified which  appeared to  be  of particular  interest to  observed foliage
damage  and post-impact responses.  These are discussed under Objectives  #3 and
#4.

                                 OBJECTIVE #3

     ANALYSIS AND SELECTION  OF INSECT POPULATIONS WHICH  HAVE  A DIVERSIFIED
      BUT UNDERSTANDABLE  INTERRELATIONSHIP WITH INDIGENOUS PLANT SPECIES
           AT THE STUDY SITES IN ORDER TO EXAMINE THE EFFECTS  OF AIR
                CONTAMINANTS ON SYMBIOTIC OR COEVOLVED  SYSTEMS

     Inventories (USDA Rangeland Laboratory, Bozeman,  Montana; USFS,  Region 1,
Missoula,  Montana;  Environmental  Studies  Laboratory,  University  of  Montana,
Missoula) of the pine and grassland ecosystems of southeastern Montana indica-
ted that a minimum  of 1,500 to 2,000 species of insects commonly are associ-
ated with  the indigenous plant species.   We elected to concentrate our efforts
on  a  few specific  life  systems,  an approach which is  believed to yield good
quality  control  and  the opportunity  to  test  feasible hypotheses.  A  "life
system" is defined as the part of an ecosystem which determines the existence,
abundance, and  evolution of a particular population.    According  to  this con-
cept, the  inherited properties  (array  of genotypes) of a population*transform
environmental resources  of  matter  and  energy into phenotypes  or  the individ-
uals of the species.  These individuals  form populations with group character-
istics  (e.g.,  birth rates, dispersal, mortality).   The  continuance  and abun-
dance of a population are determined by  interactions between inherited charac-
teristics  of  the subject  and essential  attributes  of  its environment.   The
intrinsic qualities  of  the  subject species and  of  its environment are termed
the  co-determinants  of  population  numbers.   The  co-determinants  control
population performance  (e.g.,  reproductive  capacity,  mortality,  immigration,
and  emigration  or,   in  other words, numbers  and persistence).   The  manifes-
tations  of determinant  processes  are  certain  ecological  events,   including
primary demographic  events  such  as matings,  birth  rates, mortalities,  densi-
ties,  and distributions and secondary events which  affect or modify the primary

                                      152

-------
occurrences such as resource availability and inimical  agents,  both biotic and
abiotic.   This life system  approach  to the ecology of  insect  populations was
formulated and presented by Clark et al_. ,  1967.

     Having  decided  to pursue  a life  system  approach, a  list of 28  insect
species (EPA, 1976) was drawn which were  thought to be:   (1) Major components of
the pine and grassland ecosystems of the study area; (2) dominant and/or useful
indicators  or estimators  of pollution;  (3)  economically  important,  and (4)
closely coevolved  with  dominant,  indigenous plant systems.   The rationale for
our emphasis  on the latter  point  was  expressed in a recent symposium  on the
coevolution of plants and animals (Gilbert and Raven, 1973).  They stated in the
introduction:

          (Revolutionary relationships are, by definition,  the  product of
     historical change  Devolution); yet this  historical  change is still
     proceeding.   The  relationships between  elements  in a  contemporary
     ecosystem are dynamic  in both an evolutionary and in  an  ecological
     sense,  and  depending upon  the  generation times involved,  may even
     change very rapidly during a given period of observation.  This is all
     the more  likely  in these times of extensive environmental pollution
     and the consequent alteration of relationships within ecosystems that
     must affect all of  our observations.

          Tightly   structured   organism-organism   relationships    hold
     considerable promise for the understanding of the area as a whole.

     To test  the  hypothesis  that tightly structured insect/plant interactions
may change  rapidly as  a consequence of pollution-induced alterations  of these
relationships within ecosystems, we selected from the  list of 28 species a few
herbivorous  insects of  ponderosa pines (Objective #4),  insect  pollinators and
the plants  from  which  they obtain their nectar and pollen (Objective  #5), and
ground-dwelling  beetles which  act as  scavengers  and  decomposers  and  return
nutrients to the soil ecosystems, which in turn affects plant growth.

                                OBJECTIVE #4

             SELECTION AND TESTING OF INJURY- AND DISEASE-CAUSING
               INSECTS  TO BE  UTILIZED FOR MORE INTENSIVE STUDY

     The original  objective  was  to select  and pre-test,  for use at the study
sites, insect/plant systems which not only appeared to be of greatest interest
to project  goals  but  also would be easy  to manipulate.   Towards this  goal, we
reared  (EPA,   1976)    laboratory  populations   of  grasshoppers  (Melanoplus
bivittatus) and several herbivorous pine insects such  as the pine needle scale,
(Phenacaspis pinifoliae (Fitch)), bark beetles (Ips  and  Dendroctonus spp.), and
aphids  (Cinara  spp.).   In  general,  the  insects  were  provided  with  their
preferred food such as boughs of foliage for needle feeders, bolts  of wood for
bark  beetles,  and native grasses for grasshoppers.  Work was  terminated with
grasshoppers because of the interest and subsequent work with these insects by
the investigators from Colorado (CNREL) and also because five years of personal
experience with  the grasshopper  systems of Montana's  rangelands revealed that
the population dynamics  and host relationships of these organisms were extremely

                                      153

-------
complex and that population changes were unpredictable.  In addition, a study by
Dewey (1972)  indicated  that grasshoppers  near an aluminum smelter accumulated
less  toxic  materials  (fluoride)  than pollinators,  predators,  or  many  other
foliage feeders.

     Inventories,  examinations,  and  measurements   of  foliage  pathologies
revealed the  presence of  a scale insect (identified  as  Matsucoccus  secretus
Morrison) found  under  the  fascicular sheath of needles, occurring often on or
near a  zone  of basal  necrosis (Figure 5.4)  in  foliage generally at least two
years or older.  It is not known at present if this insect is a primary factor, a
secondary  invader,  or a coincidental  occupant  in the production of all  or a
portion  of  this necrosis.   The  basal  necrosis  is  similar to  that which Dr.
Gordon  has  produced  in  the Environmental  Studies (EVST)  greenhouse  by "acid
mists."  Therefore, it is felt that  it is  essential to  clarify the relationship
of this  insect to  the necrosis and  that the implications of this relationship
are  far  reaching.  Also, Siewniak (1971)  reports that Matsucoccus pini  occurs
most frequently  on Scotch pine in areas with polluted  air.  For these  reasons,
highest  priority is  being  given to  a  study  of  the  lifecycle and relations of
this insect to ponderosa pine.

     This  insect first was  observed  in  1974 but was  seen only occasionally.
Examination of foliage samples in autumn of 1975 indicated that the insect was
more common than it was thought, probably because, as Keen (1952) states, they
are  extremely  difficult to  detect because  they push  their  way  under  the
fascicular sheath and may take on the color of their surroundings.

     A  specialist, Dr.  Douglas  R. Miller (USDA Agricultural Research  Service,
Beltsville,  Maryland),  verified  the identity of  the  insect.  The insects are
small,  less  than 1.5 mm  long, and inconspicuous.  The sheath must be dissected
from the needle  in order to find the insect; if the  sheath is pulled away from
the needle, the insect is removed with the sheath.

     Macroscopic and microscopic examination of needle bases disclosed that the
basal necrosis which occurs beneath  or just above the  fascicular sheath appears
to grade, in degree of damage, from a pale or dark brown blemish on the surface
of  the  needles  to a  small  localized  enlargement (blister) of the mesophyll
tissues; this  pushes  the hypodermal  and epidermal tissues out  into a blister.
An  eruption  or  splitting  of the epidermal  and  hypodermal  tissues  may  occur,
usually  in association with the blistering.   The original  hypothesis,^which has
since been discarded,  was  that there were  two  distinct needle  tissue disease
syndromes;   the  first  was   termed basal   necrosis for  the purpose of  needle
pathologies,  the second was  termed  basal   scale.  The reason  for  the  latter
terminology was that  cases were noticed in which  several insects, especially the
basal scale, were near or in the  areas of tissue eruption or hypertrophy.  Based
on continued histological and macroscopic examinations of the insect  and of this
disease, it  is  now  believed that there  is no  real  difference between these
damage syndromes except  of degree.

     An approach towards obtaining an understanding  and a working knowledge of
the basal scale  insect involves several procedures.   In order to determine the
extent  of   the  distribution  of  the  insect, our laboratory  staff has  been
recording the presence or absence of the insect and the life stage  of the insect

                                      154

-------
Figure 5.4.  Scale insect (Matsucoccus secretus) and the hypertrophied  or  "blistered" needle base which
             is vulnerable to attack by insects and saprophytic  fungi.  The  fascicular sheaths of the
             needles (Pinus ponderosa) have been removed.

             A.  Dark-field photomicrograph of an unidentified insect found  under the fasicular sheath.
             B.  Cross-section of erupted needle.    C.   Scales in  permanent  feeding position on a
             needle base.   D.  Motile stage of scale,  or possibly of an adelgid, in temporary feeding
             position near erupted tissues of a needle.
                                                   155

-------
insect on  each sample  of  TOO needles/tree/plot used to  quantify  the various
types of needle injury.   Examination of the needle  bases was facilitated by the
use of a 70-power zoom objective binocular dissecting scope (Bausch and Lomb).
In 1976 there  was  a search for this insect at all  of the plots for all of the
projects (EPA,  ERDA, USFS)  or on a minimum of 500,000 needles from 22 plots from
mid-March  through  mid-September.   Representative  specimens  of the  different
pathology  types  and of the  different  stages  of the insect were  prepared for
histological  analyses,  killed  and   fixed  in  FAA(formalin-aceto-alcohol),
sectioned  by a paraffin technique  (Johansen, 1940), and  stained  either with
Fuelgin's Fast  Green Schedule or a modified Fleming's triple satin method (Raske
and Hodson, 1964).   The  modified Fleming triple stain was  used to differentiate
insect feeding  tracts; using this procedure, xylem  stains blue, phloem orange,
and  mucoprotein  secretions  from feeding  red.   Observations  of sections and
photographs were performed  using a  Reichert "Zeptopan"  microscope with bright
field, dark field, and phase optics.

     The results to date concerning  the prevalence  of the  insect are summarized
in Table 5.1.   The  insect appears to be common in  the southeastern Montana area,
including  Billings,  occurs  in  relatively  low  numbers,  and  seems to prefer
younger  trees.   Macroscopic  and microscopic  examination indicated  that the
motile adults  and  larvae  frequently  were entrapped by exudates  from erupted
needle tissues  and that  feeding often occurred adjacent or  proximal  to the areas
of basal  necrosis,  but not  necessarily in the  damage zone.   In  fact, most of the
larvae in  permanent  feeding  positions  were on healthy-appearing parts of the
needle.  This,  in conjunction  with  observations  by Dr.  Gordon that the basal
damage probably occurs before the meristematic tissue is fully differentiated,
indicates that  the insect may not cause the damage  but  may be  attracted to the
area of tissue  damage.

  TABLE 5.1.  PERCENT OCCURRENCE OF  TOTAL BASAL NEEDLE  NECROSIS AND OF BASAL
              SCALE INSECTS ON NEEDLES OF PONDEROSA PINE AT SIX SITES  NEAR
              COLSTRIP AND  BILLINGS, MONTANA - 1976.

Year
of
Foliage
75
74
73
72
Number of
Foliage Samples
(trees)
48
48
46
10
Number of
Needles
Dissected
2611
2742
2514
461
Percent
Non-necrotic
Needle Bases
> 98
81.6
50.8
34.7
Percent
Necrotic
Needle Bases
< 2
*
18.4
49.2
65.3
Percent
Basal
Insects
< .5
4.9
7.5
13.9

                                                         (bibliography,  Appen-
                                                         damage  to pines  over
     A  review  of the literature  concerning  Matsucoccus
dix F)  revealed  that  some species  have caused  severe
extensive areas and that some forms cause needle galls; but the damage syndrome
discerned in- this study is not characteristic of any reported damage attributed
to  Matsucoccus.   Histological  comparisons  of  the  feeding patterns  of these
                                      156

-------
organisms in  relation  to  feeding and damage by other insects (Figures 5.5 and
5.6), fungi,  and  abiotic  factors such as frost, drought, and acid mist should
clarify  the  role of  this  insect.   There are Matsucoccus  infested  trees  in a
greenhouse at Missoula and these are being used as a source of basal scale for
further  studies  of  life  cycles,  feeding, and damage.  We  hope  to establish a
suitable population  for inoculation tests based on Koch's  Postulates.

                                OBJECTIVE #5

                 SELECTION AND TESTING OF BENEFICIAL INSECTS
                   TO  BE UTILIZED FOR MORE INTENSIVE STUDY

     The previous objective emphasized coevolved systems of insects and plants
in which the insects were detrimental both in terms of harm to the plant and in
subsequent  economic  loss  because  of plant  damage.   However, there  are  many
coevolved insect/plant systems in  which the insect  is  beneficial.   The  most
familiar  aspect of  advantageous insect-plant  interaction is  the  pollinator
relationship.   Honeybees were  chosen as  the subject species  because  of their
known sensitivity  to  pollution  stress  (Toshkov  et  al.,  1973),  ecologic  and
economic importance, availability (in terms of numbers"),  and manageability.  It
is believed  that  they  will be indicative of the population responses  of other
pollinators, and to  test  this hypothesis honeybees and  other pollinators  are
being observed at the S02 fumigation  plots for comparative purposes.  The native
bees, especially the physically smaller species, may be much more susceptible to
zootoxins than honeybees (Johansen, 1972).

     We  would  like  to  work with predatory and  predacious  insects that attack
forest pests and with the predatory ants of the grasslands;  but the initial  pre-
testing  indicated  that this exceeded study  resources,  necessitating  priority
assignments.  In 1976,  studies  of ground-dwelling beetles were added  at the ZAPS
plots on the basis  of previous surveys  and biomass studies by  the  Colorado
investigators (CNREL).

     While behavioral responses of honeybees to controlled exposures  of S02  have
been investigated at ZAPS, the studies near Colstrip have  concentrated on the
use  of  honeybees (Apis mel1 ifera  L.) as  accumulators  and  indicators  of  the
extent and magnitude of the phytotoxic and zootoxic effluents  from the Colstrip
power plants (see Objective #6).

                                OBJECTIVE #6

         CHEMICAL ANALYSIS TO  DETERMINE SULFUR AND FLUORIDE CONTENT
           OF INDIGENOUS PLANTS AND  INSECTS TO ESTABLISH BASELINE
          LEVELS AND TO MONITOR ANY  ACCUMULATION OF THESE MATERIALS
         WHICH  ARE RELEASED BY THE CONVERSION OF COAL TO ELECTRICITY

     As mentioned previously,  it was  hypothesized that honeybees would be useful
in the   collection  and detection  of  toxic  materials  from   the  environment.
Numerous reports supported this hypothesis, although many were based on studies
in Europe during the 1950's (review  by Debackere, 1972).  The  usefulness of the
published information  from these  studies is sometimes questionable because of
two problems.   One  is  the  lack of a standard expressing the concentrations of

                                      157

-------
                        t >/.,,-ir .»     .  >•• •--    *
                        rf>)  V:"^'/   ^
Figure 5.5.   Base of needle and fascicular sheath.

             A and B.   Longitudinal  sections of needle showing an unidentified
             mite under the sheath.    C and D.   Cross-sections of needles and
             sheath.
                                      158

-------
Figure 5.6.   Weevil damage to needles of Pinus ponderosa.

             A and B.   Cross-sections of damaged needles and insect.
             D.  Longitudinal sections of weevil damaged needles.
C and
                                      159

-------
toxic  substances  in insects  and in  the  vegetation they  forage.   Levels are
reported as  g/insect,  g/kg,  mg%, ppm dry  weight,  ppm  wet weight, and similar
designations so vague that one cannot determine the basic unit of measurement.
The  other  problem  is  the  assumption that  data  which correlates  changes in
insect populations, such as mortalities with tissue concentrations of a single
contaminant, may be used to estimate toxicity or harm.  This type of data may be
sufficient  to  demonstrate harm  by  air pollutants  but does  not identify the
specific agent and ignores synergistic effects.  In addition,  these  studies were
conducted near  industrial  sources  with sites distant  from  the source used as
controls.  These controls were designated as such because of their distance from
the  source.   But the  acid rain literature  indicates  that  some forms  of air
contaminants may impact areas as far as 500 km from the source, and if the air
pollution from the source was widespread,  these controls may  not  have reflected
pre-operational baseline levels of pollutants.

     Since  the  initiation of the EPA studies in August 1974, we have acquired
quantitative  analyses  of  fluorides,  sulfurs,  and  pesticides  in adult worker
honeybees,  their pollen  and  honey  food supplies,  and their  water supplies.
Bees,  honey, and pollen were obtained from commercial apiaries  at 26 locations.
Twelve  of the  apiaries are within 20  km of  Col strip,  are utilized as primary
sites,  and  have been sampled eight different times during the last two years.
The most distant beeyards are used as  (controls) and as  indicators of long range
pollutant transport.   Individual samples of bees were taken from at least four
different hives, and a combination sample was collected from a  minimum of eight
to  twelve  hives at  each  location.    Individual  honey  and pollen samples  were
pooled  from  several cells within a colony.   Individual  bee samples were defined
as 200 to 300 adult worker bees taken from the same colony.  Combined bee samples
consisted of 150 to  300 bees  per  colony  taken from each of  eight to twelve
colonies and grouped together at the time of  collection.   Individual  samples
served  as a  measure of variation among colonies; the combined samples served as
an average sample, allowed comparison with the individual samples,  and provided
a sufficient number of bees for those analytical procedures,  such as pesticide
analyses, which  require  a large number of bees (100 g or more wet weight).  A
100 g  sample of bees taken from a single colony could significantly affect the
field work force and as such affect production.

     Apiaries in southeastern  Montana are  laid out in rows of wooden pallets,
each pallet  supporting four to  six colonies  (Figure 5.7A).  The selection cri-
teria  for   hives  consisted of  sampling  all  suitable  hives  in  each  quadrant
established  by  the  rows  of pallets.   A suitable hive was one in gofld physical
condition  (e.g.,  paint  not  peeling,  supers and  covers intact  and  properly
positioned,   clean  appearance),  free  of  obstructions  and  contaminants  (e.g.,
vegetation  and  dirt at  the  entrance) and  containing  a  strong  colony (e.g.,
vigorous, intense flight activity,  populous).  Four individual  bee  samples, one
from each quadrant, and a combined sample were taken at each  apiary during each
sample period in 1975 and 1976, excluding the  spring  periods.

     An  acrylic plastic  vacuum apparatus  (Figure 5.7B)  was  developed  for
collecting  bees at the entrance to a hive to  minimize inadvertent contamination
at the  time  of  collection.   All samples for sulfur and fluoride analysis were
frozen immediately by bagging them in  plastic Whirl Paks® and  placing them under
a  block of  dry ice in a styrofoam cooler.  Since  pesticide determinations may be


                                      160

-------
                             Straight Venturi
                            Inlet (2 cm I.D.)
                                         Exhaust Outlet
                                          (5.2 cm I.D.)
                                                                                      Air
                                              Bee Barrier
                                                 (3 mm
                                              perforations)
Figure 5.7.  A.  Apiary at SU-3 by Rosebud Creek near Colstrip, Montana.    B.  Acrylic aspirator used to
             obtain honeybees for chemical analyses.  A battery (12 volt) powered squirrel cage fan pro-
             vides a vacuum and is connected to the exhaust outlet via a flexible plastic hose.  Air and
             bees are drawn through the venturi inlet; the bees are deposited in the collection jar.  The
             jar is removed by unscrewing it from the aspirator, giving a sharp downward rap to force the
             bees to the bottom, and quickly removing and capping the jar.
                                                   161

-------
confounded by residues from plastics, we used a metal vacuum collector, based on
the design of the acrylic apparatus, and stored the frozen samples in new one-
pound glass honey jars.  Honey was collected by scraping a teflon-coated spoon
across the surface of the comb.  Pollen was removed from cells by the use of a
narrow plastic pick.   Samples were stored at -10°C until they could be prepared
for analysis.

     Clean sampling and storage equipment was of the highest priority.  Cleaning
procedures  were determined by  the analyses  to  be  performed.  All  plastic
equipment, except  the clean,  sterile  Whirl  Paks®  was cleaned  by  washing in
detergent  and  tap  water,  rinsing  with 1:1 nitric  acid,  then with  tap water,
followed by a one normal rinse with hydrochloric acid, then with tap water, and
a  final  rinse with  distilled and deionized water  (Taras  et a_L , 1971).  The
equipment was  air-dried,  face or mouth downward, on  a  plastic pad.  Glass and
metal equipment was cleaned by washing in detergent and tap water, rinsing with
distilled and deionized water, and then rinsing with acetone followed by a rinse
with hexane.   Bottles were air-dried, mouth downward, on an aluminum foil sheet.
The mouths of all glass jars were sealed with clean aluminum foil  to exclude any
impurities adhering to the  inside of the covers (procedures recommended by the
Beltsville Pesticides  Laboratory).   All  chemicals  utilized  for cleaning were
reagent grade.

     Preparation  for fluoride  and sulfur analyses  consisted  of  drying  and
grinding whole  bees  and subsequent analyses using an Orion specific ion elec-
trode for fluoride and a Leco induction furnace for sulfur determination (EPA,
1976).  Pesticide analyses  were conducted  by R. Thomas of the EPA Biological
Investigations Laboratory,  Beltsville, Maryland.  The  analytical  methodology is
outlined in the FDA Analytical Manual   for chlorinated insecticides (e.g., DDT,
DDE,  ODD),  phosphate pesticides  (e.g.,  malathion, parathion),  and carbamate
pesticides   (e.g.,   sevin,   carbofuran).   FDA  methods   were  sensitive   to
approximately 0.01  ppm.  Screening procedures utilized gas  chromatography, thin
layer  chromatography,   and  gas  chromatography  in  conjunction  with  mass
spectrometry.  In 1975, 700 to 800 bee  colonies died from pesticide poisoning in
the  study  area.   Thus,  pesticides  may  be  a  serious   confounding  factor.
Choiinesterase analyses  were  performed courtesy of R. J.   Barker, Bee Research
Laboratory,  USDA,  Tucson,  Arizona, using  isolation procedures  based  on thin-
layer chromatography (Winterlin et a_L  , 1968) and a chlorimetric determination
of  acetylcholinesterase activity  (Ellman  et  aj. ,  1961).  We  have  initiated
determinations of trace metals such as  arsenic (As), cadmium (Cd), and lead (Pb)
in bees using published procedures  for atomic absorption  spectrophotfimetry.  A
Unicam SP90A  Series  2  AA  Spectrophotometer coupled to a  Van Lab strip chart
recorder  was  utilized.   The  results  will be  published  in  the  next  interim
report.

     Unless indicated, all  statistical  evaluations were performed on a Hewlett
Packard 65 calculator  or a Dec-10  computer  and utilized  standard statistical
programs,  including mean, standard deviation, standard error, F ratios based on
variances,  and  Bartlett's  test  for   homogeneity  of  variances.   Significant
differences (.05 confidence level) of variances allowed the use of a t-test for
equality of means with variances unequal, while non-significant differences of
variance allowed the use  of  a  t-test for the  equality  of  means  with equal
variances  (F ratios  and t-tests from  Sokal  and Rohlf,  1969).  All  confidence


                                       162

-------
 levels were computed  for the 95 percent lower and upper confidence limits of the
mean.

     Quality  assurance tests  of  the  analytical  procedures  utilized recovery
efficiency  studies,  and a  reciprocal exchange with  other laboratories  for a
comparison  study  was  carried  out  by  the  Boyce  Thompson  Institute.   The
procedures are precise, rapid,  and  show very good agreement with  other methods.
The  fluoride probe, based on several years of data, is accurate  to ± 0.10 ppm,
while the Leco method  for sulfur in bee tissues  is  accurate to ± 4 percent.  The
accuracy of the honeybee analyses and the pollen analyses is very  good; however,
the  samples must  be  well  mixed after grinding.   We did not  experience  the
difficulties reported by  Nation et  aj. (1971) who encountered difficulties in
obtaining a homogenous  sample  of whole bees for the atomic absorption analysis
of major and trace elements in bees, pollen, and royal jelly.  Their procedures
were similar to ours  except that they ground their bees in a mortar while we used
a Wiley mill which produces a  finely  ground powder which will pass through a 40
mesh screen.  Nation et aj_.  concluded that  the very hard structures of the head,
antennae, wings,  and  legs  made it  virtually  impossible  to  get a homogenous
sample but  thought that finely ground materials  or analyses of specific body
regions might improve recovery  efficiency.

     For consistency.,  all  levels of  fluorides and  sulfurs presented  here  are
given  on  the   basis  of  ppm  dry  weight  (pg/g  dry  weight).   To  facilitate
comparisons to  published  reports of  levels of contaminants in bees, which are
often expressed  as a  unit weight per one  bee  rather than per unit of mass or
weight  of  bee  tissue, we  present  the results  of  a  series  of experiments
conducted to determine  the weights  of  honeybees and the moisture  content of bee
tissues (Table 5.2).

     The results  from the  preliminary samples in 1974 are given in Table 5.3.
The  mean fluoride  content of adult worker  bees from 11 sites  was  7.4 ppm, while
the  mean sulfur content was 4,390 ppm.  Carlson and Dewey (1971) reported that
10.5 ppm F  was characteristic  of bees from a clean or control area in western
Montana, whiTe  bees  near  an aluminum  smelter contained 221 ppm  F  .  Obviously
the  mean 7.4 ppm F  indicated a clean area.   In addition, European investigators
have reported levels  from 0 to 40 ppm as characteristic of bees from clean areas
(Lillie, 1970).  We  found  only one  report of total  sulfurs  in bees.   Hillman
(1972)  fumigated   colonies  with  3.65 ppm  S02  for  nine  weeks  and  found  no
significant  differences  in sulfur  levels  between  fumigated  bees and  non-
fumigated adult or larval  bees.  He  reported levels  of  3,200 to 4,600 ppm in
adult bees and  1,800 to  3,900 in larval bees.

     Our interest in honeybees  extended beyond their potential  as collectors of
contaminants  from the  environment.   We  also wished  to discover  whether  the
levels of  these contaminants  would increase  in bees in the  Colstrip  area to
levels damaging to the  bees and to the bee industry in terms  of effects such as
reduced numbers, strength, and vigor,  thereby decreasing production.  In order
to test the ability to  detect  contaminants such as fluoride in bees, beekeepers
in Anaconda, Billings,  Butte,  and Columbia Falls were asked to send samples of
bees from  areas known  to  be  subjected to high ambient  air  concentrations  of
pollutants  and/or areas  in  which  beekeepers  had  repeatedly  experienced  bee
poisonings or  unexplained (e.g., pesticides, disease) losses.

                                      163

-------
TABLE 5.2.  WET WEIGHTS, DRY WEIGHTS, AND MOISTURE CONTENT ADULT WORKER
            HONEYBEES WITH STANDARD DEVIATIONS.


     G  Wet Weight/Bee       G  Dry Weight/Bee       Percent Moisture

       0.092 ± 0.004*          0.031 ± 0.002           66.85 ± 1.58
       0.088 ± 0.006t          0.028 ± 0.002           68.66 ±1.51
                    t                -                 66.53 ± 1.95


*Bees collected from  ten different apiaries in September 1974 near Colstrip,
 Montana.   Each value is a mean of ten independent samples  of 500 bees.  Each
 sample of 500 bees represents bees  from eight colonies in the  same apiary,
 5,000 bees total.

tBees collected from  14 different apiaries in August 1976 near Colstrip,
 Montana.   Each value is a mean of 15 independent samples of 142-216 bees.
 Each sample of bees from a separate colony, 2,661 bees total.

tBees collected from  15 apiaries in July and August 1976 near Colstrip,
 Montana.   Percent moisture based on samples from 64 separate colonies,
 approximately 250 bees/sample; 16,000 bees total.

TABLE 5.3.  FLUORIDE AND SULFUR LEVELS IN ADULT WORKER HONEYBEES IN PPM DRY
            WEIGHT,  WITH STANDARD DEVIATIONS*

Site
BN-2
BNE-5
BNE-4
BNE-3
BE-1
BSE-1
BS-1
BS-4
BS-5
BSE-2
BSW-1
Total for All Sites

5.
7.
5.
15.
6.
9.
5.
11.
6.
7.
5.
7.
ppm
2 ±
2 ±
5 ±
7 ±
0 ±
1 ±
1 ±
0 ±
1 ±
0 ±
2 ±
5 ±
F"
0.
0.
0.
0.
0.
1.
0.
0.
0.
0.
0.
3.

57
85
00
21
64
48
14
28
21
35
35
21
ppm
4600 ±
4700 ±
3900 ±
4300 ±
4300 ±
4700 ±
4600 ±
4600 ±
4300 ±
4400 ±
4000 ±
4390 ±
S
282
141
141
141
141
141
282
0.
141
0.
0.
299








00

00
00


*Bees collected near Colstrip, Montana in September of  1974.  Each value is
 the mean of two independent sub-samples  from a combined sample taken from
 eight to twelve colonies.
                                     164

-------
     The  results  are presented  in  Figure 5.8.   It is evident  that  bees  from
these areas had substantially higher fluoride levels than  the bees near Colstrip
and  the  bees  from  control  areas in  the study  by Carlson and  Dewey (1971).
Reportedly, the level of fluoride per bee is a function of activity and varies
with  pollution level,  geography,  season,  and  the  plant  species  in  flower
(Guilhon,  1962).    Lillie  (1970)  stated that  the presence  of  1  pg  F~/bee
indicates the  existence  of  an intermittent or permanent  pollution source and
that the lethal_dose of sodium fluoride (NaF) per bee  is  4 to  5 pg/bee.  Levels
of 4 to 5 ppb F  in the ambient air are somewhat toxic to bees (Atkins,  1966).
Based on our estimates of the dry weight of  a bee, a level of about  34 ppm would
indicate a pollution source, and approximately 135 to  170 ppm  would indicate a
toxic level.  The average dry weight of bees from a Florida study (Nation et al. ,
1971) was  32.63  mg,  which computes to  33 ppm  F   as indicative of a source of
fluoride and 123 to 153 ppm as a lethal dose of soluble  fluoride.

     In 1976 we sampled throughout the growing season, increased the number of
sample locations, sampled individual  hives as well as  taking combined samples,
and began to collect honey and pollen.   Sampling was completed  just  as the first
Colstrip power plant,  Unit 1, began to burn coal in September 1975.

     Each year, most of the  bee colonies near Colstrip  are  transported to almond
groves in  California  where  they are  rented to pollinate the groves.   The  bees
are transported from Montana in September or October and  are returned in May or
June.  Because  bees  brought back from  California  may  carry contaminants  from
that area, colonies were arranged to be left over the  winter at apiaries in the
primary study area (within 20 km of Colstrip in 1974,  1975, and 1976). Colonies
were left at six locations in the fall of 1974 (BS-1 ,  BSE-2, BNE-2, BNE-3,  BNE-
4, and BNE-10), at nine locations in 1975 (BS-1,  BS-4,  BSE-1, BSE-2, BNE-2,  BNE-
3,  BNE-4,  BNE-10,  and  BN-1),  and  at one additional  location  (BS-5)  in  1976.
Honeybee  colonies  in  the area southeast of Ashland were  never moved from the
region and as such served as further controls.

     The presence of pesticides and/or pollutants in migratory bees is a poten-
tial problem.  However, the bees are  returned with a minimum of food stores and
are  in brood boxes.   The supers,  frames, and other equipment  are not taken to
California.  If  contaminants  are  carried back  from  California,  they  should
decrease  as  food stores, equipment, and honeybees are resupplied.  The studies
indicated that this was the case (Table 5.4).   Comparison of fluoride levels in
bees over-wintered in the Colstrip vicinity compared to fluoride  levels in bees
brought  back  to  these  sites  from  California   demonstrated no  significant
difference in  fluoride content.  All  samples in this comparison were collected
after mid-June, by which time contaminants carried from California should have
been eliminated.

     In 1975 it was found that one beeyard (BNE-10) had high levels of fluoride
(20 to 108 ppm) in samples from all four  collection dates.   This same pattern was
repeated  in  1976.   Prior to  1976,  the  bees  at  BNE-10 had not  been taken to
California,  and the  hives  had been at that location for  at least three years.
Because these colonies had dwindled and were very weak by fall, they were taken
to  California  in  the  fall  of 1975  and  replaced the next   spring  by  strong
colonies;  i.e.,  the  apiculturist's  breeding stock.  As  in   1975, the  BNE-10
colonies built slowly and produced very little honey in an area that appeared to

                                      165

-------
     100  _
      80  -
      60  -
      40  .
      20  -
                      LI
             I
                la     Ib     Ic
4a     4b    5
           la,b,c   Butte-Anaconda,  Montana; near Stauffer Chemical
                    and Anaconda Smelter

           2        Billings, Montana;  near Corette Power Plant

           3        Rosebud, Montana; fluoride in water supply

           4a       Columbia Falls,  Montana; south of aluminum
                    reduction facility

           4b       Columbia Falls,  Montana; west of aluminum reduction
                    facility

           5        Col strip, Montana;  from 25 apiaries in Southeastern
                    Montana prior to power plant operation
Figure 5.8.  Mean ppm fluoride in adult worker honeybees from smelter, indus-
           trial, and urban areas.
                                   166

-------
TABLE 5.4.  FLUORIDE CONTENT OF BEES OVER-WINTERED IN THE COLSTRIP AREA COM-
            PARED TO BEES TRANSPORTED TO CALIFORNIA AND BACK INTO THE AREA.

Site
BNE-4
BNE-3
BNE-2
BSE-2
BS-1
x total
SD
SE
N
PPM
x F
Over-wintered
8.9
8. 1
10.2
6.8
9.0
8.4
2.73
.68
16
x F
Transported
8.5
10.0
8.2
7.4
6.5
8.3
2.24
.60
14

have all  the  characteristics of an  excellent  source  of honey and pollen.  By
August  1976,  the  bees at this  location  contained  74  to 108 ppm fluoride, the
highest fluoride level of any of the apiaries in the study area.

     Because we are interested  in the total  amount of pollutants in or on a bee,
we usually do not wash the samples prior to analyses.   To test whether fluoride
occurred  inside or  on the outside of the bodies of the bees, combined samples
were taken from 18 sites  and split before drying.  One sample from  each pair was
dried and  ground  as  per  normal procedure; the other was rinsed with distilled
and deionized water until the water  ran  clear.  A t-test for the difference in
means for  each  pair  (tr-jg-i  =  1.53)  was  not significant (p >.05), except in a
sample from BNE-10 in wrncn  the fluoride  content was decreased from 60.7 to 27.6
ppm by  rinsing.   The  average fluoride in bees from each of the paired samples
from the other  17 sites  did  not exceed 12 ppm.  Low concentrations of fluoride
did not easily  wash  off  or  out of the bee tissue, whereas at higher levels at
least part of  the  fluoride did so.

     Table 5.5 presents the  distribution of fluoride in the three divisions of
the  body.   Expressed  as  concentration  of  fluoride  in  the  body or per  body
region,  there was no significant difference (p >.05) in the amount of fluoride
in body regions of the bees  demonstrating elevated fluoride  levels.  Bees with
relatively low  fluoride  content contained the least fluoride (ppm) in the ab-
domen region.  However, the absolute quantity of fluoride (in ug/ body region)
in all bees was the greatest in the  thorax, which had the greatest mass.
                                      167

-------
TABLE 5.5.   FLUORIDE CONTENT AND WEIGHTS OF DIFFERENT BODY DIVISIONS (HEAD,
            THORAX WITH LEGS AND WINGS, ABDOMEN) OF ADULT WORKER HONEYBEES.
               (Each value is the mean of four sub-samples from
                  a combined sample taken from 8-12 hives.)


                      F  Concentration
Body Division         	(ppm)              Mean Dry Weight         Site
                       N     x     SD               (mg)
Whole Bees
Head
Thorax
Abdomen
Whole Bees
Head
Thorax
Abdomen
196
527
527
527
204
898
898
898
97.
97.
97.
97.
1
1
1

1.
0.
2.
7.
9
7
1
6
4
3
8
6
1.
1.
1.
0.
0.
2.
0.
0.
41
3
03
5
6
2
5
72
26.29

1

26
4
14
7
5.71
6.52
8.39
.49
.29
.02
.19
NE-10
NE-10
NE-1
NE-1
S-4
S-4
S-4
S-4
0
0





     During  1975  bees  were collected at different dates throughout the season
 and analyzed for fluorides.  The accumulation of fluoride recorded according to
 date of collection was  not statistically significant (Table  5.6).

     Recent  (within 24 hours) dead or dying bees collected in August 1975 from
 traps or from the  front of hives contained 8.0 ppm F ,  SD = 2.24, SE =  .60, N =
 14.  Bees, from these same hives, that had been dead several  days or even weeks
 and had decomposed or dried from exposure to the weather contained 4 ppm  F  , SD =
 3.51,  SE  =  1.42,  N =  6,  indicating that fluoride was fairly  stable  even in
 decomposed or dried tissue.

     The  fluoride  content  of bees, pollen, and vegetation collected near Fort
 Howes  in  the autumn of 1975 indicated that fluoride may accumulate or concen-
 trate when transferred  from plants to pollen to bees  (Table 5.7).
                                                                     *
     In August of  1975 and August of 1976, 114 samples of pollen and bees from
 the same hives were collected.  The fluoride level in bees was  four (1975) to six
 (1976)  times greater than the fluoride in  pollen.  European investigators have
 reported that contaminants are brought into the hive primarily with the pollen
 (Debackere,  1973; Toshkov  et al_. , 1974).  To test whether the fluoride in bees
 correlated with  the fluoride  level  in  the  pollen brought  into their hives,
 Pearson's product-moment correlation coefficient was utilized.  The results are
 given in Appendix G, Tables G-l, G-2,  and G-3.  However,  fluoride  in bees did not
 correlate with fluoride in pollen, and the level of fluoride  in pollen was the
 same for both years.
                                      168

-------
TABLE 5.6.  PPM FLUORIDE IN BEES COLLECTED ON DIFFERENT DATES IN 1974 AND
            1975; SAMPLES FROM EACH SITE COMBINED AT TIME OF COLLECTION.

October
1974
x 7.5
SD 3.09
SE .63
N 24
No. Sites 12
May
1975
9.3
2.3
•77
10
9
June
1975
8.7
1.08
.44
22
17
July-August
1975
7.1
1.77
.42
18
13
August-September
1975
9.5
2.38
.56
18
18

TABLE 5.7.  PPM FLUORIDE CONCENTRATIONS IN HONEYBEES, POLLEN, AND BLOSSOMS.


     Honeybees   Pollen   Sweet Clover   Alfalfa   Snakeweed   All  Vegetation
X
SD
SE
9.
1.
1.
13
74
01
1.3
.34
.17
.7
.56
.23
.4
.44
.18
1.3
.14
.10
.8
.61
.16

     Efforts  were concentrated  on  late-season  collections  of  bees and  bee
materials because during late August or early September:   (1)  Bee colonies were
at  their peak in size;  (2)  most  of the honey supplies had been  gathered  and
stored;  (3)  pollen  supplies  were at a maximum as egg laying and brood rearing
declined  in  response  to diminishing supplies of nectar;  (4)  the  alfalfa seed
crop had been pollinated; (5) colonies had had a sufficient period of exposure
to any environmental contaminants in the region, and (6) there had been several
turnovers in  bee  populations and supplies so that any contaminants brought in
from California were  likely  to be  eliminated or greatly diluted.   Because the
efforts were concentrated on ZAPS in 1976,  all of the outlying  yards did not get
sampled, as  had been done in 1975.   The results of the fluoride  analyses for bees
and pollen collected in August of 1975 and  1976 are presented in Appendix H.

     Collecting  bees   before  mid-June  proved to  be impractical  because  some
colonies had been returned from California, others had not,  and those that were
in the area were often weak,  reducing the number of bees which could be sampled
without risking harm to the colony.  In May of 1975,  there were  only ten colonies
in the Colstrip area from which to  take samples, but the average fluoride level
of 9.2 ppm was consistent with fluoride levels at these same sites later in the
season (Table  5.6).  The  most distant site, near  Biddle, Montana,  had  higher
fluoride in the bees in the spring  than later in the season  (30.2 ppm, May; 7.9
ppm, August).  Needle pathologies typical  of pollution damage  were observed and
                                      169

-------
increased sulfurs and fluorides were measured in pines approximately  .6 km from
this yard, indicating a nearby pollution source.  The Biddle site is less than 8
km from  oil  fields  which have occasional burn-offs that may impact the nearby
apiary and pine locations.   Also,  bees are medicated in the spring, and each bee
colony  consists  mainly  of over-wintered  or older bees,  all   of which  could
account for higher levels of fluoride.

     The high fluoride level in bees at BNE-10 was  surprising,  and the fluoride
in the bees did not correlate with the  level of fluoride in pollen.  There was no
known industrial  source of fluoride nearby, and other possible  sources such as
automobile  emissions  or  the  adjacent  railroad  were  encountered  at  other
locations, some of which had very low fluoride baselines.  The water supply of
the bees  was  hypothesized  as a potential source of fluoride.  The Yellowstone
River,  a  slough  approximately  20 m away, and a stock tank fed by a constantly
flowing  artesian spring  (the source of the  water  in  the slough) were readily
available water  sources.   The  fluoride  content of these waters in May of 1976
were, respectively,  0.4 ppm, 11.8 ppm, and 7.0 ppm.  Well and surface waters in
the vicinity of Colstrip (1 to 20 km)  contained 0.0 to 0.6 ppm, with a mean of
approximately  0.2  ppm  (unpublished,  Bureau  of Mines   and  Geology,  personal
communication, R. Hedges).  In August 1976, the waters were sampled where bees
were  seen drinking  at  all of the  apiaries  near  Colstrip.  Most of the bees
obtained their water from  Rosebud Creek which had a fluoride level  of  0.6 ppm at
the majority of the sites.   Mean fluoride in  bees versus mean fluoride in water
for each  site  had  a correlation coefficient of 0.99658  (Figure 5.9), which is
significant (p < .001).   However, this high correlation may result from grouping
data from a site with a high level of fluoride in bees and water with data from
all other sites at the base of the regression line,  with  no  intermediate values.
Bees from an apiary 2.5 km from the Corette power plant at Billings averaged 39
ppm F  ,  and water from the river near  that site averages about 1  ppm F , accord-
ing  to Montana  State  Department of Health  records.  Water  appears  to affect
fluoride  in bees,  and  we  shall  attempt to refine this  observation in 1977 by
obtaining bees and  water from areas where the level of  fluoride in bees falls
between the extremes of these specific cases.

     The bees at BNE-10, which is near the town of  Rosebud, Montana, averaged 7
to  12  times  higher  fluoride than any other  site in the study area;  while the
water had 16 to 20 times the fluoride as water at other sites.   These levels of
fluoride  in  these  bees  were indicative of  a pollution source and  were just
slightly  less than the level of soluble fluoride forms reported to be toxic to
bees.  The level  of fluoride in bees at all other sites was indicative  of a clean
area (Lillie, 1970)  and as such served as a baseline value.   Statistically, the
results of Bartlett's test of homogeneity of variances and appropriate t-tests
for  all  collection  dates  indicated that the  Rosebud  bees were significantly
different (tr-,-,-,-, = 9.91, p < .001) in terms of fluoride content from bees from
all other sampiejsites in  southeastern Montana.  The mean value of fluoride in
bees  at Rosebud  fell outside  of  two  standard deviations  of the  mean  for all
sites.  Therefore,  the fluoride data for the bees,  the existence of a fluoride
source  (water),  and  literature  citations  indicated  that this site  was not
representative of a  baseline level  and as such was not  directly comparable to
the other sites.   For these reasons, the data from  Rosebud has  been excluded in
treatment of the  data from the study area sites.
                                      170

-------
     oo
     LU
     LU
     CQ
     D:
     o
         50  -
         40  -
30  -
         20  -,
         10
                  r = 0.997

                  y = 5.692 + 4.044x
                                            I          I

                                            6          8

                                    FLUORIDE IN WATER
                                                       I

                                                      10
1

12
Figure 5.9.   Fluoride in adult worker honeybees vs. fluoride in water.
                                      171

-------
     Tables  I.I,  1.2,  1.3,  and  1.4  in  Appendix  I  present the  analysis  of
variance for fluoride in bees collected in August of 1975,  just before the first
power generating unit went  on-line,  and for bees  collected in  August of 1976
(both Colstrip units have been  operating at approximately one-third capacity).
The data demonstrates significant  increases  of fluoride in bees at sites S-5,
SE-1, NE-2, NE-3, and NE-4,  southeast and northeast of Colstrip, while fluoride
in bees at other sites remained about constant or decreased (Figure  5.10).  Wind
data was supplied  by the Montana State Department of Health for the period of
May through  September of 1976,  and obtained from a 300-foot tower atop a 200-
foot hill,  1,400 m  north-northeast  of the power  plants.   The  most prevalent
winds blow to  the  east-southeast,  the west-northwest, and the east-northeast,
respectively.  Note in Figure 5.11  that the wind blows least in the  direction of
those apiaries showing little or  no fluoride increase.  There are no apiaries in
the direction of the second most  prevalent winds, i.e., to the west-northwest.
Wind patterns and fluoride increases are depicted in  Figure 5.11.

     Bees were collected during two periods in 1976, late July and early August.
The fan motor of the electric sampler burned out at the first site  in July.  The
motor was  not  immediately  repairable,  so bees  were  collected  by  removing the
outside, uppermost two frames from each hive and sweeping  the bees  into plastic
bags.  Two weeks later  these same  hives and apiaries were collected using the
repaired suction apparatus.   To check whether  the altered procedure (sweeping)
affected analyses results,  hives  were sampled  at each yard using both methods.
The results  are  given in Table 5.8.  Bees inside the hives had 50 percent less
fluoride than  bees  collected at  the entrance to the hive.  Thus,  the two sam-
pling procedures gave significantly different  results, although the results of
the sweeping were consistent for both  sample periods.  This prohibited the use
of the July  1976 data for meaningful  comparisons with any collection taken with
the vacuum apparatus.

TABLE 5.8.   COMPARISON OF MEAN  FLUORIDE IN ADULT WORKER HONEYBEES COLLECTED BY
            SWEEPING BEES FROM  THE WAX COMBS AND BY VACUUM SAMPLING AT THE
            ENTRANCE TO THE HIVES.


          August, 1976          September, 1976          September, 1976
          Sweep Sample           Sweep Sample             Vacuumed Sample
            (ppm F )               (ppm F )                 (ppm F~)
x = 5.
SD = 3.
SE = 0.
N = 61
83
25
42

x =
SD =
SE =
N =
5.
3.
1.
12
87
88
12

x
SD
SE
N
= 12.
= 4
= 0.
= 45
05 *
48
67


     Honeybee colonies have a division of labor, depending for the most part on
the maturation of the worker bees.   The eldest bees in a colony are the foraging
bees,  the bees most likely to be actively flying in and out of the entrance of
the hive.   Bees  building comb,  receiving nectar  from  foragers,  and removing
                                      172

-------
Q_
s:

co
       SW-1
       SW-2
       SW-3
        S-l
        S-4
        S-5
       SE-1
       SE-2
       NE-2
       NE-3
       NE-4
        N-2
      SE-12
                                                         1975 (


                                                         1976 (
                                          12

                                      PPM FLUORIDE
                                                  16
20
24
Figure 5.10.  Mean fluoride in adult worker honeybees, each value  is the mean
             of four independent  samples with 95% confidence  interval.
                                     173

-------
                 YELLOWSTONE RIVER
Figure 5.11.  1976:75 mean fluoride ratios  for honeybees near Colstrip,
             Montana, and wind  rose.
                                     174

-------
debris, cell caps, and materials from the bottom of cells from the hive, tend to
be of intermediate age.  Differences  in the average activities of  individuals of
a  given age may  occur  among colonies or within  the  same  colony at different
times (Michener, 1974).

     In 1975  and 1976, honey samples were  collected  from  the same hives from
which pollen and bees were obtained.  Honey analyses are in process, but we are
not satisfied with the efficacy of the analyses procedures.  Honey is extremely
hygroscopic and as such requires special  handling to remove the water.  This has
been accomplished via acid  digestion  and subsequent  drying.  Hillmann (1972)
utilized a relatively complex procedure to analyze honey with the Leco furnace,
which required continual readjustments of oxygen flow and of  rheostat settings,
and which was less than satisfactory  because of occasional explosions which sent
glass  flying  around the  laboratory  (personal  communication,  R.  Hillmann).
Currently we have been searching the literature and experimenting in an effort
to improve the honey  analyses.

     Sulfur analyses  for the three years yielded  values of  4,392 ± 286 (SE =
82.3),  4,000  ± 436  (SE  =  251),  4,881  ± 497  (SE  =  143.64) ppm,  respectively.
These differences were  not  significant (p  >.05).  Hillmann (1972)  failed to
detect significant differences between bees fumigated for periods  of nine and 14
weeks with 1,  3,  and  5 ppm S02 and those from the controls.  His  explanation that
this was  due  to the comparatively higher sulfur content of  the protein in bee
tissues, which masked any additional  amounts attributable to the fumigation, may
be correct.  Honey, which is primarily carbohydrate in composition, may provide
a means to monitor sulfur accumulation in bee systems without the complication
incurred  when utilizing bee tissue.   This  is one  of the main  reasons  for
analyzing honey, especially  since a  study by Tong ^t a!. (1975)  indicates that
honey does  accrue contaminants such as  sulfur  from the environment.   Another
approach is to analyze bee tissue for specific sulfur compounds such as sulfate
and sulfite.  Gunnison (1970) was able to detect  significant increases (twofold)
of sulfur in the hemolymph of bees after fumigation with 8.2 ppm S02 for three
weeks.   However,  Gunnison  eventually decided  to  use rabbits  for  this  work
because  of  insufficient  quantities of  hemolymph,   and his  procedures  were
extremely difficult and time  consuming.

     A third possible approach to investigating the effect of  S02 on honeybees
is to  analyze for  acetylcholinesterase  (Ache)  activity,  which  reportedly is
inhibited or decreased by sulfur dioxide.  Dr.  R. J. Barker, USDA Bee Research
Laboratory, Tucson, Arizona, conducted Ache enzyme analyses  on heads or brains
from bees  (both  foraging  and  dying  bees)  from the ZAPS  and  from  one of the
southeastern Montana control sites (Table 5.9).  He found that the enzyme activ-
ity in bees from the ZAPS and the control was as high or higher than the enzyme
activity of control bees from the Tucson area, and considerably higher than the
enzyme activity  rates  of bees  poisoned by organophosphate pesticides from his
own research project.  The  failure  to detect a change  in the Ache activity of
bees  at the ZAPS may be due to the fact that many of the bees in each colony were
not exposed to  the sulfur dioxide gas because  the hives were placed 75 m north of
the plots  to  avoid conflicts with the investigators working on  the plots.  In
addition,  determinations of Ache activity levels in the thoracic ganglion may be
a more reliable and sensitive indicator than Ache activity levels in the brain
(personal  communication, W.  Stephen,  Oregon State University).  We do not intend

                                      175

-------
TABLE 5.9.   COLORIMETRIC DETERMINATION1 OF ACETYLCHOLINESTERASE ACTIVITY IN
            HEADS OF HONEYBEES FROM ZAPS AND COLSTRIP PLOTS.

Sample2
A 900
A 901
A 902
A 903
A 904
A 905
A 906
A 907
A
B
Number of
Heads
5
5
5
5
5
5
5
5
5
5
Weight of
Heads (g)
0.0428
0.0408
0.0477
0.0463
0.0475
0.0396
0.0389
0.0282
0.071
0.0323
Buffer (ml)
6.4
6.1
7.2
7.0
7.1
5.9
5.8
4.2
2.6
4.8
Rate
5.6
5.1
5.2
5.3
5.6
6.6
5.8
7.1
3.7
1.5
Rate/ml
28.0
25.5
26.0
26.5
28.0
33.0
29.0
35.5
18.5
7.5
Rate/head3
35.84
31.11
37.44
37.10
39.76
38.94
33.64
29.82
9.62
7.2

1R. J. Barker, USDA-ARS, Tucson, Arizona.

2Samples A 900 - A 907 from Eastern Montana, Samples A and B from Phoenix.
 Phoenix bees were poisoned by pesticides.

3A11 rates for bees from ZAPS are normal.

to  pursue  the feasibility of  Ache  as  an indicator of  sulfur  dioxide effects
and/or  toxicity  until  it  is  ascertained  (through methods such as  the  use of
sulfation plates or via sulfur in honey) that the bees are being exposed to the
sulfur dioxide either near Col strip or at the ZAPS (see ZAPS report).

     Bees were collected  for  pesticide analysis primarily from the area along
the  Rosebud  Creek drainage  near  Colstrip,  although  bees were also  obtained
during  1975  from  apiaries  near the communities of  Rosebud,  Broadus,  Ashland,
Fort Howes, and Biddle.   The  localities sampled in 1975 were distributed over a
broader  area  than those sampled in  1974  to provide a  means of  detecting and
monitoring pesticides  throughout  the  study region.  Bees  from  the following
sites were  analyzed for pesticides:
                                      176

-------
                              1975           1974

                              BNE-2          BNE-3
                              BNE-3          BNE-4
                              BNE-4          BNE-5
                              BNE-5          BSE-1
                              BNE-10         BSE-2
                              BSE-1          BS-1
                              BSE-2          BS-4
                              BSE-9          BS-5
                              BSE-10         BSW-2
                              BSE-12         BE-1
                              BS-4           BN-1
                              BS-5
                              BS-9
                              BSW-2

     All  of  the  1975  samples  and seven of  the  1974  samples  contained traces
(0.01 to  0.03 ppm)  of  the  DDT/DDD/DDE  chlorinated hydrocarbon complex.  Thomas
reports that this is not an unusual finding in environmental  samples.  There was
no  gross  contamination  in any of the samples.   No  residues  of carbamate or
organophosphate  pesticides were  found.  Because  pesticide analyses  require
large amounts of bee tissue,  it  was  decided to  discontinue routine pesticide
sampling  and restrict samples  utilized for pesticides to those apiaries at which
symptoms of bee  poisoning were observed.

     The  results of the analyses of honeybees and pollen  for  the  last three
years have  confirmed the original hypothesis that honeybees are useful in the
detection  and  collection of toxic materials  (fluoride)  from  the environment.
Significant  (up  to  two times) increases were detected in the mean fluoride in
bees at several  apiaries located 11 to 15 km southeast and northeast of Colstrip
over the  average  fluoride  levels for the two  years prior to the operation of the
Colstrip  power  plants.   Fluoride  levels in water apparently play a role in the
baseline  levels  of  fluoride; fluoride  in pollen may play a less  important role,
although  Maurizio (1956) thought  that  bees poisoned by industrial effluents in
Switzerland  were obtaining fluoride  via the  pollen and not  via  water.  Toshkov
et  al.  (1974)  found that  bees transported poisonous substances (e.g., copper,
zinc, and phosphorous) from the  environment into the hives mainly by means of
pollen  and,  to  a much  lesser degree,  by means of nectar.  They did not report
analyses  of  water supplies.  It is probable that the routes by which fluoride
enters a  bee colony  may be varied and rely on the forms (e.g. , gas, particulate,
liquid) of the  contaminant.   Debackere's  review (1972)  of air pollution and
apiculture makes the following observations:  (1) The gaseous forms of fluoride,
such as hydrogen  fluoride  and silicates of fluoride, are more dangerous to bees
than the  powdery compounds;  (2) there is considerable  disagreement over the
toxicity  of  industrial  fluorides  to bees;  (3) the diet  of bees  can increase
manifestly  their tolerance for  fluorine, and  (4)  fluorine compounds  cannot
penetrate the thoracic chitin and are toxic primarily by action through the gut;
for  this  reason, solubilized  forms  are said to  be  more toxic.  He describes
fluorine  poisoning as a  long-term mortality which can lead to the complete loss
of  the  colony,  but  this loss occurs  gradually,  especially  during the summer.
The action of fluorine is  that of a stomach poison;  presumably  fluorine can be


                                      177

-------
fixed by lime particles in  the epithelial cells of the digestive tract, and this
exerts  a protective  function.   The  latter action would  explain  the increased
tolerance conveyed  by albumen  feeding  and would  also explain  the increased
susceptibility  of  bees  infected  with  Nosema  disease,  which  affects  the
epithelial  cells of the intestine.  The long-term course of fluorine poisoning
of bees could account for the lack of vigor,  weak colonies, and poor production
by the  bees  at  Rosebud.  The levels of fluoride in these bees approached that
considered to be definitely toxic (LD50) if the fluoride is in a soluble form,
which is the  form of fluoride that is measured in the water tested with the Orion
specific ion  electrode.

     The mean fluoride  level  in bees from Billings,  an  area known to be sub-
jected to fluoride pollution, was 30 ppm,  which is approximately the level that
has been reported indicative of a permanent or intermittent source of fluoride
contamination (Guilhon, 1962).   This is  about five times  higher than the levels
found  in Colstrip  prior to the operation of  the Colstrip  units, and is about
three times as high as the average fluoride in bees near Colstrip  in 1976.

     Sulfur  accumulation in  bees  does not appear to be particularly useful  in
discerning accumulative effects,  probably  because of a masking  effect  by the
high sulfur content of the  protein in bee tissues.  The results of the fluoride
studies have encouraged us  to examine the accumulation of other major and trace
elements (such  as arsenic,  cadmium,  and  lead) which are  known  to  be emitted
during the combustion of coal.

     Honeybees  have  been advocated  previously (Toshkov e_t al. ,  1974)  as bio-
indicators of impurities in the environment.   Toshkov reports that honeybees may
forage  as far as 6 to 8 km from the hive or over an area of 10,000 acres.  Our
data shows significant increases  in fluoride in bees from those apiaries located
along  major  wind  pathways  from Colstrip.   If this fluoride is coming from the
Colstrip units, then  fluorides  are  being carried at least  7 to  10 km from the
power  generators,  even considering the extent of the bees'  flight range.  Based
on  the .results of  our current  studies,  honeybees  appear  to  be  excellent
indicators of fluorides in  the  environment.   In fact, they  are  sensitive  to
changes of a few  parts per million and may  be accumulators and magnifiers  of
fluoride, at least as evidenced by the magnitude of increase in  bees while the
levels of fluoride  in water and pollen remained  about constant.

     If  the  available  resources of  bees are not strained  in  an  apiary  by
conducting analyses which require  large numbers of specimens, such as pesticides
analysis, one should be able to take a greater number of  individual samples and
improve the  usefulness  of  the  data both in terms of improving the sample size
and in terms  of obtaining a better estimate of variability.  One of the possible
effects of increased fluoride in the system may be an increase in the variation
within an apiary,   since bees from each  hive  tend to forage in a specific area.
Thus, the individuals from  any one hive could be exposed to  a different level  of
fluoride in the environment.  At baseline levels, this variability may be less
as each  colony  is  exposed to a  relatively  constant  or  stable level  of the
contaminant.   It is our intention to increase the number of individual samples
and to  reinstate collections at different times  of the growing season.
                                      178

-------
                            SUMMARY AND CONCLUSIONS

      Insect  and  plant  studies  have  been carried  out  since  August 1975  to
 establish baseline  data on populations of selected  species of indigenous insects
 and plants.   The  major hypothesis  was  that one  could develop methods to predict
 the bioenvironmental  impacts  of  coal-fired power plants before damage occurred.
 These methods were  based on the  use  of indicator  species  of insects  and  plants
 as  early warning  systems  and  as  continuous monitors  of  atmospheric  pollutants.
 This  approach was unique  because it allowed baseline  studies in an area that was
 one of  the  cleanest,  most  unpolluted areas in  the contiguous United  States
 before  the Colstrip power plants began operations  (EPA, 1976; DHES,  1976; DNRC,
 1975; WES,  1973).  These studies  and  those of numerous other investigators, both
 under  this   EPA-sponsored project   or  other  federal,   state,  and private
 sponsorship,  have provided extensive  baseline  data  of  characteristics of this
 clean or pristine area.   Almost  all  published studies concerning air pollution
 impacts  on terrestrial ecosystems have  either  dealt with  post-impact studies,
 which lacked  pre-impact  baseline information, or with controlled fumigations in
 laboratory chambers.

      We  believed  (and  believe) that  biological  indicators would show the  actual
 responses  through time of organisms or  populations to individual  and synergistic
 effects  of combined pollutants or  stresses.  However, the  bioindicators  had  to
 be  sensitive  to  perturbations,   easy  to measure and  monitor, and reliable
 predictors of bionomic and economic impacts. To test our hypothesis  and to meet
 these  criteria, we  proposed and  completed these objectives:  (1) Selection and
 establishment of  permanent  study  sites  distributed   to  maximize  gradient
 exposures to emissions from the Colstrip complex; (2)  compilation of inventories
 of  insect  populations, infestations,  and  damage to indigenous plant species at
 the study  sites to select specific  groups  for use as bioindicators; (3) analyses
 and  selection of  insect  populations with  a  diversified  but  understandable
 interrelationship with indigenous plant species  in order  to ascertain  effects of
 air pollutants on symbiotic associations;  (4) selection  and testing of disease-
 and  injury-causing  insect species  for more intensive study; (5) selection and
 testing  of beneficial  insects  for  more intensive  study,  and (6)  chemical
 analyses  of  selected  plants  and  insects  to establish  baseline  levels  and to
 discern any accumulation of materials  (mainly fluorides and sulfurs) released by
 the combustion of  coal.

      For the  entomological  investigations, 17 sites in ponderosa pine-skunkbush
 ecosystems  were  established on   exposed  ridgetops  facing  Colstrip, and  25
 apiaries  on  the  river  and  creek drainages along the  interface between the
 grassland  and the  cropland  ecosystems.   The investigators  from the Colorado
 National  Resources  Ecology  Laboratory inventoried and  quantified  population
 characteristics of  arthropods in the grassland areas.  Therefore, we decided to
 compile  information concerning  the  major  or  dominant  insect populations  on
ponderosa pines,  a species which  research  has shown (Gordon et al_. , 1976; Gordon
et  a!.,  1977) to  be one  of the most sensitive plant species to  the  phytotoxic
gases"  released by  coal-burning facilities.  This  pine  is  dominant  throughout
much  of  southeastern  Montana and  is the  sole commercial  timber  resource.  In
addition,  two points  were of critical  importance:  (1)  Many reports  indicated
that conifers weakened or  injured  by pollutants  are more likely to incur attack
from insects  that prey and reproduce on weakened trees  (Haegle,  1973), and (2)

                                      179

-------
phytophagous  insects  have long-term  interactions  with  fundamental  ecosystem
processes such as  primary production  and nutrient cycling  (Mattson  and Addy,
1975).  We  quantified  pathologies of  trees and foliage  and  established data
banks concerning baseline levels of sulfur,  fluorides, and trace elements.  The
insect surveys identified prevalent species  of both pest  and beneficial insects
associated  with  the  pine, evaluated  the  status  of  insect  pests  and their
immediate potential for  outbreak or  epidemic, determined which insect species
most  likely  were  responsible  for insect-induced  pathologies,  and  provided
histopathological  demonstrations of insect damage to pine foliage.   The latter
were of particular value when juxtaposed with the histopathological  micro-slide
series  of  fungal,  pollutant,  and  abiotic   (e.g.,   frost,  drought)  damage
syndromes.  Histopathological  descriptions  allowed us  to "fingerprint" damage
from the  initial damage through each stage as  it first becomes visible and then
fully manifest.  We already have a "fingerprint" file for many types of injury
from  phytotoxic  gases and  from fungi.   We quantified foliage damage by insects
and other sources, rather than attempting to quantify the absolute  size of pest
insect populations.  Ultimately, it was the  amount of injury that was the most
meaningful measure of the effect of an insect  population on its host.  Based on
pre-operational  studies,  the  southeastern  Montana area was  considered to be
exceptionally free of serious insect problems, as evidenced both  by insect and
disease reports and by our own observations.

     One  insect, identified as a Matsucoccus scale, is of the  highest interest
because  it  resides at the base of the needles under the fascicular sheath, is
often  seen  near areas of  basal  necrosis  or basal  burn to the needles,  as is
typical of acid  rain damage, and is reported to increase in numbers in areas of
pollution  stress  (Siewniak,   1971).   The  life  system  of  this  insect  was
concentrated  on,  including   its   interactions  with  ponderosa pine,  and  the
pathological  symptoms of  its feeding and attack.  This is an ongoing objective;
its  relationships with  ponderosa  pines are still  not fully  understood.   The
results  of  this  work  should make  a significant  contribution  not only towards
understanding  the  role  air pollution  plays  in the population  dynamics of this
organism  but  also  to an understanding of the basic biology of this insect, which
is seldom studied  by researchers in the United States and Canada, although some
species have caused extensive damage in the United States  (Keen, 1952).

     The  post-operational  investigations did  not  demonstrate  any  significant
increase  in insect pests or insect damage at any of the sites, but  our chemical
analyses  and  foliage   pathology determinations  demonstrated  significant
increases in  sulfur and fluoride in plant foliage at two sites and significant
increases in  needle pathologies  of 14- to 26-month-old needles collected from
one of these sites  (see section by Gordon et a 1., this report).

     Honeybees were utilized near Col strip as accumulators and  indicators of the
extent and magnitude of the phytotoxic and zootoxic effluents  from  the Colstrip
power  units.   The 1974,  1975,  and  1976  honeybee   collections  demonstrated
significant differences among sites in the  fluoride content of bees.  However,
the pre-operational levels  of  fluoride in bees  prior  to the  operation of the
Colstrip units averaged 8.5 ppm for 22 sites in 1975 and 7.4 ppm for 11 sites in
1974.   One site was shown to have  a source of  fluoride contamination (10 to 12
ppm)  in the  bees' water supply, and the  bees  at that site averaged  96 ppm  for all
years.  Honeybees collected at apiaries approximately 15 to 20 km northeast and

                                      180

-------
southeast of  Colstrip  in September 1976 (12 months after Colstrip Unit 1 went
on-line), demonstrated a significant (p £.001) fluoride increase over the levels
found  at these  sites  during  the  previous two  years.   Increases as  high as
twofold  occurred  at  these sites, which are in the path of frequent winds from
Colstrip, while  slight  decreases in  fluoride were demonstrated at sites north
and south of  Colstrip  in the  path of the  least  frequent winds.  Both 1975 and
1976 pollen samples  averaged 1.9 ppm for  all sites.  Fluoride in water at the
sites,  which  hypothetically  could  be  responsible for significant increases of
fluoride  in  bees, was  practically constant at  0.6  ppm.   Since honeybees may
forage up to 6 to 8 km (Toshkov et a]_. , 1974) from their hives, this would mean
that if the bees are accumulating fluoride emitted by the Colstrip units, these
effluents are  being carried  at least  7  to 10  km  southeast  and northeast of
Colstrip.

     The studies indicated that honeybees may be extremely  sensitive and useful
indicators of  fluorides  in  the environment, and  this  prompted us to design a
study  for the  1977  field season investigating the presence of other major and
trace  elements.   Honeybees  do  not  appear to  be particularly  useful  for the
detection of sulfurs in the environment because the natural levels of sulfur in
bee tissues are so high that  they apparently mask any measurable increases.  The
results of analyses of honeybees for the presence of pesticide residues in 1974
and 1975 revealed only low levels (0.01 to 0.03 ppm)  of chlorinated hydrocarbons
in any of the samples, a result that is not unusual in these  types of samples.

     The  results  of the  baseline  studies  of  the accumulation  of  phytotoxic
substances of  anthropogenic  origin in vegetation,  honeybees,  and  pollen;  the
field surveys of pest insect populations and infestations on ponderosa pines and
associated measurements of foliage pathologies; and the field studies of insect-
plant  life systems  (namely  M.  secretus on ponderosa pine)  reaffirm the belief
that one should be able to follow changes in the baseline characteristics of the
pine and  grassland  ecosystems  of southeastern Montana  as  they are  exposed to
emissions from the Colstrip power plants.

     Furthermore,  it  is  believed that   these   entomological  investigations
contribute to  both  basic and  applied fields of  air  pollution research  about
changes  in  systems  of  insects  and insect-plant interactions and about  the
pathways  of air  pollutants  through an  insect  system,  honeybees  in  this  case.
The information  on  the responses of pollinators  to  effluents from  coal-fired
generating plants should be invaluable to other EPA studies on plant diversity,
plant productivity, and plant community changes, while the work on the ponderosa
pine systems adds this  prominent  ecosystem to the grassland  ecosystem studies of
the overall  project.

                                 REFERENCES

Atkins,  E. L. ,  Jr.   1966.  Well, Almost: Bees Breathe Smog with Ease.   Am.  Bee
     Jour.  106(10):374.

Bromenshenk,  J. J. and C. E.  Carlson.   1975.  Impact on Insect  Pollinators.  In:
     Air  Pollution  and Metropolitan  Woody Vegetation.  W.  B.   Smith and  L.  S.
     Dochinger, Eds.   Yale University  Printing Service.   New Haven, CT.  p. 26-
     28.

                                      181

-------
Brosset,  C.  and  A.  Akerstrom.   1972.   Long Distance  Transport  of Air  Pollu-
     tants  -  Measurements  of  Black  Air-Borne  Particulate Matter  (Soot) and
     Particle-Borne Sulphur in Sweden During the Period of September-December,
     1969.  Atm. Environ.  6:661-673.

Carlson, C. E.  and J.  E. Dewey.  1971.  Environmental  pollution by fluorides in
     Flathead  National   Forest and  Glacier National  Park.   U.S.D.A.  Forest
     Service, Northern  Region,  Div.  S & PF, Forest Insect and Disease Branch.
     57 pp.

Carlson, C. E. , W. E.  Bousfield, and M.  D. McGregor.  1974.   The relationship of
     an insect infestation on  lodgepole pine to fluorides emitted  from a nearby
     aluminum plant in Montana.  U.S.D.A.  Forest Service, Northern Region, Div.
     S & PF,  Missoula, MT.  Rpt. 74-14, 21 pp.

Ciesla, W. M.  1975.   The Role  of  Air Pollution  in  Predisposing Trees to Insect
     Attack.   In:   Air  Pollution and  Insects.   Carlson,  C.  E.  (Moderator).
     Proceedings  Joint  Meeting Twenty-sixth Annual Western Forest Insect Work
     Conference and Twenty-Second Annual  Western  International Forest Disease
     Work Conference.   Intermountain  Forest and Range Exp. Sta. , U.S.D.A. Forest
     Service, Odgen,  Utah,  and Northern Forest Research Center,  Can.  Forestry
     Service, Edmonton, Alberta,   pp. 86-96.

Clark,  L.  R. , P. W. Geier,  R. D. Hughes,  and R. F. Morris.  1967. The Ecology of
     Insect Populations  in Theory and Practice.   Methuen and Co. ,  LTD.  London,
     England.  232 pp.

Debackere,  M.   1972.   Industriele 1uchtvervuiling  en bijenteelt.   (Industrial
     pollution and apiculture). Vlaams Imkersblad.  2(6):145-55.

Dewey,  J.  E.   1970.   A  Pine Looper, Phaeoura mexicanaria (Grote), Outbreak in
     Eastern Montana.  Unpublished Report, U.S.D.S., U.S.F.S.  Northern Region,
     Missoula, Montana.

Dewey.  1972.  Accumulation of fluorides by insects near an emission source in
     Western Montana.   Environ. Ent.2(2):179-182.

DHES (Montana  Department of Health and Environmental Sciences),  Bureau of Air
     Quality.   1971.   Implementation Plan  for  Control  of  Air  Pollution  in
     Montana,  (revised January 24, 1972).

DNRC (Montana  Department of Natural  Resources and  Conservation).  1974.  Draft
     Environmental Impact Statement on Colstrip Electric Generation  Units  3 and
     4,  500 Kilivolt  Transmission  Lines   and  Associated  Facilities,  Volume
     Three-B, Power Plant Appendices,  Helena Montana.

Ellman, G. L.,  K.  D.  Courtney, V.  Andres,  Jr., and  R. M. Featherstone.   1961.  A
     New and Rapid Colorimetric Determination of Acetylcholinesterase Activity.
     Biochem.  Pharm.   7:88-95.

EPA (U.S.  Environmental Protection Agency).   1973.   Effects  of Sulfur Oxides in
     the  Atmosphere  on  Vegetation.   Revised  Chapter  5  for  the Air  Quality


                                      182

-------
     Criteria  for Sulfur Oxides.   NERL,  ORD,  U.S.  EPA,  Research Triangle Park,
     N.C.  43 pp.

EPA (U.S.  Environmental  Protection Agency).   1976.  The Bioenvironmental Impact
     of a Coal-Fired Power Plant.   (2nd Interim Report) EPA-600/3-76-013.

Ferencik,  M.   1961.   Industrial   Poisoning  of Bees  and  Its Diagnosis.  Vet.
     Casopia 10(4):377-82.

Gerdes,  R.  A.,  J.  D.   Smith,  and  H.   G.  Applegate.  1971.  The  Effects of
     Atmospheric  Hydrogen  Fluoride Upon  D.   melanogaster  - I.   Differential
     Genotypic Response.  A tin.  Environ.   5:113-116.

                    II  Fecundity,  Hatchability,  and  Fertility.  Atm. Environ.
     5:117-122.

Gilbert, L. E. and P.  H.  Raven.  1973.  Coevolution of Animals and Plants.  Univ.
     of Texas Press, Austin, Texas.   246 pp.

Gordon, C.  C.   1975.  Biological  Feasibility  of  Siting  Coal-fired  Power  Plants
     at Col strip, Montana.  Western Wildlands  2(4):28-40.

Gordon,  C.  C. ,  C.  E.  Carlson,  and P.  C.  Tourangeau.   1976.   A  Cooperative
     Evaluation  of  Potential  Air Pollution  Injury and  Damage to  Coniferous
     Habitats  on  National  Forest  Lands Near Colstrip,  Montana.   Report No.  76-
     12, U.S.  Forest Service, Missoula, Montana.

Guilhon, J. ,  R.  Truhant,  and  J.  Bernuchon.   1962.  Studies of Variations  in
     Fluoride  Levels  in  Bees with  Respect to   Industrial  Atmospheric  Air
     Pollution  in a Pyrenean Village.  C.  R.  Seances  Acad.  Agric.  Fr. 48(12):
     607-615.
Gunnison,  A.   F.    1970.   Biological  Effects of  Sulfur Dioxide  on Animals,
     Emphasizing  the Interaction of  Sulfite  with the Noncellular  Fraction  of
     Blood.  Ph.D. Thesis.   The  Penn.  State Univ., University Park,  PA. 151 pp.

Haegle, A.  S.   1973.  Interactions Between Air Pollutants and  Plant Parasites.
     Ann.  Rev. Phytopathol.  11:365-388.

Hay, C. J.   1975.   Arthropod Stress.   In:   Air Pollution and Metropolitan Woody
     Vegetation.   W.  H.  Smith and   L.  S. Dochinger  (Eds).  Yale  University
     Printing Service, New Haven,  CT.   p. 33-34.

Hillmann,   R.   C.    1972.   Biological  Effects  of  Air  Pollution  on Insects,
     Emphasizing  the  Reactions of the Honeybee  (Apis  mellifera L. )  to  sulfur
     dioxide.  Ph.D. Thesis.  The  Penn. State University, Univ.  Park,  PA.

Hillmann,  R. C. and A. W.  Benton.   1972.  Biological Effects  of Air Pollution on
     Insects, Emphasizing  the  Reactions of the Honeybee  (Apis  mel 1 ifera  L. )  to
     sulfur dioxide.  Elisha Mitchel1  Sci.  Soc.  88:195.

Johansen,  D. A.   1940.  Plant Microtechnique.   McGraw-Hill,  Inc.  New York.   523
     pp.

                                       183

-------
Johansen C.  A.  1972.   Toxicity of field-weathered insecticide residues to four
     kinds of bees.  Envir. Ent. 1(3):393-394.

Johnson,  P.   L.  (ed).   1976.   Role of  Ecology  in  Environmental Management.
     Report  of  a Workshop.  September  9-11, 1976.   Oak Ridge Tennessee.   Oak
     Ridge Associated Universities.

Kay, C.  E. ,   C.  C.  Gordon,  and P.  C.  Tourangeau.   1975.   Fluoride Levels  in
     Indigenous  Animals  and Plants Collected from Uncontaminated Ecosystems.
     Fl uoride Journal.  3(3).

Keen, F.  P.   1952.  Insect Enemies  of Western  Forests.  U.S.D.S. Misc.  Pub!.  273,
     240 pp.

Knowlton, G.  F. ,  A.  P. Sturtevant, and  C.  J.  Soranson.   1950.   Adult  Honeybee
     Losses  in  Utah  as Related to  Arsenic  Poisoning.   Bull.  Utah Agric.  Exp.
     Sta. No. 340:30 pp.

Li Hie,  R.   J.   1970.   Air  Pollutants  Affecting the  Performance of  Domestic
     Animals - A Literature Review.  U.S.D.A. Agric.  Handbook No.  380.   109 pp.

Maurizio,  A.   1976.   Bee  Poisoning  in Switzerland  by Industrial Waste  Gases
     Containing Fluorine.   XVI  Int. Beekeeping Congr.  Prelim. Sci. Meet.

Mattson,  W.  J.  and N.  D.  Addy.   1975.   Phytophagous  Insects as  Regulators  ef
     Forest Primary Production. Science.  190(4214):515-522.

McGregor, S.  E.  1976.   Insect  Pol 1ination of Cultivated Crop Plants.  U.S.D.ft. ,
     A.R.S.  Agricultural  Handbook No. 496.

Michener,  C.  D.   1974.    The  Social   Behavior  of  the Bees.   Bel knap  Press,
     Cambridge, Mass.   184-185  pp.

Mott, D.  G. ,  L. D. Nairn,  and  J. A.  Cook.  1957.  Radial  Growth in Forest Trees
     and Effects of Insect Defoliation.  For. Sci. 3(3):286-304.

Nation, J. L. and F  A. Robinson.   1971.  Concentration of Some Major and Trace
     Elements  in Honeybees,  Royal  Jelly,  and  Pollens,  Determined  by  Atomic
     Absorption Spectrophotometry.  Jour. Apic.  Res.  10(l):35-43 (1971).
                                                                    «
Ramel, C.  1967.  Genetic  Effects  of Organic Mercury Compounds.   Oikos  (suppl.)
     9:35.

Raske, A. G.  and A.  C.  Hodson.   1964.  The Development of Pineus  strobi  (Hartig)
     (Adelginae,  Phylloxeridae) on White  Pine  and  Black  Spruce.  Jour.  Can.
     Entomol.  96:599-616.

Ribbands, C.  R.   1950.   Changes in the  Behavior of Honeybees Following  Their
     Recovery from Anaesthesia. Jour. Exp. Biol.  27:302-310.
                                      184

-------
Schricker, B. and W. Stephen.  1970.  The Effect of Sublethal  Doses of Parathion
     on Honeybee Behavior.   1.  Oral Administration and the Communication Dance.
     Jour. Apic.  Res.  9(3): 155-164.

Siewniak,  M.   1971.   Uszkadzanie sosny  pospolitej  (Pinus  silvestris)  przez
     czerwca   korowinowca   (Matsucoccus  pini   Green   1925;   Margarodidae,
     Coccoidea).   Sylwan 115(12):35-41.

Sokal, R. R. and F. J. Rohlf.  1969.   Biometry.   The Principles  and  Practice of
     Statistics  in  Biological  Research.  W. H. Freeman and Co., San Francisco,
     CA.   776 pp.

Southwood,  T.  R.  1975.   Ecological  methods  with  particular reference to the
     study of insect populations.  Chapman  and Hall, Ltd., London,  England.  391
     pp.

Taras, M. J. , A.  E.  Greenberg,  R.  D. Hoak, and M.  C.  Rand (Eds).  1971.   Standard
     Methods for Examination of Water and Wastewater,  13th edition. Am. Public
     Health Assoc. , Washington, D.  C.  875 pp.

Thomas, W. A.,  G. Goldstein, and W. H. Wilcox.  1974.  Biological  Indicators of
     Environmental Quality.  A Bibliography of Abstracts.  Ann Arbor Sci. Publ. ,
     Inc.  Ann Arbor, Mich.  254 pp.

Tong, S.  C. , R. A.  Morse, C. A. Bache,  and D. J. Lisk.  1975.   Elemental  Analysis
     of Honey As  an  Indicator of Pollution.  Arch.  Environ. Health. 30:329-332.
Toshkov,   A.  S. ,  M.  M.  Shabanov,  and  N.  I. Ibrishimov.   1974.   Attempts to Use
     Bees to Prove Impurities  in  the Environment.  Comptes Rendus  de 1'Academic
     bulgare des  Sciences.   Tome.  27(5):  699-702.

Westinghouse   Environmental   Systems.    1973.     Col strip   Generation   and
     Transmission  Project.   Applicants  Environmental Analysis.  Westinghouse
     Electric Corporation.   Pittsburgh, PA.

Williams,  D.  F.   1973.   Sticky  Traps  for   Sampling  Populations  of  Stomoxys
     calcitrans,  J.  Econ.  Ent.  66(6):1279-80.

Winski,  J.  M.  1974.   If  You Don't Have Enough  Woes,  Try Fretting  About  Bees.
     Wall Street Journal ,  New York, NY.  Nov.  7, 1974.

Winterlin,  W. ,  G.  Walker,  and H.  Frank.   1968.   Detection  of  Choiinesterase-
     Inhibiting  Pesticides  Following  Separation on Thin-Layer Chromatograms.
     J. Ag. Fd. Chem.   16:808-812.
                                       185

-------
                                 APPENDIX A

  LITERATURE AND REPORTS RELEVANT TO INSECT-TREE-AIR POLLUTION INTERACTION


Anderson, R.  F.   1969.   The  insect  aspects of  damage to  Christmas  trees in
     the  vicinity  of  Oakland,  Maryland  and  Mount  Storm,  West  Virginia.
     Report submitted to U.S. Dept. H.E.W., N.A.P.C.A.,  Durham, N.C.

Anderson,  R.   F.    1970.    Relation  of  insects  and  mites  to  the  abnormal
     growth of  Christmas  trees  in  Mt.  Storm,  West  Virginia-Gorman,  Maryland
     vicinity.   Rep.  prepared for E.P.A.,  Durham,  N.C.  31  pp.

Baker, J. R.  1972.  Final  report on arthropod populations infesting Christmas
     trees in the  area  of  Mt.  Storm, West  Virginia,  and  of caged experiments
     (Modified  Moch's  Postulates).   Unpublished   Report.    North  Carolina
     State University, Raleigh,  N.C.  38 pp.

Baker, J.  R.   1973.  Final  report  on caged experiment involving  Adelgids as
     a  possible  cause  of  needle  shortening  of Scotch pine.   Unpublished
     Report.   North Carolina State University, Raleigh,  N.C.  15 pp.

Bb'sener,  R.   1969.   Zum  Vorkommen  ridenbrutender  Schadinsecten  in  rauch-
     geschadigten  Kiefern-und Fichtenbestaenden.  Arch. Forstwes.  Bd.  18:1021-
     26.

Bromenshenk,  J.  J.   1975.   Biological  Impact of Air Pollution  on  Insects.
     In:    Fort  Union Coal  Field  Symposium.   Hagen  Printing  Co.,  Billings,
     Mt.   596-607.

Bromenshenk,  J.  J.   1975.   Insects  and  Coal-Fired  Power Plants.   In:   Air
     Pollution  and  insects.   Carlson,  C.E.   (Moderator).  Proceedings Joint
     Meeting  Twenty-Sixth  Annual  Western  Forest  Insects  Work  Conference
     Twenty-second  Annual  Western  International  Forest   Disease  Work Confer-
     ence.   Intermountain  Forest   and   Range  Experiment  Station  U.S.D.A.
     Forest Service, Ogden, Utah and Northern Forest Research Center, Canadian
     Forestry Service, Edmonton,  Alberta.  86-96.

Bromenshenk,  J.  J.  1976.   Investigations of  the Effects of  Coal-Fired Power
     Plant Emissions  Upon   Insects,  Report  of  Progress.   In:   The Bioenviron-
     mental   Impact  of  a  Coal-Fired  Power  Plant,  Second  Interim  Report,
     Colstrip, Montana - June,  1975.  R.  A. Lewis,  N.  R.  Glass, A. S. Lefohn,
     Eds.  EPA-60013-76-013  Ecological Research Series.   112-129, 286-312.
                                      186

-------
Bromenshenk,  J.  J.  and C.  E.  Carlson.   1976.  Impact on  Insect  Pollinators.
     In:   Air Pollution  and Metropolitan Woody  Vegetation.   W.  H. Smith  and
     L. S. Dochinger, Eds. Yale U.  Print. Ser., New Haven,  CT.  26-28.

Carlson,   C.  E.   (Mod.).   1975.    Air  pollution  and  insects.    Proceedings
     Joint Meeting Twenty-Sixth Annual  Western Forest Insect Work  Conference
     Twenty-second  Annual  Western International  Forest  Disease  Work  Confer-
     ence.    Intermountain  Forest  and  Range   Experiment   Station   U.S.D.A.
     Forest Service, Ogden,  Utah and Northern Forest  Research Center,  Canadian
     Forestry Service, Edmonton, Alberta. 86-96.

Carlson,  C.  E.  and  J.  E.  Dewey.   1971.   Environmental pollution by fluorides
     in  Flathead National  Forest  and Glacier National Park.  U.S.D.A. Forest
     Service,  Northern Region, Div.  S&PF.   Forest  Insect and Disease  Branch,
     57PP.

Carlson,  C.   E.   and  W. P.  Hammer.   1974.  Impact of  Fluorides  and Insects on
     Radial Growth  of Lodgepole Pine  near an aluminum smelter in northwestern
     Montana.   A  preliminary  inquiry.   U.S.D.A.   Forest  Service,   Northern
     Region,  Div. S&PF, Missoula, Montana. Rep. No. 74-25.

Carlson,  C.   E. ,  M.  D. McGregor,  and N.  M.  Davis.   1974.   Sulfur  damage to
     Douglas-fir  near  a   pulp  and  paper  mill in  western Montana.   U.S.D.A.
     Forest Service,  Northern Region, Div.  S&PF, Missoula,  Mt. ,  Rpt.  No. 74-
     13, 41 pp.

Carlson,  C.  E. ,  W.  E.  Bousfield,  and M. D.  McGregor.   1974.  The relationship
     of  an insect  infestation on lodgepole  pine to  fluorides  emitted from a
     nearby   aluminum  plant  in Montana.  U.S.D.A.  Forest  Service,   Northern
     Region,  Div. S&PF, Missoula, Mt.,  Rpt. 74-14, 21 pp.

Ciesla,  W.  M.   1975.   The  role  of  atmospheric pollutants  in  predisposing
     trees  to   insect attack.   In:  Air   pollution  and  insects.    Carlson,
     C. E.   (Moderator).    Proceedings  Joint  Meeting  Twenty-Sixth  Annual
     Western  Forest   Insect  Work  Conference Twenty-second Annual   Western
     International  Forest Disease Work  Conference.   Intermountain Forest and
     Range  Experiment Station  U.S.D.A.  Forest  Service,   Ogden,  Utah  and
     Northern  Forest  Research Center,  Canadian  Forestry  Service, Edmonton,
     Alberta.   86-96.

Cobb,  F.   W.  Jr.,  D.   L.  Wood, R.  W.  Stark, and  P. R.  Miller.   1968a  II.
     Effect   of   injury  upon  physical  properties   of   oleoresin,   moisture
     content  and phloem  thickness  J_n  Photochemical   oxidant injury  and  bark
     beetle   (Coleoptera:Scolytidae)  infestation  of  ponderosa   pine.    Hil-
     gardia.  39(6):127-134.

Cobb,  F-  W.  Jr.,  D.  L.  Wood, R.  W.  Stark,  and J.   R.  Parmeter, Jr.   1968b
     IV.   Theory  on  the relationships between oxidant injury and  bark beetle
     infestation j_n  Photochemical  oxidant injury and  bark beetle  (Coleoptera:
     Scolytidae) infestation on ponderosa pine. Hilgardia.  39(6):141-157.
                                      187

-------
Compton, 0. D.
     condition
     Agri.  Exp.
,  L.  F. Remmert,  and  J.  A.  Rudinsky.  1961.  Needle  scorch  and
of  ponderosa pine trees  in the  Dalles area.  Misc.  paper 120,
 Sta. ,  Oregon State University, Corvallis, OR.  6p.
Dewey,  J.  E.   1972.  Accumulation  of fluorides  by  insects  near  an emission
     source in western Montana.  Environ.  Ent.  2(2):179-182.

Donaubauer,  E.   1966.   Secondary  Damages  of Forests  Caused  by  Industrial
     Exhaust  Fumes.  (Durch  Industrie-abgase  bedingte  Sekundaerschaeden am
     Wald).  Mitt.  Forst. Bundesvers. , no 73:101-110.

Donaubauer,  E.   1968.    Sekundarschaden  in  Osterreichischen  Rauchschadens-
     gebieten.  Schwierigkeiten  Der diagnose  und Bewertung.   Niedzynarodowej
     Knof.  Wplyw Zanieczyszczen  Powietrza   na  Lasy, 6th,  Katowice, Poland.
     277-284.

Edmunds,  G.  F. ,   Jr.  and R.  K.  Allen.   1958.  Comparison  of black pine  leaf
     scale  population-density on normal  ponderosa pine  and those  weakened by
     other  agents.   Int.  Congr.  Entomol.   Proc.  (Montreal,  10th.,  1956)
     4:391-392.
Edmunds,  George  F
     2(5):765-777.
       1973.   Ecology  of  black pineleaf  scale.   Env.  Entomol
Evenden,  J.   C.   1923.   Smelter  killed  timber and  insects.   U.S.D.A.  Bureau
     of  Ent.,  Forest  Insect  Investigations.   Typewritten  report on  file at
     R-l , Missoula, Mt.

Farrier,  M.  H.  1972.   Report  on insects and mites  in  relation  to  the long-
     short  needle  syndrome  of  Scotch pines  and  their  abundance  in Christmas
     tree  plantations   in  western  Maryland  and   northern  West  Virginia.
     Unpublished  Report.   North  Carolina  State University,  Raleigh, N.C.  57
     pp.

Fischer,  G.  W.  1950.   Second  progress   report  Spokane  County ponderosa  pine
     blight  investigation.   U.S.D.A.  Forest  Service.   Unpublished Report. 25
     P-

Gordon,  C.   C.   1976.   A  preliminary  study  of  fluoride  concentrations in
     vegetation  samples  collected  September  8 and 9, 1976  in and 'around the
     town of Kitimat,  B.C.,  Canada.

Gordon,  C.   C. ,  C. E.  Carlson,  and  P.  C. Tourangeau.   1976.   A cooperative
     evaluation  of potential  air pollution  injury  and  damage  to coniferous
     habitats  on  National  Forest lands   near  Colstrip,  Montana.   I.   Interim
     Report  of  Activities from June  1,  1975  - May 30,  1976.  U.S.D.A. Forest
     Service,  Northern Region,  S&PF and  University of Montana, ESL, Missoula,
     Mt. Report No.  76-12.
                                      188

-------
Hay,  C.  J.   1975.   Arthropod  Stress.   In:   Air  Pollution  and  Metropolitan
     Woody  Vegetation.   W.  H.   Smith and  L.   S.  Dochinger  (Eds.)  Yale  Uni-
     versity Printing Service.   33-34.

Heagle,  A.  S.   1973.    Interactions  between  air pollutants  and plant para-
     sites.  Ann.  Rev. Phytopathol.  11:365-388.

Hepting,  G.  H.   1968.    Diseases  of  forests   and  tree  crops  caused  by  air
     pollutants.  Phytopathology.  58:1098-1101.

Hibben,  C.  R.   1969.  The distinction between  injury  to  tree leaves by Ozone
     and Mesophyll-feeding leafhoppers.  For. Sci.  15(2):154-157.

Horntvedt,  Richard.   1970.   S02  Injury to  Forests  (S02-skader  Pa  skog).
     Tidsskr. Skogbruk.   78(2):237-286.

Johnson,  P.  C.   1950.    Entomological  aspects  of  the ponderosa pine blight
     study,  Spokane,  Wn.  U.S.D.A. Agric. Res.  Admin.,  Bureau  Ent.  and Plant
     Quarantine.  Typewritten report  on file at R-l.  Missoula, Mt.

Keen,  F.  P.  1952.  Insect  enemies  of western  forests.  U.S.D.A. Misc.   Pub.
     273.  240 pp.

Keen,  F.  P.  and J. C.  Evenden.   1929.   The  role of forest insects  in  respect
     to  timber  damage  in the smelter fume  area near Northport,  Wn.  U.S.D.A.
     Bureau  of  Ent. ,   Forest   Insect  Investigations.   Typewritten  report on
     file at R-l.   Missoula,  Mt.

Kudela,  M.  and  E.  Novakova.    1962.   Lesnf   skudc:   a skudy  zveri  v lesich
     Poskozovanych Kourem.  Lesnictvi. 6:493-502.

Laurent,  T.  H.  and B.  Baker.   1975.   Associations  of high defoliating insect
     populations  and   high  foliar   chemical   levels  near  two  pulpmills in
     southeast  Alaska.   In:   Air  pollution   and  insects.   Carlson,  C.  E.
     (Moderator)   Proceedings    Joint   Meeting  Twenty-Sixth  Annual   Western
     Forest Insect  Work Conference Twenty-second Annual Western  International
     Forest  Disease Work  Conference.   Intermountain  Forest and  Range  Experi-
     ment  Station  U.S.D.A.  Forest  Service,  Ogden, Utah  and Northern Forest
     Research Center, Canadian Forestry Service, Edmonton, Alberta.  86-96.

Linzon,  S.  N.  1958.   The  influence of  smelter fumes  on the growth of white
     pine  in  the  Sudbury region.  Canadian  Dept.  Agr.  Publ., Ontario  Depart-
     ment of Lands & Forests.  45 pp.

Linzon,  S.  N.   1966.    Damage  to   eastern   white  pine  by  sulfur  dioxide,
     semimature-tissue   needle   blight,   and   ozone.   J.  Air.   Pollut. Cont.
     Assoc.  16:140-144.

-------
Miller,  P.  R. ,  F.  W.  Cobb,  Jr.,  and  E.   Zavarin.   1968.   III.   Effect  of
     injury  upon  oleoresin  composition, phloem  carbohydrates  and  phloem  pH
     J_n  Photochemical  oxidant injury  and  bark beetle (Coleoptera:Scolytidae)
     infestation of ponderosa pine.  Hilgardia.  39(6):135-140.

Pack,  M.  R.  and  B. A.  Zamora.   1976.  Pre-operational  Report.   Forest  eco-
     system monitoring  for  sulfur dioxide damage  in  the area surrounding the
     Northwest  Alloys  Magnesium  Plant  at  Addy,  Washington.    Wash.  State
     Univ. Publ.,  Pullman, Washington.  148pp.

Pfeffer,  A.    1963.   Insektenschadlinge  an  tannen im bereich der gasexhala-
     tionen.  Z. Angew.  Entomol. 51:203-207.

Ranft,  H.   1968.    The  management  of young  spruce  stands  damaged  by  smoke.
     Sozialistische Fortwirtschaft (Berlin).   18:299-301, 319.

Saunders,  J.   L.   1972.   Disease and  insect  pests  of Christmas trees.   School
     for Christmas tree growers.  Cornell Univ. Col. Agric.   Proc. p. 88-90.
          j
Scheffer,  T.   C.  and  Hedgcock,  G.  G.   1955.  Injury to  northwestern  Forest
     trees  by sulfur dioxide  from  smelters.  U.S.D.A.  Tech.  Bull.  No. 1117,
     49 pp.

Schnaider,  A.  and  Z. Sierpinski.   1967.  Dangerous  condition  for some  forest
     tree  species  from  insects  in the  industrial region  of  Silesia.  Prace
     Instytut  Badawczy Lesnictwa (Warsaw) No.  316:113-150.

Sierpinski,  Z.   1966.   Znaczenie  gospodarcze  skronsnika  tuzinka (Exoteleia
     dodecella L. ) na terenach uprzemyskowionych.  Sylwan.  110:23-31.

Sierpinski,  Z.  1967.   Influence  of   industrial air  pollutants  on  the popu-
     lation  dynamics  of some  primary pine  pests.    Proc.  14th  Contr.  Int.
     Union For. Res. Organ. Vol. 5, sect.  24,  Munich.  518-531.

Siewniak,  M.   1971.   Uszkadzanie sosny  pospolitej   (Pinus silvestris) przez
     czerwca   korowinowca  (Matsucoccus  pini  Green  1925;  Margarodidae,  Coc-
     coidea).  Sylwan.   115(12):35-41.

Stark,  R.  W.  et ah  1968.   I-Incidence of bark beetle  infestation in  injured
     trees  i_n  Photochemical  oxidant  injury  and  bark  beetle   (toleoptera:
     Scolytidae) infestation of ponderosa pine.  Hilgardia.  39(6):  121-126.

Templein,  E.   1962.   On the  population dynamics  of  several pine  pests  in
     smoke-damaged  forest  stands.   Wissenschaftliche Zeitschrift  der Tech-
     nischer Universitatet Dresden.  11:631-637.

Treshow,  M.   and   M.  R.  Pack.   1971.  Fluoride.   In:    Recognition  of  Air
     Pollution  Injury  to Vegetation:   A Pictorial Atlas.   J.  S.  Jacobson and
     A.  C.  Hill,  eds.   Rept.  No.  1,   TR-7 Agricultural  Committee, Air  Pollu-
     tion Control  Assoc.
                                      190

-------
Weismann,  L.  and  L.  Svatarakova.   1973.   Influence  of sodium  fluoride  on
     behavior  of   caterpillars   Scotia   segetum   Den.   and   Schiff.   Biologia
     (Bratislavia)  Czechoslovakia.  28:105-109.

Wentzel.  K.  F.    1965.   Insekten  als  Immissionsfalgeschadlinge.    Naturewis-
     senschaft.  52:113.

Williams,  W.   1975.   Oxidant  air pollution  damage  in  California  forests.
     In:  Air  Pollution and insects.  Carlson, C.  E.  (Moderator).  Proceedings
     Joint  Meeting  Twenty-Sixth Annual  Western Forest  Insect Work  Conference
     Twenty-second  Annual  Western International   Forest Disease Work  Confer-
     ence.    Intermountain  Forest  and  Range   Experiment   Station  U.S.D.A.
     Forest Service, Ogden, Utah and Northern  Forest Research Center,  Canadian
     Forestry  Service,  Edmonton, Alberta.  86-96.

Wong,  H.  R.  and  J.   C.   E.  Melvin.   1973.   Insects  associated  with  trees
     damaged  by   hydrocarbon   condensate   in   the  Strachan Area,  Alberta.
     Canadian   Forestry  Service,  Dept.   Env.  Northern Forest  Res.   Center,
     Edmonton, Albt.,  Info. Rpt. NOR-X-74.
                                       191

-------
                                  APPENDIX B



               HONEYBEE COLLECTION SITES IN THE FORT UNION BASIN





Site No.              Location              Site No.               Location



 BN-1            T3N,  R41E,  SEC 28          BSE-6            T2S,  R46E, SEC 34



 BN-2            T4N,  R40E,  SEC 26          BSE-7*           T2S,  R47E, SEC 19



 BS-1            TIN,  R41E,  SEC 24          BSE-9            T4S,  R51E, SEC 25



 BS-4            T1S,  R42E,  SEC  8          BSE-9a*           T4S,  R51E, SEC 30



 BS-5            T1S,  R42E,  SEC  3          BSE-10           T8S,  R51E, SEC 20



 BSE-8           T3S,  R45E,  SEC 18          BE-1              T2N,  R43E, SEC 16



 BSE-11          T7S,  R47E,  SEC  9          BNE-2            T3N,  R43E, SEC 33



 BSE-12          T7S,  R46E,  SEC  6          BNE-2a           T3N,  R43E, SEC 21



 BSW-]           T1S,  R41E,  SEC 13          BNE-3            T3N,  R43E, SEC  8



 BSW-2           T1S,  R41E,  SEC 22          BNE-4            T4N,  R43E, SEC 32



 BSW-3           T2S,  R41E,  SEC  4          BNE-5            T4N,  R42E, SEC 13



 BSE-1           TIN,  R42E,  SEC 26          BNE-6            T5N,  R42E, SEC  5



 BSE-2           TIN,  R43E,  SEC 19          BNE-10           T6N,  R42E, SEC 16






*Wild honeybee colonies.
                                     192

-------
                                APPENDIX C

                DIRECTIONS  FOR  PONDEROSA PINE-INSECT STUDY
I.  Field Procedures

   A.   Record  the  plot  location,  date  of  collection,  and  number of  the
        tree being sampled in the field diary.

   B.   Prepare  a  collection  card  with  the  same  information  as  in  the
        previous step, and place the card in a new plastic collection bag.

   C.   Classify  the tree based  on  condition  (modification  of the procedure
        of Carlson et aj., 1974), and record in diary.

        Tree Class

             0     No apparent  insect feeding  or pollution damage.  Needle
                   retention  appears  normal.

              1     Some  damage  evident:   light  insect activity, foliar burn
                   restricted to older  needles,  or  crowns  becoming thin.
                   Needle  retention below normal.

             2     Damage  by  insect  feeding quite evident, foliar  burn con-
                   spicuous,   or  thin crown.  Needle  retention below normal
                   or poor.

             3     Heavy   insect  damage  evident.   Excessive  insect  feeding
                   and/or  foliar burn.   Crowns  thin  and  needle   retention
                   poor.

             4     Tree is dead or dying.

    D.    Estimate   height  of  tree  and   measure  diameter  at  breast  height;
         record  in  field diary.

    E.    Inspect the  base,   trunk,   crown,  branches,  and  twigs  of the tree
         for  insects  and  injury  (includes  damage by storms,  porcupines, or
         other  agents).   Record  observations  and place  any  insects  into a
         labeled vial  filled with   80%  ethanol   to  which has  been added a
         drop of glycerol.
                                    193

-------
APPENDIX C (continued)

     F.    Beat both the  upper  and lower foliage  of  the tree with a sweep net
          (50  sweeps/level).   Preserve  any   insects   obtained  in  alcohol
          (paper  Lepidoptera);  list  any insects  in field  diary.   If insect
          has  not  been   identified,  give  a  brief  written description  and
          indicate order  or  family.

     G.    Examine  100  mature  cones  (fewer  if  necessary)  on  the tree  and
          record  the   percent  of  the  cones  dwarfed  or damaged  by  insect
          feeding; identify  the damaging insect(s).

     H.    Remove  four  branches  (five  internodes/branch) from  top one  half
          and four  branches  from  the lower one half of the  foliage from side
          of  tree  facing Colstrip.   Place  the branches  from each  of  the two
          levels in separate plastic  bags.   Keep  the samples out  of  the sun,
          covered  by  a tarp or  in a canvas bag.   Process  or  refrigerate as
          soon as possible.

     I.    If  any  dead or  dying   trees  are visible from  the   marked  sample
          tree, inspect the  trees and record the probable cause(s) of death.


 II.  Laboratory Procedures for each Sample

     A.    Cut branches from  a  given  tree into  internode  segments  to  separate
          the foliage by age.

     B.    From the  four  most  recent  ages,  select four internodes  for deter-
          minations of  percent needle  retention.

     C.    Remove all fascicles end maintain by age.

     D.    Randomly  select  100  fascicles  of  each   age,  remove the  sheaths,
          and save 100  needles.

     E.    For  each  of these 100  needles,   inspect  for  each  of  the  various
          types  of  injury;  including basal  necrosis,  basal  scale (unidenti-
          fied),  pine  needle scale (Phenacaspis  pinifoliae,  defoliators,  tip
          necrosis, chlorotic  mottle, and  weevil)  and record  the percentage
          of needles affected by each agent.

     F.    Save representative  specimens  of different  pathology  types;  kill
          and fix in FAA (formalin-aceto-alcohol).

     G.    Save any  insects in 80% ethanol, label, record  any comments.

     H.    Complete  measurements  of cross-sectional  area, needle  length,  and
          moisture percentage.

     I.    Prepare all stripped needles for chemical analyses.
                                      194

-------
                                  APPENDIX D
                   CHECKLIST OF PREVALENT INSECTS ASSOCIATED
                 WITH PONDEROSA PINES IN SOUTHEASTERN MONTANA
Order
Family
Species
Neuroptera
Homoptera
Hemiptera

Coleoptera
Mantispidae

Chrysopidae

Aphidae

Coccidae

>i     n

Coreldae

Buprestidae

Cerambycidae
                     Coccinellidae
                     Curculionidae
ClimacieTIa brunnea (Say)

Chrysopa sp.

Cinara ponderosae (Wms.)

Matsucoccus secretus Morrison

Phenacaspis pinifoliae  (Fitch)

Leptoglossus sp.

Bupresti s spp.

Arthopalus  foreicol 1 is  (Lee. )

Batyle Thomson  sp.

Monochamus  maculosus Hald.

Neoclytus muricatul us  (Kby.)

Prionus imbricorm's (L. )

Anatis quindecimpunctata  (01iv. )

Cleis pi eta (Rand. )

Hippodamia  convergens  Guerin

Neomysia  Casey  sp.

Hagdalis  Germar spp.

Pissodes  probably  fasciatus  Hopk,
                                       195

-------
APPENDIX D (continued)
Order Family
Coleoptera Curcul ionidae
n n n ii
" " Scarabaeidae
n n n M
" " Scolytidae
n n n n
n n n n
n n n n
n n n n
n n n n
n n n n
n n n n
n n n n
Diptera Cecidomyi idae
" " Tachinidae
Lepidoptera Arctiidae
n n M n
" " Gel echi idae
" " Geometridae
" " Lasiocampidae
" " Olethreutidae
ii ii n n
n n n n
" " Pyralidae
Species
Scythropus elegans (Couper)
Scythropus probably albidus Fall
Cotalpa lanigera Wick.
Polyphylla decimlineata (Say)
Conophthorus Hopk. probably
ponderosae
Dendroctonus ponderosae Hopk.
" " valens LeConte
Dryocoetes probably confuses Sw.
Hylurgops subcostulatus (Mann.)
Ips calligraphus (Germar)
Ips p i ni (Say)
Pityophthorus sp.
Scolytus sp.
Unidentified species
Euphorocera sp. near edwardsii
(Williston)
Dasychira near grisefacta Dyar
Hypantria cunea (Drury)
Coleotechnites sp.
Phaeoura mexicanaria (Grote)
Malacosoma disstria Hbn.
Laspeyresia sp.
Petrova sp.
Rhyacionia (Busck) sp.
Dioryctria abietella (D.&S.)
                                     196

-------
APPENDIX D (continued)

Order	Fami 1y
                         Species
Lepidoptera
Hymenoptera
Pyralidae

Pieridae

Yponomeutidae


Diprionidae

 ii       ii

Ichneumonidae

Pteromalidae


Vespidae
Dioryctri'a spp.

Neophasia menapia Felder

Zellaria probably haimbachi
 Busck

Dlprion sp.

Neodiprion fulvlceps complex

Ichneumon pulcherior (Heinrich)

Coeloplsthia suborbicularis
 (Provancher)

Vespula spp.
                                      197

-------
                                  APPENDIX E

                     PRE-OPERATIONAL ON SITE MEASUREMENTS
                            Diameter   % Dwarfed   Tree   Tree    Porcupine
F Number Site
F-1300 N #4
F-1300.5
F-1305
F-1305.5
F-1310
F-1310. 5
F-1315
F-1315. 5
F-1320 "
F-1320.5
F-1850 NE #1
F-1850. 5
F-1855
F-1855.5
F-1860
Approx.
Height
(ft)
22
24
18
27
24
30
27
23
25
24
42
42
43
41
24
Breast
Height
(in)
11.9
10.8
14.6
12.6
18.0
17.7
15.4
15.3
12.1
11.1
21.3
15.9
15.8
18.1
11.1
Cones Class Class Attack
(50-100 (0-4) (0-4) (trunk and/
cones/tree) 1974 1975 or limbs)
2
7
3
0
32
0
6
2
3
0
20
18
20
34
42
0 0
0
0 0
0
0 0
0
0 0
0
0 0
0
0 0
0
0 0
0
0 0
NE*
Pt
NE
NE
NE
NE
NE
NE
RJ
P
NE
KE
NE
NE
NE
*Not evident
tPrevious
tRecent (within last 12 months)
                                     198

-------
APPENDIX E (continued)
F Number
       Approx.
       Height
Site    (ft)
Diameter   % Dwarfed   Tree   Tree   Porcupine
 Breast      Cones     Class  Class   Attack
 Height     (50-100    (0-4)  (0-4)  (trunk and/
  (in)    cones/tree)  1974   1975    or limbs)
F-1860.5 NE #1
F-1865
F-1865.5
F-1870
F-1870.5
F-1575 NE #3
F-1575.5
F-1580
F-1580.5 "
F-1585
F-1585.5
F-1590
F-1590.5
F-1595
F-1595.5
F-1775 NW #3
F-1775.5
F-1780
F-1780.5
F-1785
F-1785.5
30
45
45
27
30
36
29
38
32
23
20
19
30
23
20
45
42
39
32
40
50
14.3
16.7
16.4
11.0
11.5
16.6
14.8
11.9
15.4
9.5
9.2
10.5
11.9
11.6
9.4
21.6
20.7
16.6
14.3
17.0
20.2
37
10
14
26
26
3
1
3
5
0
0
0
0
0
0
76
-
45
60
68
-
0
0 0
0
0 0
0
0 0
0
0 0
0
0 1
1
0 0
1
0 1
1
0 0
0
0 0
1
1 0
0
NE
NE
NE
P
NE
R
R
P
R
P
P
R
NE
NE
P
R
P
P
P
NE
NE
                                     199

-------
APPENDIX E (continued)
F Number
       Approx.
       Height
Site    (ft)
Diameter   % Dwarfed   Tree   Tree   Porcupine
 Breast      Cones     Class  Class   Attack
 Height     (50-100    (0-4)  (0-4)  (trunk and/
  (in)    cones/tree)   1974   1975    or limbs)
F-1790 NW #3
F-1790.5 "
F-1795 "
F-1795.5
F-1350 NW #4
F-1350.5
F-1355
F-1355.5
F-1360
F-1360.5
F-1365
F-1365.5
F-1370
F-1370.5
F-1750 W #3
F-1750.5
F-1755
F-1755.5
F-1760
F-1760.5
F-1765
33
35
37
41
28
29
44
26
32
31
34
20
38
28
35
32
42
31
54
48
26
14.0
11.1
20.1
18.8
10.2
11.3
20. 1
13.2
15.3
15.9
14.3
17.3
20.1
16.2
13.7
12.6
19.4
13.5
10.2
10.5
10.0
60
62
73
37
-
30
-
13
55

48
44
60
33
25


-
23
32
-
0
-
0
-
1
-
1
-
1
-
1
-
0
-
0
-
0

0
-
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
0
p
p
p
p
NE
p
NE
P
NE
NE
NE
NE
NE
NE
NE
NE
R
P
P
NE
NE
                                     200

-------
APPENDIX E (continued)
                  Approx.
                  Height
F Number   Site    (ft)
Diameter   % Dwarfed   Tree   Tree   Porcupine
 Breast      Cones     Class   Class    Attack
 Height     (50-100    (0-4)   (0-4)   (trunk and/
  (in)    cones/tree)  1974   1975    or limbs)
F-1765.5 W #3
F-1770
F-1770.5 "
F-1200 W #4
F-1200.5
F-1205
F-1205.5
F-1210
F-1210.5
F-1215
F-1215.5
F-1220
F-1220.5
F-1475 SE #1
F-1475.5 "
F-1480 "
F1480.5 "
F-1485 "
F- 1485. 5
F-1490
F-1490.5
25
40
30
28
26
36
30
28
24
42
20
28
30
29
30
30
32
26
28
26
20
13.2
13.8
12.3
13.5
9.9
14.0
11.0
11.5
11.5
18.1
15.9
14.6
12. 1
12. 1
13.2
15.9
12.3
13.5
14.0
10.5
8.6
64
24
-
2
0
12
3
5
3
10
3
0
5
2
1
22
21
77
96
28
29
0
0 0
-
0 0
0
0 0
0
0 0
0
0 0
0
0 0
0
0 0
0
0 0
0
0 0
0
0 0
0
NE
p
NE
NE
NE
NE
NE
NE
NE
NE
NE
P
NE
NE
NE
NE
NE
NE
NE
NE
NE
                                     201

-------
APPENDIX E (continued)
F Number Site
F-1495 SE #1
F-1495.5
F-1050 SE #3
F-1050.5
F-1055
F-1055.5
F-1060
F-1060.5
F-1065
F-1065.5
F-1070
F-1070.5
F-1100 S #3
F-1100.5
F-1105
F-1105.5 "
F-1110
F-1110.5
F-1115
F-1115.5
F-1120
Approx.
Height
(ft)
26
26
30
35
38
35
26
33
26
26
36
33
42
31
30
34
28
35
45
29
34
Diameter
Breast
Height
(in)
18.8
12. 1
13. 1
17.3
19.3
15.6
14.5
18.1
11.5
14.6
16.2
18.0
15.8
9.9
12.1
13.5
12.7
15.1
16.8
11.1
19.1
% Dwarfed
Cones
(50-100
cones/tree)
26
26
9
10
4
6
10
8
4
10
12
-
-
-
-
-
-
-
-
-
-
Tree
Class
(0-4)
1974
0
-
0
-
0
-
0
-
0
-
0
-
0
-
0
-
0

0
-
0
Tree
Class
(0-4)
1975
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Porcupine
Attack
(trunk and/
or limbs)
NE
NE
P
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
                                     202

-------
APPENDIX E (continued)
                            Diameter   % Dwarfed   Tree   Tree   Porcupine
F Number
F-1120.5
F-1825
F-1825.5
F-1830
F-1830.5
F-1835
F-1835.5
F-1840
F-1840.5
F-1845
F-1845.5
F-1275
F-1275.5
F-1280
F-1280.5
F-1285
F-1285.5
F-1290
F-1290.5
F-1295
F-1295.5
Approx.
Height
Site (ft)
S #3 32
E #1 45
51
45
42
24
27
32
25
38
34
E #3 24
25
30
24
38
37
27
37
35
33
Breast
Height
(in)
9.9
15.8
14.0
25.1
20. 1
15.0
13.0
20.0
11.0
20.3
13.7
16.1
14.2
17.8
18.3
14.8
15.6
13.7
15.9
20. 1
16.6
Cones
(50-100
cones/tree)
-
29
31
32
46
41
39
47
62
30
94
5
0
1
6
3
1
2
2
0
1
Class
(0-4)
1974
-
0
-
0
-
0

0
-
0
-
0
-
0
-
0
-
0

0
-
Class
(0-4)
1975
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Attack
(trunk and/
or limbs)
NE
NE
NE
R
NE
NE
NE
R
R
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
                                     203

-------
APPENDIX E (continued)
F Number
       Approx.
       Height
Site    (ft)
Diameter   % Dwarfed   Tree   Tree   Porcupine
 Breast      Cones     Class  Class   Attack
 Height     (50-100    (0-4)  (0-4)  (trunk and/
  (in)    cones/tree)  1974   1975    or limbs)
F-1250 E #4
F-1250.5
F-1255
F-1255.5
F-1260
F-1260.5
F-1265
F-1265.5
F-1270
F-1270.5
F-1225 E #5
F-1225.5
F-1230
F-1230.5
F-1235
F- 1235. 5
F-1240
F-1240.5
F-1245
F-1245.5
26
18
28
30
30
24
15
17
18
16
36
33
33
25
37
40
29
34
35
28
21.0
9.5
14.3
11.6
12.3
12. 1
9.4
13.2
10.7
8.9
21.3
21.1
20.7
12.7
23.6
18.3
11.6
18.8
26.1
17.2
0
3
3
6
3
2
0
1
0
0
25
20
3
3
4
3
4
5
5
6
0
-
0
-
0
-
0
-
0
-
0
-
0
-
0
-
0
-
0
-
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
p
p
p
R
p
p
R
p
p
R
R
R
P
R
P
P
R
NE
R
R
                                     204

-------
                                 APPENDIX  F

                       BIBLIOGRAPHY TO THE MATSUCOCCUS
                     (Homoptera,  Coccoidea, Margarodidae)


Carle,  P.,  J.  P.  Carde,   and  C.  Boulay.   1970.   Comportement  de  piqure  de
     Matsucoccus  feytaudi  Due.   (Coc.   Margarodidae)   caracterisation   histo-
     logique   et  histochimique   des   desorganisations  engendrees  dans   le
     vegetal   (Plnus  pinaster   Ait.  var.  mesogeensis).   Ann.  Sci.   Forest
     27(1):89-104~:

Herbert,  F.  B.   1921.  The  Genus  Matsucoccus with a  new species.   (Hemip. -
     Homop.).  Proc.  Entomol. Soc.  Washington 23:15-22.

Keen,  F.  P.   1928.   Insect  enemies  of  California pines and  their control.
     Calif.  Div.  Forestry.  Bui. 7,  113pp.

Keen,  F.  P.   1952.   Insect Enemies of Western Forests.   Revised ed. , U.S.D.A.
     Misc. Publ.  273.  280  pp.

McCambridge,  W.   F.  and  D.  A.  Pierce.   1964.   Observations  of  the  Life
     History  of   the  Pinyon  Needle  Scale, Matsucoccus acalyptus  (Homoptera,
     Coccoidea, Margarodidae).   Ann. Ent. Soc. Am.  57:197-200.

McKenzie,  H.  L.  1941.   Injury  by  Sugar Pine Matsucoccus  Scale  Resembles
     That of Blister Rust.  Jour. For.   39:488-489.

                   1941.    Matsucoccus  bisetosus  Morrison,  a  Potential   Enemy
     of California Pines.   Jour. Econ. Ent.  34:783-785,  illus.

	.    1943.   The  Seasonal  History  of   Matsucoccus   vexillorum
     Morrison  (Homoptera:    Coccoidea:    Margarodidae).   Microentomology   8
     (Part 2):42-52,  illus.

	,   L.   S.  Gill  and Don  E.  Ellis.   1948.  The  Prescott  Scale
     (Matsucoccus  vexi1lorum)  and  Associated  Organisms that  Cause  Flagging
     Injury  to Ponderosa  Pine in the  Southwest.   Jour. Agr.  Res.  76:33-51,
     illus.

Schvester,  D.   1974.   Bio-ecologie   des  Matsucoccus  (Coccidae  Margarodidae)
     en  particulier   de Matsucoccus  feytaudi  Due.   In:  Ecologie  Forestiere.
     P. Pesson, comp.   241-256.
                                      205

-------
Siewniak, M.   December, 1971.   Uszkadzanie sosny pospolitej (Pinus silves-
     tris) przez czerwca korowinowca (Matsucoccus p i m' Green 1925; Margarodi-
     dae, Coccoidea).   Sylwan.   115(12):35-41.

	.   1972.  Potrzeby uwzglednienia zjawiska sprzezenia zwrotnego
     w badaniach entomologicznych ukladu pasozyt-zywiciel.   Ekol. Pol., Ser.
     B. 18(l):29-37.

Taketani, A.  1972.   Studies on a margarodid scale, Matsucoccus matsumurae
     (Kuwana) (Hemiptera;  Coccoidea).   Jap. Forest Exp. Sta. Bui. 246:1-9.

Washburn, R.  L.   1962.   Forest Insect Condition in the United States, 1961.
     Intermountain States.   U.S.  Forest Service, Wash., D.C.  16-20.
                                     206

-------
                                  APPENDIX G

TABLE G.I.  PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENT FOR PPM FLUORIDE
            IN ADULT WORKER HONEYBEES VERSUS POLLEN, 1975 AND 1976 DATA


  N = 114, Chi Square =       3.841, F-Value =      3.0700000, Alpha = .0500
                     The Transformation Codes are 0 and 0

                                                  Yl           Y2
	(Honeybees)    (Pollen)	

               Mean                             9.57        1.89

               Variance                        16.311       0.840

               Standard Deviation               4.0386      0.9167

               Standard Error                   0.3782      0.0858

               Covariance                             -0.395

The product-moment correlation coefficient is -0.10674.

The 95.0 percent confidence limits are LI = 0.2851 and L2 = 0.0788.

The eigenvalues are 16.32101 and 0.83035.

The equation of the principal  axis is:   Yl = 83.89246 + -39.17078 Y2.

The equation of the minor axis is:  Yl  = 9.52261  + 0.02553 Y2.

The 95.0 percent confidence limits to the slope of the principal axis are:
     LI = 55.20636    L2 = 14.43632
Following Pairs Show the Coordinates of
A to H for Plotting Confidence Elli
Number One Yl
A
B
C
D .
E
F
G
H
10.5071
8.6350
9.5711
9.5711
8.6296
10.5125
9.5765
9.5656
Points
pse.
Y2
1.8974
1.8974
2.1099
1.6849
1.9214
1.8733
2.1097
1.6850
                                     207

-------
TABLE G.2.   PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENT FOR PPM FLUORIDE
            IN ADULT WORKER HONEYBEES VERSUS POLLEN, 1975 DATA


  N = 66, Chi Square =       3.841, F-Value =      3,1400000, Alpha = .0500
                     The Transformation Codes are 0 and 0
                                                  Yl           Y2
	(Honeybees)    (Pollen)	

               Mean                             8.13        1.90

               Variance                         7.346       0.776

               Standard Deviation               2.7103      0.8806

               Standard Error                   0.3336      0.1084

               Covariance                             0.305

The product-moment correlation coefficient is:  0.12801.

The 95.0 percent confidence limits are LI = 0.1177 and L2 = 0.3590.

The eigenvalues are:  7.36038 and 0.76134.

The equation of the principal axis is:  Yl = 32.91300 + 21.55098 Y2.

The equation of the minor axis is Yl  = 8.22019 + -0.04640 Y2.

The 95.0 percent confidence limits to the slope of the principal axis are:
     LI = 7.47355        L2 = 24.81377
Following Pairs Show
A to H For Plotti
Number Two
A
B
C
D
E
F
G
H
the Coordinates of Poi
ng Confidence Ellipse
Yl
8.9675
7.2962
8.1318
8.1318
8.9743
7.2893
8.1192
8.1444
nts
Y2
1.9045
1 . 9045
2.1761
1.6330
1.9436
1.8655
2.1755
1.6336
                                     208

-------
TABLE G.3.  PEARSON PRODUCT-MOMENT CORRELATION COEFFICIENT FOR PPM FLUORIDE
            IN ADULT WORKER HONEYBEES VERSUS POLLEN, 1976 DATA


  N = 48, Chi Square =      3.841, F-Value =      3.1900000, Alpha = .0500
                     The Transformation Codes are 0 and 0

                                                  Yl          Y2
	(Honeybees)    (Pollen)	

               Mean                            11.55        1.88

               Variance                        22.148       0.948

               Standard Deviation               4.7061      0.9736

               Standard Error                   0.6792      0.1405

               Covariance                             -1.338

The product-moment correlation coefficient is:   -0.29208.

The 95.0 percent confidence limits are LI = -0.5321 and L2 = 0.0086.

The eigenvalues are:  22.23181 and 0.86378.

The equation of the principal axis is:  Yl = 41.56816 +  15.90366 Y2.

The equation of the minor axis is:  Yl = 11.43132 + 0.06288 Y2.

The 95 percent confidence limits to the slope of the principal axis are:
     LI = 210.50580      L2 = 8.23499
Following Pairs
A to H For
Number One
A
B
C
D
E
F
G
H
Show the Coordinates of Points
Plotting Confidence Ellipse
Yl Y2
13.2087
9.8913
11.5500
11.5500
9.8158
13.2842
11.5715
11.5285
1.8875
1.8875
2.2306
1 . 5444
1.9965
1.7785
2.2293
1.5457
                                     209

-------
                   APPENDIX H
MEAN FLUORIDE  IN  ADULT WORKER HONEYBEES
1975-1976.
August 1975
Each value represents
a different hive 	
Site
SW1
SW2
SW3
SI
S4
S5
SET
SE2
NE2
NE3
NE4
N2
NE10
SE11
S9
SE10
SE12
SE9
NE5
NE2a
SE6

Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
Bees
Pollen
1
8.6
1 .5
7.6
1 .1
8.5
6.3
2.9
7.2
0.5
8.8
2.3
8.2
1.7
7.8
1 .1
6.5
1.2
6.8
1 .3
6.6
1 .9
4.5
1.5
55.7
2.9
8.5
1.0
10.6
2.4
5.6
1 .8
9.8
1.6
10.2
1 .3
6.2
2.0
11 .1
10.6
2
4.7
1 .5
6.2
1.8
10.7
8.9
4.1
6.2
0.6
7.0
2.5
9.7
1.5
6.3
1 .1
7.5
2.5
8.9
3.2
1 .4
4.6
2.0
58.8
1 .5
11 .1
1.2
15.2
3.0
6.7
2.6
10.6
1.8
7.2
1.2
6.3
2.6
6.9
18.3
3
9.9
2.2
5.3
2.3
7.5
6.3
2.3
8.1
0.4
7.6
2.4
10.0
1.4
5.4
1.9
8.1
1.5
9.6
1.9
9.1
1.9
3.4
2.1
33.5
1.9
7.8
1.6
20.5
2.8
10.5
3.6
7.2
1.7
12.1
1.3
8.4
1 .8
7.4
8.2
4
9.3
2.0
12.2
2.8
7.9
5.1
2.9
9.7
0.9
5.4
2.2
8.7
1.4
10.4
1 .6
9.7
1.4
8.5
2.4
6.9
0.6
6.2
1.4
44.9
1 .6
--
10.5
3.1
6.2
2.0
8.0
1.6
11.2
1.8
7.8
0.6
7.3
11.4
Each value
represents
a combined
sample from
8-12 hives
5
8.5
8.5
10.0
8.2
9.7
9.3
12.0
9.7
10.2
9.2
8.1
7.1
47.6
--
16.5
8.3
7.8
10.6
9.3
10.6

6
8.3
8.2
10.9
8.8
9.6
8.4
9.7
11.2
8.5
8.6
7.0
6.3
60.7
--
19.2
7.4
9.4
9.9
6.9
8.2

August
September 1976
Each value represents
a different hive
1
8.8
2.2
12.0
5.0
6.7
2.1
6.4
2.5
7.8
1.7
15.2
2.7
15.8
1.2
6.7
0.8
16.7
1.6
12.6
1.6
16.2
2.1
3.9
2.3
77.6
2.9
3.8
0.6



12.5



2
11.8
2.5
10.2
1.1
8.9
1.7
5.8
2.9
8.0
2.2
10.4
2.2
17.1
0.6
8.4
1.2
15.4
1.4
14.4
0.8
16.8
1 .2
4.0
110.8
2.7
5.9
2.8



20.6



3
7.8
1.2
7.8
1 .5
14.2
0.7
5.8
3.3
11.4
2.8
16.2
2.1
14.7
0.7
13.1
1.0
21.5
1.3
23.1
1.5
10.4
0.9
3.0
99.8
1.9
4.0
3.4



--



4
13.1
2.6
9.9
1.7
9.1
1.3
5.9
4.1
12.4
1.5
15.4
3.0
20.2
0.8
10.5
0.8
17.5
1.5
13.8
2.8
11.0
3.1
--
104.8
1.7
3.2



--



Each value
represents
a combined
sample from
8-12 hives
5 6 H20
0.6
10.1 8.4 0.2
11.0 10.6 0.6
5.7 6.3 0.1
13.6 12.1 0.6
12.0 13.3 0.6
11.0 10.5 0.6
12.4 13.5 0.6
15.8 14.1 0.3
11.7 11.7 0.6
12.5 12.7 0.6
1.7 2.3 0.1
98.3 95.4 10.5
—







                       210

-------
                                  APPENDIX I

TABLE I.I.  BASIC STATISTICS FOR FLUORIDE IN ADULT WORKER HONEYBEES, 1975.

N
Mean
Median
Variance
Stand. Dev.
Coeff. Var.
= 79
Statistic
8.36
8.00
7.88
2.80
33.55
0 Classes
Standard Error
0.315
0.395
2.954
Transformation Code = 0
Confidence Limits
95%
7.73 8.99
7.21 8.78
27.67 39.43

TABLE 1.2.
BASIC STATISTICS
FOR FLUORIDE IN
ADULT WORKER HONEYBEES, 1976.

N
Mean
Median
Variance
Stand. Dev.
Coeff. Var.
= 53
Statistic
11.27
11.00
25.39
5.03
44.67
0 Classes
Standard Error
0.692
0.867
5.133
Transformation Code = 0
Confidence Limits
95%
9.88 12.67
9.24 12.74
34.36 54.99

F = 3.2216, which is highly significant.

The exact probability of F = 3.2216 with 78 and 52 D.F.  is 0.00003; therefore
the variances for 1975, 1976 are unequal.  (H.:af ^ a|).
                                     211

-------
APPENDIX I (continued)
TABLE 1.3.  PPM FLUORIDE IN ADULT WORKER HONEYBEES, 1975 AND 1976, TWO LEVEL
            NESTED ANOVA.

Level
2 (years)
1 (sites)
0 (hives/site)
Level
2
1
0
*Exact probability
** Exact probability
SS
269.004
1339.043
596.458
Anova Table
DF MS FS
1 269.003 6.3928*
32 1.845 6.8753**
31.9 42.079
98 6.086
VARIANCE COMPONENTS
Variance Component Percent
3.
9.
6.
of FS = 6.3928; DF
of FS = 6.8753; DF
577 18.947
215 48.813
086 32.238
1, 31.9; is 0.0125.
32, 98; is less than 0.001.

TABLE 1.4.  BASIC STATISTICS FOR FLUORIDE IN ADULT WORKER HONEYBEES.
                                 1975 VS. 1976
                    (Approximate test of equality of means
                       when variances are heterogenous.)




Sample Number
(1975) 1
(1976) 2
Mean
8.36
11.27
Variance
7.883
25.396
F* = 14.648
DF = 1 and 73.8
 t = 3.83
F* is greater than F ^^  ?3 ^ = 11.9; we reject the null hypothesis that
samples were drawn from populations with equal  mean fluoride content.
The exact probability of F = 14.648 with (1, 73.8) DF is 0.00051.
                                      212

-------
                                  SECTION 6

        THE EFFECTS OF* COAL-FIRED  POWER PLANT EMISSIONS ON VERTEBRATE
            ANIMALS IN SOUTHEASTERN MONTANA (A REPORT OF PROGRESS)

                                     by

   R. A. Lewis, M.  L. Morton, M. D. Kern, J. D. Chilgren, and E. M. Preston


                                INTRODUCTION

     Vertebrate responses to air pollution vary seasonally with the quality and
quantity of  ecosystem  resources,  with  the sex, age,  or physiological state of
the  organism,  and with  secondary  stressors,  such  as   disease,  competitive
interactions,  or  other  pollutants.   Hence,  field  work in this project has em-
phasized  the  description  and evaluation  of annual   and  life cycles  and the
elucidation of mechanisms  that  regulate these cycles.

     The preoperational (baseline)  phase  of the investigation is nearing com-
pletion, although  analyses of  this work will  continue  for some time.  Major
objectives for this phase of the study are to:

     1.   Measure  and predict  changes  in  population structure and/or dynamics
     of grassland birds and small  mammals as a function of  annual, seasonal, and
     life  cycles  as well  as other  environmental  information  including biotic
     interactions  and  physical   factors  that  influence  the structural  and
     dynamic processes under study;

     2.   assess the specific effects of air pollution, if possible,  upon avian
     and mammalian  population  structure  and dynamics as well  as upon specific
     organ systems, and

     3.   evaluate   selected   physiological,   biochemical,  and   behavioral
     functions  that  may   have potential  for  sensitive  assay  of  pollution
     challenge.  Hopefully, low levels  of  pollution  stress will  be identified
     before serious or irreversible  effects occur.

     Although  considerably  less  substantial  than  that for birds, the baseline
work on mammals  is more complete at this  time and is therefore emphasized in
this synopsis.  The preoperational components of the study will be completed for
mammals in 1977, but not until 1978 for birds.  The bulk of the  remaining bird
work will  be  performed  under grant by appropriate  investigators.   The major
components are:

     1.   Reproductive and developmental biology;

                                      213

-------
     2.    measures  of condition, physiologic  stress,  homeostasis,  and adapta-
          tion;

     3.    population biology, and

     4.    histological  cycles  of  organ-systems of potential  or  probable con-
          cern.

     Field experiments,  now completed,  on the effects of sulfur dioxide on the
population movements  and  physiology  of  small  mammals will  be  treated  in  a
subsequent report.

     Avian species  being studied are the Western Meadow!ark (Sturnella neglec-
ta).  Mourning  Dove  (Zenaidura macroura),  Lark Bunting  (Calamospiza melano-
corys),   Vesper  Sparrow  (Pooecetes  gramineus),  and  Lark  Sparrow  (Chondestes
grammacus).  The mammals of greatest importance are the Deer Mouse (Peromyscus
maniculatus)  and Prairie Vole (Microtus ochrogaster).

     The reproductive and developmental  portions of the study emphasize growth
and development  of the young, bioenergetics,  productivity,  and  the regulation of
reproductive processes  and  molt.   The  physiological  component  treats  those
functions  that  reflect  condition and  vigor  of  the  species  and  their stress
responses.  Population parameters,  such  as abundance  and dispersion of species,
sex, and age ratios, and some of the mechanisms that  effect population adjust-
ments  are also  of  concern.   Finally,   the  study of  histological  cycles  is
designed to help interpret mechanisms at all levels and to identify direct and
indirect (e.g.,  intercurrent disease) effects of pollution.

                                   MAMMALS

INTRODUCTION

     The  study  of  small  mammals  is  largely  confined  to  the mice  and  voles
(Cricetidae).  These are  the  most  abundantly distributed  and easily trapped
mammals  in southeastern  Montana.   Some  information is  also  being  gathered on
members  of the  families  Heteromyidae  (including  pocket  mice),  and Sciuridae
(squirrels, chipmunks, etc.).

     The use of small mammals as indicator species of air pollution is not new
(e.g., Kay, 1974),  but  it has not received much attention.   Although the impact
of  small  mammals on the character of grassland  ecosystems is reasonably well
documented (Golley  et al_. ,  1975),  their potential importance with respect to
ecosystem responses to  air pollution has rarely been  considered.  Small mammals
in the Northern  Great Plains contribute to maintenance of the health and vigor
of the grasslands.  Mechanisms include the following:

     1.    Grazing,  within limits,  promotes the growth of plants.  Thus, grazing
          by rodents may  have  a significant effect on the productivity of the
          range.

     2.    Burrows of fossorial  mammals may increase water  storage and retention
          in  the  soi1.


                                      214

-------
     3.    Burrowing increases transport of minerals from deeper to upper soil
          layers and may  increase the rate of decomposition of organic matter.

     4.    Feces may enrich soil biota.

     5.    Specific  effects  depend in  large part  upon  habits  and population
          structure of the  predominant species in a community.  For example,
          some species consume  insects (many of which are harmful from man's
          point of view)  or weed seeds, whereas others can damage agricultural
          crops.

     6.    Rodents are  important  food of many avian and mammalian predators.

     Many changes in rodent population structure would be expected to have an
appreciable impact upon grasslands.

STUDY AREAS AND METHODS

     The information on small mammals presented in this  report was gathered at
selected sites east or  southeast of Colstrip, Montana,  between the spring of
1974 and the  fall  of  1976.   Investigators were  in  the field each month.  Thus
seasonal changes  as well  as instantaneous evaluations of many functions were
assessed.  In  1976  only the McRae site remained  under intensive investigation.

     Five  locations were selected,  surveyed,  and  staked  in  mixed sagebrush-
grassland habitat  in  the spring of  1974.  Four of these sites (Kluver North,
Kluver East, Kluver West, and McRae1s) were chosen because of their proximity
to exclosures constructed for grassland studies (see other reports, this vol-
ume).  The  remaining  site  was  located on  Pony Creek  about  one km  west of
Gar-field's  ranch  house.   Square  grids  150 m  on a side  (2.25  ha)  were con-
structed.   Trapping stations were 15 m apart  (121  stations  per grid).  Each
month during the 1974 and 1975 field seasons, Sherman live traps (7.6 x 7.6 x
25.4 cm) were baited with rolled oats for three or four  nights with free entry
and  exit  (prebaiting).   This was  followed by  a one-day capture,  mark, and
release  (CMR)  trapping  session.  In  1976  the  traps  were prebaited  for two
nights followed  by three successive  sessions  of  capture, mark,  and release.
This procedure was  repeated at three-week  intervals..  Traps were set at dusk
and picked up  at dawn.   In cold weather  tissue paper for nest material and  large
quantities  of  bait  were provided to  enhance survival.  The toes (no more than
one per foot)  of  trapped mice were clipped to permit individual identification.
Mice were sexed, aged, and weighed to the nearest 0.1 g on a  pan  balance.  They
were also examined externally for signs of sexual activity and pelage changes.
All were released at their station of  capture.

     Specimens were collected at weekly intervals in live traps and transported
to the laboratory  at Fort Howes Ranger Station.  Linear  dimensions (body, ear
to  notch,   hind   foot,  and  tail)  were  measured with  precision  calipers.
Following ether anesthesia,  two  blood samples were taken  from the orbital  sinus
in heparinized capillary tubes.  Hematocrit was determined from these samples
following high-speed centrifugation  for 10  minutes.  Hemoglobin concentration
was  measured  spectrophotornetrically  by the cyanmethemoglobin  technique, and
plasma protein  concentration was measured  with a diffraction meter.  Heart,

                                      215

-------
lungs, kidneys, adrenals, spleen, liver, and reproductive tracts were removed
and  fixed  in Bouin's  solution  or  10%  neutral  formalin, and  after  one week
weighed to  the  nearest 0.2 mg on a torsion balance.  After weighing, tissues
fixed  in  Bouin's  were  transferred  to  70%  ethanol.   Reproductive  tracts
collected  in  September and October  1975  were supplied by  Don  Dodge from C.
Gordon's  laboratory.   Gonadal  weights  given  are  always those  of the paired
glands.

     It was noted that reproductive readiness, large testes, or pregnancy were
not  observed  in  P.  maniculatus  weighing less than  14 g.  Therefore,  the term
"immature"  is applied to animals below 14 g body weight and  "adult" to those 14
g or heavier.

REPRODUCTIVE AND DEVELOPMENTAL BIOLOGY

     The  reproductive  and  developmental portions  of the study  emphasize de-
scription of the annual reproductive cycles of a small  set  of indigenous spe-
cies.  The evaluations  are not complete at this time.

Females

     Pregnant deer mice occurred in the samples from early spring  1974 through
late summer 1975 (Table 6.1).  The first two females  trapped in March  1975 were
pregnant.   Embryo lengths j_n utero were 60 mm and  110 mm,  the latter probably
near  term.   In  1975  the  proportion  of pregnant females decreased in July and
August  and in September only one of  20 had embryos.  None of  the 11  adults
trapped  in  October  were  pregnant.   In September 1974,  however,  one  of three
adults  was pregnant and in  late October  four of five  adults  were pregnant,
although  none  of  five  taken in  early October were pregnant.  A resurgence of
breeding in October 1974 may have been related to  unusually  wet weather during
which there was new growth of grasses.  Precipitation in October  1974 was more
than double that of the long term average.

     Litter sizes of all  species, based on  dissections of pregnant  females, are
presented  in Table 6.2.  In P_.  maniculatus litter  size (1975) based on counts
of uterine  scars was not significantly higher (0.10  < P < 0.20) than counts of
visible embryos.  Similarly,  litter  size did not vary significantly throughout
the season.

     Litter  sizes between  1975  and  1976 differed significantly  for *both the
deer mouse  (P  <  0.05)  and the prairie  vole.  Data for the  voles, however, is
based  upon  a small  sample  size  and might merely reflect  variation  in area-
specific populations.  Nearly all pregnant voles in 1976 were obtained from one
region of about 250 m2.  One  major  difficulty in the study  of the  prairie vole
is its erratic  distribution.

     A chi-square test revealed  no  difference in  distribution of embryos be-
tween the two uterine horns (P >  0.98)  in P. maniculatus.

     The weight of the  reproductive  tract  of female  mice changes seasonally in
association with the production  of  young.   Thus,  in P_. maniculatus there wso
about a four-fold difference between monthly mean  ovarian weight  in winter and

                                     216

-------
TABLE 6.1.   PERCENT OF PREGNANT FEMALES IN SAMPLES OF Peromyscus  maniculatus
COLLECTED NEAR COLSTRIP.

Month
July 1974
August
September
October
November
December
January 1975
March
April
May
June
July
August
September
October
April 1976
May
June
July
August
September


Number in
Sample
4
8
14
10
8
10
3
2
9
6
8
9
10
20
11
18
4
13
7
10
3
Percent
Pregnant
75.0
62.5
28.6
40.0
0.0
0.0
0.0
100.0
77.8
66.7
75.0
33.3
40.0
5.0
0.0
33.3
25.0
46.2
57.1
40.0
33.3

TABLE 6.2. LITTER SIZE
IN MICE COLLECTED NEAR COLSTRIP.

Species
Peromyscus maniculatus




Microtus ochrogaster

Perognathus fasciatus

Reithrodontomys megalotis

Size
1975
1976
1975
1976
1975
1975
1976
1975
1976
1975
determi nant
uterine scars
uterine scars
embryos
embryos
total
embryos
embryos
embryos
embryos
+ 1976
N
9
2
42
22
24
11
6
1
2
2
Mean
5.
6.
4.
5.
5.
3.
5.
6.
5.
4.
33
50
95
68
75
00
00
00
00
00
0

0
0
0
0
0



SD
.87
--
.99
.72
.74
.85
.89
--
--
--
Range
4-
6.
3.
6
7
7
4-7
4-
2-
4-
--
--
--
7
4
6
-
-
-
                                    217

-------
summer and a  50-fold  difference  in uterine weight (Figure 6.1).  The largest
average  ovarian  weights were  recorded in  April  and May  and the  lowest  in
November,  December,  and January.  Mean  uterine weight  in October  1975  was
strongly biased by one  female  with seven near-term  implanted embryos,  whose
uterus weighed  11.9  g.   Mean  uterine  weight  of the other 19  females  in  the
sample was  only 29.8 mg.

     Ovarian weight in  £.  maniculatus  varied seasonally and in proportion to
body weight (P < 0.005).  The least squares regression line of ovarian weight
on  body  weight  in pregnant  females  was  Y = 3.13 ±  0.69X  (standard error  of
estimate of Y  on X = 3.72;  r = 0.574).

     M. ochrogaster is reproductively active throughout the  year except in mid-
winter (see Table 6.3).   A pregnant female and a male  with scrotal testes taken
on December 5 had just reached sexual  maturity, as judged by body weight,  and
were probably engaged in their first breeding effort.  The  data do not permit
accurate determination of the annual reproductive period  in  M. ochrogaster,  but
reproductive  tracts were  definitely  reduced in size during winter months  and
only one male with scrotal testes was collected from November through January
(Table 6.3).

Males

     There  is a  marked  seasonal  variation in testicular weight of P. manicu-
latus  (Figure 6.2).  Maximum weight and presumably peak  functional status were
maintained in the  samples from March through August.

     A cycle  in  seminal  vesicle  weight paralleled that of the testes during
phases  of  peak and  decreasing testicular size but  lagged during testicular
recrudescence.  This expected  relationship  illustrates  the trophic effect of
testosterone  on  sexual  accessory  organs  (McKeever,  1964a,  b;  Cinq-Mars  and
Brown,  1969;  Chapman,  1972).   The  decision   to  assign adult status  to  £.
mam'culatus with a body weight of 14 g or more was  based  upon visual inspection
of  reproductive  tracts.   Flake  (1974)  found  minimum  weight of  pregnant  £.
maniculatus to be  18  g, but investigators in this research found it to be at
least  as low as  14.3 g.   Individual  males of  intermediate  size  exhibited a
seasonal change  in  tendency  to reach sexual maturity.   Males in May and June
tended  to  have larger  testes  and seminal vesicles  than individuals of com-
parable  size  later in the reproductive season in July and August (Figure 6.3).
Testicular  weights  differed significantly  (P  < 0.05)  among  all  body weight
groups  except 14.0 to 14.9 g, as  did  seminal  vesicle weights  (P  < 0.01)  in
animals weighing  16.0  g or more.

     The breeding season in eastern Montana has not been previously documented
for any species of mouse,  but in  Colorado pregnant deer mice (£. maniculatus)
were recorded primarily from April through September (Lechleitner, 1969),  and
in Wyoming from April  through August (Brown, 1966).  Data show that the repro-
ductive period of deer mice in  the  Col strip area extends  from mid-March to mid-
September.

     The distinct  breeding  cycle  in  £. mam'culatus results in pronounced and
regular variation in the number and proportions of immatures captured (Figures

                                     218

-------
      20
    £
        0
                                     (2)
(8)
 kf >  'J' (20,
   Xi^t\l JI3)
\^ I



t
                                                                                          1.4
                                                                                        -  1,2
                                                                                          1.0
                                                                           0,8
                                                                                          0,6
                                                                           0,4
                                                                                         0.2
                                                                                         0.0
                                                                                 CD

                                                                                 k-
                                                                                 0>
                                                                                5
Figure 6.1
JASONDJFMAMJJASO
Seasonal  changes in  mean  ovarian weight  (solid  circles) and  uterine weight  (squares)  in
Peromyscus  maniculatus  near  Colstrip,  Montana, July  1974 to October 1975.   Vertical  lines
indicated ± 2 SE for ovarian weight.  Sample size for both organs shown in parentheses.

-------
        500
ro
ro
CD
        100
      o>
      E
      
-------
     400
350
                     May,June



                     July, August
                                                 (23)
300
  o>
  E
                                     (7)
  
                                                        100-5
                                                             a>
 E

(/)

 c
 o
 0>
Figure 6.3.    Early (May, June) and late (July, August) summer differences in

             testes weight (histograms) and seminal vesicles (triangles) in

             young   Peromyscus   maniculatus.    Sample   size   shown   in
             parentheses.
                                 221

-------
      TABLE 6.3.  BODY AND REPRODUCTIVE ORGAN WEIGHT IN Hicrotus ochrogaster COLLECTED  NEAR COLSTRIP,  MONTANA
ro
no
July 1974 to August 1975.

July 28
August 9

August 20




September 3



November 27



December 5




December 17








Sex
1974 M
M
M
F
F
F
M
M
F
F
F
F
F
F
M
M
F
M
M
M
M
F
M
M






Body
Weight
g
25.4
23.4
28.8
28.3
13.3
34.2
38.9
30.5
28.4
44.5
34.1
42.3
19.1
24.5
26.0
23.7
28.9
21.0
31.7
26.8
25.3
24.4
36.2
20.3




*

Testes
mg
154.2
73.8
273. 81
--
--
--
534. 61
391. O1
--
--
--
--
--
--
82.4
66.8
--
18.8
124.8
38.0
230. 21
--
130.0
32.8






Ovaries
and uterus
mg
--
--
--
133.
34.
221.
--
--
384.
4450.
453.
1820.
14.
44.
_
--
130.
--
--
--
--
58.
--
--









8
8
0


0
O2
6
O2
8
4


6




0








January 9 1975





March 13





April 15

April 18

May 6

May 16
June 10
June 17

August 5


August 7



August 12
Sex
F
F
F
M
M
M
F
F
F
F
M
M
F
M
F
M
F
M
M
F
F
M
F
M
M
F
M
M
M
M
Body
Weight
g
25.6
29.5
30.9
22.8
34.0
22.9
25.4
23.6
24.2
27.8
36.0
28.3
38.0
36.2
28.0
35.1
33.3
30.3
30.8
32.7
27.1
31.2
39.2
32.0
37.2
40.3
37.6
31.0
20.7
23.6
Testes
mg
--
—
--
80.0
84.0
34.0
--
--
--
--
598. I1
288.4
--
710. 21
--
590. 21
--
387. 21
505.4
--
--
780. 31
--
104. 91
76. 91
--
920. 21
214. 61
362. O1
38.6
Ovaries
and uterus
mg
48.
58.
84.
--
—
--
94.
306.
72.
44.
--
--
1080.
--
200.
--
388.
--
--
408.
187.
--
207.
--
--
266.
--
--
--

0
0
8



6
72
0
8


O2

2

22


22
62

22


O2





     xTestes  in scrotum

     2Embryos visible in uterus

-------
6.4 and 6.5).  All  deer  mice captured in March were adults,  but one caught on
April  18 was 12 g.  In 1976 the smallest immatures captured in April were 7.0,
8.5, and 11.9 g captured on April  9, 13, and 23,  respectively.  The proportion of
immatures in the  trapped  sample increased thereafter, peaking  in October of 1974
and September  or thereafter  in  1976.   Immatures captured in  mid-winter were
undoubtedly from  the last litters of autumn.

     During much  of the  reproductive  period  females  are  heavier  than  males
(Figure 6.6).  This weight increase was undoubtedly associated with pregnancy.
No  differences  in  linear  body  dimensions  between adult males and  adult non-
pregnant females  were  found  during this time  (Tables  6.4  and 6.5).  When all
animals sampled  for the  duration  of  the  study were  compared,  however, tail
length was  significantly greater in  females (t = 1.90, P < 0.05) whereas hind
foot length was considerably greater in males (t = 3.09, P  <  0.005).  The sexes
did not differ in body length (t =  0.98,  P < 0.10) or in ear length (t = 0.87, P
< 0.10).

MEASURES OF CONDITION,  PHYSIOLOGICAL STRESS, HOMEOSTASIS, AND  ADAPTATION

     The blood vascular  system  of  mammals is  complex  but  well  characterized.
Blood  carries  erythrocytes that are involved  in  oxygen transport,  leucocytes
that mediate responses to  insult,  hormones, enzymes and other proteins,  and a
host of other  chemicals  that are characteristic  of  a  given  functional  state.
Some of the most easily assessed characteristics  include hematocrit (percentage
of packed red blood cells occupied  by a given volume of  whole blood), hemoglobin
content, erythrocyte count,  and plasma protein analysis.  The first three are
highly regulated  in vertebrate vascular systems.   Plasma protein  varies more in
relation to other factors such as blood osmotic pressure and nutritional  state.
Hematocrit,  blood hemoglobin content,  and  plasma  protein  concentration were
measured in rodents trapped in all  three years of the study.   In addition, serum
acetylcholinesterase assays have been performed on blood samples taken in 1976.

     The hematocrit ratio  and hemoglobin concentration of  M. ochrogaster were
higher in winter  (December and January) than  in  summer (April  to September),
whereas plasma protein concentration  was higher  in summer (Table 6.6).   Blood
data for  P.  maniculatus  were analyzed by sex in both winter (November to Jan-
uary)  and  summer  (April  to  September)  (Table  6.7).   As  in  M.  ochrogaster,
hematocrit tended to be higher in winter than in summer. Unlike M. ochrogaster,
however, hemoglobin of males was higher in summer than in  winter (Table 6.7).
The  data  indicate  that  mean  corpuscular hemoglobin  concentration  (MCHC)
decreased during  winter  (Table  6.8).   There was  a statistically significant
decrease in  plasma  protein concentration (PPC) during winter in both male and
female £.  maniculatus  (Table 6.7).

     In 1976 seasonal  changes in blood functions were not determined since the
sampling period spanned  only  six months  (April to September).  However,  suffi-
cient samples were  taken to determine further sexual  differences  in hematocrit,
hemoglobin, and plasma protein for both the deer mouse and  prairie vole (Tables
6.9, 6.10,  6.11).   In the vole, males had significantly higher values of hema-
tocrit and  hemoglobin than females (Table 6.9).  There was  also a tendency for
higher plasma protein  readings in males.
                                      223

-------
     50
    40
 o

 g  30

«*-
 O
 p
  . 20
 1   io
 E
       0
 rigure 6.4.
JASONDJFMAMJJA

Seasonal  change in percent of Peromyscus maniculatus captured as immatures (body weight less
than 14 g) near Colstrip, July 1974 to August 1975.

-------
IV)
rvi
en
   80



   60



   40

.c
o
o 20
    c
    OJ
    O


    I  20
      40
       60
            iill Adults

                 Immatures
                                     Males
                             (24)
             (57)  HZ!        (82)  (4|)

                       (78)  "~
                                       (36)
    Figure 6. 5.
                                                       (29)
                                    (49)

                                    ill
                                          Females
                                                                  (32)
              J
         S    0

          1974
N
                                                     M
A    M

 1975
Seasonal change in sex and age of Peromyscus maniculatus captured near Colstrip, July 1974
to August 1975.  Sample size shown in parentheses.

-------
       26
       24
       22
     o»20
ro
01
 o
GO
        16
        14
        12
        10
         8
              (4)
                                                                                      (37)
                                                                                                     (29)
                                                                                                            (18)
                          ji*—t—k\*\J II IVI %J I ^t ^
                                                                                                             (16)
              JASONDJF
                                                                    M
A     M
J      A
     Figure 6.6.     Seasonal changes in mean body weight in live Peromyscus maniculatus.   Vertical bars indicate

                    ± SE.   Sample size shown in parentheses.

-------
           TABLE 6.4.   SEASONAL CHANGES IN LINEAR MEASUREMENTS IN MILLIMETERS OF ADULT, NON-PREGNANT FEMALE
                       Peromyscus maniculatus CAPTURED NEAR COLSTRIP,  MONTANA, July 1974 to August 1975.
IX)
ro


Body Length

July 1974
August
September
October
November
December
January 1975
April
June
July
August
Total Sample
Mean
89.8
89.9
84.8
81.0
85.1
79.7
70.0
88.7
83.0
88.7
86.5
86.4
S.D.
3.2
6.4
4.9
8.1
5.8
3.1
0.0
5.1
	
11.5
3.6
7.1
(N)
12
37
33
6
11
7
3
6
2
9
5
131
Tan"
Mean
63.7
63.9
61.5
56.3
57.0
56.0
48.0
58.2
65.0
58.7
58.0
60.9
1 Length
S.D.
4.0
4.1
5.4
6.2
4.9
4.2
2.0
4.3
5.7
8.6
3.0
5.9
(N)
12
37
33
6
11
7
3
5
2
9
5
130
Hindfoot Length
Mean
19.0
18.4
18.6
18.5
18.5
18.7
17.7
18.7
19.5
19.6
19.4
18.7
S.D.
0.9
1.2
1.2
1.0
0.7
0.5
0.6
0.5
—
0.5
0.5
1.0
(N)
12
37
33
6
11
7
3
6
2
9
5
131
Ear
Mean
16.0
15.5
15.5
15.8
16.7
15.3
14.3
16.3
17.0
16.8
16.6
15.8
Length
S.D.
1.2
1.2
1.4
1.3
1.2
0.5
0.3
1.2
	
1.6
0.7
1.2
(N)
12
38
33
6
11
7
3
6
2
9
5
132

-------
     TABLE 6.5.  SEASONAL CHANGES IN LINEAR MEASUREMENT IN MILLIMETERS OF ADULT MALE Peromyscus CAPTURED NEAR
ro
co
COLSTRIP, MONTANA
, July 1
Body Length

July 1974
August
September
October
November
December
January 1975
March
April
May
June
July
August
Total Sample
Mean
87.6
90.6
82.0
85.2
83.2
79.4
73.4
84.7
88.1
83.0
85.8
87.1
89.9
85.7
S.D.
4.2
5.0
5.5
5.9
4.4
4.9
5.8
6.3
7.0
3.4
5.3
6.0
«3.7
6.6
(N)
22
65
61
17
15
11
8
7
10
10
16
26
12
280
974 to August
Tail
Mean
62.0
63.1
60.4
58.2
53.9
56.3
52.1
57.3
58.7
58.4
61.3
57.9
58.1
59.8
1975.
1 Length
S.D.
4.2
4.2
4.7
5.9
5.9
7.4
5.5
4.8
5.1
4.7
4.9
3.8
4.2
5.5
(N)
22
65
61
16
15
11
8
7
10
10
16
26
12
279



Hindfoot Length
Mean
19.4
18.6
18.8
18.8
18.8
19.0
18.5
20.9
18.7
19.5
19.5
19.7
19.4
19.0
S.D.
0.9
1.3
1.1
0.8
0.9
1.0
0.9
3.2
0.8
0.6
0.9
0.7
0.5
1.2
(N)
22
65
61
17
15
11
8
7
10
10
16
26
12
280

Ear
Mean
16.2
15.8
15.2
16.3
15.7
15.5
14.6
15.2
15.9
16.6
17.1
16.9
16.9
15.9


Length
S.D.
1.3
1.1
1.2
1.3
0.9
1.4
0.7
1.3
1.1
0.6
0.8
0.8
0.9
1.3
(N)
22
65
61
17
15
11
8
7
10
10
16
26
12
280

-------
TABLE 6.6.   HEMATOLOGICAL VALUES FOR Microtus  ochrogaster COLLECTED NEAR
COLSTRIP.
July
1974 to
August 1975
•



Hematocrit, (%)
Summer
Winter
Hemoglobin, (g%)
Summer
Winter
Plasma protein, (g%)
Summer
Winter
N
20
13
13
6
20
13
Mean
49.80
52.23
15.42
17.20
6.22
5.97
S.D.
4.42
4.90
1.26
1.18
0.44
0.43
t

1.65

2.97

1.67
P

>0.05

<0.005

>0.05

TABLE 6.7. HEMATOLOGICAL VALUES FOR
COLSTRIP- July 1974 to
Peromysus
August 1975
maniculatus

COLLECTED NEAR


Hematocrit, (%)
Summer females
Winter females
Summer males
Winter males
All females
All males
Hemoglobin, (g%)
Summer females
Winter females
Summer males
Winter males
All females
All males
Plasma Protein, (g%)
Summer females
Winter females
Summer males
Winter males
All females
All males
N
60
18
114
25
78
139
37
5
75
6
42
81
43
18
86
22
61
108
Mean
48.48
50.06
49.88
52.24
48.85
50.30
15.85
14.45
15.87
15.29
15.68
15.82
6.39
6.03
6.39
6.17
6.28
6.35
S.D.
3.44
5.58
4.42
3.07
4.05
4.30
1.31
2.64
1.40
0.46
1.55
1.36
0.75
0.60
0.56
0.46
0.68
0.52
t

1. 14
--
3.19
--
2.50

1.17
--
2.32
--
0.50

1.98
--
1.90
--
0.70
P

>0.10
--
<0.005
--
<0.01

>0.10
--
<0.025
--
>0.30

<0.05
--
<0.05
--
>0.20
                                    229

-------
 TABLE  6.8.   MEAN  CORPUSCULAR  HEMOGLOBIN  CONCENTRATION  IN  RODENTS  COLLECTED
             NEAR  COLSTRIP.  July  1974  to August  1975.


          Species  and  Season                            MCHC1

          Microtus ochrogaster
                    Summer                              31.0
                    Winter                              32.9

          Peromyscus mam'culatus
                    Summer  females                      32.7
                    Winter  females                      28.9
                    Summer  males                        31.8
                    Winter  males                        29.3


MCHCi =  100 (mean hemoglobin  concentration),         ted
                    mean  hematocnt                  K

  TABLE 6.9.   HEMATOLOGICAL VALUES FOR  Microtus ochrogaster COLLECTED NEAR

COLSTRIP. April
to September 1976.

Function
Sex
Hematocrit (%) M
F
Hemoglobin (g
%) M
F
Plasma protein (g %) M
F
N
11
12
11
10
11
12
Mean SD t P
43:73 z:?6 2-54 <0-02
15l9 0.77 2'45 <°-05
I'll ?'«n J-IS <0.40, NS
O . O / 1 . o U

TABLE 6.10.
HEMATOLOGICAL VALUE
COLSTRIP. April to
FOR Peromyscus maniculatus COLLECTED NEAR
September 1976.

Function
Hematocrit (%)
Hemoglobin (g
Plasma protein
Sex
M
F
%) M
F
(g %) M
F
N
93
49
79
37
106
55
Mean SD t * P
48^2 3' 10 4'34 <0-001
1 c no I oc
leios \zl 3-19 <0-01
6.20 1.06
6.21 0.72 "" Nb
                                     230

-------
 TABLE 6.11.   HEMATOLOGICAL VALUES OF Perognathus fasclatus COLLECTED NEAR
COLSTRIP. April
to September
1976.


Function
Hematocrit (%)
Hemoglobin (g %)
Plasma protein (g %)
N
4
5
4
Mean
55.75
17.78
5.48
SD
2.36
1.74
0.92

TABLE 6.12.   MEAN CORPUSCULAR HEMOGLOBIN CONCENTRATION IN RODENTS COLLECTED

             NEAR COLSTRIP.  April to September 1976.

Species
Peromyscus mam'culatus
Microtus orchrogaster
Perognathus fasciatus

Sex MCHC1
M
F
M
F
M+F

33.6
33.4
32.8
32.6
31.9

       100 (mean Hb concentration)          ,   , ,    -,- , ,       , „  ,,
     =	TT—r	-, as computed from Tables 9, 10, 11
                m63n ncT.
                                    231

-------
     As in 1974-75, male deer mice showed higher hematocrit readings compared
with females (Table 6.10).  Males also showed higher hemoglobin values, which
was not found in the 1974-75 data (Table 6.7).  Plasma protein  was  identical in
both sexes.  The  same  values  for P.  fasciatus are  tabulated in  Table 6.11.
Although the sample size is too small  for adequate comparison, hematocrit and
hemoglobin are a little  larger than for either the deer mouse or the prairie
vole, while plasma  protein is somewhat smaller.

     The mean  corpuscular  hemoglobin  concentration of voles and mice in 1976
was the same for both sexes (Table 6.12) and was 4-5% higher  in 1976 than for
summer voles and mice  in 1974-75.

     Weights of several  organs are presented in Table  6.13.   In  addition to
their basic reference  value, these data provide indications  of  functional state
as well as annual  cycle and life cycle trends and adjustments  to environmental
factors.  For  example,  greater liver  and kidney size  in  females  is probably
related to the increased metabolic demands imposed by reproduction.   Mean liver
weight  in  13  lactating  females was  1567.6  mg  (S.D.  = 277.5)  and  in  9 non-
lactating  females  also  collected in summer  it  was  1213.3  mg (S.D.  = 257.2).
The  difference was highly  significant (t = 3.08, P  < 0.005).   Mean kidney
weight was also larger  in these same lactating females, 363.3 mg vs. 233.1 mg
(t = 1.93,  P  < 0.05).

     The significance  of sexual dimorphism  in adrenal gland  size is unclear,
but  it is a condition common  to many  eutherians  (Christian, 1952; Chester-
Jones,  1957).   As  in  this study, McKeever (1964b) found that female P_. mani-
culatus  had  larger  adrenals  than  males.    He  showed  that there was  little
seasonal change in gland size,  indicating that the  difference is independent of
sex  hormone  levels,  unlike  the situation in  M.  montanus  (McKeever,  1959).
McKeever  (1964b)   also  found  adrenal  size  to  be  independent  of  population
density,  in  agreement  with  Bendell  (1959)  but contrary  to  the  findings of
Christian (1959) on P.  leucopus.

POPULATION BIOLOGY

     Five  species  of  mice were  captured in live traps:  the deer mouse (Pero-
myscus maniculatus), prairie vole (Microtus ochrogaster), Wyoming  pocket mouse
(Perognathus fasciatus),  northern  grasshopper mouse (Onychomys leucogaster),
and  harvest  mouse  (Reithrodontomys megalotis).   A few  immature  cottontails
(Sylvilagus  auduboni) were  captured in fall  or early winter.  Occasionally  a
thirteen-lined ground squirrel  (Spermophi1 us  tridecemlineatus) or least chip-
munk (Eutamias minimus), both diurnal  species, were taken if traps were opened
slightly before dusk.

     This investigation is aimed primarily at nocturnal mice because of their
abundance  and  diversity.   Of  the species captured,  P.  maniculatus  formed an
overwhelming portion of the catch.   Members of this species  were  taken in every
trapping session in which animals were captured.  They accounted  for  77.4% (721
of 931) of the individuals captured in  the period and 76.4% (220 of  288)  for the
period  April  to June 1976.   Furthermore, they represented 79.8%  (1178 of 1476)
of the  total  captures  made at the grid  sites.  At the end of  September 1976 this
                                      232

-------
  TABLE 6.13.   ORGAN WEIGHTS  OF  ADULT  Peromyscus  maniculatus  COLLECTED  NEAR

COLSTRIP. July 1974
to August
1975.1




Heart
Femal es
Males
Lungs
Females
Males
Liver
Females
Males
Spleen
Females
Males
Kidneys
Females
Males
Adrenal s
Females
Males
N

28
47

15
21

26
45

26
46

33
42

34
43
Mean

159.9
160.7

267.6
292.7

1427.0
1190.1

122.0
84.3

345.6
315.3

14.32
10.47
S.D.

28.5
27.3

65.9
59.7

317.6
271.8

151.4
48.7

51.5
37. 1

5.84
4.42
t

	
0.11

	
1.17

	
3.19

	
1.24

	
2.85

	
3.21
P

	
>0.45

---
>0.10

—
<0.005

—
<0.10

	
<0.005

—
<0.005

1mi11igrams
                                     233

-------
figure was  71.8%,  including  recaptures  of marked  animals (Tables  6.14 and
6.15).

     A capture summary  of  animals  by species,  sex and trapping location shows
that the five  species  were distributed in about the same proportions through-
out the study area, the Kluver East grid being an exception.  At that location
no R.  mega!otis were captured  but 68.8% (11 of 16.) of the 0. leucogaster indi-
viduals taken in the 1974-5 study were captured there (Table 6.16).

     Table  6.16  includes  only  individuals whose sex was  determined.   These
data indicate that  in  all  species except  P. maniculatus (Table 6.17) the sex
ratios were not  statistically  different.  The  sex  ratio of  P.  maniculatus
during the  spring  and  summer  was similar to that found  in 1974-75 at the same
site.   However, 3-1/2 times as  many male P.  fasicatus were trapped as females.

     During  1974-75,  59.3% (210  of  354) of P.  maniculatus  captured on grids
were males.   Recapture  rate was  58.6% in  favor of males (268 of  457).  There
was  no indication  of  differental  survival as  seen in  recapture tendencies
(Figure  6.7)  or   in  overwinter  survival   (Table  6.18).   Collectively,  this
suggests that  more males  were  trapped because more  were present.  The rela-
tively low  percentage  of  females suggests also that  the £.  maniculatus popu-
lation was  relatively  low during the time  of this study.

     Trapping success  is  summarized  in Table  6.19.  Overall success never ex-
ceeded 10%  in a given month.  Having two  traps per station undoubtedly contri-
buted  to the relatively low catch per trap;  still,  there is every indication
that mouse  populations  were low.  During  a population high, for example, trap-
ping  success can be  at least as  great  as 44% for  P.  maniculatus (Hoffman,
1955).

     There  was  considerable seasonal  variation in  the  relative abundance of
species  captured  on the  various  grids  (Figure 6.8).  P.  maniculatus was in-
variably the most abundant species,  comprising  from 52 to 95% of the total
monthly  catch  (Figure 6.8).   0.  leucogaster and  £.  fasciatus  are hibernators
and their  absence from samples  during winter  was expected.   R.  mega!otis is
probably a  hibernator  also.   However,  one was  taken  in  January 1975 and none
thereafter.   A  sharp  decrease  in  the catch of M. ochrogaster occurred after
October  1974.   Prior  to January  1975  voles comprised 13 to 33% of the catch
(Lewis and  Morton, 1975).  This decrease  in numbers of M. ochrogaster is unex-
plained but  it  could  be part  of a normal  population  cycle.  Density-dependent
cycles  occur with a  periodicity of  about four  to  five years  (Krebs, 1966;
Batzli and  Pitelka,  1970;  Christian,  1971; Krebs et al_. ,  1973) in microtine
rodents.   A  period of  unusually  severe  weather  coincided with  the  onset of
decreased catch of M.  ochrogaster and R.  megalotis.   Winter temperatures were
generally mild  in comparison to  long-term averages  through most of January
(Figure 6.9).  From late January and through early March, however,  temperatures
were at  the seasonal  low  or  oscillating  enough  to  cause a daily freeze-thaw
cycle.   Consequently,  in four  attempts at trapping in February and early March,
not a single animal was captured.  The presence of fresh  tracks in the  snow at
trap entrances indicated that  mice were abroad and active, but trap mechanisms
were invariably frozen and inoperative.  Poorly  functioning traps  also affected
trapping success during  periods of low temperature in  November and December.

                                      234

-------
     90
     80
     70
     60


  "o


  I  5°
  °>

  •o

  -  40
  O
     30
     20

                           E3 Males  (N = I86)

                           g] Females (N = I30)
                         345

                        Times Captured
Figure 6.7.     Recapture frequencies  of  Peromyscus maniculatus captured on

             grids near Colstrip, July 1975 to July 1975.
                              235

-------
IND
CO
CT>
           o
           O
           0>
           o
           V.
           0>
           0.
 10
  0
 IO
  0
 EO
 10
  0
 40
 30
 20
 90
  0
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
  0
                                         Onychomys leucogaster
                 r
                              Reithrodontomys megalotis "I
                                               Perognathus fasciatus
                                                            Microtus  ochrogaster
Peromyscus maniculatus
                           ASOND    J     FMA
    Figure 6.8.     Seasonal change in relative numbers of mice captured on grids near Colstrip.

-------
JVi
CO
       Figure 6.9.     Mean  daily  high  and  low ambient temperatures summarized at 5-day intervals (shaded area) and
                      monthly  precipitation (histograms) from July  1974  through  August 1975.   Long-term monthly
                      averages are shown by solid symbols.  Data recorded at Miles  City, Montana, airport, which is
                      about 75 km from study sites and provided by Miles  City  Flight Service.

-------
TABLE 6.14.   TOTAL AND RELATIVE  NUMBERS  OF  MICE CAPTURED IN COLSTRIP STUDY,
             July 10,  1974 to  August 12,  1975.

Gri
Individuals
Peromyscus
maniculatus
Microtus
ochrogaster
Perognathus
fasciatus
Onychomys
leucogaster
Reithrodontomys
megalotis
TOTALS
354
70
26
12
12
474
ds
Recaptures
457
63
14
13
1
548
Collect!
367
63
11
4
9
457
Total
ons captures
1178
196
51
29
22
1476
% of total
captures
79.8
13.3
3.4
2.0
1.5
100.0

TABLE 6.15. TOTAL
12 to
AND RELATIVE NUMBERS
September 10, 1976.
OF MICE
CAPTURED NEAR
COLSTRIP, April

Grids
Indi
Peromyscus
maniculatus
Microtus
ochrogaster
Perognathus
fasciatus
Onychomys
1 eucogaster
Reithrodontomys
megalotis
TOTALS
vidual
38
13
14
0
1
66
Recaptures
127
49
18
0
1
195
Collect!
182
31
6
0
3
222
Total
ons captures
347
93
38
0
5
483
% of total
captures
71.8
19.3
7.9
*
0
1.0
100.0
                                    238

-------
 TABLE 6.16.   SUMMARY OF CAPTURE RECORDS OF MICE ACCORDING TO  SPECIES,  SEX,  AND  TRAPPING  LOCATIONS  DURING
              1974-75.

Species
Peromyscus maniculatus
Mai es
Females
Microtus ochrogaster
Males
Females
ro , . „
co Perognatnus Tasciat.us
Males
Females
Onychomys 1 eucogaster
Mai es
Females
Reithrodontomys megalotis
Males
Females
McRae
XI
79
57
20
14
4
11

0
0
2
3
2R
121
94
18
21
4
6

0
0
0
1
Kluver North
I
40
20
4
6
6
2

0
1
0
4
R
61
26
3
4
1
3

0
0
0
0
Kluver East
I
45
44
5
4
0
1

5
6
0
0
R
54
51
4
1
0
0

6
7
0
0
Kluver West
I
26
15
1
1
1
0

0
0
0
1
R
13
11
0
0
0
0

0
0
0
0
Garfield's
I
20
8
9
6
0
1

0
0
0
2
R
19
7
5
7
0
0

0
0
0
0
Col lections
I
220
142
33
30
4
6

3
1
6
3

1Number of individuals
2Number of recaptures

-------
  TABLE  6.17.   SEX  DISTRIBUTION AND  SEX  RATIOS  OF  ALL MICE CAPTURED DURING
               1974-75.
Species
Peromyscus maniculatus
Microtus ochrogaster
Perognathus fasciatus
Qnychomys leucogaster
Reithrodontomys megalotis

Males1
430
72
15
8
8
Females1
286
61
21
8
13
Ratio
1.50:1
1.18:1
0.7:1
1.00:1
0.62:1
P
<0.005
NS
NS
NS
NS

1Numbers of recaptures  not  included
          TABLE 6.18.   OVERWINTER  SURVIVAL  OF  Peromyscus  maniculatus.
Age
  Period of
first capture   Males
            Recaptured
         March-August 1975  Survival
Females   Males  Females       %
Adults
July-August 74   31

                 20

                  9
             September-
               October 74
             November-
               December  74
             January 75
             TOTAL
                 68
Immatures    July-December  74  24
  23

  16

   3
  _2
  44
  16
 2
_4
17
 0
_2
 7
16.7

19.4

16.7
60.0
21.4
 7.5
                                    240

-------
                                  TABLE 6.19.   SEASONAL CHANGES IN  SPECIES  COMPOSITION  AND  TRAPPING  SUCCESS  ON  COLSTRIP  GRID.1
ro

1974
Species
Peromyscus maniculatus
Microtus ochrogaster
Perognathus fasciatus
Onychomys 1 eucogaster
Rei throdontomys megalotis
Total mice
Trap nights
% success, all mice
% success, Peromyscus only
July
51
13
0
0
0
64
1452
4.4
3.5
Aug
130
23
2
8
0
163
3388
4.8
3.8
Sept
95
19
19
2
8
143
1452
9.8
6.5
Oct
76
21
11
1
3
112
1452
7.7
5.2
Nov
33
11
0
0
1
45
968
4.6
3.4
Dec
36
14
0
0
1
50
968
5.2
3.7
Jan
31
16
0
0
1
48
484
9.9
6.4
Mar
28
2
0
1
0
31
720
4.3
3.9
Aor
98
10
0
2
0
110
1210
9.1
8.1
1975
May
76
2
5
4
0
87
1210
7.2
6.3

1976
June July May 19 Jun 10 Jun 30 Jul 21 Aug 1
77
1
1
6
0
85
1210
7.0
6.4
80 30 33 29 18 25
10 9 10 17 4
2 8 11 7 2 1
100000
002000
84 38 55 46 37 30
2178 --2 --2 — 2 -2 -z
3.9 5.2 8.3 6.6 5.4 4.2
3.7 4.1 5.0 4.2 2.6 3.5

Sep 1
30
6
3
0
0
39
	 2
5.5
4.2

       'Total numbers of animals.
       Approximately 720 each session.

-------
Other  investigators  have  reported   similar  difficulties  (Gunderson,   1950;
MacKay, 1962; Bergstedt,  1965).

     Only a  few mice of any species except P_.  maniculatus were captured  during
a given trapping session.  In such cases there is negligible value in  applying
procedures or models for predicting population size; the best estimate of pop-
ulation  size is the  minimum  number  of animals  present,  that is, the  number
actually captured  during  a trapping  session.   This number, the minimum esti-
mate,  is shown for all grid trapping efforts (Tables 6.20 - 6.23) along with an
estimate of  population size for £. maniculatus by application of the method of
Jolly  (1965).  The Jolly stochastic model was the best of four methods  tested
for predicting population size (French et a_L  , 1971), but it may underestimate
the population (O'Farrell et a]_. ,  1975).  Only at Kluver North and  McRae's were
sufficient  capture records  obtained  to  predict a  population substantially
above that of the minimum estimate.

     The similarity of  Jolly's  estimates to  minimum estimates in many  of the
trapping sessions indicates that nearly all P_.  maniculatus present on the grids
were being captured and supports the conclusion from data on  trapping success
and  sex ratios  that the  population  was  low.   The  same  might  be  inferred
regarding the  numbers  of other rodent species, but the minimum estimates are
suspect  for  several reasons.  First,  rolled oats may not be an effective bait
for primarily insectivorous species such as 0.  leucogaster;  even £. maniculatus
probably  prefers   insects  when  they  are  readily available  (Williams,   1959;
Brown,  1966;  Kritzman,  1974).  In fact seasonal changes in acceptance of bait
by rodents may influence trapping success more than population density (Fitch,
1954).  Second, weather  conditions were variable.   The unquantifiable effect of
frozen  trap  mechanism has been mentioned, but other factors  such as  rainfall
can be implicated.   For example,  only 38 captures of  P_.  fasciatus were made
during  the  entire  study but 14 (37%) of these occurred on one rainy  night in
September on the McRae grid (Table 6.20).  Weather has also been identified as
affecting the  response  of rodents to traps (Getz, 1961, 1968; Gentry et. al_. ,
1966;  Weiner  and  Smith,  1972;  Gurnell,  1976).  The  suitability  of   sage-
grassland  as  rodent  habitat  is  also  questionable.   In  running  collection
traplines  in 1975, it  was noted  that the tops  of  steep  little  knolls with
outcrops  of  rocks  such   as  those  at  Hay   Coulee  were  comparatively more
productive for trapping P. maniculatus, although in 1976 flat areas were also
highly  productive.  Microtus runways were noted in numbers in old hay fields,
creek bottoms, and roadside ditches but not in grid areas.  The best  success in
catching M.   ochrogaster on grids was at McRae's where a grassy swaTe  bisected
the grid and at Garfield's where a swale flanked one side and where both Pony
Creek and Rosebud Creek bottoms were within a few hundred meters.  Such  grassy
areas probably effect a patchy distribution of M. ochrogaster because  of their
suitability  for runways  rather than as food sources (Martin,  1956;  Fleharty and
Olson, 1969;  Grant  and Morris, 1971).

HISTOLOGY

     The study of histological  cycles is designed to  help  interpret  biological
change at the levels previously discussed (e.g.,  reproduction  and  development,
physiological condition, and population dynamics) and to aid  in the identifi-
cation of direct and indirect effects  of pollution.   Tissues  under  investiga-

                                      242

-------
TABLE 6.20.  NUMBER OF MICE CAPTURED ON KLUVER NORTH GRID.   MINIMUM ESTI-
             MATE (ME) EQUALS TOTAL CATCH ON A GIVEN TRAPPING DATE AND
             JOLLY ESTIMATE (JE) WITH STANDARD ERROR EQUALS POPULATION
             SIZE PREDICTED BY JOLLY'S (1965) MODEL.

Date
Peromyscus
Microtus
Perognathus Onychomys
nianiculatus ochrogaster fasciatus

July 15 1974
August 2
August 13
August 23
September 13
October 8
November 5
December 5
March 20 1975
May 6
May 13
May 16
July 11
July 15
July 18
ME
5
9
9
4
16
12
11
15
15
14
11
11
5
3
3
JE

18.
16.
6.
23.
17.
21.
41.
33.
16.
12.
10.
9.
12.
--


9
7
7
1
5
4
2
8
2
3
0
0
0
-


9
5
3
6
5
7
23
17
4
3
3
4
-
-
SE

.0
.4
.7
.8
.4
.8
.5
.0
.4
.6
.2
.3
--
--
ME
0
1
0
0
4
1
1
6
0
1
1
0
0
0
0
ME
0
0
0
1
0
5
0
0
0
1
2
2
0
0
0
leucogaster
ME
0
3
0
0
0
0
0
0
1
0
0
0
0
0
0
Reithrodontomys
megalotis
ME
0
0
0
0
1
2
1
0
0
0
0
0
0
0
0
                                   243

-------
TABLE 6.21.   NUMBER OF MICE CAPTURED ON KLUVER EAST GRID.   MINIMUM ESTIMATE
             (ME) EQUALS TOTAL CATCH ON A GIVEN TRAPPING DATE AND JOLLY
             ESTIMATE (JE) WITH STANDARD ERROR EQUALS POPULATION SIZE PRE-
             DICTED BY JOLLY'S (1965) MODEL.

Date
Peromyscus
maniculatus
ME
July 23 1974
August 6
August 16
August 30
September 20
October 12
November 15
December 13
March 25 1975
May 28
May 29
June 3
June 6
July 25
July 30
August 1
1
1
1
3
8
1
3
13




8
7
2
8
23
1
1
1
1
1
1
7
0
1
1
0
3
JE

9.
14.
15.
16.
8.
15.
6.
8.
23.
20.
12.
24.
43.
10.


7
2
3
5
0
2
0
0
0
5
9
4
3
0

Microtus
Perognathus Onychomys
ochrogaster fasciatus leucogaster
SE

3.
4.
4.
5.
2.
11.
5.
2.
4.
4.
3.
13.
40.
--


3
5
4
2
8
5
8
8
8
9
8
4
5
-

ME
2
1
0
4
2
1
4
0
0
0
0
0
0
0
0
0
ME
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
ME
0
3
2
3
2
1
0
0
0
2
2
3
3
0
1
0
Reithrodontomys
megalotis
ME
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
                                   244

-------
 TABLE 6.22.   NUMBER OF MICE CAPTURED  ON  KLUVER  WEST GRID.  MINIMUM  ESTIMATE
              (ME) EQUALS TOTAL CATCH  ON  A  GIVEN TRAPPING DATE AND JOLLY
              ESTIMATE (JE)  WITH STANDARD ERROR  EQUALS  POPULATION SIZE PRE-
              DICTED BY JOLLY'S (1965) MODEL.


     Date     Peromyscus   Microtus    Perognathus  Onychomys  Reithrodontomys
              maniculatus ochrogaster   fasciatus  leucogaster    megalotis

             ME   JE    SE    ME         ME          ME             ME

July 15 1974  4   —   ---0          0          0             0

August 23   —   —    0          0          0             0

August 13     4   12.0 11.5    000             0

August 23     6   10.5  5.9    0          0          0             0

September 17  8    8.0  2.8   0          0          0             1

October 11   12   12.0  3.5    0           1          0             0

November 12  12   16.8  6.3    1          0          0             0

December 10   9   17.1  ---    1          0          0             0

May 20   1975 5   —   -—0          0          0             0
                                     245

-------
TABLE 6.23.   NUMBER OF MICE CAPTURED ON MCRAE'S GRID.   MINIMUM ESTIMATE
             (ME) EQUALS TOTAL CATCH ON A GIVEN TRAPPING DATE AND JOLLY
             ESTIMATE (JE)  WITH STANDARD ERROR EQUALS POPULATION SIZE
             PREDICTED BY JOLLY'S (1965) MODEL.

Date
Peromyscus
Mi
crotus
maniculatus ochrogaster

July 10 1974
July 28
August 9
August 20
September 10
September 27
October 1
October 22
December 3
January 25 '75
April 1
April 4
April 15
April 18
June 10
June 17
June 24
July 2
ME
11
14
16
21
28
18
13
23
9
20
17
21
21
28
23
16
18
14
JE
-•
20
17
21
28
27
24
28
26.
36.
24.
28.
28.
30.
33.
22.
25.
17.
--
.5
.3
.0
.0
.3
.0
.7
,7
2
,5
9
3
4
2
6
7
9
SE
--
6.
4.
4.
5.
6.
6.
7.
10.
10.
5.
6.
6.
6.
8.
5.
6.
--
• -
4
2
6
3
7
8
0
8
6
8
2
0
3
6
3
5
-
ME
2
6
7
7
3
7
5
8
7
10
2
4
2
1
1
0
0
0
Perognathus
fasciatus
ME
0
0
0
0
3
14
1
4
0
0
0
0
0
0
0
1
0
2
Onychomys
leucogaster
ME
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Reithrodontomys
mega! otis
ME
0
0
0
0
2
2
1
1
0
0
0
0
0
0
0
0
0
0
                                     246

-------
tion can  be categorized  as  indicators of  stress (e.g.,  adrenal  and thyroid
glands),  indicators  of  immunological  change  (e.g.,  spleen),   indicators  of
toxicological stress (e.g., lung, liver, and kidney), and indicators of repro-
ductive function  (e.g.,  gonads and accessory glands).  Only  the major repro-
ductive structures of immature and adult £.  maniculatus and M. ochrogaster are
treated in  this  report.   Portions  of the testes and associated structures are
described synoptically in Tables 6.24 to 6.31;  these are further described and
discussed in the associated text.

Seminiferous tubules

     The body of the testis contains many pyramidal compartments which contain
one to several  seminiferous tubules,  each bounded by a thick tunica albuginea of
regularly arranged  collagen  fibers.   Within these tubules the  spermatozoa or
male germ  cells  are  formed  and are  visible in  the tubule   walls  along with
developing  gametes  and Sertoli  cells,  the major  supporting  component of the
seminiferous epithelium.   A regular feature of the tubules,  especially when they
are spermatogenetically inactive, is one-to-few round eosinophilic droplets the
size of mature primary spermatocytes.  These occur in the tubular lumen or among
the layers of gametes lining its wall.  They have been previously described and
are interpreted as degenerate gametocytes (e.g., Clarke and  Kennedy, 1967).

     Spermatogenic cells at various stages of development also line the tubule.
These  are described in order of development:  spermatogonia  or gonia, primary
spermatocytes,  secondary  spermatocytes,  spermatids,  and spermatozoa.   Sperma-
togonia  multiply by  mitosis.   Frequent mitotic  spermatogonia  appear  in  the
seminiferous tubules.  They contain chromosomes, are small in  size, and line the
margins of the seminiferous tubule.   However, these dividing gonia are distinct
from primary or secondary  spermatocytes that undergo meiosis.

Interstitial tissue

     The interstitial tissue or Cells of Leydig represent the endocrine compo-
nent of the testis.   They are located in areas  around  the  convoluted semin-
iferous   tubules.    They  elaborate  several   steroid   hormones,   especially
testosterone,  without  which  the  seminiferous  epithelium  does not  sustain
proliferation and differentiation of the germ cells.   Testosterone also supports
the normal  development  of accessory glands of the reproductive tract, such as
the seminal  vesicles.  When  active,  Leydig cells are easily  recognized.  When
inactive, they cannot be distinguished from other connective  tissue cells that
lie between  the seminiferous tubules.  Leydig cells classified on the basis of
cytoplasmic characteristics are as follows:

     1.   inactive:    appear  as  other connective  tissue cells,  e.g.,  fibro-
          blasts;

     2.   moderately  active:   round  nucleus  and cells, but with  only a small
          amount of cytoplasm;  nucleus occupies most of the  cell  volume;

     3.   active:    round  nucleus and cells; cytoplasm  more  abundant  and fre-
          quently filled with secretory granules or small vesicles;
                                     247

-------
                                       TABLE  6.24.   GENERAL SUMMARY -- IMMATURE TESTIS Peromyscus maniculatus.
Seminiferous Tubules
Month
July
Aug
Sept
Oct
Nov
-£> Dec
Jan
Mar
May
June
July
Aug
Year
1974
1974
1974
1974
1974
1974
1975
1975
1975
1975
1975
1975
Average Diameter1
(u ± SE) (n)
121.5 (1)
75.8 (3)
73.2 ± 10.6 (10)
56.4 ± 4.6 (10)
56.3 ± 3.2 (8)
72.0 ± 8.6 (6)
105.6 ± 8.2 (5)
152.0 (2)
139.2 ± 13.9 (4)
145.0 (3)
78.4 ± 7.6 (14)
73.6 ± 18.6 (6)
Most
Advanced
Stage2
Spermatozoa
Primaries (2)
Spermatozoa (1 )
Gonia (1 )
Primaries (5)
Secondaries (2)
Spermatozoa (2)
Primaries (9)
Spermatozoa (1 )
Primaries
Primaries (1)
Secondaries (4)
Spermatozoa (1)
Spermatozoa
Spermatozoa
Spermatozoa
Spermatozoa
Primaries (2)
Secondaries (1)
Spermatids (6)
Spermatozoa (5)
Primaries (4)
Spermatozoa (2)
Mitotic
Gonia
Many
Some
None (1)
Some (2)
Many (7)
None (1)
Many (9)
None (1)
Many (7)
None (4)
Some (2)
None (2)
Many (3)
Many
Some (1)
Many (3)
Many
None (8)
Some (3)
Many (2)
None (2)
Some (2)
Many (2)
First
Me i otic
Figures3
None
None (1)
Many (2)
None (6)
Many (4)
None (3)
Some (1 )
Many (6)
None (1)
Some (3)
Many (4)
Some (2)
Many (4)
None (4)
Many (1)
None
None (3)
Many (1)
None (1)
Some (1 )
Many (1)
None (4)
Some (3)
Many (6)
None (4)
Many (2)
Degenerating
Gametes
None
None (2)
Some (1)
None (3)
Some (7)
None (2)
Some (7)
Many (1)
None (2)
Some (7)
Many (2)
None (2)
Some (4)
None (4)
Some (1)
None
None
Some ( 1 )
None
None (6)
Some (5)
Many (3)
None (1)
Some (3)
Many (2)
Activity of
Leydig Cells
Mod. Active
Inactive (2)
Mod. Active (1)
Inactive (9)
Mod. Active (1)
Inactive
Inactive (6)
Mod. Active (2)
Inactive (4)
Mod. Active (1)
Active (1)
Active
Very Active
Active (3)
Very Active (1 }
Active (2)
Very Active (1)
Inactive (10)
Mod. Active (2)
Active (2)
Inactive (4)
Mod. Active (1)
Very Active (1)
'Microns ±  standard error; number of samples in parenthesis.
2Primaries = primary spermatocytes; secondaries = secondary spermatocytes.
3Second Meiotic figures in immature testis are not included.

-------
                                              TABLE  6.25.   GENERAL  SUMMARY  — ADULT  TESTIS  Peromyscus maniculatus.1
ro
Seminiferous Tubules
Month
July

Aug



Sept



Oct



Nov
Dec

Jan
Mar


Apr


May

June


July



Aug



Year
1974

1974



1974



1974



1974
1974

1975
1975


1975


1975

1975


1975



1975



Average Diameter
(y ± SE) (n)
159.8 ± 6.1 (4)

148.5 ± 9.6 (5)



133.7 ± 9.4 (8)



90.8 ± 12.8 (8)



64.0 (1)
99.2 ± 8.9 (4)

102.0 (1)
157.0 (3)


163.2 ± 14.0 (7)


196.0 ± 18.0 (10)

162.7 ± 4.6 (18)


162.8 ± 5.8 (14)



162.9 ± 7.3 (10)



Most
Advanced
Stage
Spermatozoa

Spermatozoa



Primaries (1)
Spermatozoa (7)


Gonia (1)
Primaries (3)
Spermatozoa (4)

Primaries
Primaries (1 )
Spermatozoa (3)
Spermatozoa
Spermatozoa


Spermatozoa


Spermatozoa

Spermatozoa


Primaries (1 )
Spermatozoa(lS)


Spermatozoa



Mitotic
Gonia
Many

Many



None (1)
Some (1)
Many (6)

None (2)
Some (1 )
Many (5)

Many
None (1)
Some (3)
Many
None (1)
Some (1 )
Many (1)
None (2)
Some (2)
Many (3)
None (6)
Many (4)
Many


None (3)
Many (11)


None (2)
Many (8)


First
Meiotic
Figures
None

None (4)
Many (1)


None (1 )
Some ( 1 )
Many (6)

None (3)
Some (3)
Many (2)

Many
Many

Many
Many


None (6)
Some (1)

None (6)
Some (4)
None (10)
Some (1 )
Many (7)
None (8)
Some (2)
Many (4)

None (7)
Some (2)
Many (1)

Second
Meiotic
Figures
None

None



None



None



None
None

None
None


None


None (9)
Some (1)
None


None



None (9)
Many (1)


Degenerating
Gametes
None (3)
Some (1)
None (4)
Some (1)


None (4)
Some (4)


None (6)
Some (2)


	
None (3)
Some (1)
None
None


None


None (9)
Some (1 )
None (17)
Some (1)

None



None (9)
Some (1)


Activity of
Leydig Cells
Active (2)
Very Active (2)
Inactive (1)
Mod. Active (1)
Active (1)
Very Active (2)
Inactive (1)
Mod. Active (2)
Active (4)
Very Active (!)
Inactive (5)
Mod. Active (1)
Active (1)
Very Active (1)
Inactive
Active

Active
Active (2)
Very Active (1 )

Active (1)
Very Active (6)

Active

Mod. Active (1)
Active (7)
Very Active(lO)
Inactive (1)
Mod. Active (4)
Active (5)
Very Active (4)
Inactive (1)
Mod. Active (1)
Active (2)
Very Active (6)
       'Explanations  as  in  Table 6.24.

-------
           TABLE  6.26.   GENERAL SUMMARY -- TAIL OF EPIDIDYMIS OF IMMATURE Peromyscus  maniculatus
Month
Aug
Mar
June
July
ro .
en Aug
Epithel ium
Year
1974
1975
1975
1975
1975
Type
Col1
Cub2
Col
Ps3
Cub
Col
Ps
Cub
Col
Cub

(1)
(2)
(2)
0)
(3)
(3)
(1)
(1)
0)
Height
(M)
10.5
18.0
19.5
10.5
17.8
(n)
(1)
(2)
(3)
(1)
(2)
Cilia
Many
Many
Many
Many
Some (1)
Diameter
(M) (n)
222.5
113.5
147.0
226.5
260.5
(D
(2)
(3)
(1)
(2)
Tubule
Contents
Sperm4
HD
2(2)
3(2)
HD
2(1)
Empty5 Other6
None
None
1(0)
None
HO)
None
None
Fluid (1) (0)
None
None

     = simple columnar epithelium.
2Cub = simple cuboidal epithelium.
3Ps = pseudostratified columnar epithelium.
4lst number indicates number of animals with some sperm.   Number in parentheses are those with sperm in
 all tubules.
5lst number indicates number of animals with some empty tubules.  Number in parentheses are the number
 of mice with fully empty tubules.
6This column describes tubule contents other than sperm.   The first number indicates number of animals
 having some of the contents described, while that in parentheses indicates those having all of the
 contents.
7Degenerating.
8Lumen contains granular material.

-------
                             TABLE 6.27   GENERAL SUMMARY -- HEAD OF EPIDIDYMIS OF IMMATURE Peromyscus  maniculatus!


Month Year Type
Aug 1974 Col1
Sept 1974 Cub2
Col
Ps3


Oct 1974 Cub
Col
Ps
Nov 1974 Col
Ps
ro
— • Dec 1974 Col
Ps
Jan 1975 Ps

Mar 1975 Col
Ps
May 1975 Cub
Col
Ps
July 1975 Cub
Col
Ps
(\ug 1975 Col
Ps

(1)
(2)
(7)


(3)
(4)
(9)
(1)
(3)
(2)
(2)


(2)
(2)
(2)
(4)
(2)
(1)
(14)
(11)
(3)
(3)
Epithelium
Height
(y ± SE) (n)
15.5 (1)
15.5 ± 5.6 (9)




14.1 ± 0.8 (9)


14.6 (3)

19.0 (3)

18.0 (1)

22.8 (2)

20.8 ± 3.0 (4)


17.8 ± 1.0 (14)


22.0 (3)

Tubule
Cil
None
?
None
Many


1
Some
Many
None
Some
None
Many
Ma ny

Many

Many


None
Some
Many
Some
Many
la

(1)
(3)
(5)


(1)
(2)
(6)
(2)
(1)
(1)
(2)







(2)
(1)
(11)
(1)
(2)
Diameter
(y ± SE) (n)
50.5 (1)
45.0 ± 7.4 (9)




48.6 ± 6.8 (9)


42.5 (3)

59.0 (3)

71.0 (1)

108.0 (2)

108.8 ± 12.6 (4)


78.4 ± 11.6 (14)


91.6 (3)


Sperm''
0(0)
2(1)




0(0)


0(0)

0(0)

1(0)

2(1)

4(0)


3(0)


3(0)


Empty5
1(0)
8(4)




9(9)


3(3)

3(2)

1(0)

None

4(0)


14(8)


3(0)

Contents
Other6
Fibroblasts
Deg.7 lymphocytes

Epithelial cells
or lymphocytes
Granules8
Fluid
None


None

Granules

Granules
Deg. Sperm
Sloughed epith.
cells
None


Granules
Heterophils

None



(D
(1)

0)
(1)
(2)





(1)

(1)
(1)

(1)



(2)
(1)





(0)


(0)







(0)

I n \
\
-------
                       TABLE 6.28.  GENERAL SUMMARY -- TAIL OF EPIDIDMYIS OF ADULT Peromyscus maniculatus.1

Month
Sept


Oct
Mar

Apr
ro
en
May

June


July


Aug
Year
1974


1974
1975

1975


1975

1975


1975


1975

Type
Cub (1)
Col (5)
Ps (3)
Col
Cub (1)
Col (1)
Cub (5)
Col (6)

Cub (3)
Col (5)
Cub (6)
Col (7)
Ps (1)
Cub (3)
Col (8)
Ps (4)
Cub (2)
Epithel ium
Height
(y ± SE) (n)
17.9 ± 0.7 (5)


15.8 (2)
11.0 (1)

12.8 ± 0.6 (6)


12.7 ± 0.7 (5)

13.4 ± 1.0 (7)


20.4 ± 2.1 (8)


13.3+1 .5 (5)
Tubule
Cilia
Many


Many
Many

Many


Many

Many


Many


Many
Diameter
(v ± SE) (n)
165.1 ± 12.1 (5)


200.5 (2)
113.5 (1)

219.5 ± 12.7 (6)


226.1 ± 7.0 (5)

261.5 ± 11.8 (7)


170.8 ± 9.5 (8)


192.6 ± 18.6 (5)

Sperm
5(5)


2(2)
KD

6(6)


5(2)

7(6)


8(1)


5(4)
Contents
Empty Other
None None


None None
None None

None None


3(0) Fluid (1) (0)

1(0) None


7(0) None


1(0) None
'Abbreviations and explanations as in Table 6.26.

-------
                        TABLE 6.29.  GENERAL SUMMARY -- HEAD OF EPIDIDYMIS OF ADULT Peromyscus maniculatus,1
Month
July

Sept


Oct




Dec
ro
en
oo
Mar
Apr

May


June



July

Aug

Year
1974

1974


1974




1974


1975
1975

1975


1975



1975

1975



Type
Col
Ps
Col
Ps

Col
Ps



Col
Ps

Col
Col
Ps
Cub
Col
Ps
Cub
Col
Ps

Col
Ps
Col
Ps


(5)
(6)

(2)
(8)







(6)
(4)
(1)
(6)
(4)
(1)
(18)
(17)

(7)
(7)
(6)
(5)
Epithel ium
Height
(y ± SE) (n)
20.5 ± 1.8 (1)

22.8 ± 2.1 (7)


19.1 i 1.8 (8)




24.5 (1)


23.5 (1)
18.8 ± 3.9 (6)

21.5 ± 0.3 (6)


22.6 ± 1.5 (18)



25.4 i 0.8 (7)

23.9 ± 1.3 (6)

Tubule
Cilia
Many

Many


? (1)
Many (7)



Many


Many
Many

Many


Many



Many

Many

Diameter
(y i SE) (n)
122.5 ± 3.4 (1)

80.6 i 7.6 (7)


66.3 i 8.0 (8)
Contents
Sperm
1(0)

5(2)


2(1)
Empty
1(0)

5(0)


7(4)
Other
None

Deg. sperm
Granules
Monocytes
Deg. sperm



(2)
(1) (0)
(1)
(1)
Prim. spermatocytes(l ) /n^



77.0 (1)


123.0 (1)
109.3 ± 10.6 (6)

116.2 t 7.5 (6)


124.4 i 6.0 (18)



108.9 ± 4.0 (7)

123.8 ± 27.5 (6)




1(0)


KD
6(2)

6(2)


17(7)



7(0)

6(1)




1(0)


None
4(0)

4(0)


19(0)



7(0)

5(0)

Pycnotic nuclei
debris
Fluid
None


None
None

None


Deg. sperm
Monocytes
Fluid
Cell debris
None

None
-
& *• '
(1)
(2)









(1)
(]) (0)
(2) (0)
(1)




Abbreviations  and explanations  as  in Table 6.26.

-------
                                           TABLE 6.30.   GENERAL  SUMMARY  -- HEAD OF THE EPIDIDYMIS OF ADULT Microtus ochrogaster!
ro
en

Month



Aug
Nov
Dec

Aug






Aug

Sept

Mar

Apr


May

Aug


Year



1974
1974
1974

1975






1974

1974

1975

1975


1975

1975



Type



Ps
Col
Ps

Col
Ps





Col
Ps
Col
Ps
Col
Ps
Cub
Col
Ps
Col
Ps
Ps








(1)
(1)





(2)
(3)
(1)
(1)
(2)
(2)
(1)
(3)
(3)
(2)
(2)


Epithelium
Height
(y ± SE) (n)


13-.0 (1)
17.0 (1)
20.5 (2)

20.5 (2)






25.3 (3)

27.0 (1)

26.0 (2)

18.7 (3)


20.2 (2)

30.0 (2)

Tubule
Cilia Diameter
(y ± SE)
Immature Microtus

Many 47.0
Many 67.0
Many 63.5

None (1) 74.8
Many (1)



Adult Microtus

Many 112.5

Many 151.0

Many 95.5

Many 85.2


Some (1) 107.0
Many (1)
Some (1) 120.2
Many (1)
(n)


(1)
(1)
(2)

(2)






(3)

(1)

(2)

(3)


(2)

(2)

Sperm


None
1(0)
None

None






3(1)

KD

2(1)

2(2)


2(0)

2(1)

Contents
Empty


KD
None
2(0)

2(0)






2(0)

None

1(0)

KD


2(0)

1(0)

Other


None
Fluid
Gametes
Deg sperm
Debris
Deg sperm
Fluid
Monocytes
Heterophil
Debris


Fluid

None

None

None


None

None





(1)
(1)
(1)
(1)
(1)
(1)
(1)
(1)



(1)
















(0)
CO)



(0)




(0)












                         'Abbreviations and explanations as in Table 6.26.

-------
                TABLE 6.31.   GENERAL SUMMARY — TAIL OF EPIDIDYMIS Microtus ochrogaster.1

Epi the! ium
Month Year Type Height



Dec 1974 Ps




cn
01 Mar 1975 Col
Ps
Apr 1975 Col
Ps
May 1975 Col
Ps
June 1975 Col
Ps
(M)


22.0





(1) 12.0
CD
(2) 24.8
(2)
(1) 18.5
(1)
(1) 18.0
CD
(n)


CD





(1)

(2)

(1)

(1)

Tubule
Cilia Diameter Contents
(u) (n) Sperm Empty Other
Immature Microtus

Many 206.5 (1) None None Granules (1)
Cell debris (1)
Heterophils (1)
Adult Microtus


Some 244.0 (1) 1(1) None None

Many 198.0 (2) 2(2) None None

Many 298.0 (1) 1(0) None Red Droplets (1)

None 263.5 (1) 1(1) None None







(1)








(0)




Abbreviations and explanations as in Table 6.26.

-------
     4.   very active:  cell  round,  but  larger than active or moderately active
          cells;  cytoplasm  very  abundant and generally  filled  with  secretory
          granules or small  vesicles.

Epididyim's

     The epididymis is an organ comprised of a single duct formed by the fusion
of many smaller excretory ducts from the testis.  It stores spermatazoa, and is
believed to  enable  the sperm to develop the ability to  fertilize ova and to
move normally.  The epididymis is thus seen as an organ that is critical to the
final  stage  of  sperm development.   The  epididymis is comprised of head,  body
and  tail.  The  head and tail  differ markedly in histological  structure.   They
conform to descriptions of two regions of the epididymis of field voles (Lecyk,
1962) and are thought  to be equivalent to these regions.

     The tubules  in the head  region are relatively small  and  round.   They are
lined  by a simple high columnar to  pseudostratified columnar  epithelium  rein-
forced by a  thin  capsule of connective  tissue.   Epithelial cells in reproduct-
ively  active males  have  basal  oval  nuclei and much  apical  cytoplasm.   The
latter is filled  with  secretory granules and sometimes contains a single  large
unstained apical  vacuole as well.  The cells generally feature many long stero-
cilia.  The  lumen and tubule are round  and  regular in outline, and the tubules
are  separated  from  each  other  by variable amounts of vascular connective
tissue.  In  a nonreproductive state, the tubules shrink in size, the  simple
epithelium diminishes  in height,  and the cells are densely crowded. The tubule
lumen  is  reduced  and often disappears  entirely.  The  cilia are  difficult  to
locate and are completely absent in  many cases.

     The tubules  in the tail region  are  very large, irregular in shape, and are
generally  filled  with  sperm.   The  epithelium  is  simple  low  columnar  to
cuboidal.   Cell   nuclei are  centrally  located  and  the  cytoplasm is  scanty.
Cilia are difficult to discern.

     Degenerating sperm or degenerate cells resembling primary gametocytes are
sometimes  observed  in  the   epididymal  tubules.    Occasionally,  the tissue
contains focal  infiltrations  of  lymphocytes and sometimes  heterophilic  poly-
morphonuclear leucocytes in the connective tissue between the tubules.

     Epithelial height  and  tubule   diameter  of adult mice in any gjven  month
exceed those  of immatures in the head but not the tail of the epididymis.   Fur-
thermore, seasonal  changes  in epithelial height and  tubule diameter  are  not
always in the same direction.  Because  diameter changes  in tubule lumen from
month to month are larger than changes  in epithelial height, diameter may be a
more sensitive measure of functional changes throughout the year.

                                    BIRDS

                                INTRODUCTION

     Birds'are  the  most  visible  vertebrates of the grassland ecosystem  under
study during  the  period of spring to  late fall.  The dominant species are the
Western Meadowlark  (Sturnella  neglecta), Mourning Dove  (Zenaidura macroura),

                                      256

-------
Lark Bunting  (Calamospiza  melanocorys),  Vesper Sparrow (Pooecetes gramineus),
and Lark Sparrow (Chondestes  grammacus).  The Mourning Dove occurs near riparian
habitats, where drinking water and the presence of trees for nesting are found,
but the other species can be found in a variety of grassland or sage-grassland
areas.   Within these habitats,  only the meadowlark seems to be well distributed.
The others often show discontinuity in their distribution.

     Birds are  potentially  important indicator species because they are rela-
tively  long-lived  secondary  consumers.   Such forms are highly vulnerable when
ecosystem damage occurs (e.g., Stickel, 1975).  Furthermore, they are unlikely
to show the acute effects  of  pollution and most likely to suddenly and belatedly
exhibit severe  damage  when  pollutants have accumulated to toxic levels in the
organ-systems (Stickel, 1975).  Their  long lives enhance potential for accum-
ulation of pollutants.  Even low-level, non-lethal accumulations of pollutants
which do not directly debilitate may have important effects on later generations
of birds and mammals  (summarized by Stickel, 1975  and Truhaut, 1975).

     Birds are day-active and typically roost above ground,  in contrast to the
nocturnal fossorial mammals that are under study.  The two groups will thus have
very different  pollutant  exposure histories.  Investigation of the effects of
diurnal  variation   in  power plant  emissions is  one  of special  interest  and
potential importance.   Mourning Doves, for example, feed most intensively during
the  first few  daylight  hours  and  again  late in  the  day.   Clearly,  this  may
influence both their rate of  exposure and susceptibility to air pollutants.  The
species  of  birds  under  study employ  different  feeding strategies,  namely,
granivory (dove),  insectivory and carnivory (meadowlark), or both (bunting and
sparrows).   If different food categories are differentially  affected by pollu-
tion damage, then  it should be possible to determine the effects from assay of
physiological conditions.  For example, sulfur dioxide reduces pollination and
pollen  germination and therefore  the quality of  seeds  (Garber, 1967).  Of the
birds in question  such an  effect would be expected to  affect seed-eaters select-
ively.   Examination of these species at histological  levels  and integration of
tissue  and  gross data may reveal  diet-related effects  of stack emissions and
hence  the dietary  group  which is most useful as  a  bioindicator  of pollution
stress.

     Members of  higher trophic levels, such as birds,  may accumulate pollutants
for long periods before overt  responses are seen.  As indicated in  a recent Oak
Ridge  Workshop  (Johnson,   1976),   there   is  a  general  need for  information
concerning  the  sublethal  and  chronic,  cumulative effects of pollutants.  The
"threshold"  exposure rates that might be expected to  produce various biological
effects (Lewis, Glass, and Lefohn, 1975;  p.  142) on organisms are not yet known,
especially  in  areas  such  as  Col strip  where the  impacts  of  pollutants  are
continuous,  but  small.    It   is  thus  desirable  to  determine which  tissues
selectively  accumulate  pollutants  (Truhaut,  1975).    By  identifying  avian
tissues that selectively accumulate or react strongly to pollutants, it may be
possible  to  anticipate when or with how much exposure  a dramatic and sudden
effect will be overtly  manifested in a species.

     Some birds  are sensitive indicators of pollution (Lewis,  Glass, and Lefohn,
1975, Ch.  8).   For example,  pigeons living  in highly  industrialized areas of
Japan develop lung pathologies when aerial pollution levels  are only one-tenth


                                      257

-------
the value of  those  necessary to produce analogous pathologies in human lungs.
Birds appear  to be  particularly  useful for  measuring  levels  of air quality
because of the  design  of their respiratory system and their broad and regular
dispersion throughout the ecosystem.   In contrast to  bioindicating organisms of
lower trophic  levels,  birds  provide  a mechanism of separately assessing the
effects of aerial and dietary pollutants.  The lungs and respiratory tract are
particularly  valuable  for  monitoring  air  pollution.   In  contrast,  ingested
pollutants probably  affect the liver and perhaps the kidney selectively.

     Intensive work  is proceeding on the avian  lung.  This organism is struct-
urally analogous to a high volume sampler involved in aerial pollutant scaveng-
ing.  It is a primary target for air-borne  particulates because the nasal pass-
ages have little ability to  remove them  from inhaled air and  because flow across
the  respiratory  surface  is  in only one  direction  (Bretz and Schmidt-Nielsen,
1971).  As the studies of Japanese pigeons suggest, the avian lung is damaged by
rather  small  amounts of  toxicants  from coal-fired power plants.  It should be
possible to determine even low emission  rates and types from stacks of this kind
by  periodically examining the  lungs  of birds  in  impacted  areas.  Coal  dust,
silica,  and  the  fugitive  dust  of agricultural  activities are  all  somewhat
different  in  appearance  and likely different with respect to the inflammatory
responses which they produce  in avian respiratory tissues.

     The avian  Tung  also seems to offer unusual potential for use in pollution
gradient analysis.   Stack emissions settle out at distances from power sources
which vary with the  type of pollutant.  Heavy particles settle  in  the immediate
vicinity of the plant.  However, particles dispersed  in the form of aerosols are
transported  great  distances  from  the source and carry  toxic  gases  and trace
metals.  This, coupled with  the broad and fairly regular dispersion of birds in
the  grassland  ecosystem,   should  permit  (1)   distinguishing  the effects  of
pollution  from the  effects  of other environmental  stresses,  and (2) through
pollution gradient analysis, evaluating the effects of different concentrations
of stack emissions acting on birds over an extended period.

POPULATION BIOLOGY

     Census data can yield information on the relative spatiotemporal abundance
of  bird species.  When  a census is repeated  at regular intervals,  changes in
species  frequencies  and dispersion  patterns can be  deduced and these can be
related to changes  in concurrently measured environmental parameters.„

     By  application  of  the  method  described below,  the investigators hope to
relate  changes  in easily measurable descriptive parameters  such as species di-
versity, equitability, and species richness to  changes  in functional relation-
ships between  species  and their habitats.   Of particular interest are changes
attributable to the  operation of the coal-fired power plants at Colstrip.  This
report  summarizes the baseline data and suggests a format for the analysis and
interpretation of future  census data.
                                      258

-------
Methods

     The method  employed  is patterned after the North American Breeding Bird
Survey  (Robbins  and Van Velzen, 1970).  In  application,  the observer starts
one-half hour  before  local  sunrise and makes 60 three-minute stops at 0.5 mi
(0.8 km) intervals along a predetermined  route (Figure 6.10).  At each  stop the
number of birds of each species seen in a 400 m radius and heard, regardless of
distance,  is  recorded.   Since  bird  species  have  differential  individual
detectability, the  census data provide  no  information  on absolute abundance
(Emlen,  1971).  However,  it should be possible to  detect changes  in species
relative abundance patterns  by  comparing data obtained during a baseline period
with those  obtained using  the same census  technique  at other times.

     The census  route was  selected to  provide sampling stations  at various
distances from the Colstrip power plants along an anticipated gradient of pol-
lution  impact.   Individual  stations span a wide range  of habitats including
open  grassland,   streams,  rolling hills  with  ponderosa  pine and  juniper
coverage,  cliffs and  habitats  affected by  a  wide  range  of  human  impacts
including  urban   activities,  railroad right-of-way,  farming,  ranching,  and
mining.

     The 1975  breeding season will be considered the baseline period for pres-
ent purposes  since  the power plants were  not  operational at that time.  The
route  in Figure 6.10 was censused on nine dates during the baseline period (May
6,  May 23,  June  11,  June  30,  July  14,  July 28,   August 11,  August  26,  and
September 8).
                                                                        s
     Species diversity was measured by the Shannon-Weaver function (H'=-l p.
                                                                        i=l
Iog2p.)  (Shannon  and  Weaver,  1949), species richness (S) was measured by the
number  of   species  present,  and equitability  (E)  was  measured by  H'/log2S
(Pielou, 1977).  To inspect overall temporal trends in these  parameters, total
H1, total S, and  total E  were calculated from lumped census data for all sites
on  each sampling  date.   Spatial  and site  specific trends  were inspected by
calculating  total across  season H1 , S, and  E from lumped data from all sampling
dates for each site.

Results and Discussion

     Evaluation of pollution impact depends heavily  upon a comparison of future
species  frequencies and dispersion patterns with baseline patterns established
prior  to power plant operation.  It is assumed that  the parameters measured are
not in  the process of reaction  to the environmental  influences extrinsic to the
ecosystem  under   scrutiny.   Limited  data from  the  U.S.   IBP Grassland  Biome
program  indicate  that  grassland  bird species diversity,  equitability,  and
species  richness  as measured in the  present  study  do not vary substantially
between  years, but  generally decrease along a gradient  of decreasing primary
production (Wiens,  1974,  Table 6.3).  H', E, and S are apparently buffered from
response to  normally  encountered environmental  variations but may respond to
changes  in  productivity  of   the  ecosystem  brought  about by chronic  air
pollution.    If so, the roadside bird census may provide  an inexpensive method
of indexing long-term ecosystem response to chronic  pollution.

                                      259

-------
  I06°45
T. 2 N.
                                                            R. 43 E.
                                                                     T. 2 N.
   I06°45'
R 42 E
 6       !06°30
                                                      R. 43 E.
   Figure 6.10.   Map of the  Rosebud-Col strip roadside census  route.

                                   260

-------
     Temporal  trends  in  species  frequencies  are  presented  in  Table 6.32.
Though  63  species were  recorded on the census  route  during 1975,  most were
relatively rare.  Meadowlarks were clearly predominant and the six most common
species  contributed  nearly  62%  of  total  bird abundance.  Diversity (H1) peaks
in  late  May,  decreases until late  July,  and  then remains fairly stable into
September.  Seasonal  trends in E and S are not  as clear£ut, but the trend in H1
reflects the interaction of  these two components of diversity.

     There is  no  statistically  significant correlation between distance from
the power  plant and  H1 , E,  or  S in the baseline  data  (Table  6.33).  For all
three  least  squares  regressions, the coefficient  of correlation  is less than
0.7 (p  > 0.05).  Total across-season diversity ranges from  1.28  at site 5 to
3.92 at  site  39.   In temperate  regions  bird  species  diversity is correlated
with  foliage  height diversity  (MacArthur and  MacArthur,  1961).  Diversity
increases  with the  number  of layers in vegetation and with the  evenness of
foliage  apportionment  among layers.  Site  5 is an  open grassland,  one-layered
habitat.  Site 39 is riparian.  Grassland,  shrubs and  trees  occur  within 400 m
providing a three-layered terrestrial habitat as well as an aquatic habitat for
shore  and  water birds.  It may be possible to account for  much  of  the  site
variation  in  bird species  diversity by  variations in habitat diversity.   An
attempt  is being made  to quantity  habitat  diversity at the study  sites during
the 1976 field season.

     Average cross season H1  and S  for birds in the pine savannah  southeast of
Colstrip are intermediate between  typical  values for grasslands and those for
shrublands  (bottom  of Table  6.33).  This  is expected since the sampling sta-
tions range from open grassland  to forest.

     Average  E is  lower than would  be  predicted  by the trends for H' and S.
This  may be  an artifact of the  census  method  which tends  to  overestimate
meadowlark  abundance  relative  to  other  species  yielding generally  lower  E
values  than  would otherwise be found.   Variability in H',   S, and  E  is  very
similar  to that found in other grassland and shrubland censuses  (Tramer, 1969).

     H'  is dependent upon both  species richness and equitability.  In general,
bird species diversity correlates  strongly with species richness while equit-
ability  tends  to  be  high with  little variability  regardless of species rich-
ness.   For 267 breeding bird censuses from many habitat types, Tramer (1969)
demonstrated that:   H1  = 0.941   (log2S)  -  0.251  (r =  0.972).  The same trend
occurs in the  baseline across season diversity data in the present study.  The
least squares  regression line (H1 = 0.999  ± 0.075  S.E. x log2S - 0.889 ± 0.255
S.E.) is similar to  the general  trend demonstrated by Tramer, but  H1 and log2S
are not as  strongly correlated (r = 0.754).

     The components  of H1 (S  and E) may  respond differently  on exposure to en-
vironmental stress.   In  general, high  species richness and high equitability
are characteristics  of late successional  stages (Margalef,  1963,  1968; Odum,
1969).   Stressful  environments generally permit less diversity to  develop than
favorable environments do (Sanders,  1969; Slobodkin and Sanders, 1969).

     Environmental stress is primarily determined by (1) the degree of temporal
predictability  in environmental  conditions and (2) the degree  of physiological

                                     26]

-------
TABLE 6.32.   SPECIES FREQUENCY  ALONG THE CENSUS ROUTE DURING 1975.
Proportional Abundance (x
Common Name 5/6 5/23 6/11 6/30
Mallard 50
Red-Tailed Hawk 17
Golden Eagle 17
Marsh Hawk 66
Prairie Falcon
Peregrine Falcon
Sparrow Hawk 446
Sharp-Tailed Grouse 66
Ring-Necked Pheasant 826
Killdeer 17
Common Snipe 17
Mourning Dove 446
Long-Eared Owl 17
Poor-Will
Common Nighthawk
Chimney Swift
Belted Kingfisher
Red-Shafted Flicker 66
Red-Headed Woodpecker
Eastern Kingbird
Western Kingbird
Cassin's Kingbird
Say's Phoebe 50
Horned Lark
Barn Swallow
Cliff Swallow
Black-Billed Magpie 33
Common Raven
Common Crow 132
Pinon Jay
Black-Capped Chickadee 413
White-Breasted Nuthatch
House Wren
Winter Wren
Brown Thrasher
Robin 33
Wood Thrush
Mountain Bluebird
Cedar Waxwing
Loggerhead Shrike 50
Starling 132
Yel low Warbler

108

27


108
108
755


296





81

243
189

27

243
81
27

135

108




243

54



162

27
27


20
160
20
747 343
27

427 625



242

53 121

107 81
347 81
53

27
187 161
293 444
80

133 60
323




27 40
107 60

240 20

27

53 60
101*) for the given date Seasonal
7/14 7/28 8/11 8/26 9/8 Total

77 27


40

40 51 23 93 301
27
80 26 68 56 164
26 37 27

783 918 1847 935 902

26
102

20
100 26 68 56 301
82
120 204 225 150 55
141 77 90 224 109
20 19
26 109

80 153 113 93
482 332 473
23
40
40 102 93
180 1458 27
26 90 ' 56 273
23
20 23
23 55
20
221 51 37 27
37
20 77
45
20 45

80 51 90 19 27
7
24
5
12
5
2
140
24
339
15
2
787
2
2
10
29
2
92
7
123
131
10
22
2
107
228
17
5
77
249
114
2
5
7
10
196
5
39
5
17
19
56
                                                                          (continued)
                              262

-------
TABLE 6.32.   (continued )
Common Name
Yellow Throated Warbler
Ovenbird
Yellow-Breasted Chat
American Redstart
Western Meadow! ark
Red-Winged Blackbird
Bullock's Oriole
Brewer's Blackbird
Common Grackle
Brown-Headed Cowbird
Lazuli Bunting
American Goldfinch
Red Crossbill
Rufous-Sided Towhee
Lark Bunting
Savannah Sparrow
Vesper Sparrow
Lark Sparrow
Chipping Sparrow
Clay-Colored Sparrow
White-Crowned Sparrow
N(xlO°)
H'
E
S
Proportional Abundance (x
5/6 5/23 6/11 6/30
17

4645
413

1058





33

331
430
132


33
605
3.14
0.65
28
27

3396
593

350





108
755
189
539
943
108


371
3.66
0.76
28
27
27
3627
640
27
243
53
107




987
160
187
747
27


375
3.59
0.71
33
60
20
3347
444

383
20
262

81


1815
60
343
444
20


496
3.47
0.71
29
101*) for the given date
7/14 7/28 8/11
20

3775
482

1104
20
261
80
20

60
863
40
321
522
60


498
3.40
0.67
33
26

3163
77

2168

102
26
179


510
77
306
995
26


392
3.29
0.67
30
23

3176
90



203
45
90

68
1779
135
405
541



444
3.22
0.67
28
8/26
19
37

2991
467



280
37
75

56
2056
75
393
93
56


535
3.23
0.67
28
Seasonal
9/8 Total


2787


137
27


82
410

2022
219
1667
137

55

368
3.29
0.70
26
10
2
17
5
3448
361
2
605
12
140
22
56
36
36
1165
157
479
465
31
5
5

3.77
0.63
63
                                               263

-------
TABLE 6.33.   SPECIES DIVERSITY,  EQUITABILITY, AND SPECIES RICHNESS AT THE
             SAMPLING STATIONS DURING 1975

Site number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Distance from
Power Plant
(km)
15.8
15.0
14.2
13.6
12.8
12.0
11.4
10.6
9.9
9.3
8.6
9.0
9. 1
8.5
8.2
7.5
6.7
5.9
1.3
1.9
2.7
3.5
4.5
5. 1
5.8
16.2
15.7
15.2
14.7
14.4
14. 1
14. 1
14.1
14.1
13.9
13.8
13.3
13.0
13.0
13.0
13.0
13.1
13.4
Species
Diversity
(H1)
2.70
1.35
2.47
2.36
1.28
1.56
2.74
2.97
2.89
3.47
3.29
2.83
1.84
1.82
1.61
2.44
2.24
1.69
1.89
1.31
2.20
2.40
2.69
2.03
1.69
1.67
3.38
2.97
2.88
1.82
2.03
3.05
3.29
2.95
2.62
1.64
2.50
3.33
3.92
2.11
1.95
1.63
2.70
Equitabil ity
(E)
.73
.58
.65
.66
.55
.60
.70
.78
.78
.85
.77
.79
.58
.71
.57
.81
.80
.66
.60
.46
.64
.72
.85
.72
.84
.60
.87
.78
.76
.61
.61
.78
.82
.70
.87
.71
.75
.80
.89
.64
.76
.52
.75
Species
Richness
(S)
13
5
14
12
5
6
15
14
14
17
19
12
9
6
7
8
7
6
9
7
11
10
9
7
4
7
15
14
14
8
10
15
16
19
8
5
10
18
21
10
6
9
12
                                                                 (continued)
                                   264

-------
TABLE 6.33.  (continued)

Site number
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Distance from
Power Plant
(km)
13.9
14.2
14.6
14.9
15.2
15.7
15.8
16.2
16.8
17.3
17.4
17.6
17.8
17.9
18.2
18.2
18.2
Species
Diversity
(H1)
3.09
2.70
2.78
2.26
2.32
2.45
2.40
2.47
3.56
2.84
2.91
3.03
2.80
2.48
2.62
2.95
2.22
Equitability
(E)
.89
.78
.78
.65
.67
.77
.69
.69
.94
.86
.81
.84
.78
.72
.76
.80
.64
Species
Richness
(S)
11
11
12
11
11
9
11
12
14
10
12
12
12
11
11
13
11

Average ± 2 Standard Errors   2.47 ± 0.16    0.73 + 0.02    10.95 ± 0.98



Typical Grassland Values1     1.9310.24    0.84 ±0.034    5.74 ±1.00



Typical Shrubland Values1     3.14+0.16    0.85+0.024   14.08 ±2.31




1from Tramer  (1969)
                                      265

-------
stress  imposed   (Slobodkin   and   Sanders,   1969).    Either   a  temporally
unpredictable  or   a   temporally   constant   but   physiologically  stressful
environment  should  constrain  the development of  high  species  richness since
relatively few species would  be able to successfully  colonize the environment.
However, equitability  in the distribution of individuals should  be different in
the  two environments.  In  a constant  but severe environment,  equitability
should  be  relatively  high  in  late  successional stages.  As the species which
have  successfully  adapted  to  the  extreme environment begin to  realize full
exploitation of their  available resources, competition may become the dominant
selective pressure and this should lead to greater equitability (Emlen, 1973).
In an  unpredictable environment,  however, catastrophic disturbances may keep
the  ecosystem  at a low  state of  maturity and  should lead to  a log-normal
distribution  of  species  abundances (MacArthur,  1960).    As   a  result,  both
species richness  and equitability would be low.

     Figure  6.11  illustrates  the  combinations  of cumulative  S  and  E values
which  occurred  at  the  60  sampling  stations during 1975.   The  quadrants
represent the trends in species discrimination of their environments at climax
expected  from the  previous   discussion.   This  format may  prove useful  in
interpreting  shifts  in  positions  of  the plotted points  that occur  in  the
future.  Vectors  drawn from the baseline point positions  for a  particular site
to its  new position at some future time can be used to quantify  the reaction of
the  avian  community at that site to changes in environmental conditions.  The
vector  sum for all 60  stations can be used to quantify the new  shift away from
baseline of the avian  community along the  census route and may provide a useful
index of initial  response to operation of the Colstrip power plants and permit
tracking of subsequent ecological succession.

     Table 6.34  illustrates  the  interpretation that can  be given to various
temporal shifts in vector direction. Vector length can be used  to quantify the
magnitude  of the vector  shift.   This  technique  permits  evaluation  only  of
quantitative  shifts in  species numbers and relative abundances.  Qualitative
changes  in  species  composition  may  be  of  equal   or   greater  importance.
Comparisons  of  the  life  history  strategies  of those  species  favored  by
pollution  with  those   less  favored may lead to  better understanding of more
general effects manifested by  the ecosystem of interest.  Thus where generalist
species  tend to  be replacing or  increasing  at  the  expense  of specialists,
retrograde succession  may be indicated.

HISTOLOGY

Avian Tissues As  Indicators of Pollution Challenge

     Histological evaluation  of  selective tissues and organs  of birds is in
progress.   Baseline definition will  be complete  in  about  a year.  Only those
tissues that are  likely to  respond  to the  low levels of emissions  expected from
the Colstrip  plants  are examined histologically.  These include  the following:

Tissues that  are  indicators  of "stress."  The responses of mammals to stress
are well documented  (e.g.,  see Christian,  Lloyd, and Davis, 1965)  and generally
fit the "general adaptation syndrome" originally promulgated by  Hans Selye (von
Faber, 1964).  Early responses occur in the adrenal gland, particularly in the

                                     266

-------
     CD
     O
     UJ
     c
     O
     t»
     O
     0)
     (0

     0)
     tA
     O
     k_
     O
     O
     UJ
                  CONSTANT OR
                                  CONSTANT-FAVORABLE
S
E
uu
60
40(
— r-rrcvns IMOI-C. O//TCOO
. • 1
. :••!*!*
• *
• *
UNPREDICTABL E*S TRESS
* I
UNEXPECTED
I I I
•) 5 10 !5 20 25
                         TOTAL  NUMBER OF SPECIES
                      OBSERVED AT SITES DURING 1975
     o
Figure 6.11.
      Species  richness versus  cumulative  equitability  for  the 60
      sampling stations during the baseline period (1975).


                          267

-------
                   TABLE 6.34.  INTERPRETATION OF TEMPORAL SHIFTS IN VECTOR DIRECTIONS
                                ON PLOTS OF EQUITABILITY VS.  SPECIES RICHNESS
       A                 A              Directional Shift
Species Richness    Equitability            in Vector                      Interpretation
      (S)               (E)                 (Degrees)

       o                 +                      0           numerically dominant species becoming less
                                                            dominant

       +                 +                   >0, <90        new species are being added, but all  species
                                                            are becoming more even numerically

       +o                     90           new species are being added but relative
                                                            abundances have not changed
ro
S

new rare species are being added
>90, <180 numerical dominance
i ncreasing
by abundant species

is

                                               180          numerically dominant species are becoming
                                                            more dominant
                                            >180, <270      rare species decreasing and/or dropping out,
                                                            dominant species may be increasing
                                                            no differential effect on relative abundance
                                               270          all species decreasing numerically, rare
                                                            species dropping out
                                                            (1)  rare species drop out, others remain
                                           >270,<360             abundant
                                                            (2)  rare species drop out, dominant species
                                                                 become less abundant

-------
adrenal cortex; this tissue hypertrophies and produces  unusually large amounts
of  glucocorticoids.   Lymphatic  organs  involute secondarily  and  predictable
changes  also   take   place  in  the  blood  (including  lymphocytopenia  and
neutrophilia).  These functional changes are accompanied by changes in tissue
structure.

     1.  Adrenal Gland—There  are  surprisingly few studies of stress-related
changes in the adrenal glands of birds, but those available are in harmony with
the above picture of stress on the mammalian adrenal gland.  Stressors such as
oil, kepone, and high population density are known to immediately increase the
weight  of  the gland  (Flickinger,  1961; Gorman  and Milne,  1971;  Eroschenko,
1973).   However, the adrenal gland of the bird is a mixture of interrenal and
chromaffin tissue and changes in weight indicate little about which component
is  affected.   In  fact,  changes  in  the  weight  of  the gland  are  sometimes
misleading.  Seasonal changes in the adrenal gland of the Wood Pigeon provide
an  instructive example (Ljunggren, 1969).  To begin with, there are no marked
seasonal changes in  the  weight of the gland.   Yet, on a histological level,
there are rather pronounced seasonal variations in the activity of interrenal
and chromaffin tissue.  Furthermore, the male adrenal is generally heavier than
that of  the  female.   On  a weight basis, it might be concluded that males are
stressed more  than  females.   However,  just the opposite appears  to  be  true:
the fractional volume of cortical tissue in the female's gland always exceeds
that in the male's gland.

     Among the  few  histological  studies that exist concerning birds, stress-
related changes in the adrenal include (1) hypertrophy and hyperplasia of in-
terrenal and chromaffin cells, (2) changes in the fractional volumes of inter-
renal  and  chromaffin  tissue,  (3) increases  in  the  number  and size of nuclei
within  interrenal  cells,  and  (4)  increases in the vascularity of  the  gland
(Stoewsand  and  Scott,  1964;  Ljunggren,  1969;  Gorman  and  Milne,  1971;
Bhattacharyya   and   Ghosh,  1972;  Bhattacharyya,  Ray,   and  Manna,   1972;
Eroschenko, 1973).  The subject is well reviewed by von Faber (1964).

     Few seasonal studies  of the adrenal gland exist and the functional signi-
ficance  of  seasonal  changes in this gland  are  not  well established.  In all
species examined to date except the Brant, adrenal activity is reduced during
periods  of molt, but  elevated at breeding  (summarized  in  Gorman  and Milne,
1971).   Changes in the adrenal weight of Western Meadowlarks  and Mourning Doves
during  1975 at Colstrip (Lewis, unpublished) are consistent  with these earlier
studies—  adrenal weight  increased during the breeding season and diminished
during the postnuptial  molt.

     2.  Thyroid Gland—Manifestations of stress at the level  of thyroid gland
include  losses  of  colloid  and increases  in  the  height of  the  follicular
epithelium.   Both  changes are  indicative  of  elevated glandular  activity.
Little,  if any, concurrent change takes  place in the weight of  the  gland
(Ljunggren, 1968) which underscores the need for its  histological evaluation.

     The adrenal and thyroid glands at the tissue level  are potentially sensi-
tive and quick-responding  bioindicators  of (1) the normal  stressors  in the
Colstrip ecosystem  and  (2)  the additional  stress  of  power plant emissions.
Blood levels of corticosterone in key avian species will be  measured  in future

                                      269

-------
summers at the site.   It will  be interesting to compare changes in the histol-
ogy of the interrenal tissue and alterations in  corticosterone production.

     Tissues that belong to the immune system.   Tissues in this category react
directly  to  foreign  materials  in the body or are secondarily  affected by the
induced activity of the adrenal gland.

     1.  General—The immune system  of  the bird includes  lymphoid cells that
originate  in the thymus  (T  lymphocytes), bone  marrow (BM lymphocytes),  and
bursa of Fabricius (B lymphocytes).   These cells disseminate to lymphatic tis-
sue throughout the body and are responsible for immune responses.  Descendants
of  T  cells  produce  substances  which destroy  foreign bodies, such  as  skin
grafts,  outright.   Descendants  of   BM  and B  cells  produce  nonspecific  and
specific antibodies,  respectively.  There is accumulating evidence that T cells
are also involved in activities traditionally ascribed to BM and B cells.  The
entire subject of avian immunity was  reviewed recently by Moticka (1975).

     2.  Bursa of Fabricius—The bursa is a lymphoepithelial gland which deve-
lops as an outgrowth of the proctodaeum.  It is necessary for  the development
of  antibody-mediated  responses.   It  normally  involutes when  a bird reaches
sexual  maturity.  However,  it will  involute earlier  if  the  younger bird is
subjected  to  stress  (von  Faber,  1964).   It will, for example, regress in the
presence  of  glucocorticoids  (Click,  1964;  Siegel ,   1971).  Consequently,  it
should  be  very  useful  for identifying stressors  that affect  young  birds, as
opposed  to  those which  affect  the avian  population  as  a  whole.   Sudden
regressive  changes  in  the  structure  of  the  bursa  are  more likely  to be
translated quickly into  histological  changes than into reductions in the weight
of the  gland.  This underlines the importance of examining the tissue structure
of the gland.

     3.  Thymus—The thymus  is  a multilobed gland which is distributed along
the  jugular  veins in the  neck.  It is diffusely arranged and difficult to re-
move totally, though we are generally able to do this successfully.  The thymus
of many avian species, including the Mourning Doves and Lark Buntings collected
at Colstrip  (Lewis,  unpublished), enlarges during molt (Hb'hn,   1956;  Anderson,
1970;  Kendall,  1975;  Ward and Kendall,  1975).   Periodic  rejuvenation of the
gland at this time appears to provide erthrocytes to meet the  demands of molt
and perhaps  additional lymphocytes to deal with stresses additional to that of
the  molt  itself  (Ward  and D'Cruz, 1968; Ward  and Kendall,  1975).  T«he thymus
involutes if the adrenal gland becomes hyperactive.   Immediate  and predictable
thymic   responses  include   lymphopenia,  eosinopenia,   and  heterophil ia.
Consequently, it may be particularly sensitive to the  stress of chronic,  low-
level  pollution  challenge.   Stress-related effects  are  more likely  to be
registered first as  alterations in tissue structure and only later as changes
in organ weight.

     4.  Spleen—The  spleen has been  largely ignored  by avian biologists (Ewart
and McMillan, 1970).   Since it is not a storage organ for blood, changes  in its
weight  may be  true  reflections of different states  of its  activity.  In  this
respect, some literature concerning seasonal changes in its weight is available
for  reference  (Riddle,  1928;  Oakeson,  1953).   Its  major   functions  are
haemopoiesis, antibody production, and filtration of  the blood (as part  of the

                                     270

-------
 mononuclear phagocyte  system).   Splenic leucopoiesis  is  influenced  by  the
 activity of the adrenal gland and by antigen challenges.   Foreign  proteins  and
 carbohydrates,   for  example,  stimulate  its production  of plasma  cells  and
 antibodies  (Jankovic and Isakovic, 1966). The  phagocytes  of its red  pulp also
 selectively engulf blood-borne particulate matter.   The  number  and activity of
 the  germinal centers and  the activity of the  follicular tissue in its white pump
 are  directly related to its antibody production.   In addition,  the activity of
 parafollicular  areas of its white pulp  varies directly with the number of cell-
 mediated immune  responses taking place  within  the birds.   In   other  words,
 splenic  tissue  provides information  about the magnitude of immune  responses  and
 also permits one to  further  determine  if the responses  are  primarily humoral,
 cell-mediated,  or both.

      The value  of the  spleen  as  an  early indicator  of  pollution  challenge is
 suggested by the studies  of Zarkower (1972) with mice.  He demonstrated that  air
 containing  carbon, sulfur dioxide, or combinations of the two, suppresses  immune
 responses.   These pollutants  destroy cells of  the  immune  system selectively
 (Piliero, 1970) and their  effects should be expressed relatively  early  in  the
 structure of lymphoid tissues.

      Detoxifying tissues.   The major site of detoxification  in the vertebrate
 body are the liver, kidney, and lungs.

      1.   Liver—The  liver  has  a large number of functions, only one of which is
 the  detoxification  of endogenous  and foreign substances  which pass through  it.
 These other functions  probably reduce  its utility  as a specific  indicator of
 pollution challenge.  This is  illustrated by the fact that liver weight  varies
 with age, diet, season,  stage of the reproductive  cycle,  and even time  of  the
 day   (Farner,   1960;  Hartman  and Browne!1,  1961;   Ljunggren,  1968).   Weight
 variations  such as  these  underscore  the importance  of examining hepatic  tissue
 as a possible indicator of pollution effects, rather than relying  on changes in
 its  weight which are relatively insensitive at best.

      Changes in hepatic structure of birds in the Col strip area in response  to
 chronic  exposure to  trace metals and  other pollutants may occur.  Toxicants such
,as ethanol  and  carbon tetrachloride  produce  cirrhosis; and  kepone  causes focal
 necrosis and marked congestion in the liver  (Huber, 1965).   Trace  metals, such
 as arsenic  and  copper, that occur  in  stack emissions, produce fatty  degeneration
 and  necrosis of hepatic parenchyma (Smith and Jones, 1961).  Gases such  as  CO,
 also found in   stack  emissions, directly suppress the detoxifying  functions  of
 rat   liver  (Cooper  et  al. , 1965; Conney et aj_. ,  1968),  whereas  chlorinated
 hydrocarbons such as DDT enhance the  catabolic functions of avian liver (Peakall
 1970).    In addition  to   direct pollution-related alterations  of  hepatic
 parenchyma,  the liver's population of Kupffer cells (as part of  the mononuclear
 phagocyte system) may  accumulate foreign particulate  material  at rates that
 correspond  to environmental pollution levels.

      In   summary,  the  possibility that  the liver  will  exhibit   histological
 changes  in  birds from  chronically impacted  areas is very  real and needs exam-
 ination.  In addition,  the liver  may be useful for distinguishing between  the
 effects of ingested pollutants and those acquired via the  respiratory system.
                                      271

-------
     2.   Kidney—The kidney is the second major site of detoxification in the
vertebrate organism as  well  as the major  site  of excretion.  It is directly
damaged by pollutants such as  copper, fluorine, lead, and mercury, all of which
occur in  stack emissions  (Smith  and Jones,  1961).   Damage  to  the kidney's
filtration apparatus is  indicated by the presence of  protein (casts)  within
renal  tubules.   There  may  be complete  loss  of renal  tubules,  or  simply
disintegration  of  their  epithelia.   Congestion,  the accumulation  of fluid
between tubules, and the presence of fibrin in the renal corpuscule are other
common pathologies.  Reliable and thorough histological  descriptions  of the
avian kidney are now available for reference (Johnson  and Mugaas,  1970 a,b).

     3.   Lung—The  respiratory tract  is a third major site  of  detoxification in
the  vertebrate organism.   In  addition,  its  alveolar  epithelium contains
phagocytes which remove particulate matter from inhaled air.   The lung  is sub-
ject to direct irritation  from inhaled substances; and because of  its unusually
large blood  supply,  it  reacts to these  substances immediately with  tissue
proliferation,  pneumonia,  focal  accumulations of leucocytes, congestion, and
visible  damage  to  the alveolar  wall  (Smith  and Jones,  1961).   The  lung's
unusual  sensitivity to  air-borne pollutants is suggested by Lewis, Morton, and
Jones  (1975).   The  histology  of  the  tissue  should  consequently  be  a
particularly sensitive  and selective bioindicator of air quality.

     Reproductive tissues.  The reproductive system is one of the first to be
affected deleteriously  by  chronic  stress (Christian and Davis, 1964;  von Faber,
1964; van Tienhoven, 1968).  Because this  system, particularly the histological
events of  spermatogenesis,  has been  well studied  in  a large number of birds
(Lofts and Murton,  1973),  any  deviations from normal in  the Col strip population
should be readily appreciated.

     The  importance  of evaluating reproductive  tissues  is  twofold.   In the
short  term,  they   should  be  relatively  sensitive,  although   secondary,
indicators of the stress associated with pollution impact on the ecosystem.  In
the  long-term  ecological   context, they  are  equally  important predictors of
population  growth  in   chronically  polluted  areas,  since  upsets   in  the
reproductive  function  of  one  generation directly influence the size  of the
following generation.

                               ACKNOWLEDGEMENTS

     The authors wish to thank the several people who have served as technical
assistants over the past two years.  A special thanks is extended to Larry Doe
and Tom Gullett  for their major assistance during  field  operations.

                              LITERATURE CITED

Anderson,  W.  L.  1970.   Seasonal changes  in thymus  weights  in Ring-necked
     Pheasants.  Condor. 72:205-208.

Batzli,  G.  D.  and F.  A. Pitelka.  1970.   Influence of meadow  mouse populations
     on  California grassland.   Ecology. 51:1027-1039.
                                      272

-------
Bendell, J.  F.   1959.  Food as a control of a population of white-footed  mice,
     Peromyscirs leucopus noveboracensis (Fischer).  Can. J.  Zool.  37:173-209.

Bergstedt,  B.   1965.   Distribution, reproduction, growth  and dynamics of  the
     rodent  species  Clethrionomys  glareolus  (Schreber),  Apodemus  flavicollis
     (Melchior)  and  Apodemus  sylvaticus  (Linne) in  southern Sweden.  Oikos.
     16:132-160.

Bhattacharyya, T. K. , and A. Ghosh.  1972.   Cellular modification of interrenal
     tissue  induced  by corticoid  therapy and  stress  in  three avian  apecies.
     Amer. J. Anat.   133:483-494.

Bhattacharyya,  T.  K. ,  B.  Ray, and  C.  Manna.   1972.   Participation of adreno-
     cortical  and  medullary hormones  during stress  in intact and  amphenone-
     treated pigeons.   Endocrinol.  Exp.  6:179-186.

Bretz, W.  L. ,  and  K.  Schmidt-Nielson.   1971.  Bird respiration:   flow  patterns
     in the duck lung.  J. Exp.  Biol.  54:103-118.

Brown, L. N.  1966.   Reproduction of Peromyscus maniculatus in the Laramie  Basin
     of Wyoming. Amer. Midi. Nat.   76:183-189.

Chapman,  D.  I.   1972.   Seasonal  changes in the  gonads  and accessory glands of
     male mammals.  Mamm.  Rev.  1:231-248.

Chester-Jones,  I.  1957.   The adrenal cortex.  Cambridge Univ.  Press,  Cambridge,
     England.  315 p.

Christian,   J.  J.   1952.   The  relation of  adrenal  weight to  body weight in
     mammals.  Naval  Med.  Res.  Ins.  Memo. Rep.   52-7:941-946.

	.  1959.   The roles  of endocrine and behavioral  factors in  the
     growth  of  mammalian  populations.   In:  Comparative  Endocrinology,  A.
     Gorbman, ed. ,  Wiley,  New York.  pp. 71-97.

                  1971.   Population density  and fertility  in  mammals.  In:  The
     Action  of Hormones:  Genes  to  Population,  P.  Foa, ed.   Charles Thomas,
     Springfield, 111.   pp.  471-499.

Christian, J. J. and D. E. Davis.   1964.  Endocrines,  behaviors,  and population.
     Science. 146:1550-1560.

Christian, J. J. , J.  A. Lloyd,  and  D. E. Davis.  1965.  The role of endocrines in
     the self-regulation  of mammalian  populations.  Recent Progr.  Hormone  Res.
     21:501-571.

Cinq-Mars,   R.  J.   and   L.  N.  Brown.   1969.    Reproduction  and  ecological
     distribution  of   the   rockmouse,  Peromyscus  difficilis,   in   northern
     Colorado.  Amer. Midi.  Nat.  81:205-217.
                                      273

-------
Conney, A.  H.,  W. Levin, M. Ikeda, R. Kuntzman,  D. Y. Cooper, and 0.  Rosenthal.
     1968.    Inhibitory  effects  of  carbon  monoxide  on  the hydroxylation  of
     testosterone by rat liver microsomes.  J. Biol. Chem.  243:3912-3915.

Cooper, D.  Y. ,   S.  Levin,  S.  Narasimhulu, 0.  Rosenthal, and R. W.  Estabrook.
     1965.    Photochemical  action  spectrum of  the terminal  oxidase of  mixed
     function oxidase systems.  Science.  147:400-402.

Emlen, J.   1971.  Population Densities  of Birds  Derived from Transect  Counts.
     Auk.  88:323-342.

              1973.   Ecology:  An Evolutionary Approach.   Addison-Wesley Co.,
     Massachusetts.  493 p.

Eroschenko, V.   1973.  The weight and histology  of  gonads,  livers,  and adrenal
     glands in the Japanese Quail as affected  by photoperiods and the insecti-
     cide  kepone.  Ph.D.  Dissertation.   Davis:  University of California.   155
     pp.

Ewart, D.  F. , and  D.  B. McMillan.  1970.  The spleen of  the cowbird.  J. Morph.
     130:187-206.

Farner,  D.  S.   1960.   Digestion  and the  digestive  system.   In:   Biology  and
     comparative  physiology of  birds,  Vol.  1,  A.  J.  Marshall,  ed.   Academic
     Press, Inc., New York.  pp. 411-467.

Fitch,  H.  S.   1954.   Seasonal  acceptance of  bait by small  mammals.  J.  Mamm.
     35:39-47.

Flake,  L.  D.  1974.  Reproduction of four rodent  species in a shortgrass prairie
     of Colorado.  J.  Mamm.  55:213-216.

Fleharty,  E.  D.  and  L.  E.  Olson.   1969.   Summer  food habits  of Microtus ochro-
     gaster and Sigmodon hispidus.  J. Mamm.  50:475-486.

Flickinger, G. I.  1961.  Effect of grouping on adrenals and gonads of chickens.
     Gen. Comp.  Endocrinol.  1:332-340.

French, N. R. , C.  D.  Jorgensen, M. H. Smith and B. G.  Maza.  1971. Comparison of
     some  IBP population estimate methods  for small mammals.  Special Report,
     USNC/IBP.

Garber,  K.   1967.   Luftverunreinigung  und ihre  Wirkugen.   Berlin-Nikolassee:
     Bebruded Born Traeger. 279 p.

Gentry, J.  B. ,  F.  B.  Golley,  and  J.  T.  McGinnis.  1966.   Effect of weather on
     small mammals.  Amer.  Midi. Nat.  75:526-530.

Getz,  L.  L.   1961.   Responses  of small  mammals  to  live-traps and weather con-
     ditions.   Amer.  Midi.  Nat.  66:160-170.
                                      274

-------
                1968.   Influence  of weather on the  activity  of  the  red-backed
     vole.  J.  Mamm.   49:565-570.

Click,  B.   1964.   The bursa of Fabricius  and the development of  immunologic
     competence.   In:   The thymus  in immunobiology.   R.  A.  Good  and A.  E.
     Gabrielsen, eds.  Hoeber Medical Division, Harper and Rowe, New York.   pp.
     343-359.

Golley, F. B. ,  L.  Ryszkowski,  and J. T. Sokur.  1975.   The role of small  mammals
     in  temperate  forests,   grasslands  and cultivated  fields.   In:   Small
     mammals:  their  productivity  and population dynamics.  F.  B.  Golley,  K.
     Petrusewicz,  and L.  Ryszkowski,  eds.   Vol  5,   International  Biological
     Programme.  Cambridge University Press,  England,  pp. 223-241.

Gorman, M.  L.  and H. Milne.   1971.   Seasonal  changes in the adrenal  steroid
     tissue  of the Common Eider Somateria mol1issima and  its  relation to or-
     ganic metabolism in normal and  oil-polluted birds.  Ibis.  113:328-228.

Gunderson, H. L.   1950.  A study of  some small  mammal populations at Cedar Creek
     Forest, Anoka County, Minnesota.  Occ.   Pap.  Minn.  Mus.  Nat.  Hist.,  No.
     4:1-49.

Gurnell,  J.  1976.   Studies  on the effects  of bait  and  sampling intensity  on
     trapping and estimating Wood mice, Apodemus sylvaticus.  J.  Zoo!.   178:91-
     105.

Hartman,  F.  A.,  and K.  A.  Browne!1.   1959.   Liver  lipids  in  hummingbirds.
     Condor. 61:270-277.

Hoffman,  R.  S.   1955.  A  population-high for  Peromyscus maniculatus.   J.  Mamm.
     36:571-572.

Hb'hn,  E.  0.   1956.  Seasonal   recrudescence  of the  thymus in sexually  mature
     birds.  Can.  J.  Biochem.  Physio!.  34:90-101.

Huber,  J.  J.   1965.   Some  physiological  effects  of  the  insecticide  kepone  in
     the  laboratory mouse.  Toxicol.  Appl.  Pharmacol.  7:516-524.

Jankovic,  B.D.,  and  K.  Isakovic.    1966.    Antibody producing in  bursectomized
     chickens given repeated injections of  antigen.  Nature. 211:202-203.

Johnson,  0. W.  , and  J. N.  Mugaas.   1970 a.   The histological  features of avian
     kidneys.  Amer.  J. Anat.   127:423-436.

                  1970b.  Quantitative and  organizational features of the avian
     renal medulla.  Condor.  72:288-292.

Johnson,  P.  L.   (ed.).   1976.   Role of  ecology in environmental management.
     Report of a Workshop. September 9-11,  1976.  Tenn.:   Oak Ridge Associated
     Universities.  22 pp.
                                     275

-------
Jolly, G. M.  1965.  Explicit estimates from capture-recapture  data  with  both
     death and immigration-stochastic model.  Biometrika, 52:225-247.

Kay, E.  1974.  An inquiry into the distributions of fluoride  in  the environ-
     ment of Garrison,  Montana.  Fluoride.  7:7-31.

Kendall, M.  D.   1975.   Sizes  and  numbers of nuclei  in  the cortex  of  thymus
     glands  of  Red-billed  Weavers  Quelea quelea.  Cell Tissue  Res.   164:233-
     249.

Krebs, C. J.  1966.  Demographic changes in fluctuating populations of Microtus
     californicus.   Ecol. Monographs.  36:239-273.

Krebs, C. J. , M. S.  Gaines, B. L. Keller, J.  H. Myers, and R. H.  Tamarin.  1973.
     Population cycles  in small  rodents.  Science, 179:35-41.

Kritzman, E. B.   1974.   Ecological  relationships of Peromyscus  maniculatus and
     Perognathus parvus in  eastern Washington.  J. Mamm.  55:172-188.

Lechleitner,  R.  R.    1969.   Wild  mammals  of  Colorado.    Pruett  Publ.  Co.,
     Boulder, Colorado. 254pp.

Lecyk, M.   1962.  The  effect of the  length of daylight on  reproduction  in the
     field vole (Microtus arvalis Pall.).  Zool.  Poloniae.  12:189-232.

Lewis, R. A., N.  R. Glass,  and A. S. Lefohn (eds.).  1975.   The bioenvironmental
     impact of a coal-fired power plant.  2nd interim  report. EPA-600/3-76-013.
     U. S.  Environmental Protection Agency, Corvallis,  Oregon.   315 pp.

Lewis, R. A.  and M.  L. Morton.  1975.  The effects of coal-fired power plant
     emissions  on vertebrate animals.   In: Proc. Fort  Union Coal  Field  Symp.,
     Montana Acad.  Sci., W. F. Clark, ed.  5:579-595.

Lewis, R. A. , M. L.  Morton, and S.  C. Jones.  1975.  The effects of coal-fired
     power  plant emissions on vertebrate animals  in  Southeastern Montana  (a
     report  of  progress).   In: The  bioenvironmental  impact  of  a  coal-fired
     power  plant,  2nd  interim  report.   R.  R.   Lewis,  N.  R. Glass,  and A.  S.
     Lefohn,  eds.,  EPA-600/3-76-013.   U.  S.  Environmental  Protection Agency,
     Corvallis, Oregon, p. 140-187.

Ljunggren,   L.   1968.   Seasonal  studies  of Wood  Pigeon populations.   I.  Body
     weight, feeding habits,  1iver and thyroid activity.  Viltrevy. 5:463-504.

	.   1969.   Seasonal  studies  of Wood  Pigeon populations.   II.
     Gonads,  crop  glands,  adrenals  and the hypothalamo-hypophysical system.
     Viltrevy. 6:41-126.

Lofts, B. ,  and R. K. Murton.  1973.   Reproduction in birds.   In:   Avian biology,
     Vol. III.   D. S.  Farner  and  J.   R. King, eds.  Academic Press,  Inc., New
     York.   pp.  1-107.
                                      276

-------
MacArthur,  R.  H.   1960.  On the  Relative  Abundance of Species.  Amer.  Nat.
     94:25-36.

MacArthur, R. H. and J. W. MacArthur.  1961.  On Bird Species Diversity.   Eco-
     logy, 42:594-598.

MacKay, A. A.  1962.   An easy method of trapping small taiga mammals in  winter.
     J. Mamm.  43:556-557.

Margalef, R.   1963.  On  Certain  Unifying Principles  in  Ecology.   Amer.  Nat.
     97:357-374.

Margalef, R.  1968.  Perspectives in Ecological Theory.  University of  Chicago
     Press, Chicago.   Ill p.

Martin,  E.  P.   1956.  A population  study of the prairie vole  (Microtus ochro-
     gaster)  in northeastern Kansas.  Univ.  Kans. Publ. Mus. Nat.  Hist.  8:361-
     416'.

McKeever, S.  1959.  Effects of reproductive activity on the weight of  adrenal
     glands in Microtus montanus.  Anat. Rec.  135:1-5.

                   1964a.  The  biology of the golden-mantled  ground  squirrel,
     Citell us lateral is.  Ecol.  Monogr.  34:383-401.

	.  1964b.  Variation in the weight of the adrenal,  pituitary and
     thyroid gland of the white-footed mouse, Peromyscus maniculatus.   Amer. J.
     Anat.  114:1-15.

Moticka,  E.  J.   1975.   Development of  immunological  competence in  chickens.
     Amer. Zool.   15:135-146.

Oakeson,  B. B.   1953.  Cyclic changes in liver and spleen weights in  migratory
     White-crowned Sparrows.  Condor.  55:3-16.

Odum,  E.  P.   1969.   The Strategy of Ecosystem  Development.   Science.  164:262-
     270.

O'Farrell,  T.  P.,  R.  J. Olson,  R. 0.  Gilbert,  and J. D.  Hedlund.   1975.   A
     population  of Great Basin  pocket mice,  Perognathus parvus,  in the  shrub-
     steppe of south-central Washington.  Ecol. Mongr.  45:1-28.

Peakall,  D.  B.   1970.   p,p'-DDT:   Effect on  calcium  metabolism and concentra-
     tion of estraodil in the blood.  Science. 168:592-594.

Pielou, E. C.   1977.  Mathematical  ecology.   John Wiley  and Sons, New York.  385
     pp.

Piliero, S. J.  1970.  Immune suppressants.   BioScience,  20:710-714.

Riddle, 0.  1928.  Sex and seasonal differences in weight of liver and spleen.
     Proc. Soc.  Exp.  Biol.  25:474-476.


                                      277

-------
Robbins,  C.  S.  and  W.  T.  Van  Velzen.   1970.   Progress Report on  the  North
     American Breeding  Bird Survey.   pp.  22-30.   In:   Bird Census  Work and
     Environmental  Monitoring,  S.  Svenson,  ed.   Swedish  Natural  Science
     Research Council, Redaktionstjansten.

Sanders,  H.  L.   1969.   Benthic  Marine Diversity  and  the Stability-Time  Hypo-
     thesis.  Brookhaven Symp.  Biol.  22:71-81.

Shannon,  C. E. and W. Weaver.  1949.  The mathematical theory of communication.
     University of Illinois Press,  Urbana.  117 p.

Siege!, H.  S.   1971.  Adrenals,  stress,  and  the  environment.   World's Poultry
     Sci.  J.  27:327-349.

Slobodkin,  L.  B.  and  H.  L.  Sanders.    1969.    On  the  Contribution  of
     Environmental  Predictability  to  Species  Diversity.   Brookhaven  Symp.
     Biol.  22:82-95.

Smith, H.  A.  and T. C. Jones.  1961.  Veterinary pathology.  2nd ed.  Lea and
     Febiger, Philadelphia.  1521 pp.

Stickel,  W.  H.   1975.   Some effects of  pollutants  in  terrestrial  ecosystems.
     In:  Ecological Toxicology Research.  A. D.  Mclntyre and C. F.  Mills, eds.
     Plenum Press, New York.  pp. 24-74.

Stoewsand,  G. S.  and M. L. Scott.   1964.  Influence  of  high  protein diets on
     vitamin A metabolism  and adrenal  hypertrophy in the chick.  J.  Nutrition.
     82:139-144.

Tramer, E. J.  1969.  Bird  Species  Diversity:  Components of Shannon's Formula.
     Ecology. 50:927-929.

Truhaut,  R.  1975.  Ecotoxicology—a new branch of toxicology:  a general sur-
     vey  of  its  aims,  methods,  and  prospects.   In:    Ecological  Toxicology
     Research, A. D. Mclntyre and C. F. Mills, eds.   Plenum  Press, New York.  p.
     3-23.

van Tienhoven,  A.   1968.   Reproductive  Physiology  of  Vertebrates.  W.  B.
     Saunders Co., Philadelphia.  498 pp.

von Faber,  J.   1964.   Stress  and  general   adaptation  syndrome  in  poultry.
     World's Poultry Sci.  J.  20:176-182.

Ward, P.  and D.  D'Cruz.   1968.    Seasonal  changes  in  the thymus gland  of a
     tropical bird.   Ibis.  110:203-205.

Ward, P. and M. D. Kendall.  1975.   Morphological  changes  in  the thymus of young
     and adult Red-billed Queleas Quelea quelea (Aves).   Phil. Trans.  Roy. Soc.
     London Ser.  B.  Biol. Sci.  273:55-64.

Weiner,  J.  G.  and  M.  H.  Smith.   1972.   Relative efficiencies of  four  small
     mammal traps.  J. Mamm.  53:868-873.

                                      278

-------
Wiens, J.  A.   1974.   Climatic  Instability and the  'Ecological Saturation1 of
     Bird Communities in North American Grasslands.  Condor.  76:385-400.

Williams, 0.   1959.  Food habits of the deer-mouse.  J.  Mamm.   40:415-419.

Zarkower,  A.   1972.  Alternations  in antibody response induced by chronic in-
     halation of S02  and carbon.  Arch.  Environ.  Health. 25:45-50.
                                     279

-------
                                   SECTION 7

                    REMOTE SENSING OF THE BIOENVIRONMENTAL
                      EFFECTS OF STACK EMISSIONS IN THE
                               COLSTRIP VICINITY

                                     by

                      J. E. Taylor and  W. C. Leininger


                                INTRODUCTION

     Procedures  for  ground  and  aerial  photography  have  been  developed  to
monitor air  pollution  effects in  a grassland  ecosystem.   Photography offers
several strong  advantages  for  biological   monitoring.   Permanent,  complete,
periodic records  can be obtained, stored, exactly reproduced,  and compared over
time.   In  this  way trends  and changes  can be recognized and documented.  Also,
the relatively  unselective nature  of  photography permits  re-examination  of
past imagery for previously unnoticed features which later studies show to  be
important.

     The  photographic   studies   described  here   include  groundlevel   and
relatively  low  elevation   aerial  imagery.   Periodic  flights  allow aerial
assessment of rates of  vegetation changes.  Ground studies help discriminate
between normal  and stress-induced  changes.  Pollution stress is  detectable, but
not yet quantifiable, in low-level aerial photography.

                           GROUND  LEVEL PHOTOGRAPHY

     Ground  level  photography provides  a  detailed  record of  plant  species,
phenology, and  pathologic signs.   All of this contributes to the total  data set
and assists  in  the interpretation of aerial  imagery.  Also,  vertical ground
photo plots may be measured and analyzed for cover, number,  frequency,  pattern,
and plant  volume.

PROCEDURES

     Photo plots  are  established  on  all  study sites  and  on  representative
examples  of  the  various  communities  present  in  the area.  Permanent meter-
square plots are  photographed stereoscopically in color and black-and-white.
Details are  given  by Taylor et aj_.  , 1976.

     Aspect  photographs  are made from vantage  points  within and overlooking
plot areas.  These are taken  with color and  infra-red color,  the latter to
compare with aerial coverage.
                                     280

-------
     Ground photography  is  collected periodically through the growing season
so that any changes will be recorded.

DISCUSSION

     A sample stereoscopic pair of photo plots appears as Figure  7.1.  Species
identification, canopy cover, litter, bare ground, and plant  heights are among
the kinds of data which can be derived from these photographs.

     Plant volume, which  can be used to estimate biomass, may be obtained by
combining  canopy  coverage and  height,  the  latter  measured  with  a  parallax
wedge.   Plant density and pattern also can be  studied from these pictures.

     For each photo plot an index of species  identification has been prepared.
An example is shown as Figure 7.2.  Species which are diffused throughout the
quadrat are termed "matrix species," and are listed but not precisely located
on  the  indices.   Species which  are  readily  distinguishable  in the  plot-
photographs are mapped.  The combination of plot photographs  and plot indices
makes  a permanent record  of species presence  and distribution.  Sequential
records allow the  evaluation  of temporal changes.

                              AERIAL PHOTOGRAPHY

     Low  level  aerial  photography  gives  a more  generalized view  of  plant
species  and  community distributions,  pathology,  and  cover  than does  ground
photography.   However, it yields more detail  than higher level imagery and so
represents  a  useful  compromise  between   detail  and comprehensiveness  of
coverage.

     Low level  aerial imagery is most practical  for making detailed vegetation
maps, sensing population-level stresses, or any other purposes requiring large
scale  synoptic  views.   It also aids in developing interpretations of smaller
scale, high level  photography.

PROCEDURES

     In  this  project,  a  Cessna  182  airplane  which  can easily  handle  the
elevation  range of 500 to 7000 feet above ground, was used.   The  plane, leased
from Miles City Flying Service, Miles City, Montana, has been modified by the
addition of a 12" diameter belly hole, which accepts a special mount, designed
and manufactured  by  W.E.  Woodcock of Miles  City (Woodcock,  1976).  The mount
supports a Hasselblad EL/M motor-driven camera (70 mm  format) and a manually-
operated  Leica MDa  (35 mm).   Both  are fitted with  50 mm  lenses,  yielding
negative  scales  from 1:3000 to  1:40,000.  Other  focal  lengths  are sometimes
used to  obtain changed  scales without changing flight  elevation.  For special
purposes, 80,  150, 250,  and 500 mm lenses are  used.

     In  addition  to  the mounted camera, oblique photographs  from the air are
taken with a second Hasselblad used as a hand camera.   This  kind of photography
supplements  the   more   traditional   vertical   imagery,  since  it  is  more
representative of familiar aspects of scenes for interpretation, display, etc.
                                     281

-------
ro
8
                Figure 7.1.   Example stereogram of photoplot (McRae Knoll B, June 16, 1975).

-------
ro
oo
OJ
0 / 2 3 4 5 6 7 8 9 IO


1


2

3
4

S

6

7

8

9

in
i i i i i i i i PK
.zz x~"^
•K / ^
.C F )
.z \ ' -z
\_x

- .c

• s -z
/ \ 	 »A
/ Ns
/ 1 *Z >Z
/ 1
"\ F '
\ 1
~ \ ; B*
•K\ /
- \ /
~^~ — — ' *z







SYMBOL
—
Z
C
	
F

K
S

B
A



                                              Agropyron spicatum
                                              Allium  textile
                                              Artemisia cana
                                              Artemisia frigida
                                              Car ex fill folia*
                                              Ceratoides lanata
                                              Koeleria  cristata*
                                              Leucocrinum montanum
                                              Sphaeralcea coccinea
                                              Stipa comata*
                                              Taraxacum  officinale
                                              Tragopogon dubius
                                                   (*MATRIX SPECIES)
Figure 7.2.   Photoplot  index of McRae Knoll B, June 17, 1975.

-------
DISCUSSION

     Three  primary  film  types  are  used with  other  materials  for  special
purposes.   The main films are color,  color infra-red and black-and-white (H&W
VTE Pan).  Each has advantages for  particular  uses.

     Color  is realistic  and  many  plant species are  easily recognized.  It is
easy and fast  to process, and is  visually acceptable to lay viewers.

     Color  infra-red  is excellent for  delineating  plant  species and stress
effects.  Although  it appears strange to inexperienced users, it is useful for
many aspects of this project.

     Black-and-white is most practical  for mapping photography, especially at
relatively small  scales.  It makes a  good,  inexpensive base map  for overlaying
ground  information  (vegetation,  geologic  features,  soil  types, drainages,
etc.).

     The  Hasselblad  format  (6x6  cm)  allows  greater coverage  than  the  35 mm
(2.4x3.6  cm)  frame.   Especially at low elevations (<1000 ft),  this increases
the  chance  of  covering  ground  targets completely.   The  35  mm photography
permits processing  in  the  field,  so  that  pictures  taken  one  day can  be
interpreted the next.   Also,  if  exposures or coverage are not acceptable, a re-
flight can be made immediately.  When the larger  format photography is returned
from processing,  the interpretations on the 35 mm imagery can be applied.

     An example   of  low  level  aerial   photography   is  given  in  Figure  7.3.
General plant pattern, landforms,  deferment effects, and sampling impacts are
apparent.   This kind of photography can be analyzed for plant cover, pattern,
and identification of  conspicuous  species.

                                   COLSTRIP

      Investigators working on this aspect of the Colstrip project have gained
enough experience to make  detailed vegetation maps  from aerial  photographs with
only  limited  ground  checking.   Not only can they separate gross  vegetational
groupings  (trees, shrubs, grasslands),  but they are also able to  discriminate
among grassland types  at a high degree of  resolution.

     Figures 7.4, 7.5, and 7.6 are  vegetation maps of the McRae Knolls. Thirty-
six mapping variables are indicated;  some are plant community types and others
are individual species.  Table 7.1   is an index to map units.

     The  base for  this  map was   1:660  color  aerial photography.   A ground
transect  network provided criteria for mapping units and identification.  This
map constitutes  a  baseline for mapping other sites  in  the general area.

     The  investigators  conclude that four overflights (early spring, peak of
green  for cool-season species, peak of green  for  warm-season species, and
summer dormancy)  using  both  color and color  infra-red  film types provide an
adquate base for detailed vegetational mapping in heterogenous situations.
                                     284

-------
ro
8
                         Figure  1.6.   LOW levei  aeriai  pnotograph or Hay Coulee exclosure.

-------
CD
CT)
                         Figure  7.4.   Detailed vegetation map, McRae Knoll A, Summer,  1976.

-------
INi
                        Figure 7.5.   Detailed  vegetation  map,  McRae Knoll  B,  Summer, 1976.

-------
ro
CO
CO
                        Figure 7.6.  Detailed  vegetation  map,  McRae Knoll C, Summer, 1976.

-------
TABLE 7.1.  INDEX TO MAP UNITS.

SYMBOL
1
2
3
4
5
6
7
8
9
10
n
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
36
27
28
29
30
31
32
33
34
35
COMMUNITY
Tree types
Acer negunda
Fraxinus pennsylvanica
F. pennsyl vanica/Prunus virginiana
Juniperus scopulorum
Populus deltoides
Primus americana
P. virginiana
Prunus/Acer/Fraxi nus
Salix amygdaloides
Shrub types
Artemisia cana
A. cana/mixed grass
Artemisia dracunulus
Ceratoides lanata
Rhus trilobata
R. trilobata/P. virginiana
R. trilobata/Ribes/P. virginiana
Ribes
Rosa woods ii
Shepherdia canadensis
Symphoricarpos occidental is
S. occidental is/A, cana
S. occidental is/Ribes
S. occidentalis/Ribes/R. woodsii
S. occidental is/R. woodsii
S. occidentalis/R. woodsii/Ribes/P. virginiana
Forb types
Yucca glauca
Mel i lotus officinal is
Grass types
Bromus inermis
B. japonicus/B. tectorum
Carex filifolia/B. japonicus
Calamovilfa longi folia
C. longifolia/C. filifolia
Poa pratensis/Agropyron smithii/Stipi viridula
P. pratensis/Carex
P. pratensis/S. viridula/A. cana/A. smithii
Schizachyrium scoparium

               289

-------
     In more  homogenous  vegetation,  additional  ground  studies are  needed  to
elucidate the more subtle community patterns.  The Kluver sites and Hay Coulee
fall  in  this   category.    Larger  scale  photography  and   further   ground
observations are planned  for these locations.

                                     ZAPS

     Even though the aerial photography shows distinct tonal  differences among
the ZAPS plots,  the  reasons  are still  to  be determined.  Early  signs  of S(L
stress  can  be  detected,  but  the magnitude  of pollution  damage cannot  be
quantified.   Additional  ground  studies and more detailed  photographic analysis
should lead  to an accurate and fast method for damage assessment.

                                    SUMMARY

     Photo  plots  have been  established throughout  the  study  area,  and are
examined  periodically during  the growing  season.   They  provide records  or
changing species composition, cover, and stress effects.

     Aerial   photography  techniques  have  been  developed  to   monitor  air
pollution effects as shown by plant changes.  Color,  color infra-red and  black-
and-white films  each  have particular uses.

     Pollution  stress is  detectable  in  aerial  photography,   but cannot  be
quantified without additional analysis.

     A procedure for  vegetational mapping has been developed and tested.

                                  REFERENCES

Taylor, J.  E. ,  W.  C.  Leininger,  and  R.  J.  Fuchs.   1976.  Monitoring plant
     community  changes   due  to  emissions  from  fossil  fuel power plants  in
     eastern  Montana.  Section  II  of the bioenvironmental  impact of a coal-
     fired  power plant,  second  interim  report.  USEPA  Ecological Research
     Series  EPA-600/3-76-013. pp.  14-40.

Woodcock,  W.  E.   1976.   Aerial  reconnaissance and photogrammetric with small
     cameras.  Photogrammetric  Eng. and Remote Sensing.42:503-511.
                                     290

-------
                                    SECTION 8

                INTEGRATED AEROSOL CHARACTERIZATION MONITORING,
                               COLSTRIP, MONTANA

                                       by

    N. L. Abshire, V.  E.  Derr, G. T. McNice,  R.  Pueschel, and C. Van Valin


                                 INTRODUCTION

GENERAL

      In  addition  to   other  objectives, the  Colstrip  Coal-Fired  Power Plant
Project  requires  the  concentration, dispersion and identification of background
particulate matter and particulate  matter resulting from the activities of the
mines  and  coal-fired  power  plants.   This  project  started  in  May  1975.
Observations  were performed at Hay Coulee  from  May 18 to June 16, 1975; from
August 17 to  September 15,  1975;  and from May 23  to June 5,  1976.  In addition,
aircraft flights  were made on April 20 and 21,  1976.   The observation periods
were  chosen to give maximum density of observation in the Spring and  Fall at the
beginning and end of  the growing  season.  Beginning in  October 1976 a sampling
flight  was  made  between  the  intense  observation periods.   Further  intense
observations  were made August  14  to August 28, 1976, and in  the Spring and Fall
of 1977.

      This research program  is designed to  help the  Environmental  Protection
Agency Laboratory at  Corvallis,  Oregon, assess the impact of coal-fired power
plants on a grassland  ecosystem and develop a valid and effective environmental
impact assessment protocol.  Atmosphere particles  emitted by coal-fired plants
(e.g., ash) or formed  from gases  emitted (e.g., sulfates or chlorides) produce
an array of  effects  on  the  ecosystem.  These particulates  may produce direct
biolgical effects by  contact through leaves and soil.   They  may also affect the
radiation  reaching the  biosphere  by   absorption and  blanketing  mechanisms.
Further, by adding cloud  condensation  nuclei  and  ice nuclei  to the atmosphere,
they  may significantly  modify precipitation, either  near  the source  or  far
downwind.  The  effects  of  particulate matters  are strongly  dependent  on  the
temperature structure  (inversions) and mixing and  diffusion characteristics of
the atmosphere downwind  from the  source.  The parameters needed to characterize
the effects of particles should be examined  over  a wide geographical  area and
over sufficient time to obtain  representative averages.

     The program seeks to characterize  the impact  of air pollutants on the total
ecosystem.  Small particles are a significant part of the pollution emitted by
fossil fuel  power plants, even when the stack emission  has been carefully fil-

                                      291

-------
tered.   These particles  from the stacks and the natural  background of particles
from the earth's  surface  are called "aerosols."  For this  report, clouds and
fogs are excluded  from the term aerosols.  Aerosols range in size from a  clump of
a few molecules to tens of micrometers.  The larger aerosol particles  fall out
rapidly  and  are  not  as  frequently  observed  in  the  atmosphere.    However,
particles  up  to  several  micrometers  in diameter may remain  suspended in the
atmosphere  for  long periods,  up  to weeks  or months.  Thus  if filters which
remove  large  particles  are  not as effective  in  removing  small particles, the
plume  from a smoke  stack may produce significant  pollution  which can remain
suspended  in  the  vicinity,  or be swept by  winds over distances up to several
hundred miles.  In  order  to  evaluate  the amount of such material, its source,
its distribution in the atmosphere,  the effects of local weather conditions and
terrain, and to identify the  composition and size and shape  distributions of the
particles, it is necessary to use special  remote, i_n situ and  airborne  sensors.
These sensors are  described  in some detail in the  Discussion  section of this
paper.   The principles of such measurements  are discussed below.

MEASUREMENT RATIONALE

Particles

     The remote measurement of the distribution of particles  in the atmosphere
is  accomplished by  the  use  of  lidar  (Lujht Detection and  Ranging).  Lidar
operates very much  like  radar.   A powerful short pulse  of light  (10"8sec) is
transmitted.  The light scattered back from small particles  is collected by a
telescope  (70 cm diameter) and detected by  photomultipliers.   Lasers, operating
at near visible wavelengths are required to detect particles whose radii are of
approximately the same size as  the  electromagnetic radiation (light)  waves
scattered.   If  the  particles are small  compared with  the wavelength of the
radiation  (as would be the case for microwave radar) the scattered radiation is
too weak to  detect.   Detection  of particles by  a  calibrated lidar yields two
pieces of  information.   First, the backscatter  cross section is obtained.  It is
dependent  on  the  size,  shape and index of  refraction of the particles and is
proportional to the number of particles of a given  size, shape  and index.  The
chief  asset of lidar  remote sensors  is  the  ability  to obtain  signals  from
particles  at  large  distances,  up  to  20  km,  and  by  scanning  the  vertical
hemisphere, with  good  spatial  resolution, to determine the distribution of
particles.   Second,  the  polarization of the scattered light is measured.  Since
the laser transmitter is linearly polarized, the degree of depolarization caused
by the  scatterer may be determined.                                 •

     The degree of  depolarization is  indicative of the shape of the particle:
spherical  particles  produce  no depolarization in backscatter;  non-spherical
particles  can  cause  significant  depolarization.   Since  newly-disturbed dust
particles  in  the  atmosphere are  frequently very non-spherical, the depolari-
zation is frequently a useful way of discriminating between dust from the mines
and the  more spherical particles from the stacks.

     The lidar  cannot  measure  mass  or particle concentrations without cali-
bration.   For  that purpose,  aircraft sampling is employed.   Suitable nucleopore
or millipore membrane filters capture particles for  later elemental, size, and
shape  analysis   with  an   x-ray  fluoresence  spectrometer   and   an   electron


                                     292

-------
microscope.   The  aircraft   also  contains  an Aitken  particle  counter  which
determines  the  number of  particles per  liter    With  these calibrations, and
since the  variation  of aerosol types in time and space is small compared with
aircraft  sample times,  the  lidar  may  be  used  efficiently  to  measure mass
concentrations   over  very  large  volumes   and   observe   changes  in  those
concentrations under the  influence of local weather conditions.

     The very important identification of particle characteristics  is performed
by  several  instruments located on  the  ground in a trailer  next to the lidar
system.   This system,  of  fundamental  importance to the project, measures many
parameters  of the  particles.  The trailer is equipped to continuously measure
the  total   aerosol  population,  or  "Aitken  nuclei,"   (active  at  300  percent
supersaturation),   cloud   condensation   nuclei    (active   at   0.1   percent
supersaturation),  aerosol  light  scattering  and  standard  meteorological  para-
meters such as wind speed, direction, temperature, and  humidity.

     The  direct measurement of  solar  radiation  at   visible  wavelengths  can
provide an independent determination of  aerosol loading.  Radiation from the sun
in  the  visible  range  is known  to  remain constant to within  1  percent over
periods of  decades.  Observation  of solar  irradiation by instruments at ground
level can therefore be used to accurately measure the attenuation imposed by the
atmosphere.  The major  causes of atmospheric attenuation (turbidity)  are clouds,
aerosols, ozone, water vapor, and Rayleigh scattering by the air.  The effect of
Rayleigh  scattering  can  be  accurately  calculated  and  allowed  for   Ozone and
water vapor absorption occurs  in  specific  wavelength bands  and  can  thus  be
avoided  or corrected  by  proper  choice  of wavelength band  for the observing
equipment.  Therefore, when  clouds  are  not  between the sun  and observer,  the
attenuation by atmospheric aerosols  alone can be determined.

Radiation Effects

     Two  radiation measurements  are of  importance.  First,  the rate  at which
solar energy  reaches  the  earth as  a function of  wavelength must be measured
while  the  power plants  are  not in operation  (baseline  measurements).   The
changes  registered  during   succeeding  measurements  while  the  plants  are
operating  may then be correlated with  the particulate loading and changes  in
cloud cover.  Second, the net heat loss of the earth must be measured by infrared
radiometry  by measuring  the  upward and downward radiation,  employing aircraft
and ground-based instruments, operating  from  8 to 12 pm.

     Measuring  the  IR  fluxes  within  and  outside the plumes  gives  the  IR
"shadowing" effect by  the plume,  inversion techniques permit the determination
of  the  IR  volume  absorption  coefficient.   In  conjunction  with  the  measured
aerosol  size distribution,  the aerosol absorption index can be evaluated, aiding
in the evaluation of  long-term climatic effects.

Meteorological Conditions

     Standard  meteorological variables   are  measured  by standard  instruments.
The   meteorological   conditions  which   trap  pollutants   below  temperature
inversions  are  best  measured by  an acoustic sounder.   The acoustic sounder is
also a radar-like  device, which uses a pulse  of sound.  The  scattering of sound
back  to  the  receiver  is dependent  on  the  degree  of  homogeneity in vertical

                                      293

-------
temperature structure.  The device is especially sensitive  then  to  temperature
inversions and can determine their height and persistence.

ENVIRONMENTAL EFFECTS

     Environmental  and health effects of aerosols produced  by  human activities
and by industrial activities in rural Montana are foreseen in four areas:

     1.   Inadvertent weather modification:   Size,  shape and  water solubility
(chemical composition) determine the cloud and ice nucleating capabilities, and
thereby effects of the atmospheric aerosol on cloud and precipitation processes.
The proposed experiments are designed to scrutinize a cause-effect relationship
between weather modifying particular matter and human activities in Montana.

     2.   Climate   modification:    Size,   shape   and  chemical   composition
(refractive index)  determine the light scattering and  absorption capability of
the atmospheric aerosol.  The backscattering to absorption ratio, in  relation to
the  surface  albedo,  is the  critical  parameter that  determines  whether the
aerosol's effect leads to cooling or warming of the earth-atmosphere system.  IR
extinction  can cause  surface  temperature  changes  by  thermal blanketing and
affect  radiometric measurements from satellites.  All  of these parameters will
be evaluated by means of the experimental data.

     3.   Ecological effects:  The  damage  to vegetation  by sulfur  and  halogen
gases and to  structures by corrosive particulate sulfates and  halides downwind
from their  sources is greatly affected by the gas-to-particle  conversion rate.
Ir\ situ measurements of this conversion rate as  function of the amount of water
vapor  and  third-body  aerosols  is  planned  by  adding  proper gas analysis
techniques to the measurements proposed.

     4.   Health effects:  Particle sizes determine the depth of penetration of
an aerosol  into the animal respiratory system.  Particle solubility (chemical
composition)  determines their  retention times.   The data  from the proposed
research  will  enable  health  officials  to  better  evaluate  health  hazards
resulting from environmental pollution.

GENERAL DISCUSSION OF OBSERVATIONS

     Observations were planned for Spring and late  Summer of 1975 to provide  a
baseline against which to judge changes in atmospheric aerosols aft*r the coal-
burning plants began operation.   Further observation was planned for Spring and
late Summer 1976 and 1977.   The  observation  of May 23 to June 15,  1976, provided
opportunity only for measurements of unit #2 operating under initial tests with
gas  rather  than coal.   Interesting differences from  1975 observations were
observed which can  only be partially reported here  until the data processing is
complete.
                                      294

-------
                            MEASUREMENT TECHNIQUES

GENERAL APPROACH

     In  order to  adequately  characterize the  aerosol  content, a combination
of  observation techniques  was  used.  The method was  to make well calibrated
in situ  and  airborne  measurements  to   provide  benchmarks  for  some  of the
remote  sensing devices,  primarily  the lidar  (laser  radar),  enabling them to
extend  the  calibrations  throughout  the volume  of  interest.   The  various
devices  and techniques  are described below and  are  summarized in Tables 8.1
and 8.2.

SURFACE MEASUREMENTS

     1.   Aerosol  collection:   Aerosols  were  collected on  membrane  filters
and  later  examined for  number, shape  and size distribution in a transmission
electron  microscope,  for  elemental  composition  (atomic  number   >  11)  in  a
scanning  electron microscope with  energy dispersive  x-ray analysis,  and for
ice nuclei in a thermal  diffusion chamber.

     2.   Aitken  Nuclei  (AN)  count:   AN are  defined  as  the total  aerosol
population,  and their  concentration  was measured with a  continuously operating
Environment-One   particle  counter.    This  unit  subjects  a   humidified  air
sample  to  rapid  adiabatic expansion  that results  in  a water saturation of
300  to 400  percent.   In this condition,  water droplets  form on  all  aerosol
particles   present   in   the   chamber.    The   particle   concentration  is
proportional  to  the  attenuation   of  a  light  beam  propagated   through  the
cloud.

     3.   Cloud  Condensation  Nuclei  (CCN):   Many of the  AN  are  effective as
CCN.  Their  concentration was  measured in a thermal  diffusion chamber by the
photographic  technique  of Allee   (1974)  at  .5  percent  supersaturation,  a
valve  typically  found  in nature.   The cloud  droplets  found were  photographed
and counted.

     4.   Ice  Nuclei  (IN):  IN comprise  a small but  important subset of the
AN.  These  were   counted with  an NCAR acoustic ice  nucleus  counter  (Langer
et al_. , 1967) at -20°C  (Bigg e_t al_. , 1963).

     5.   Gas  analysis:  Sulfur  dioxide  was  monitored  with   a Meloy Sulfur
Gas  Analyzer.   A  Theta  Sensors,  Inc. sulfur  dioxide  analyzer was  used for
the 1976 airborne measurements.

     6.   Meteorology:    Wind  speed  and  direction,   temperature,    relative
humidity and insolation were measured continuously.

     7.   Insolation:    The  total   radiant  energy  density,   including  both
direct  and  scattered  sunlight,  was  continously  monitored  with  an Eppley
pyranometer.

     8.   Light  scattering:  The   coefficient  of  scattering  of  atmospheric
aerosols was  measured continuously with  the Meteorology Research, Inc.  model
1550 nephelometer.

                                     295

-------
                 TABLE 8.1.   PARAMETERS MEASURED IN THE NOAA EPA COLSTRIP INTEGRATED ATMOSPHERIC

                             CHARACTERIZATION PROGRAM.
ro
to
01
Parameter
Temperature
Wind, Direction, Vel.
Cloud Cover
Relative Humidity
Solar Energy, Direct
Solar Energy, Direct
Spectral
Solar Energy, Total
Turbidity
Lidar Aerosol Scatter
Radar
Atmospheric
Temperature
Turbulence
Radiometer
Photography
Units
"C
M/S, Degrees
%
%
Cal/cm /min
Cal/cnWmin
Cal/cm /min
Volts
Backscatter
Intensity
Backscatter
Intensity
(Uncalibrated)
Backscatter
Intensity
(Uncalibrated)
watts/cm'

Method
of
Recording
Recorder
Recorder
Log
Recorder



Log
Log, Digital
Tape, Photos
Photos
Facsimile
Chart, Tape
Film
Instrument




Pyrheliometer
(no filters)
Pyrheliometer
(w/3 filters)
Pyranometer
(no filters)
Volz
Turbidity meter
Lidar System
Radar System
Acoustic
Sounder
Barnes PRT-5
Camera
Frequency
of
Recording
60, (00)
Continuous
1. 2 times
daily or when
changes occur
2. Whenever
insolation meas.
made
Continuous
Continuous
during day
1000,1300,
1600
Continuous
during day
1000,1300,
1600
In accordance
with variability
Clouds only
Continuous
Chart-contin-
uous tape-on
each shot
As needed
Precision
0.5 C
10%
Accuracy
0.5 C
10%
See FMH - 1
10%
5%
5%
5%
Unknown
±10%
—
—
10%
--
10%
5%
5%
5%
Unknown
±30%
—
—
10%
-
Remarks
....



Wide Band
3 wavelength bands, "when
sun's disk is visible
APCL+WPL
....
Polarization Ratio Available
Test for Cloud Observation
Mark time on Facsimile
Chart
Mark time on Chart
Targets of Opportunity

-------
TABLE 8.2.  PARAMETERS MEASURED IN THE NOAA EPA COLSTRIP INTEGRATED ATMOSPHERIC
            CHARACTERIZATION PROGRAM.(continued)
Parameter
Cloud Condensation
Nuclei
Cloud Condensation
Nuclei
Ice Nuclei
Ice Nuclei
Light Scattering
Aitken Nuclei
Aitken Nuclei
Cloud Cover
(Insolation)
Sulfur Gases
Elemental Composition
Of Individual Particles
Aerosol Size
Distribution
Atmospheric
Transmissivity
(IR Extinction Coefficient)
Units
CM'3
CM'3
CM'3
CIVT3
bscat<1°-4m-1>
CM'3
CM'3
CM'3
ppb-ppm



Method
of
Recording
Recorder
Photographic
R
L
R
R
I
R
R
PandL
Pand L
R
Instrument
MRI
Allee
Photographic
NCAR Acoustic
Counter
Langer Diffusion
Chamber
MRI
Env. -1
Gardner Counter
Eppley
Pyranometer
Meloy
Electron
Microscope
(SEM-EDX)
Electron
Microscope
(SEM-EDX)
Barnes
Radiometer
Frequency
of
Recording
Continuous
Point
Sample
C
PTI
C
C
Point Sample
C

Preselected
Time
Intervals
Preselected
Time
Intervals
Airborne
Measurements
Precision
±20%
±10%

±50%
± 5%
±10%
±10%
•-
±0.01 ppm
For Heavy Metals
Approximate
Sensitivity 10"15gm
±10%
0.5 C
Accuracy
Unknown
Unknown
but high
Unknown
(State of the Art)
Unknown
(State of the Art)
± 5%
±10%
....
....
±0.01 ppm

±10%
0.5 C
Remarks

Visual Counting-
Absolute Measurement

Including Airborne
Collection by
Membrane Filter

—
Ground and Airborne
Measurements
—
Including Airborne
Measurements
Membrane Filter Collection--
Ground and Airborne
Membrane Filter Collection-
Ground and Airborne
Downward Looking

-------
AIRBORNE MEASUREMENTS

     A Cessna  182  single  engine aircraft was  utilized for the  measurement of
temperature,  AN  concentration  (using a  Gardner Small  Particle Counter) and
the  collection  of  aerosols  on  membrane  filters  as  described  earlier.  A
Barnes  PRT-5   radiometer  was  utilized  for  the  measurement  of  atmospheric
transmissivity looking down.

REMOTE SENSING MEASUREMENTS

     1.   Lidar:    The  remote  measurement of  atmospheric particle density was
accomplished  through the  use  of Lidar.  A  short powerful  pulse  of  light is
transmitted  into  the  atmosphere.   Some  of  this light is  scattered by the
atmosphere  back   into  a  telescope,  detected  and  processed, yielding the
optical  backscatter  coefficient of  the scatterer.   This  quantity  is  propor-
tional  to  the  number of  scatterers  of  given  size,  shape  and  refractive
index,  but cannot produce either  mass  concentration,  size distribution or
chemical  composition without  the support of the  i_n  sj'tu and  airborne  measure-
ments.  For a  given  situation,  however, it  is  very  sensitive  to  any  changes,
particularly  in  concentration,  and  has  proved  very useful  in alerting the
other observers as  to a change in the character of the atmospheric aerosols.

     The  lidar system consists  of a  semi-trailer and a  van  (Figure  8.1).  The
lasers  and  telescope arejocated in  the  semi-trailer.   The laser generates a
5  joule pulse,  at 6943 Angstroms  wavelength,  30 nanoseconds  in duration at
repetition  rates  up to  1  pulse  per  second.   The  beam  is  collimated  to 1
milliradian  beamwidth and aligned  parallel  to  the  telescope within  1 milli-
radian.   The  telescope  is  a  folded Newtonian  configuration  with  a  70 cm
diameter  and  a  2  meter  focal  length,  and  can  be positioned to  cover the
vertical  hemisphere.  The backscattered  light  is collected in  the  telescope,
transmitted to photomultipliers, detected  and sent to the van for processing.

     The  detected  signal  is  digitized every  10 nanoseconds   and  stored on
magnetic  tape  for  data reduction in  a  large  digital computer.  The digitizer
also  stores the  signal  return  and  presents  it  on  an  oscilloscope  for  real
time  evaluation of  the  raw  data.   Auxiliary  information  such as telescope
position, date and time,  laser power, and photomultiplier voltage  are  recorded
along with the data.

     The  lidar system is  calibrated  through  the use of an  8 x 8 f^oot target
of  known  optical   properties.   These  properties  (reflectively,  uniformity,
polarization   properties)   are  measured in  the  laboratory  before  and  after
each field trip.

     2.   Acoustic  Sounder:    The  degree  of  pollutant trapping  is  strongly
influenced  by  the  vertical  temperature structure, particularly the  existence,
height  and  strength of a temperature inversion  layer.  The  vertical  tempera-
ture  structure was continuously monitored by  an acoustic sounder of  the type
described  by  Owens  (1975).   This  unit also operates  in a fashion  similar to
radar.  It  emits a 100 watt,  100 millisecond  sound  pulse every  10  seconds in
a  vertical  beam.  Variations  in temperature  scatter some  of  the  sound back
into a five  foot  acoustic collector, where it is detected and presented  on  a


                                      298

-------
                                     24 GHz Radar
Retractable
 Telescope
               Data Processing
                  Equipment
Control Van
                          Figure 8.1.  NOAA lidar system.

-------
facsimile chart.  This information  has  proved very  useful  in the location of
inversion  layers,   so  that  the  aircraft  could be  properly  positioned for
maximum  utility.    Quantitative  information  regarding  the  strength  of the
inversion was  not obtained.

     3.   Insolation:   An  independent  measure   of  the  total  aerosol  content
between  an  observer  and the  sun  is  through  the  use  of  several  insolation
(incoming  solar radiation)  devices.   The  method  is  to  measure  the  total
incident  radiation with an  Eppley Model  2 pyranometer  as  well  as direct
solar  radiation  at several  wavelengths.   The  sun's  radiation has been  shown
to  be  constant (within  1%)  over decades  of time,  so  that variations can be
attributed  to aerosols  in  the  solar  path.   The  effects  of optically thin
clouds can be eliminated through the use of the  lidar data, which  is  extremely
sensitive to  clouds,  being  capable  of detecting cirrus layers too thin  to be
seen visually.  The method  fails when  optically thick clouds block the sun.
Aerosol  size  distributions   can  be  estimated by comparing  the  data  obtained
at  various  wavelengths,  since  light  scattering  is  dependent on the  size-to-
wavelength  ratio.   These  measurements  were  made  with  an  Eppley  Model  8
normal   incidence   pyrheliometer  operating  alternately unfiltered,  and  with
Schott OG-1,  RG-2  and  RG-8 filters.  The pyrhel iometer was mounted on a sun-
tracker  along  with  Barnes  ITS  radiometer to  extend  the measurements  into the
8 to 12 urn region of the infrared.

     Turbidity measurements  were made  using a Voltz turbidity meter  operating
at  the  wavelengths  .502 and  .382  urn.   Measurements  were  made  three  times
daily with  some determinations  being  made  several times per hour to  test the
consistency of the  measurements.

RECENT CHANGES

     Following  the  first year's experience and  findings from monitoring the
pregenerating  plant  atmosphere  at Colstrip,  considerable  improvements and
additions  to   the   monitoring  capability  were  accomplished.   Of particular
note was the "scaling-up" of airborne instrumentation to meet  the  requirements
of  measuring  the   power  plant  plume   diffusion,   composition  and   chemical
reactions,  effect   on  atmospheric  optical  properties,  and  consequences  in
terms of possible precipitation attention.

     Eleven parameters were measured from a Cessna 206  single-engine  airplane.
These  are:    1)  Aitken  nuclei  with  the  continously  operating €nvironment,
Inc. particle  counter; 2)  Cloud condensation  nuclei  with  the  Allee  thermal
diffusion chamber-photographic method;  3)  Ice nuclei by means of  the  membrane
filter method;  4), 5),  and 6)  Total  oxides of nitrogen,  nitric oxide, and
nitrogen  dioxide concentrations  with the Monitor Labs, Inc.  chemiluminescent
instrument;   7) Ozone  with  the  Monitor  Labs,  Inc.   chemi1uminescent  ozone
monitor;  8) Sulfur  dioxide  with  the Theta  Sensors,  Inc. monitor;  9)  Tempera-
ture,  continuously recorded;  10)   Relative humidity,  continuously  recorded;
and  11)  Light  scattering with  the  Meteorology Research,  Inc.  continuously
operating Model  1550  nephelometer.   In  addition, membrane filter  collections
of atmospheric aerosols were done for subsequent  analysis by  Scanning Electron
Microscope  (SEM)  and  Transmission  Electron  Microscope  (TEM).   The SEM with
energy dispersive  x-ray analysis,  yields  the  elemental composition  of  indi-

                                      300

-------
vidual particles  of  diameter > 0.1 |jm,  and  for particle topography.  The TEM
is  utilized  for  the  aerosol size distribution, which  is  crucial  to calcula-
tions of possible atmospheric heating or cooling, and transmissivity.

     The  mobile  laboratory  was  operated  in  the  same location  as  for  the
previous year's  study.   Parameters measured were:  1)  Sulfur  dioxide with the
Meloy Laboratories, Inc. flame photometric sulfur gas analyzer; 2) Atmospheric
light  scattering with  the  Meteorology  Research,  Inc.  nephelometer;  3)  Ice
nuclei by  the  NCAR acoustic  counter method (continuous); 4) Ice nuclei by the
membrane filter  method;  5) temperature;  6) relative  humidity;  7)  wind speed;
and  8)  wind direction.  Membrane filters were  also  collected  according  to a
pre-determined  schedule,  for  examination  by   SEM  and TEM.   Pilot  balloon
releases and tracking were done once each hour for wind direction and velocity
aloft while the aircraft was  operating.

     The  overall measurement capabilities  are  summarized  in Tables  8.1  and
8.2.

                                  DISCUSSION
LIDAR DATA

     The  system was  on  site  for  30  days in the Spring (May  18  through  June
16)  and  the Fall  of 1975.   The  Spring  was an  unusually  wet  one,  with  some
rainfall  occurring nearly  every  day.   The Fall  was  more  typical—mostly hot
and  dry  with   occasional   showers.   The  power  plant  was  not  in  operation
during  either  period,  so  the data  gathered  will  be  used as  a  baseline for
future measurements.

     Figure  8.2  is  representative of the  lidar data  for  the  Spring period.
The  signal  begins at zero  as  the laser beam  is  initiated  outside  the tele-
scope's  field  of view.   As  the beam  comes  within  the  field  of   view  the
signal  increases until  all  of the  beam is totally  contained  and is said to
have  passed "crossover."  Data  recorded at heights  lower  than crossover are
not  valid,  since  not all  of the  laser  energy  is  used.  This  curve  indicates
a  nearly  uniform atmosphere with  an  aerosol  loading  approximately  15 times a
theoretically  clear  atmosphere.  The aircraft was  operating  900  meters above
the  surface when  these  lidar data  was  recorded and  the  optical  backscatter
coefficient  calculated  from the particles collected  in the membrane filter at
this altitude is also plotted.

     Figure  8.3  shows  a typical  acoustic sounder  record  for  the  1975 Spring
experiment.  A  weak  inversion structure  is seen  operating  from 0000 to about
0900 when  thermal activity  becomes the  dominant feature.  The dark bands from
1200  to  1400  are due to  the aircraft  operating  overhead.   Thermal activity
continues  until   2000 when the evening  temperature inversion reappears.   The
Spring  inversions,  if   present,  were  usually very  weak,  and would  not  have
played a significant role in  the trapping of pollutants.

     Figure  8.4  is representative of  the 1976 Fall  lidar data.  It is essenti-
ally the  same  as the Spring data except for a  sharp layer at a height of 500
meters due  to  aerosols  trapped by  a  weak inversion  layer.  The corresponding
acoustic sounder record (Figure 8.5) barely shows the temperature inversion.

                                     301

-------
    10
      -i)
    10
      -5
CO
°-    £
o in-b
    10
      -7
                                   COLSTRIP, MONTANA
                                 DAY 151  MAY 31, 1975
                -MEMBRANE FILTER
                                                                 LIDAR
                                                    THEORETICAL CLEAR AIR
          1000                      2000

                  HEIGHT (M)

Figure 8.2.   Optical  backscattei—Spring 1975.
3000

-------
     00
CO
O
                        0400
0800
   1200
Time, (hrs)
1600
                                   COLSTRIP, MONTANA
                             DAY 149           MAY 29, 1975
2000
0000
                         Figure 8.3. Acoustic sounder record—Spring 1975.

-------
    10'
                                    COLSTRIP, MONTANA
                                  DAY 241  AUG.  29,  1975
    10
      -5
 _
ct.
cc
                                  -MEMBRANE FILTER
                                                                   LIDAR
fe  lfl
      -6
                                                      THEORETICAL CLEAR AIR
    10
      -7
                              1000
2000
3000
                                        HEIGHT (M)
                Figure 8.4.   Optical backscatter--Fal1  1975.
                                       304

-------
GO
o
     CU3
     OJ
     o
            1200
                                            Time, (hrs)
1600
                                            Time, (hrs)
                                  COLSTRIP, MONTANA
                            DAY 241       AUGUST 29, 1975
                          Figure 8.5.  Acoustic sounder record—Fall 1975.

-------
In general, the  Fall  experiment generated much stronger inversion structures,
frequently containing  multiple layers at several  different  altitudes (Figure
8.6).

     During the  Spring experiment  the  EPA-Las Vegas  airborne  lidar overflew
the  site  allowing  the two  systems to  probe  the same  airspace at  the  same
time.   This allowed the transfer  of the calibration of  the  NOAA lidar to the
airborne system, which  then  flew  a series of  criss-crossed  patterns in a box
30 miles  on  a  side,  centered  about  Colstrip.   The airborne  lidar  data  have
not been received.

     The system  was on site  for   14  days in 1976.  The weather  was warm and
dry,  but No.  1  power  plant had been  shut down for maintenance.  However, No.
2  was  operating  intermittently during the last several  days.   Although there
has  been  insufficient  time   to  analyze  these  recent  data,  some interesting
features have  been noted.  The  1976  data were  taken  using  the dual-polari-
zation  detector,  to separate  the  stack  signals  from  ordinary  dust signals.
Figure  8.7 shows  a tracing of an  oscilloscope photograph of raw data in which
the  system  was simultaneously  observing smoke from the power  plant and  dust
from  the  mines.   The  return   from  the  smoke appears  only in  the  "parallel"
channel, which  means   its  plane of polarization  is  parallel to  the plane  of
polarization  of  the   laser   beam.   This  indicates  the  smoke  particles  are
spherical.    The  return  from   the  dust,  however,  appears  in  both  channels,
indicating  that  the dust  particles are  irregular.  This  is consistent  with
the  shape  observed in  an  electron microscope of  particles  obtained in other
areas.

IN SITU DATA

Cloud and Ice  Nuclei

     The portion  of the  aerosol  that  has  a direct bearing  on the  formation
and dissipation  of clouds  and precipitation are termed  cloud and ice nuclei,
i.e.,  centers around  which   water  droplets  or  ice crystals develop  at low
supersaturations.

     Before start-up  of the  Colstrip  generating  plant it was  found that the
cloud  condensation nuclei  (CCN)  concentrations  at  Colstrip   were a  large
fraction, 50  percent or even  more for  most  of  the time, of the total aerosol
population.    The  reason  for   this  is  the large  proportion  of  chlorine- and
sulfur-containing  particles  that  was  found  to exist  in the  Colstrip aerosol
(see Figures  8.8  and  8.9)._  Sulphur and chlorine  usually  are found  in nature
in their anionic  forms,  SO^  and Cl , that  form water  soluble substances, and
hence are efficient as CCN.

     Also prior  to operation  of  the  power generating plant,  ice nuclei  (IN)
concentrations were at  least  double those measured in  most clean air regions.
The  investigators  believe this is  a  consequence of  the high  incidence  of
heavy metals  in  the Colstrip  aerosol.   Zinc is present in many more  particles
there  than  in  aerosol  samples  from  most  other  locations  that  the authors
have   investigated.   Some  other  metallic  elements  that  occur  unexpectedly
often  are  titanium,  cadmium,  chromium  and  nickel.    Copper  was found to  be

                                     306

-------
CO
o
       0)
       00

       53
                                                Time, (hrs)
                               1600
    2000
Time, (hrs)
                                                                               1200
                                      COLSTRIP, MONTANA

                                DAY 231       AUGUST 19, 1975

            Figure 8.6.   Acoustic sounder record showing multiple inversion structure.

-------
OJ
o
CO
                       •  0
15
                                          RANGE  (KM)
       Figure 8.7.  Raw lidar return showing difference in depolarization between dust and stack effluent.

-------
                      COLSTRIP,  MONTANA
c 100
o>
E
UJ
® 80

c
« ^~
c
o 60
"c
o
0
jS 40
o

I.
o
Q_
- 20
o
__
o>
o
o> n






—




—



^^^_





__


iq^piq

:•§•:•:•:
•I'X'X'
£•:??:
•:|:j:jj:j

iis
•$•:•:•:•
'XvX*
II
'X*!*X*
1

xi$
x*x*x
•:•:•:•:•:•
igijgi
vX*X*

•x-ivi-
III


III
iSi'Si
1^
XvX*i
•x*x¥
11

*!*!*!
i-i'x''-''
ililil
'X*Xv
11
IP

X*X*X'
x*x*x
11
•x*x*S
:|:-:-S:
XvX'I
gig:
X'l'X*'

III
'X*Xv

1
'XyX1
H

I;X;";X
j-SjSj
SS-x
B
IP
'*•!":*;'*•
X;X;X
XwX
;XvX;
SSi?
•i'i'i'i-i*
vX;X;

X;X;X

;XvX;
•X'X'J1
-::::::::
X'X'M
1
•X*Xv
I
IVIMI UMU UUINt-, l^lsj


















>x*i*x-
•X*X;X
1
Xv??!
X"X'J
?S?:5
XvX;!

II
II

n
::::::::¥
^
||
x*x*x
SSS
III

ww
•g'-:^
•i-iwi
•X''*X'
%'**"'l'
•I'X'X'
xwS

i'C'I'X'!
•i'S'B
II?
•X'**X'
X'X'X
X;X;X
1

•x'x':'
::::::::::i
•:•:•:•:•:•
III
Si&i

sis
:•$•:•:•:
III
111
X;XgX
III


•i'i'i-i'i-
X'X'X
X'X'X
Sv?vi

li?
:•:•:•:•:•:
111
1
X'X'X
1








Hi
II
|xix:






Wf
!S!R?
•SHii;
:?:?S

Hi
•:•$•$:

III
y-x':-:
jXvX;
«







n P
•I;X*X*i %*Xv!
:-:-x*-:v ISSSS

555S-!
Ill
|g
|:|:|:gi

Bl
:i:i;x%
'X"I*X*i
;X;I;X;
IjXvX;
1!
x*x*x
11
           No   Al    P   Cl   Co   V   Ba   Cr   Zn
             Mg  Si    S    K   Ti   Fe   Cd  Ni
Figure 8.8. Percent of aerosol particles that contain the elements shown,
        averaged over May and June 1975.
                         309

-------
                      COLSTRIP,   MONTANA
                               26 AUG.,
r- inn
c IUU
E
Ld
_£ 80
o>
c
c
o 60
c
o
o



(/)
® 40
o
^-
^^
o
CL

M- 20
o
c
o>
o
k_
x*X'S
X'X'X*!
§iw!
•X'X'X
*.'.'.*.V.
•.•«•.*.*.•.
I'X'I'V"
i:||:S:
O
A
I

m
II

XvX'I
§wi§
X'X'X



X'X'X
x*x*x
wlvl;

|||

•I'l'I'X*

•:•:•:•§•
•:•:•:•:•!•
=1:1

nr
\*i •


:5:?S
•:§:$:•
•X'I'X'
|:|x$

•X'Xv


vivi'i*
x*x*x
i'l'X'X
wil?
w^v'
:iv'iv*i:

ft'X'X*
:•:*!*:•:*:
i*x¥x
>X*XV.
r* i
in
• ^M








::?:*:•:•;

$:*i*i:i*:
X'X'X1
•I'i'i'i*!*'
•:•!•:•!*:•
Ill

vX'X'l
II
ills
111
III
il
ill

i ;
• ^








Siwi

•x'lvi'i
X'X'X'
•I'i'i'i'i'!
5:::S:-i
•X'X'I*!
lil
•X'Xv!
:::::::::::;
•X'X'I'I
X'X'X'
•X'X'I'!
1
r»_
")
*^














!v!vP!
;:•;$:•:•;
'I'X'X*!
||
i-SSS

i
•
m

0
^^ 1



















X^IS
•
11
•
V /
r
•• •







!|JWB!
!•:•!•!•!•:•
vX'X'l

X'X'X*
•i'i'i'i'i'i
III
H
;?•:?•!§
'X'X'X
xw:=
'.%V.Vi
illlx

^S

. 3/ » L)
• • m • %^ • ^^

















p«
|:¥:j:|:j; Sfei-
|^ n ^

l~>_ /^ . "7 _
            Mg  Si    S    K   Ti    Fe   Cd   Ni
Figure 8.9. Percent of aerosol particles that contain the elements shown,
        averaged over the three days indicated.
                         310

-------
occasionally  present.   Titanium  dioxide and  copper sulfide,  in particular,
have  been  identified  as  active  ice  nucleating materials.   In  addition,  the
elements  aluminum  and silicon  have  been  identified  in almost  all   of  the
aerosols  that  were  investigated.   These  elements  are widespread   in  the
earth's  crust in  the  form of  the common  clay mineral,  Kaolinite, which  is
known, in the aerosol form, to be a moderately active ice nucleant.

     There  exists  one  other class of  atmospheric  freezing nuclei.   These  are
biogenically derived and efficient at relatively warm (T > -5°C) temperatures.
While  the  presence of  such nuclei  is  anticipated,  there is  no  means  in  the
investigators' present capability for their detection.

     Figures  8.10  and 8.11  are  summaries  of  aerosol  properties  found  during
the  1975  field  investigations  at Colstrip.  The light scattering coefficient,
b    .,  is   typical  of  rural  western  atmospheres;  this  is  discussed  further
bliow.  The  total  particle population  density  (AN, or  Aitken nuclei)  is also
typical of clean atmospheres.  The CCN and IN are discussed above.

     A  cursory  look  at   the  record  from  the airborne  operation was  done
following  the May-June  1976  field investigation  at Colstrip.   From this  it
is  estimated that  in  the  Colstrip power plant plume  at short distances from
the  stacks,  the  AN and CCN concentrations  and b    .  are  increased by factors
of 1.5 to 4.                                      scat

AEROSOL OPTICAL PROPERTIES

Visible-light Optical Effects

     Atmospheric  aerosols  affect the shortwave portion  of solar  radiation  in
two  ways:    (1)  They  can  reduce  the  amount  of solar  radiation  reaching  the
ground  by  scattering  a significant portion backward  into space,  resulting  in
surface  cooling.   (2)  Absorption  of  a significant portion of solar radiation,
on the  other hand, can lead to atmospheric warming.  Which of these processes
predominates  depends  on  the  size and  chemical  constitution of  the  aerosol
particles.

     Results from  the analyses of sized and elemental composition (determining
the  scattering  portion of  the  particle  refractive  index)  were  used in  a
computer  program   calculating  the   extinction,   scattering  and  absorption
properties  of the Colstrip aerosol.   From  the  result  it   can  be concluded
that  the  presently existing  aerosol   has  a  net  cooling effect  at  surface
albedos  less than 0.5, which is  the  case  for  the  largest portion  of  the
year.   Only during  the winter  months when  the  ground  is  snow-covered will
the situation be such that  a warming effect can  be anticipated.

     The calculated  scattering coefficient was verified with in situ measure-
ments using an integrating  nephelometer.

Infrared Optical Effects

     Aerosols absorb IR radiation in the water  vapor window, thereby affecting
the  accuracy of radiometric measurements of,  e.g.,  temperature  performed  in

                                     311

-------
co
ro
                bscat
               ICT4m
                       0

                       6
                 AN
               I03crrf3
                 CCN

                I03cm-3
                IN £
                     -i
                       2
                       •
                       0
                                             COLSTRIP, MONTANA
	I  I I  I I  I I I  I I  I I  I I  I I  I I  I I  I I I  I I  I I  I I  I
  15     20     25     30  1    4      9      14
  	MAY	—H	JUNE I975	
                           Figure 8.10.  Aerosol properties—Spring 1975.

-------
OJ
CO
                       21-
                                             COLSTRIP,  MONTANA
                 scat
                       0L
                                    I I I  I I  I 1 I  I I  I I
                 AN
               I03cm~3
                 CCN
                I03cm~3
                IN S
-i
                       0
                          15     20     25
                           	AUG.  —
                           I I  I I  I I I  I I  I I I  I I  I I I  I
                          30  1    4      9       14
                          —*4+	SEPT. 1975	
                            Figure 8.11.  Aerosol properties--Fal1  1975.

-------
the 8-12  pm band from  aircraft  and satellites.  At the  same  time they alter
the flux  of  terrestrial  radiation  into  space,  a  phenomenon  that  directly
affects the earth's  heat  budget.   The magnitude of  this  effect can be evalu-
ated by measuring with  a  radiation thermometer TJI situ at different altitudes
the brightness temperature  of  ground surfaces, which  are  in thermal  equilib-
rium with the atmosphere.   Varying the  optical  path  length  by  a  factor of
two,  the   investigators  were  unable to  detect  any  effect  of  the  Col strip
aerosol on the earth's equivalent black body temperature.

GASES

     The  background  concentrations   of  nitric  oxide,  nitrogen  dioxide  and
sulfur  dioxide outside  of the plume  are  below instrument detectability.  The
background concentration of  ozone is around 30 parts per billion by volume.

     Unit  No.  2  of the  Colstrip  power plant  was   in  intermittent operation
during  June  1976.    The   following  values  of gaseous  concentrations  were
observed  in  the  plume  shortly  after it  exited into  the  atmosphere:   NO  and
N02 were  both around 0.5  parts per million.   S02 amounted  to around one part
per million,  and the  ozone concentration  was reduced to  about  8 parts  per
bi11 ion.

SOLAR IRRADIANCE  DATA

     The  solar  observing instruments  were operated  essentially continuously
during  the  1975  Spring and Fall  campaigns and during the  recently completed
1976 Spring campaign.

     The period August 28-September 15,  1975 was chosen for detailed analysis.
This period was  characterized  by mostly clear and partly cloudy weather which
permitted  measurements  of  the aerosol-related attenuation  almost every day.
The solar  intensity  data  from the normal incidence pyrheliometer and the Volz
turbidity  meter   were  corrected  for  sun-earth  distance  and  normalized  to
obtain  the  logarithmic (decadic)  extinction coefficient with  correction  for
Rayleigh  scattering  and ozone  absorption.   Figure 8.12 shows hourly values of
pyrhelimometer decadic  extinction  for 09-17 hrs.   It  is seen that the extinc-
tion tends  to increase  during the morning  hours  and level  off or decrease in
the later  afternoon.  Figure 8.13  shows this  trend more clearly.

     This   is  probably due  to  dust  raised  by  mining  activity'which  would be
carried upward by thermal  updrafts and drift to the vicinity of the observing
station.   The  acoustic  sounder  records  (Figures  8.4 to  8.6)  show  thermal
plume   activity  beginning  most   mornings  and  lasting  until   near  sundown.
Another  interesting  feature  illustrated   by   Figure  8.12   is  that the  noon
extinction values tend  to remain nearly constant for  3 to 4 days then become
changeable  or  switch to  a  new  level.   The cross-hatched  areas  indicate the
stable  periods.   This  behavior  is  no doubt related to the  movement  of large
scale  weather patterns and further study of this feature is planned.

     Another way of  looking at the variability of  atmospheric attenuation is
by  means   of  Langley  plots  (Figure  8.14),  which   display  the  logarithm of
intensity   versus  the airmass.  The  latter  is a measure of  the  amount of air

                                      314

-------
                                           rot,STRIP, MONT-
   .15  --
   .12 --
   r
   y
   r.
   T
   V
   r
   T
   T
   .1
   H
    .06--
    .03- -
    0.0
       239
241
243
245
              247
                                         249
                                   251
                                                       253
                                                              255
                                                                     257
                                                               259
                                     PAY
Figure 8.12.  Hourly values of pyrhel iometer (0.3-2.8 urn) decadic extinction
               (09-17 hrs) for day  numbers 240-258  (Aug.  28-Sept.  15, 1975).
                                        315

-------
                           DFCAHTC rXTU'CTTON COLSTFIF, MONT-  1<575
     .15"
     .12
     .09
     .06
     .03
                                           4-
                         06
                                          12

                                        TIMF
                                              18
24
Figure 8.13.
Hourly values  of pyrhel iometer  (0.3-2.8 |jm) extinction  vs  time
 of day for day  numbers 240-258,  1975.
                                        316

-------
      .2
        240
          244
   z
   u
      .1
      .2
      .3
      .4
247
                           248
                                    252-
                                          251     243
                            5                   10

                              AIRMASS
                                           15
Figure 8.14.  Superposed pyrhel iometer (3-28 urn) Langley plots for afternoons
             during period Aug. 28-Sept. 15, 1975  (Day No.  240-258).
                                   317

-------
through which  the solar  radiation  must pass  relative to  what  it would pass
through  if  the  scan were  directly  overhead.   The airmass  is  approximately
equal  to sec6,  where  6  is  the angle  between  the   vertical  and  the  solar
direction,  viewed  from  the  observing  site.    Stable,   horizontally uniform
atmospheric conditions  would  result  in a straight line on a Langley  plot with
slope  related  to the amount  of vertical  attenuation.   Figure 8.14  therefore
illustrates both  the  day-to-day  and  the short-term variability  of the atmos-
pheric attenuation  due  to aerosols.   The relationship between decadic extinc-
tion  as   determined  by  the  narrow  band width  Volz   sun  photometer  and  the
normal  incidence pyrheliometer  is  shown  in  Figure   8.15.   The  minimum  and
maximum values  determined  for each day by each instrument are plotted.

     The  extinctions  obtained by  the two instruments correlate  very well  on
the  average.   This  supports the  use of Volz  turbidity  data, which are rou-
tinely collected at  some 50 stations in the United States.

     The  reduction  of the direct solar  irradiance, as measured  by the pyrhe-
1iometer  and  Volz sun  photometer  and  indicated  by Figure  8.15,  ranged from
about  7%  to  25%.  The total  irradiance  from the  entire  sky (i.e., the sum of
direct and diffuse  radiation  as  measured by the pyranometer) varies generally
over  a small   range  because  the scattered  diffuse radiation  compensates  for
the  reduction   in  the  direct  component.  In  fact,  under partly  cloudy con-
ditions,  with  certain types  of  clouds  near  the solar direction,  the sum  of
direct and  diffuse  radiation  can  at times  exceed the  levels expected  for a
clear day.

     However,  most  aerosols  absorb  as  well  as scatter  light  with the result
that  increased  aerosol  loading in  the atmosphere results in the diminution of
total  irradiation incident  at  ground level.   This reduction in total  irradia-
tion ranged from near zero to about  10% for the data of  Figure 8.15.

     The  turbidity  measured at  Colstrip in the  Fall  of  1975 (average about
0.07)  agrees  closely   with  turbidity  values  published  by  Flowers et  aj.
(1969)  for  Missoula, Montana,  and  Huron, South  Dakota, which  are, respec-
tively, 300-400  miles to  the  west and  to  the  east of Colstrip.   By  contrast,
turbidity  values published  by  Flowers  et  ajL   for  many  of  the  observing
stations  in  the eastern half  of  the  United  States and  for   Los   Angeles,
exceed 0.20, which  means  a diminution of the direct component of greater than
37%.   Under  the latter  conditons  a  reduction  of total   irradiation  by  15  to
20% does not  seem unlikely.

     Another  factor  which  should  be  considered  is  the possibility  that
introduction  of  additional  aerosols  may  provide  cloud  condensation nuclei
leading to the  formation of more clouds than would otherwise develop.

                                 CONCLUSIONS

     This section presents  a  summary of the conclusions  at this stage of the
observations.   The  principal  conclusions  of  this  study  will  come only after
observations are made while  the power  plants  are  burning coal.   During 1975
observations there  was  no plant  operation.    In  Spring  1976,  only  sporadic
operation of Plant No. 2 occurred in a gas-fired mode.

                                      318

-------
                                                RANGE(MIN.andMAX.) OF  DECADIC EXTINCTION

                                                09:00-16:00 MDT AT COLSTRIP, MONTANA
                                                           BY  VOLZ TURBIDITY METER
                                                              AT  A=0.5nm
                                                           OF WIDEBAND 0.3-2.8Mm
                                                              PYRHELIOMETER
                                                (RAYLEIGH SCATTERING LOSS AND OZONE
                                                 ABSORPTION SUBTRACTED IN BOTH CASES)
           .151
          .121
          .091
     O
     z
     H
     X
     U
     O
     5
     <
     o
     u
          .061
.031
             239
                       243
                                  247
                                            251
                                                       255
                                                        259
                                DAY NO.    1975
Figure 8.15.
     Minimum and maximum  values of decadic extinction  as  determined
     by pyrheliometer (dashed curve) and  by Volz sun photometer
     (solid  curve) for day numbers 240-258,  1975.
                                          319

-------
     Observations  in 1975  showed  a very pristine atmosphere.   Total  particle
counts  below  500  per  liter were  observed.   The  lidar  returns showed  very
little  aerosol  content  over the whole  range of the  lidar.   Occasional,  mild
short-lived intrusions of dust from  the mines were observed.

     The  acoustic  sounder  showed  that  many temperature  inversions  occur  at
altitudes  above 500  feet.   In  such cases there  would be a tendency  of  stack
emissions  (from  500 ft.  stacks)   to  remain near ground level.  Analysis  of
acoustic sounder records is  not complete.

     The  depolarization ratio of  the lidar return  is an  important,  although
not  completely unambiguous, characterization index  of the type  of  particles.
It  is  found  that  newly risen  dust  (from  wind  on  explosions)  has a  large
depolarization ratio while stack  emission has no discernible depolarization.

     Solar  radiation measurements  showed  marked differences due  to  aerosols.
It  may be  expected  that significant  changes  in  solar flux rates will  occur
as the stack emissions increase.
                      LIST OF ABBREVIATIONS AND SYMBOLS

 Lidar     --   JJ_ght detection and ranging
 IR        --   infrared
 SEM       --   scanning electron microscope
 TEM       --   transmission electron microscope
 CCN       --   cloud condensation nuclei
 IN        --   ice nuclei
 T         --   temperature
 AN        --   Aitken nuclei
                                 REFERENCES

Allee,  P.   A.   "A  description  of  the  ESSA-APCL  portable  thermal   diffusion
     cloud  chamber."   Proceedings  of  the  Second  International  Workshop on
     Condensation  and  Ice  Nuclei,  1974,  Dept.  of Atmos.  Sciences,  Colorado
     State University, Fort Collins, CO.  pp. 39-41.

Langer,  G. ,  Rosinski,   J. ,  and  Edwards,   C.  P.   "A  continuous  ice  nucleus
     counter."  J.  Appl.  Met.  6, pp.  114-125 (1967).

Bigg,  E.  K. ,  Mossop, S. C., Mead,  R.  T.,   and Thorndike,  N.  S. C.   "The  mea-
     surement  of  ice nucleus  concentrations by  means of 'Mi 11ipore.filters."
     J. Appl.  Met.  2, pp. 266-269 (1963).

Owens,  E.  J.   "NOAA  Mark  VII   Acoustic  Echo Sounder."   NOAA  Technical Memo-
     randum ERLWPL-12,  U.S. Dept. of Commerce, Boulder, CO (1975).

Flowers, E..  C. ,  McCormick,  R.  A., and  Kurfis,  K.  R.   "Atmospheric  turbidity
     over  the  United  States   1961-1966."   J.  Appl.   Met.   8,   pp.   955-962
     (1969).

                                     320

-------
FIELD AND LABORATORY
     EXPERIMENTS
       321

-------
                                   SECTION 9

                          ZONAL AIR POLLUTION SYSTEM:
                           DESIGN AND PERFORMANCE

                                     by

                           J.J.  Lee and R.A.  Lewis


                                INTRODUCTION

     A major objective of the  Montana Coal-fired Power  Plant  Project is the
generation of  a  protocol that will  improve  ability to  assess  the  impact of
energy generation  on  the  environment  prior  to  critical land  use  decisions
concerning the initiation of power plant construction.  The field experimental
component  of  this  project  is  expected  to  provide  data  essential  to  the
construction of  such  a  protocol.   In  particular,  the  investigators  hope to
establish  the  specificity,  thresholds,  and temporal  interrelations of  the
effects of sulfur dioxide on well-defined components of a grassland ecosystem.
Also, by  correlation  of  results  of the field experiments with those from the
field studies  near Colstrip, Montana,  and perhaps elsewhere,  they  hope to
evaluate the site-specificity of these responses. Finally, during a five-year
investigation the  investigators hope  to generate predictive models that will
link short-term low threshold effects to those  that  occur only after prolonged
low level  exposure.

     The  experiments   are  being   conducted   within  two  27-acre  grassland
exclosures  in  the  Custer National  Forest in  southeastern  Montana.   Sulfur
dioxide (S02) fumigation  of four one-acre plots within the first exclosure was
initiated in May  1975.   Fumigation  of  the second set of four plots within a
second exclosure  was started in April 1976.  Livestock are excluded to protect
them from injury and also to protect equipment from damage.

     The  experiments  are designed  to  test the  effects of  SO., upon biomass
dynamics  (plant,  arthropod,  and  small  mammals), plant  and  animal  community
structure, insect and  fungal  diseases of plants,  pollination systems, lichens,
and upon a number of physiological and biochemical functions.  Dominant plants
on  the   study  plots  are  western  wheatgrass  (Agropyron  smithi i),  prairie
junegrass (Koeleria cristata). and Sandberg bluegrass (Poa secunda).

     By  using  a  gas  delivery  system  developed   for  the project  (Zonal  Air
Pollution System--ZAPS),  the investigators  are  able  to maintain a selected
median concentration of sulfur dioxide on each plot during the entire growing
season (circa April  through October).  Continuous  monitoring  and  control of gas
concentrations insure maintenance of the desired levels.  Since the plots are
ecologically and  physiographically similar,  observed differences in effects
among plots can be attributed to S02 concentration.   Fumigations  are planned to

                                     322

-------
continue for at least four growing seasons for the first set of plots, and at
least three for the second set.

     A prototype system was designed and tested during September 1974.  Results
indicated the feasibility of this type of control (Lee, Lewis and Body, 1975,
1976b; Lee and Lewis, 1976a).  Predictions were made  regarding the performance
characteristics of the  system (see Table 9.4).  Analysis of S02 data for the
1975  season  confirmed  these  predictions  (which were  based  upon  modeling
solutions  and  tests  of  the  prototype) and provided  further  evidence of the
adequacy of the system design.

                             DESIGN REQUIREMENTS

     The  ultimate  design goal  was to  provide a system  for  the unambiguous
assessment  of  the  sulfur dioxide  impact  on  otherwise undisturbed grassland.
The designers approached this ideal by establishing realistic and appropriate
ecological and physical  design criteria.

     Disturbance  of  biota  and  of micro-climates  by the  system was  to  be
minimized.   Effects  on  incident  radiation,  prey  refuges,  ground  level
obstructions and pathways,  temperature,  humidity, wind and other features of
the  micro-habitats  were to  be   kept  as  small as  possible.   The area  to  be
stressed  had  to be  large,  on  the spatial  scales  of the populations  to  be
sampled, to  reduce  edge effects and to ensure adequate population and sample
sizes.   Furthermore,  rational  application  of the system  would require that
areas  chosen  for comparisons  (i.e.,  "treatment" and  "control"  plots)  be  as
nearly uniform as possible in habitat, edaphic, and terrain features.

     The  logarithms   of pollutant concentrations  were  to  follow  a  normal
distribution as occurs in polluted areas (HEW, 1970).  The  distribution was to
be  spatially  uniform or  nearly so,  at  least on  a  time  average  basis,  and
concentrations  were  to  be  controllable  for  a  range of  selected  averages.
Concentrations outside the plots and cross-fumigations among  the plots were to
be minimized.   Finally, cost, maintenance, and operation had to be reasonable.

                           THE GAS DELIVERY SYSTEM

     The   system  that   was   designed  and  placed  in  operation  for  this
investigation  generally  meets the above  criteria.   Each  gas  delivery system
consists of a network of one inch aluminum pipes (schedule 40, alloy 6061-T6)
set parallel to the ground and supported at twenty foot intervals by  five foot
pipes driven 2  1/2  feet into the  ground  (Figure 9.1).   Release points (1/32
inch horizontal holes) are situated at 10 foot intervals, so that no location
within a plot  is  more than  18  feet  from a source.  A continuous flow of air
through the lines is provided by a helical compressor (Becker model SV 80-1).
On all but one of each  set  of  four plots,  S02  is bled  into  the air stream.
Equipment is housed in a heated  shed with the  temperature kept high enough so
that the vapor  pressure  of the  S02  in  the  tanks is always above the desired
release  pressure.   Pressure regulators  on  the  tanks  and a  flow controller
(Brooks model  8944)  and  flowmeter (Brooks model 1110) at the injection point
assure the desired flow rate of S02. Air pressure in the lines  (about  one pound
per square inch, gauge (psig)) varies by less than 5% across the system.

                                      323

-------
                                          M

T ]
280'.
\-















^r\















- ' 1











-


;


                200'


  M  Central  Monitoring  Station
  D  Delivery  Stations
  G  6 KW Diesel  Generator
WOO Three Conductor 0/0
        Copper Wire
W2  Three Conductor 2
        Aluminum Wire
                                                     W 2
                                           WOO
                                           IIOO'
                                          C±1G
 Figure 9.1.   Schematic of the first set of Zonal Air Pollution Systems show-
              ing the individual delivery systems and the common monitoring
              and electrical systems.  CONTROL,  LOW,  MEDIUM and HIGH indicate
              relative S02 exposures on the experimental plots.

     The  only ground  level  obstructions within  the plots  are the  supports.
These should  cause minimum interference  with animal  movements, wTiile  the pipe
network should  negligibly  influence the  micro-climate.  The  relatively small
size of the plots precludes the study of  large animals, although insects, other
arthropods  and  small  rodents  are  included.  The  contiguity of plots  allowed
selection for nearly uniform habitat conditions.

     The  four plots of each  set are  located  along a line, with  intervening
buffer zones  to minimize interference  between  plots (Figure 9.2).  To  achieve
desired ambient concentrations during  1975 (Table 9.4), three of four plots of
the first set of plots received S02 at the constant rates of 1.5,  3.5, and  7.0
standard  cubic  feet  per  hour  (SCFH)  per plot.  This  implies that  the  S02
concentration in the delivery  lines is approximately  1/2   1% (volume) for  the
largest S02 rate;  it  is  correspondingly  lower for  plots  with lower  S02 flow
                                      324

-------
   70'
   70'
            a
             70'
                                                    28O'
                                     C    COMPRESSOR
                                     S    SULFUR DIOXIDE TANKS
                                     H    I KW HEATER
                                     V    VALVE
                                    	  I" A! PIPE, 1/32" HOLES
                                            EVERY  10*
                                    	I" Al PIPE, NO HOLES
                                     A   SAMPLERS
Figure 9.2.  Schematic of  a single Zonal Air Pollution treatment plot.
                                325

-------
rates.   Concentrations  on  the  plots  varied with meteorological  conditions.  The
fourth plot  received no  direct  input of  S02 but  did,  however, occasionally
receive some S02  due  to drift from other  plots.   Concentrations on this plot
exceed 5 pphm only 2% of the time.

     The authors'  application of a  Gaussian  dispersion  model (Turner, 1969)
indicated  that  even  under favorable  dissipation  conditions, concentrations
about 200  feet  from  a  40  pphm plot would remain below 5 pphm.  The effects on
surrounding areas were thus expected to be minimal, and significant  levels were
not expected to occur outside  the study exclosure.  The actual  impacted area was
somewhat larger  than expected, however,  primarily  due  to channeling  of  the
pollutant  by  terrain features, and  to a  greater sensitivity of  some  of  the
vegetation than  was expected.

     By utilizing many  small,  elevated, dilute point sources (over 250 per 1  1/4
acres), adequate dilution  of S02  at  ground level is insured and, in effect, an
area source is created.  This  prevents step-function changes in concentrations
in time and space ("hot spots"),  except at very short distances from a source.
Testing of the system,  as  described  below,  confirmed  expectations regarding the
nature of this distribution.

     Knowledge of the physical parameters  (pressure,  density, orifice diameter)
has permitted M. Shirazi of The Corvallis  Environmental  Research Laboratory to
model certain features of the  plume  near a release point.   The mixture of gases
is slightly heavier  than  the  ambient air at  the  point  of discharge.  For the
analysis of this type of jet,  it has been  found that an  important parameter is
the densimetric  Froude  number, defined by
                              F = U/VAp_ D g
                                     P

     This  dimensionless  number  represents  the  ratio  of inertia!  to bouyant
forces.   For the values  of  F greater than 40, the plume is neutrally bouyant;
i.e., the  S02  can  be  considered to  be a trace  constituent  whose density is
completely determined by the fate of its ambient  air parcel.  For the calcula-
tion of F, one needs the discharge velocity U, the jet diameter D,  the ratio of
excess  mass  density  Ap  to  the  ambient  density  p,  and the  gravitational
acceleration constant g.

     The parameters and  calculations used to determine F are  summarized in Table
9. la.  The calculated discharge  rate  per jet is  1.8  x 10~3 ftVsec  (about 27
ftVhr per plot),  the discharge velocity is  330  ft/sec,  and F is  larger than
10,000.   The S02 concentration  is  thus controlled by wind and turbulence, and
the S02  will  not simply  settle to earth and accumulate.  This  conclusion is, of
course,  substantiated  by the observation that S02 concentrations did  not build
up  during the growing season,  and did not ever approach the concentrations in
the lines.
                                      326

-------
    TABLE 9.la.   SUMMARY OF PARAMETERS AND CALCULATIONS USED TO ESTIMATE
                 DENSIMETRIC FROUDE NUMBER,
Mass Flow Through Jet:

  W = 223.8 APo (C  [(P_)2/k - (P_) (k+1)/k])1/2 ib/sec = 1.4 x 10-4 Ibs/sec
             R   =£   Po        Po
                 I o

Where      A = area of 1/32 in discharge jet, ft2

          Po = pressure in line, 15.7 psi

           P = pressure of ambient air, 14.7 psi

          To = absolute temperature, 540° R

          Cp = specific heat of air, constant p, .24 BTU/lb - °R

           k - Cp/Cv where Cv is specific heat at constant volume,  k = 1.4

           R = universal gas constant divided by molecular wt.  of air,
               53.3 ft-lb/lb - °R

Volume Flow Through Jet:

  V = WRTo = 1.8 x 10-3 ftVsec
       Po

Discharge Velocity:  U = V/A = 332 ft/sec

Excess Density of 1% (volume) Gas Mixture:
    pmix           (molecular wt) -x       _gg x 28_9? + _Q1  x
1.012
  pambient         (molecular wt)ambient               28'97

  Wp   .012
     The analyses of this type of problem indicate that jet dilution in still
air can be estimated by
                         0/0  = .32 X/D
                            o
where 0 is the total  entrained ambient air within the plume at a distance X/D
jet  diameters  away from  the  point of  discharge.   Examples  of  dilution for
several  distances  are given  in Table  9.1b.   The actual  observed  dilutions
(Table 9.8)  were at least 15 times greater than those predicted by this rough
model.
                                      327

-------
 TABLE 9.1b.  PREDICTED DILUTION FACTORS AT VARIOUS  DISTANCES  FROM DISCHARGE
              POINT.

          Distance  (ft)                            Dilution  Factor

            0.5                                           62
            1.0                                          123
            3.0                                          369
            5.0                                          615
                             SYSTEM PERFORMANCE

     A full size prototype of the system was constructed  and  employed  to test
the feasibility of this experimental approach to the field during October 1974.
Results of these tests (Lee, Lewis and  Body, 1975, 1976b; Lee and Lewis, 1976a)
demonstrated that concentrations across a plot are adequately controlled at the
low levels desired, but also suggested  an improved design  to  minimize  spatial
variation of concentrations.  This new  design (Figure 9.2) has been utilized in
all  experimental  gasing to  date.   Analysis of  data  for  one entire  growing
season indicates that the conclusions regarding the prototype were correct, and
that spatial variability is minimized by the new design.

     S02 concentrations  on  the plots were monitored  by recording the  output
from  a  Meloy model  SA160  sulfur analyzer  operating  in  a logarithmic mode.
Samples were drawn through Teflon tubing by a time-share  device  (Adgo  Co.) so
that each sample location was monitored at the rate of approximately 8  minutes
per hour.

     Data on  S02  concentrations were  obtained one foot above ground level at
central  location  c  (Figure  9.2) for  each  plot  for  the  entire  1975  growing
season.   Locations  a and b  (Figure 9.2)  were also  monitored  for prolonged
periods  for various plots.   Location  d was  sampled  occasionally; these data
have not yet been fully analyzed.  In  addition, specialized,  short-term tests
of S02 distribution have been performed.  These have been analyzed in terms of
temporal and  spatial  variations  of S02.  Comprehensive  characterization of
horizontal  and vertical  distributions will continue during 1977.

                          TEMPORAL VARIATION OF S02

     The frequency distributions of 8-minute medians (standard location c) are
shown in Figure 9.3, while Figures 9.4 and 9.5 show the frequency  distributions
for  1-hour  and  3-hour  averages  obtained  from  the  8-minute  medians  by
interpolation and  averaging.   All  the distributions  are approximately log-
normal,   typical  of  pollutant  frequency distributions  (HEW,  1970;   Larsen,
1969).   Such distributions  are characterized by  the  geometric mean (GM) and
standard geometric  deviation  (SGD).   These  parameters summarize the temporal
frequency of  occurrence of  concentrations,  and  thus  represent a measure of
dose.   Typical values  for  actual pollutant distributions  are given in Table
9.2.   Any area with similar GM and SGD  can be said to  have  received  a  similar
dose.


                                     328

-------
1 UU
E
f-
a.
CL
c
o
TD
E
c
E
CO
- 10

0
i—
a:
i-
z
LJ
O
z
o
o
CM
0
1
	 « 	 	 	
- D •
D •
- A a •

A a •
A D •

A D •
A D •


— O A O •
O A D •
O A D •
o A a •
O AD*
O A D ®

0 A D •

0 A 0 •


0 A D f>

- , , , o , A |D «
      0.01   O.I      I          10           50          90
                    CUMULATIVE  FRACTION  ABOVE X
99
Figure 9.3.     Cumulative frequency distributions of 8-min.  median S02 concen-
              trations, 1975 season,  o = plot A; A = plot B;  n = plot C; •=
              plot D.

                                  329

-------
1 UU
E
a.
CL
C7>
o
a>
o
_c
oes
^JTRATION ( 1
o
UJ
o
z
0
o

-------
  TABLE 9.2.   GEOMETRIC MEANS (GM, pphm) AND STANDARD GEOMETRIC DEVIATIONS
              (SGD) FOR SEVERAL URBAN AREAS.  Five Minute Averages 1962-67
              (source:   7)


          City                     GM                       SGD






Chicago 1
Cincinnati
Denver
Los Angeles
Philadelphia
St. Louis
San Francisco
Washington
0.4
1.6
1.3
1.3
5.5
2.8
0.5
3.9
2.2
2.9
2.1
2.3
2.4
2.8
2.9
2.2






The values of these parameters
for the entire growing season, and
lower concentrations in October were
for the
in Table
probably
four plots are given in Table
9.3b for individual months.
due to frequent rain.
9.3a
The
     Due to  the  logarithmic mode of the S02 monitor, all readings of zero on
the  strip  chart  are  recorded as concentrations  of  .9  pphm.   The minimum re-
cordable GM  is thus also .9 pphm.  This effect was only important for plot A,
which was below .9 pphm more than 75% of the time.

     In models of  pollutant dispersion source strength  appears  as  a normal-
ization  constant,  with  the patterns  of  distribution  determined  by meteor-
ological conditions (Turner, 1969).   The  resultant low  variability of SGD over
a range of GM's is evident for the urban areas summarized in  Table 9.2 and for
the  experimental  plots in Table 9.3.  The only dissimilar SGD is for the plot
which did not receive direct S02 input.  The similarity of the SGD's  in Tables
9.2  and  9.3  indicates  that the fine-scale  variability  of S02 concentrations
throughout the study plots is realistic and reproducible.

     Further evidence of  the  similarity  of variability  among study  plots is
given in Figure 9.6.   In this graph the S02 concentrations have all  been nor-
malized to have the same GM as the highest concentation plot,  plot D.  This was
done by multiplying the S02 concentrations for plot "i" by the ratio GM (plot
D):    GM  (plot  "i").   The coincidence  of  the   resulting  curves for  plots
receiving direct  S02  input shows that they differ in "amount" of S02 (GM) but
not  in S02 "variability" (SGD).  The curve for the fourth plot is affected by
the fact that the true GM is lower than the measured GM.   The  similarily of the
shape of this curve to the other curves  is obvious  in Figure 9.6.

     The authors  predicted S02 distribution parameters  and peaks prior to the
introduction of S02  (Lee,  Lewis and Body, 1976b).  These are compared to the
observed values in Table 9.4.   The seasonal 3-hour peaks are the most important
values from  the  standpoint  of the present Federal Secondary Standard for S02
(Table 9.5).   The differences  between medians and geometric means are measures
of deviation  from true log-normality.
                                      331

-------
       TABLE 9.3a.   PARAMETERS OF S02 FREQUENCY DISTRIBUTIONS FOR 1975
                    GROWING SEASON.
For 8-min
Plot Medians
GM
A 1.0
B 2.2
C 3.8
D 6.8
SGD
1.5
2.4
2.7
2.7
For
1-hr
Averages
GM
1.0
2.1
3.6
6.4
SGD
1.5
2.4
2.7
2.6
For 3-hr
Averages
GM
1.0
2.2
3.9
6.2
SGD
1.6
2.2
2.5
2.6

 GM:   Geometric mean,  pphm (minimum recordable is .9 pphm).
SGD:   Standard geometric deviation.

  TABLE 9.3b.   GEOMETRIC MEANS AND STANDARD DEVIATIONS OF 8-MIN. MEDIAN S02
               CONCENTRATIONS ON A MONTHLY BASIS.

Month
June



July



August



September



October



Plot
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
GM
(pphm)
1.18
2.54
4.40
7.52
0.96
2.43
4.47
7.59
1.02
2.31
3.54
6.38
1.02
2.32
4.42
8.37
0.97
1.43
2.32
4.02
SGD
1.50
2.05
2.40
2.75
1.38
2.29
2.49
2.86
1.62
2.44
2.56
2.52
1.59
2.61
2.89
2.69
1.45
2.17
2.63
2.46
                                     332

-------
        TABLE 9.4.   COMPARISON OF PREDICTED1 AND OBSERVED S02 PARAMETERS, 8-MIN.  AVERAGING TIME,
                    1975 GROWING SEASON.


Plot


Geometric
(pphm)

Mean

Predicted Observed
A
B
CO
co r
co L
D
0.0
2.0
5.0
10.0
1.0
2.2
3.8
6.8

Median SGD
(pphm)



Predicted Observed Predicted Observed
	 	 	 1
2.0 1.7 2.1 2.
5.0 3.4 2.1 2.
10.0 5.8 2.1 2.
5
4
7
7
Seasonal
3-hr. Peak
(pphm)



Predicted Observed
0
15-20
40-50
100-200
11
20
47
69

1 Predicted by Lee, Lewis, and Body, 1975.

-------
               TABLE 9.5.   MONTANA AND FEDERAL STANDARDS,  pphm.

Montana
Federal Primary
Federal Secondary
1-hr. 3-hr. 24-hr.
25a 10b
14C
c
50°

anot to be exceeded more than  once  per four days (approx.  1%)



 not to be exceeded more than  1% of time



cnot to be exceeded more than  once  per year



TABLE 9.6.   PEAKS OBSERVED  DURING 1975 SEASON,  VARIOUS AVERAGING TIMES,  pphm.

Plot
A
B
C
D
8-min
14
44
71
140
1-hr
11
32
54
89
3-hr
11
20
47
69
16-hr
3
13
24
33

TABLE 9.7.

PERCENT OF TIME
WERE EXCEEDED.
MONTANA (MT)

AND FEDERAL

(FED) STANDARD LEVELS


Plot
A
B
C
D
MT 1-hr
0
0.3a
3.8
10.2
MT 24-hr
0
1.2
27.5
59.3
Fed 3-hr
0
0
0
1.1
Fed 24-hr
0
0
10.0
34.6

 Not frequent enough  to  constitute  a  violation  of standard
                                     334

-------
              100
            a.
            a
            c
            o
            6
            c
            'i
            co
            -  10
            OL
            \-
            Z
            UJ
            <_>
            •z.
            o
            o
                 - O  A  D •
                 - O      Cb 9
                     O    Ok •
                       O    /X|]
                                  -MO
                                     4Q
                                   o
                                    o
                                          A»
                            1
                                    I
                                             1
0.01   O.I    I        10        50        90
          CUMULATIVE FRACTION ABOVE X
                                                              99
Figure 9.6.     Cumulative frequency distributions of 8-min.  median S02 concen-
               trations, normalized to same geometric mean,  1975 season.  o =
               plot A;  A = plot B; n =  plot C; • = plot D.

     The peaks observed during the season for various averaging times are pre-
sented in Table 9.6, and compared with relevant State and Federal Standards in
Table 9.7.  Sixteen-hour averages  were  used to approximate  24-hour averages.
The only plot on which the Federal Secondary Standard was exceeded was plot D,
while the Montana Standards were exceeded on all  plots receiving S02 input.

     The  diurnal  variation of  S02 for  a  typical  24-hour  period  during the
summer is shown  in Figure 9.7.   S02 concentrations  usually  peaked during the
night and were low during the day.  A different pattern was evident during late
summer and early autumn (Figure  9.8), with peaks occurring early in the morning
and  late  in  the  afternoon.   This pattern is similar to  that observed for the
prototype system (Lee,  Lewis and Body, 1975).

     The  variation of  the GM of  8-min. medians with  time  of day is shown in
Figures 9.9a-e for individual months and in Figure 9.9f for the entire season.
These graphs again show that concentrations were  generally  lowest at mid-day,
and that the pattern of S02 variation was  similar across the plots.
                                      335

-------
       40
co
OJ
cn
     Q.
     Q.
     C
     o
     O)
     E

     c

     E

    CO
    o
    or
    UJ
    o
    z
    o
    o
     
-------
     Diurnal  patterns  of  pollutant S02  concentration  from stationary point
sources and in urban  environments  are  frequently similar to that produced by
this system  (Holzworth,  1973;  Le  Quinio,  1973;  Raynor   et al_,  1974; Smith,
1968;  Saito and  Mizoguchi,  1973).   Such  variations  in  air pollutant concen-
trations are  due  to (a) variations  in source  strengths that in  turn may result
from daily cycles  in  human  activity  (Garnett,  1973; Holzworth,  1973);  (b)
variations in transport wind speeds and directions, atmospheric diffusion and
interactions  (Cormier,  1974;  Fukuoka,  1973;  Garnett,  1973; Holzworth, 1973;
Lomaya and Tsintsadze,  1974; Martin, 1974;  Smith,  1968).   All  of these vary
with weather and season (Balabuyev et a_l,  1973;  Druilhet and Fontan, 1973;
Fukuoka, 1973; Sandig  and  Sandig,  1973).   Atmospheric dilution is frequently
greatest during the  day and least  at  night.   This  may result in one or more
daytime minimal  and a nocturnal maximum in pollutant concentration (Holzworth,
1973).
                                       JUNE
           0
5.0        10.0        15.0       20.0       25.0

        TIME  OF  DAY  (hr)
Figure 9.9a.    Variation  of  geometric  means  of  S02 concentration with time of
               day for the months  June through  October and for the  entire 1975
               season.    o=  plot  A; A = plot  B;  o =  plot  C;  *  = plot D.
               Horizontal line  at  0.9  pphm represents the detection  threshold;
               all concentrations  at or below this level were assigned a value
               of 0.9 pphm for the  purpose of computing geometric  means.

                                     337

-------
                               JULY
 Q.
 Q.
<
UJ
o
o:
l-
UJ
S
O
LU
   10.0       15.0

TIME OF DAY (hr)



   Figure 9.9b.


      AUGUST
                                              20.0
25.0
                        TIME OF DAY (hr)
                                                         25.0

-------
                           SEPTEMBER
 E  20.0
 Q.
 Q.
UJ
o
ct:
h-
UJ
O
UJ
CD
    15.0 -
0
                 5.0
          10.0       15.0      20.0


       TIME OF DAY  (hr)



          Figure 9.9d.



            OCTOBER
5.0
                           10.0      15.0


                        TIME  OF DAY  (hr)

                           Figure 9.9e.


                              339
20.0
           25.0
25.0

-------
                                WHOLE  SEASON
                                                          20.0
25.0
                               TIME OF  DAY (hr)

                                 Figure 9.9f.

                          SPATIAL VARIATION OF S02

      Tests  on  the prototype system  showed that there were no  "hot-spots" of S02
 on  the  plot, and that S02 concentrations varied smoothly over the plot (Lee  et
 a]_. ,  1976b).    Investigations  on the  ZAPS  plots  during  1975 and  1976  have
 provided further insight into the nature of the spatial variation of S02.

      Numerous  short-duration  samples of S02  were  taken  near a  S02  release
 point.  An example of  such a series  is given in Table 9.8.  These indicated that
 within  three  feet concentrations   become  comparable  to   locations  halfway
 between the delivery  lines.  This conclusion was upheld by data obtained over
 five  days  during  June 1976 for locations approximately  six feet  and 18 feet
 from  a  release  point  (Table 9.9).  Concentrations near the relea.se point were
 typically 48% higher than those far  from the release point.

     Various points on the first set of four plots, and several  points  on the
 second set, showed signs of excessive concentrations in the  immediate vicinity
 of release points.   This was  caused  by  jets  being directed in a near-vertical,
 rather than horizontal, direction.   The data given  in Table  9.9 is for  a mis-
 directed  plume.  It  thus  demonstrates  the localness  of  the  effect,  and
 indicates  the maximum gradient to be expected.   The misdirected  jets will  be
 left  as they   are  since  the  affected  areas  are  not  extensive  enough  to
 invalidate  the  current experiments, and since the S02  gradients near these jets
provide unique  opportunities for research.
                                     340

-------
TABLE 9.8.   VARIABILITY OF S02 CONCENTRATION IN THE VICINITY OF A S02 RELEASE
            POINT NEAR LOCATION £.

Horizontal  Distance
   Downwind from              Vertical Distance           S02 Concentration
   Release Point             Below Release Point           (10-min. median)
       (cm)                         (cm)                       (ppnm)
0
0
25
25
25
25
50
100
200
300
30
15
65
30
15
0
0
0
0
0
4.3
18.2
3.1
3.1
185.0
5.5
4.2
14.6
5.1
6.1

Notes:
      1.  June 11, 1976.
      2.  Winds light and variable, 1-2 mph.
      3.  S02 concentration at standard location c was approximately 5 pphm.
      4.  Measurements are for consecutive  10 min. periods separated by 10 min.
         periods during which the S02 monitor was zeroed with a S02 filter.

      TABLE 9.9.   COMPARISON OF S02 PARAMETERS 2 METERS AND 6.1 METERS FROM A
                 DELIVERY LINE.   June 12-16, 1976.

          Distance
          From Line           G.M.           Median          S.G.D.
          (meters)           (pphm)          (pphm)

             2.0              16.7            13.0            2.6

             6.1              11.4             8.8            2.5
                                     341

-------
     A comparison of  S02  concentrations at locations a, b, and c (Figure 9.2)
for plot B is given in Table 9.10.  These are based on data obtained  during most
of the 1975 season, and illustrate within-plot variability.  The GM's agree to
within ±  5%,  and  the  SGD's to within + 10%.  The standard location (c) has an
intermediate  GM,  but  exhibits somewhat more  variability (highest SGD).  This
location has  the  lowest  1-hr peak, an intermediate 3-hr peak, and the highest
16-hr peak.   The  S02  concentration at the standard location is typical of S02
concentrations across  the  plot.

           TABLE 9.10.  COMPARISON OF S02 CONCENTRATIONS ACROSS PLOT B.

Sample
Location

a
b
c
G.M.
(8-min medians)
(pphm)
2.0
2.2
2.1
S.G.D.
(8-min medians)

2.0
2.2
2.4
1-hr
Peak
(pphm)
43
33
32
3-hr
Peak
(pphm)
26
18
20
16-hr
Peak
(pphm)
8
8
13

                           SUMMARY AND CONCLUSIONS

     The exposure system  behaved  in a predictable and realistic manner during
the 1975 season.   Different S02 distributions were maintained on the four plots,
and these  had  the desired relationships to the Federal Secondary Standard for
S02.  Maintenance was  routine and  not excessively time consuming.  The system is
a  useful  and practical tool  for  determining the effects  of S02 on naturally
growing ecosystems.

     The behavior of the system is similar in several respects to a system that
was developed  independently  by  a  French research team (de Cormis,  Bonte1 and
Tisne, 1975).  In the  French  experiments,  trees planted in a 2,000 m2 area have
been fumigated through 128 release points with S02 at an average concentration
of  3.4 pphm.  The  control  consists  of  a  similar  area  without  a pollution
delivery system located 100 m from the fumigation plot.  The French system uses
a S02 flow rate of approximately .85 SCFH, or about 80% that of  the ZAPS after
allowance  for   differences   in   plot  areas  and  S02  concentrations.   The
concentration  varies  with meteorological  conditions;  high S02  concentrations
are,  however,  prevented  by   discontinuing fumigation when  the-concentration
reaches a pre-determined value.  They conclude that "The discontinuous nature of
pollution  related to  the  effect  of climatic conditions  is  not a handicap.  On
the contrary, this phenomenon is quite similar to those encountered in certain
polluted sites  which have provided the basis for this study."

                               ACKNOWLEDGEMENTS

     The assistance of  Denis  Body in the  construction,  Ted Fletcher and Eric
Preston  in the  operation and  on-line  testing  of  the  system,  and  of Brian
Satterfield in  data gathering  is gratefully acknowledged.
                                      342

-------
                                  REFERENCES

Balabuyev, A.  G. ,  0.  V.  Lomaya, and D. G. Tsintsadze.  1973.   Annual  and Diurnal
     Course of  the Concentration  of Atmospheric  Solid  Aerosols Over  a City
     (Godovoy  I   sutochnyy   Khod   Kontsentratsii  Atmosfernykh  Aerozoley  V
     Gorodkikh Usloviyakh).  Soobshch. Akad. Nauk Gruz.  SSR,  69(3):585-589.

Cormier,  R. V.  1974.  The Nature and Variability of Integrated Boundary Layer
     Winds.  Preprint, American  Meteorological  Society,  p.   244-249 (Presented
     at the  Conference  on Weather Forecasting and Analysis,  5th,  St.  Louis,
     March 4-7,  1974).

de Cormis, L. , J.  Bonte1,  and  A.  Tisne.  1975.  Technique experimental permet-
     tant  1'etude  de  1'incidence   sur  la  vegetation  d'une pollution  par le
     dioxyde de soufie applique en permanence et a dose subnecrotique.  Pollut.
     Atmos.  17(16):103-107.

Druilhet,  A.  and  J.  Fontan.   1973.  Determination of the  Vertical Diffusion
     Coefficients  Between 0  and 100 m by  Using Radon  and THB.  (Determination
     des  Coefficients de Diffusion Verticale entre 0 et 100 Male Aide do Rudun
     et due THB) Boundary-layer Meteorol.  3:468-498.

Fukuoka,  Y.  1973.   Meteorological  Study  of Air Pollution:   (1)  The  General and
     Specific Cycles  for  Air  Pollution.   (Fukushima daigaku kyoikugakubu rika
     hokoku).  Sci. Rep. Fac.   Educ.  Fukushima Univ.   23:51-64.

Garnett,  A.   1973.  Emissions,  Air Pollution and the Atmospheric Environment.
     J. Inst.  Fuel  46:39-45.

Health, Education  and Welfare,  Department  of.  1970.  Air Quality Criteria for
     Sulfur Oxides, U.S. Govt.  Printing Office.  178pp.

Holzworth, G. C.   1973.  Variations of Meteorology, Pollutant Emissions, and Air
     Quality.  American Chemical  Society,  American Inst.   of  Aeronautics  and
     Astronautics,   American   Meteorological    Society,    U.S.    Dept.   of
     Transportation, Environmental  Protection Agency,  Inst. of Electrical and
     Electronic Engineers, Instrument Society of America, National  Aeronautics
     and  Space  Administration,  and National  Oceanographic  and  Atmospheric
     Administration,  2nd  Joint  Conf.  Sensing  Environ.   Pollut.,  Washington,
     D.C.  , p.  247-255.

Larsen, R.  I.  1969.  A new  mathematical  model of air pollutant  concentration
     averaging time and frequency.  J. APCA.  19:24-30.

Lee, J. J., and  R.  A.  Lewis.   1976a.  Field  experimental component:  the bio-
     environmental effects of sulfur dioxide.  Iji R. A. Lewis and A. S. Lefohn,
     ed., The Bioenvironmental Impact of a Coal-Fired Power Plant,  First Interim
     report, EPA-600/3-76-002.

Lee, J. J., R. A.   Lewis,  and  D. E.  Body.  1975.  A field experimental system for
     the evaluation of the bioenvironmental effects  of sulfur dioxide.   In W. S.
                                      343

-------
     Clark,  ed. ,  The Fort  Union Coal  Symposium,  MT.  Academy  of  Science,  E.
     Montana College, Billings.

	.  1976b.  The Field  experimental  component:   evaluation of the
     Zonal Air Pollution System.  I_n. R. A. Lewis,  N.  R.  Glass,  and A. S. Lefohn,
     ed. ,  The Bioenvironmental  Impact  of  a  Coal-Fired  Power Plant,  Second
     Interim Report, EPA-600/3-76-013.

Lomaya, 0. V. and D.  G. Tsintsadze.  1974.  Analysis of the Diurnal Course of Air
     Pollution.   (Analize  sutochnogo  khoda  zagryazneiya Vozdukha).   Soobsch.
     Akad. Nauk.  Gruz SSR.  73:329-332.

Martin, D.  E.   1974.  Some  Air  Pollution  Climatologies of  the St.  Louis Urban
     Complex.    Preprint,   American   Meteorological   Society,  p.   180-182.
     Presented at the conference on Weather  Forecasting and  Analysis,  5th,  St.
     Louis, MO, March 4-7, 1974.

Le Quinio, R.  1973.  Concentrations sur une  Heure  de pollutants a des emissions
     ponctuelles  pres   du  sol-presentation  probabiliste.    Atmos.   Environ.
     7(4):423-428.

Raynor, G.  S. ,  M.  E.  Smith, and  I.  A.  Singer.   1974.  Temporal  and Spatial
     Variation  in Sulfur Dioxide Concentrations  on Suburban  Long  Island,  New
     York.  J.  Air Poll.  Contr. Assoc.  24:586-590.

Smith, M.  E.  1968.   The Influence of Atmospheric Dispersion on the Exposure of
     Plants to Airborne  Pollutants.  Phytopath.  58:1018-1088.

Saito,  K.  and T.  Mizoguchi.   1973.   Measurement of  Sulfur Oxides  and  Floating
     Dust  by the Air Contamination  Automatic Measuring  Recorder.   (Taikiosen
     jido sokutei kirokukei ni yoru  10  sankabutsu oyobi  Fuyu bijin no sokutei  ni
     tsuite ken eisei nura kenkyusho nenpo.  6:145-147.

Sandig, R.  and R.  Sandig.  1973.  (Zur Edv-gestutzten Bestimmung  der Schwef-
     eldioxid-immissionen an einer Stationaeren Messstelle  im  Stadtzentrum von
     Zwickau).  Z. Ges. Hyg. Grenzgebite (Berlin).   19:890-896.

Turner, D. B.  1969.  Workbook of Atmospheric Dispersion Estimates.  U.S. Dept.
     of Health, Education and Welfare.  84 pp.
                                      344

-------
                                  SECTION 10

       FIRST-YEAR EFFECTS OF CONTROLLED SULFUR DIOXIDE FUMIGATION ON A
                         MIXED-GRASS PRAIRIE ECOSYSTEM

                                     by

      J.  L.  Dodd, W.  K.  Lauenroth,  R. K.  Heitschmidt  and  J. W.  Leetham


                                 INTRODUCTION

     At the beginning  of  the Col strip  Coal-fired Power  Plant  project  it was
concluded that sulfur dioxide (S02) was probably the  single most important gas
of the several effluents expected from  the power plant.  A field experiment was
designed to  expose  four  0.5 ha grassland  study sites  to  different ambient
concentrations of sulfur dioxide on a continuous basis  throughout the growing
season.  The objectives of the experiment were to determine the  effects of 2, 5
and 10 pphm monthly median sulfur dioxide concentrations  on key structural and
functional  attributes of the grassland.  These concentrations were selected to
bracket  Federal  and   State  of  Montana  primary and  secondary  air  quality
standards.

     The fumigation experiment  is  located on a  grassland  site  in  the  Custer
National  Forest 85 km  southeast of Colstrip.  At this distance  it is presumed
to be  out  of  the influence  of the Colstrip plume.  The site is located at an
elevation  of  about  1,200  m  on  the  divide between the  Powder River and the
Tongue River drainage basins.

     Soils  of the study site are Farland silt loam.   These are  well developed
soils and on the site, have an A-horizon about 25 cm thick  over a B-horizon that
is  about 70  cm  thick.   The climate  in  this  part of  the  Great  Plains  is
continental with about 75% of the precipitation occurring during the growing
season.  The annual rainfall at the  study  site is about 350-400 mm.

     The vegetation  of the site is typical of the northern mixed-grass prairie
in that it  is  dominated by cool  season  grasses,  mainly  western  wheatgrass and
junegrass,  and has  smaller quantities of other grasses  and a variety of forbs
and half-shrubs.

     Data  on  various  abiotic variables  including air  and  soil  temperatures,
relative  humidity,  wind velocity,  soil water content,  and precipitation were
collected for  most  of the  growing season.  Growing conditions in the early part
of the 1975 season  were very favorable with May and June each  receiving over
100 mm of precipitation.  This is probably well above normal  for  the  area.  Soil
                                     345

-------
water was near field capacity until about the first of July and then decreased
rapidly and remained low throughout the remainder of the season.

                                   PHENOLOGY

     Previous  investigations  concerning the effects  of S02 upon the primary
producer  component  of  an  ecosystem  have virtually  ignored  reproductive
processes.  The possible effects of air pollutants on pollination, flower set,
seed development, and seed dispersal should not be overlooked. Any changes in
the  time  of  occurrence  of  these  events  as  a  result of exposure  to  air
pollutants may significantly alter  the plant's reproductive capabilities.

     The objective of the  phenology study was to identify any major differences
between treatments in the phenological development of the major plant species.

     Fourteen  species were observed on a weekly time  interval throughout the
growing  season (Table 10.1).   As  can be seen, a  majority studied  were cool
season species and, for a considerable portion of these  species, phenological
development occurred prior to initiation of the continuous  fumigations  in late
May.

    TABLE 10.1.  SCIENTIFIC AND COMMON NAMES OF THE 14 MAJOR  PLANT SPECIES
                 SELECTED FOR PHENOLOGICAL STUDY


     Scientific Name                                    Common Name

Cool season grasses
     Agropyron smithii                            Western wheatgrass
     Bromus japonicus                             Japanese  brome
     Koeleria cristata                            Prairie junegrass
     Poa secunda                                  Kentucky  bluegrass
     Stipa comata                                 Needle-and-thread
Warm season grasses

     Aristida  longiseta                           Red-three-awn
     Bouteloua gracilis                           Blue  gramma
Warm season forbs

     Antennaria rosea                             Rose  pussytoes
     Psoralea argophylla                          Silverleaf  scurf pea
Cool season forbs

     Achi1 lea mi 11ifolium                         Western yarrow
     Sphaeralcea coccinea                         Scarlet  globe mallow
     Taraxacum officinale                         Common dandelion
     Tragopogon dubius                            Yellow salsify
Half shrubs

     Artemisia frigida                            Fringed  sagewort
                                      346

-------
     A  14-stage  phenological  classification  (Table  10.4  in Section  I)  was
utilized to describe  the  observed phenological status of the selected species
within each treatment on each observation date.  The recorded status reflected
the mean of all plants of the species within a treatment.

     Statistical    analyses   indicated  no   significant   differences  between
treatments in the  time of occurrence of major phenological events.  However, a
definite trend was noted indicating an advancement in phenological development
with increasing  S02  concentration.  This  trend was evidenced by  10  of the 14
species observed  (Appendix).

     An  example  of a species  displaying  this trend is  false salsify,  a cool
season  forb  (Figure  10.1).  The  low  and  high treatments reached  the mature
floral bud stage,  stage  9,  one week prior to the control and medium treatment
while all three  treatments  reached the open  flower  stage,  stage 9, two weeks
prior to  the  control.   In  addition,  the medium  and  high concentrations began
seed dispersal approximately one week prior to the control and low treatments.
This trend was reflected in all groups except the half-shrubs in which fringed
sagewort was  the  only  species observed.

     Although  it  was  concluded  from  statistical   analyses that  no  major
differences occurred  in  the timing of phenological events  as a result of S02
fumigation,  the  trend  toward  advancement  with increasing  S02  concentration
suggests  minor  differences may  have  occurred.  Since observations  were only
conducted approximately every 6 to 7 days, treatment differences of only 2 to 3
days  would  probably  not  have  been revealed  except on  every third  or fourth
observation date.  Furthermore,  since  continuous fumigation was not initiated
until after a considerable portion of the phenological progression of the cool
season  species   had  occurred,   a  more  dramatic effect  might be  expected  if
fumigation were initiated at the beginning of the growing  season.

                                VISIBLE  INJURY

     Visible  injury  to  plant  tissue  resulting  from S02  exposure  has been the
subject  of  numerous  studies  since the early  1930's  (Hill  and  Thomas, 1933;
Thomas  and  Hill,  1935;  Nat.  Res. Council Con.,  1939).   The earliest studies
primarily focused  on describing characteristic  patterns  of  visible  injury  as
exemplified by  different species  and the effect  of this  injury  on yield  in
various agronomic  crops (Briesley £t al.,  1950, 1959).  More  recently, emphasis
has been  placed  on determining the "threshold concentration" inducing visible
injury (Thomas, 1956;  Daines, 1968).  These "threshold concentrations" have been
found to vary with changes in concentration, duration of exposure, and abiotic
variables such  as temperature,  relative  humidity, and  light (Daines,  1968).
However, a large  majority of these studies have been conducted under controlled
laboratory  conditions  and  normally  evaluated  only acute effect of  high
concentration  short-term  exposures.   Field-observed  chronic visible  injury
induced  by  low  concentration  long-term  exposures  has  not previously  been
reported for  native range species.

     The objective of the  following  research was to identify and describe any
visible injury incurred by western wheatgrass as a result  of S02 fumigation.
                                      347

-------
CO
-p*
co
           14-,
          icH
           e'
o
o
        o   6-

        ID
        C.
        CL
          0
                  EJ Control

                  tD Low

                  [HI Medium

                  m High
                                               False  Salsify
               26 May   3Jun    9 Jun   ISJun   22Jun   3OJun   7Jul    ISJul    21 Jul   29Jul    4 Aug
         Figure 10.1.  Phenoloqical  development of false salsify  (Traqopogon dubius) for Taylor Creek,  1975.

-------
     Western wheatgrass was selected for this facet of the study since it is the
dominant  species within  the  fumigation  plots  and because  its morphological
characteristics  are  such  that field collection,  mounting,  and visible  injury
determinations are simplified.

     Four plants were collected outside  the perimeter of each harvest  quadrat on
the  June,  July, August,  and  September  sample  dates.   After  the  quadrat was
randomly located, the four closest western wheatgrass plants to the  corners of
the  quadrat  were clipped  at ground  level.  Thus, 20 plants were collected per
replication and 40 plants per treatment.

     After clipping, the leaves were removed  and mounted in order of age between
two  strips of clear adhesive acetate.  Each leaf's total surface area and total
injured  area was  then recorded  from  visual determinations  using  a   guided
overlay.   Visible   injury  was  defined  to   include  any portion  of  the leaf
exhibiting  necrotic  condition, including  normal  leaf  senescence.   The total
surface of missing  leaves was considered injured.

     An increase in the degree of leaf senescence has previously been reported
as the only visible injury detectable in perennial ryegrass  plants when exposed
throughout  a growing season  to  low concentrations  of S02  (Bell  and Clough,
1973).  Examination of the leaf surfaces of our sampled plants  also revealed no
distinguishable  pattern of  visible injury   other  than an  increase  in leaf
senescence.  Consequently,  statistical analyses were run in an attempt to  detect
differences between the percent of leaf surface necrosis in the control  plants
and  the  fumigated  plants.   Visible  injury,  resulting from exposure  to S02, as
used here,  is  defined  as  the necrotic portion of the  leaf surface  exceeding
normal senescence as  reflected by comparison of the fumigated plants with the
control  plants.  Leaf  injury  means for  western  wheatgrass are  presented  in
Figures  10.2 (by  data treatment  and  leaf   age)  and  10.3  (by leaf age  and
treatment  date).  The  Analysis  of  Variance  (ANOVA)  utilized  to  test  the
differences  between  treatments across  the four  dates  indicated a significant
difference  between  the LOW concentration and the HIGH concentration with no
differences between  all other means (Table  10.2).  As  expected, a significant
difference (P < 0.05) in total injury between the four dates was also  found, but
this is a reflection of the normal  aging  process.  Since the last sample date was
very late in the growing season,  it was felt  that normal fall  senescence might
be masking  the  effects  of the S02.   Therefore,  an  ANOVA  was  run encompassing
data only  from  the  first  three  sample  dates.   This  analysis   indicated
significant treatment differences between the  CONTROL and the LOW treatment when
compared to  the HIGH  treatment and between  the  LOW  and the MEDIUM  treatments
(Table 10.3).   It is  felt  that the LOW treatment exhibited a  lower  percent of
leaf  necrosis  than  the control  because  of  a significantly  higher  number of
leaves per  sampled  plant  on the  sample date  of 6 August.  Thus, with a  larger
proportion of the total  leaf  surface composed of younger tissue, a  lower mean
percent  of injury  was  reflected  by each plant.  It  has been  reported that
middle-aged  leaves  are  more  susceptible  to  S02  damage than  either older or
younger leaves  (Guderian  and  Van  Haut,  1968).  In an attempt  to examine this
phenomenon, an ANOVA of only the first four leaves was undertaken.
                                      349

-------
to
on
O




"7 FL
f O

>»
L.
3
50-
+-
c
Q)
O
k_
Q)
Q.
25-
o

















'-^ •"
">. _1
""-•>•>
jhl~
--"-_ "-
iXI "








1








nr^
4 4
* *.'
« •
*








?si



- *
~,

^~
,±-_
- * •
— **.-
*•*•
~_.».
-•*-








1
^
^
1
§








1-r
' ,
• a
«
V
< »








*s
' *• a
^3
'^ii

;.--•
6---,
* -*
\^

C
J.
- ,J* "
-=£-•"
\ *
~* _




^
^
^
1
1








» c.
A
• 4
»
b t.
r «,
V ^
t* *
>•
«
» «
rV
«








^
^
^
1
&










. : -
. "="".
_^-
-\ -
^
.-•
^-
^"*
^
^
f
^
^
^
^
P



•o
4
A
» «
fr-*
« ^
4
*
»
r, •
»
•»
» j
» >
A
A
* »
V
» *
»
.%*
» »
!» r
!' »
4 A
Ifr C
» 1
7 A
14 '
<%
*
V
4 «



b
r.>
e^
fe
^,-•1
S!
fe
«
^
**•£
^
I^<
Sb
(JG
^
V^
1
Su
^
^
S
te
=:*- Leaf 1

^ Leaf 2

v- Leaf 3

S Leaf 4


    Figure  10.2.  Seasonal  change  in percentage of  necrotic leaf surface for first, second, third, and fourth

                  leaves  of western wheatgrass, Taylor Creek, 1975.

-------
                      lOO-i
                       75-
                    •E  50-
CO
CJ1
                    c
                    0)
                    O
                       25-
                        0
                                  Leaf I
                              Leaf 2
Leaf 3
Leaf 4
    Figure 10.3.
Effect of S02  treatments on leaf injury (percentage of  leaf  surface necrotic averaged
across three dates—June, July, and August) by leaf age,  Taylor Creek, 1975.

-------
   TABLE 10.2.   TREATMENT MEANS FOR PERCENT LEAF INJURY SUSTAINED BY WESTERN
                WHEATGRASS OVER FOUR SAMPLE DATES.   Means underscored by
                same line are not significantly different (P < 0.05).

LOW
38.91

CONTROL
42.14

MEDIUM
44.16

HIGH
48.61
  TABLE 10.3.   TREATMENT MEANS FOR PERCENT LEAF INJURY SUSTAINED BY WESTERN
               WHEATGRASS OVER THREE SAMPLES DATES.   Means underscored by
               same line not significantly different (P < 0.05)

LOW
26.43

CONTROL
29.74


MEDIUM
34.63

HIGH
38.09
     Both the 3-date  and  the  4-date ANOVA indicated a significant difference
between  all  leaves  in  the percent  of the  total  leaf surface  necrotic.   A
significant increase  in leaf  necrosis  occurred with increasing leaf age,  but
this  is  as expected  since  normal   leaf  senescence increases  with  leaf age.
However,  an  important difference  in  the two  analyses was  noted  in  that no
significant treatment by leaf interactions was found to be significant in  the
3-date  analyses.    Tukey's  Q  values  (P  =  .05)   indicated  no  significant
differences between the percent injury exhibited by the four treatments' first
leaves, but a significant  difference occurred between the CONTROL and the  two
HIGH concentrations for the second, third, and fourth leaves.  The significance
of  this  interaction cannot  be overemphasized because  it  clearly establishes
that  an  increase   in  leaf  senescence  did occur  as a  result  of the  S02
fumigation.

     Analyses of the number of leaves lost per plant, the  age  of the lost leaf,
and the age  of  the plant  losing the leaf indicated no treatment differences.
In general, the oldest leaves of the plants were lost first in  all treatments.

     In  summarizing the  above  findings,  three  conclusions  have been drawn
concerning the effects of the S02 fumigation  treatments  on foliar injury:

     1.   Chronic  visible  injury for western wheatgrass is primarily expressed
          by  an  increase in leaf senescence without  specific pattern.

     2.   A significant increase in visible injury  occurred with  increasing S02
          concentrations.

     3.   A significantly  greater  amount  of injury was incurred by the older
          leaves than  by the younger leaves.
                                     352

-------
     It  should  be  noted  that  the visible  injury conclusions  are  based  on
statistically  significant  differences between the CONTROL and  the MEDIUM  and
HIGH treatments  and no differences were found to occur between the CONTROL  and
the LOW treatment.

     The  significance of  these  conclusions,  as  they  relate  to  a  range  of
ecosystems,  remains  to  be seen.  Guderian and Schoenbeck (1971)  attempted  to
distinguish  between the two synonyms  "damage" and  "injury" as they pertain  to
plant responses  resulting from the action  of air pollutants.   They suggest that
"injury" should  be considered to  include  all plant responses  to  air pollutants
while "damage" should only refer  to those responses which significantly alter
the  desired  use of  the  plant.   After  one  season  of  data  collections   and
analyses, the  investigators feel  the  injury has been detected as  reflected  by
the increase in leaf senescence  with increasing S02 concentration.  However,  it
remains to be seen whether or not damage has occurred.

             PLANT BIOMASS DYNAMICS AND NET PRIMARY  PRODUCTIVITY

     The objective of this part of the study was to characterize  the effects  of
three  levels of sulfur  dioxide  fumigation  on  above-  and  belowground plant
biomass dynamics and net primary productivity.

     Aboveground biomass was sampled by the harvest method and separated in the
field by species and  into three categories:  live, recent dead, and old dead.
Live and recent dead represent growth  during the current growing  season  and old
dead is biomass  remaining from previous growing seasons.  On each of six sample
dates, ten randomly located, circular  0.5 m2 quadrats were  harvested from each
treatment.

     Total  biomass accumulation by all species was similar with  no significant
differences  among treatments at any time during the season (Figure 10.4).  The
average growth rate  from 15 April to 15 July was 1.4 g/m2/day.

     Fumigation was initiated on 1 June and had no measurable  effect on  biomass
accumulation.  The point at which we measured  a change in the rate of growth was
6 August.  This decrease in growth rate appears to be related to a rapid  decline
in  soil  water between  15  July and  6  August  rather than a  response  to  the
fumigation treatments.

     The two most important functional  groups contributing to  total biomass are
cool season  grasses  and  cool  season  forbs.  Biomass  accumulation by  both  of
these groups was more variable among  treatments  than  total  growth,  but  no
significant  differences  or  explainable  trends  were evident  (Figure  10.5).
Average growth rate  of  cool  season grasses  from  15  April  to 15 July was  1.1
g/m2/day.   At the time that  fumigation was begun, approximately one-third  of
the cool  season  grass  growth  had occurred.   Cool   season   forb  growth  was
greatest for the HIGH concentration treatment and lowest for  the LOW treatment
with the control  falling approximately  midway between  the two.

     Stepping down to a finer level of resolution one can see the  responses  of
the two most important  species in the cool season grass group.  The first  of
these  is  western wheatgrass (Figure 10.6).  Western wheatgrass is  an important

                                      353

-------
          150 -,
GO
in
          IOO -
        E
       \
        o>
 (O
 to
 O
 e
 o
00
           50 -
            0 -
                                                                                               25.E.
	 Control
	Low
—  — - Medium
	High
                    Apr
                          May     '    Jun      '     Jul      '     Aug     '    Sept'
    Figure 10.4.    Seasonal change in current year1s production for all species combined,  Taylor Creek,  1975.

-------
         150 -i
CO
CJ1
en
          100 -
CO
o

o
GO
          50 -
                                                          Cool Season Grass
            	  Control
            	Low
            — —  —  Medium
            	High
                                                          Cool Season  Forb
                                                                                              2S.E.
                                                                                              2S.E.
                   Apr
                         May
I     Aug
Sept
     Figure 10.5.   Seasonal changes  in current year's production for cool  season grasses and  forbs, Taylor
                  Creek, 1975.

-------
GJ
cn
CT)
            IOO  -i
        CM
         o>
10
(/>
CJ

I
             80  -
             60 -
             20  -
              0
                                                        Western  Wheatgrass  Growth-—
               	  Control
               	Low
               	Medium
               	  High
                                                                                             25-E.
                                                western  Wheatgrass Dead
                                                                             ^^—- I2SE
                  2SE
                                   May
                                        Jun
Jul
Aug
Sept
    Figure 10.6.  Seasonal  changes in current year's production and  previous year's dead for western wheat-
                 grass,  Taylor Creek, 1975.

-------
dominant  over much  of the  northern  Great Plains  and is  probably  the most
important species  to  the  livestock  economy  of  the  region.  The growth rate of
western wheatgrass averaged  0.6  g/m2/day  from  15 April  to  15  July and 0.5
g/m2/day for the entire period of measurement.

     While these data do  not demonstrate treatment effects, they do  represent a
solid  characterization of  the  dynamics of growth  and disappearance to dead
material  during  the  1975  growing  season.  Additionally the May sample of live
and recent dead  biomass and the April  and May samples of old dead  indicate high
degree  homogeneity  of western  wheatgrass biomass across  treatments.   The
importance  of  this  homogeneity  relates  first  to  the  question  of whether
significant effects of fumigation on the growth of western wheatgrass occurred
during  1975  that  were  offset  by  initial  differences  in populations  among
treatments.    These data  indicate that this was  not the  case.  Secondly,  the
homogeneity of  western wheatgrass biomass  during the first year of treatment
will strengthen  our conclusions about differences should they occur during the
second or third years  of treatment.

     Prairie  junegrass is  the  second  important  cool   season  grass on  the
experimental  plots  (Figure   10.7).   Several  of  the differences  in prairie
junegrass biomass  accumulation among treatments are significant although these
differences have not been explained in terms  of treatment effects.  The data
for old dead  biomass on the  first sample date  indicate  inherent variability in
prairie junegrass  populations among treatments and at this time differences in
prairie junegrass biomass  accumulation are attributed to this variability.

     Belowground biomass was  sampled to a  depth of 20 cm by the harvest method.
Included  in  these samples  were crowns,  rhizomes, and  roots.    Crowns  and
rhizomes were separated by hand and represent primarily live tissue while root
biomass includes all  belowground  organic material greater than 0.5 mm2.

     Total  root and  rhizome biomass  were  not significantly different among
treatments  on any of  the sample dates  (Figure 10.8).   The  most perplexing
aspect  of these data are  the  two different  trends.   All of the fumigated
treatments represent  one  trend  while the control exhibits a  different trend.
There  is  evidence  from the  literature that exposure to ozone may  reduce root
reserves.   The   action of ozone  and  S02  may  be  similar  (Tingey,  1974)  and
therefore explain  these   trends,  but  at this  point conclusions  based  on  one
year's data cannot be made.  Crown biomass was  similar  among treatments  and
essentially constant at approximately 60 g/m2.

     Total roots and rhizomes were separated into two depth increments of roots
and rhizomes  alone (Figure 10.9).  It is evident that the major differences in
belowground biomass occurred  in the 0-10 cm  layer and did not occur  until after
fumigation was initiated.   Trends  in root biomass were similar  among treatments
between the  April  and May  sample  dates  and   different  for all  other dates.
Again,  these  differences  cannot be  satisfactorily explained with  a  single
year1s data.

     Root biomass  in the  10-20 cm  layer was essentially  constant throughout the
growing season  at  approximately  100 g/m2.   Rhizomes  also exhibited only minor
fluctuations and averaged  25  g/m2.

                                     357

-------
CO
en
CO
           75 -,
          5O -
      CM
O
E
o
CD
          25 ~
           0  -
                    	  Control
                    	Low
                    — — —  Medium
                    	  High
                                            Prairie  Junegrass  Growth
                                            Prairie  Junegrass  Dead
                                                                                              J2S.E.
               2S.EJ
                                         1      Jun
                                                     Jul
"I    Aug     I    Sept
    Figure  10.7.  Seasonal  change  in current year's production and previous year's dead  for prairie June-
                 grass, Taylor Creek, 1975.

-------
          9OO
 S!

^  70O

 D

CM
 E

 en
       O

       O
       CD
       en
       o
      -d
       O)
      I
 c
 6
 s_

 i1
 _o
 CD
 CD
          5OO -
          200 -
                                                                                Total Roots  Rhizomes
                                                                                 2 S.E.
                                                                                 	 Control
                                                                                 	Low
                                                                                 — — — - Medium
                                                                                 — •	 High
                                                                              I
                                                                           Crowns

                                                                           2 S.E.
                   Apr
                        May
                                        Jun
                                                   Jul
Aug
Sept
Figure  10.8.  Seasonal  changes  in biomass  of crowns  and roots plus  rhizomes  (0.020 cm),  Taylor Creek,
               1975.

-------
                  aoo ~i
CO
er>
o
               o>
               o>
               I
                  6OO -]
               o
               o
              bo

               8.
              .o 40O
               \_
               CD
              TD
               C
               •
               o
               0}
               CD
                                                                                          	  Control
                                                                                          	Low
                                                                                          	  Medium
                                                                                          	  High
                                                                                         Roots  0-IOcm
                                                                                         2S.E.
                 2OO
                    O
^— "= -^T-J =-_
Apr 1 M
i.-=.-^" —--="•"
ay ' Jun

1 Jul
=^^_- -. 	 • •— —
1 Aug

1 Sept
I
1
                                                                                          Roots  IO-2Ocm
                                                                                          2S.E.

                                                                                          Rhizomes

                                                                                       I  25.E.
               Figure  10.9.  Seasonal change in biomass  of roots  at 0-10  cm  and 10-20-cm depths.

-------
     Since  the  results  of the  belowground  biomass  sampling  are  still  in
question  estimates  of   belowground  net  accumulation  have  not  been  made.
Aboveground net primary productivity was calculated by summing peak live plus
recent dead biomass  by  functional  groups (Figure 10.10).  At first glance it
appears that primary productivity increased  with  increasing S02 concentration.
More careful observation reveals  that the higher values for the MEDIUM and HIGH
treatments  were  the  result  of  composition  differences.   Both  of  these
treatments  had  measurable  contributions  of  half shrubs  while they were only
sampled in  trace  amounts for the CONTROL and  LOW  treatments.  Comparison of
aboveground productivity excluding  half shrubs  shows   no  major differences
among treatments.

     In  summarizing  these  results  it  is   emphasized   that no  significant
differences in seasonal  biomass dynamics or net primary productivity were found
as  a result  of one  year  of sulfur dioxide  fumigation.  There  are  several
reasons why this was not an  unexpected result.  First,  fumigation was not begun
until approximately one-third of the total  plant growth  had occurred.  Second,
the  major  primary producers  on  the  experimental  plots   are  grasses  and as a
group  they  are generally considered to  be  less sensitive to S02  than  other
plants  (Hill  et  a!.  ,  1974).   Finally  since  the  sample methods  only  yield
interpretable data on  the more abundant species, many significant changes could
have  occurred  in  populations  of species comprising  the  forb group  that the
investigators  could  not  detect.  In  reference to this last point,  sample
techniques  have been modified for the 1976 growing season and a greater degree
of precision for these data is anticipated.

                               LITTER DYNAMICS

     As shown  in  Table  10.4 the influence  of  season  and treatment on litter
standing crop estimates are unclear.   It appears that  litter  increased between
April  and  May  and again  between  July  and  August.   Differences  in  litter
standing crop resulting from the  S02 stress are not  consistent and are probably
not statistically significant.

                              SOIL RESPIRATION

     Soil  respiration was estimated utilizing four  samples  per replication
with  two  replications   per  treatment  on  four  sample  dates (Table  10.5).
Difficulties  encountered in the  laboratory processing  of  samples prevented
early  season  sampling.   The  higher  rates encountered on  15  August at Taylor
Creek and 18 August at Colstrip are attributed to rainfall events immediately
preceding the sampling period.
                                     361

-------
                                       Aboveground  Net Primary  Productivity
             20O-
oo
01
no
e
\
o»


co

O
E
o
im
             !OO-
                Cool Season Grasses


                Cool Season Forbs


                Warm Season Grasses


                Warm Season  Forbs


                Half- Shrubs
                         Control
                                     Low
Medium
                                                       Treatment
High
    Figure  10.10.
         Aboveground net primary productivity for Taylor Creek study sites,  1975.
         by functional groups represent peaks of current year's production.)
                          (Contributions

-------
TABLE 10.4.   INTRASEASONAL DYNAMICS OF LITTER STANDING CROP (x ± SE,  Oven-dry  Ash-free  g/m2)  FOR  TAYLOR
             CREEK SITES (A = CONTROL, B = LOW,  C = MEDIUM, D = HIGH).

Treatment
co A
CO
B
C
D
Average
April 20
112
112
135
39

± 6
± 21
± 16
+ 7
100
May
102 ±
174 ±
168 ±
179 +
156
5
16
21
24
23

June 1
154 ±
137 ±
178 ±
124 +
151
5
8
14
17
12

July 13
123 ± 5
155 ± 17
155 ± 18
154 ± 16
147
August
172 ±
192 ±
193 ±
189 ±
187
13
17
15
17
22

September 18
145
149
193
100

± 20
± 15
± 17
+ 33
175
Season
Average
138
153
170
148

-------
       TABLE 10 5   SOIL RESPIRATION TREATMENT MEANS (mg C02/m2/day) AND
                    STANDARD ERROR OF ESTIMATE FOR THE FOUR TAYLOR CREEK
                    TREATMENTS ON FOUR DATES, 1975 (A - CONTROL, B = LOW,
                    C = MEDIUM, D = HIGH).
Date
2 August
8 August
15 August-7'
27 August

5.
4.
6.
5.
A
12 ±
38 ±
92 ±
97 ±

.28
.19
.20
.56

5.
4.
6.
4.
B
43 ±
46 ±
43 ±
72 ±

.21
.16
.12
.20

5.
4.
6.
4.
C
72 ±
00 ±
69 ±
01 ±

.23
.38
.29
.34
D
5.79 ±
4.10 ±
6.81 ±
4.18 ±

.50
.31
.26
.52

- Mean of one replication.

     Although final analysis  is  not complete the  27  August sample at Taylor
Creek indicates a significant  difference exists between the control  (A) and the
S02  treatments.   If   this  difference  is  found  to  be  real,  two  obvious
explanations  exist:    (1)  reduction  in  microbial  activity  because of  S02
treatment; and  (2)  a  reduction in root respiration because of S02 treatment.
More intensive sampling is scheduled for 1976 in anticipation of ascertaining
the seasonal  dynamics  of all sites and treatment  differences at Taylor Creek.

                          LITTER BAG DECOMPOSITION

     The  rate  of decomposition  of native  litter  was investigated utilizing
approximately 5  g  of  dead western wheatgrass, Agropyron smithii, placed in a
mesh  bag  on the  soil surface and retrieved at various time throughout the
growing season.   The means of the percentage of organic matter decomposed from
10 samples per replicate (two  replicates per treatment) are presented  in Figure
10.11.

     Although statistical  analysis  has not been completed, two trends  are noted
from  the  data.   As expected,  the maximum  rate of  decomposition  within all
treatments  occurred  in   the   early  part  of  the   growing season.   This  is
attributed to the higher moisture  conditions and, thus,  favorabje for decomp-
osition.  The second  trend noted is the higher amount  of decomposition within
the  control  when  compared to the  S02  treatments.   This  trend  suggests  a
reduction in microbial  activity  may have occurred because  of the S02 fumiga-
tion.  However,  since the  differences  between treatments  are essentially the
same on all  dates it  is questionable whether the differences can be attributed
to S02 fumigation.  Since  the fumigations were not initiated until  1  June, the
differences  found on  10 June  may be attributable to  site differences.  If one
hypothesizes that the decomposers  are very sensitive  to  S02 these differences
may  have  developed  from  the S02 fumigation.  This hypothesis is supported by
the minimal  difference noted between the S02 treatments by the 9 August sample
date.   The  significance of  replication  of  the above  investigation in 1976
cannot be  overemphasized in light of the 1975 results.
                                      364

-------
6-n
           CD
           CD
           in
           D
           CD
           c
           o
           o
           w_
           CD
          Q-
                 30 r
20
   Figure 10.11.
 10
                                                 	  Control
                                                 	Low

                                                 	Medium

                                                 	High
                   15
                     Apr
              May
                           10
Jun
Jul
Aug
Sep
                                           19
 Decomposition of Agropyron  smithii dead shoot material in nylon mesh litter  bags for
 Taylor Creek study sites,  1975.[Letter bags were  placed in field 15 April  and subsets
 were  retrieved on 10 June,  9  July, 8 August, and  19 September.)

-------
                 ARTHROPOD POPULATION AND BIOMASS DYNAMICS

     Above- and belowground  arthropods  were collected on  six dates in 1975.
Harvest and  extraction techniques  were the same as  those described for the
Colstrip site monitoring  program.   The  arthropod data have large statistical
errors resulting from sampling limitations.  Since these data have not yet been
subjected  to  analyses  of variance,  the  inferences  made  in  the   following
discussion are tentative.

     Total arthropod  numbers  minus  ants,  which have  been  removed because of
erratic  population  estimates  arising  from  clustered  distribution,  were
variable throughout the period sample (Figure 10.12).  The LOW treatment had
the fewest number  of  arthropods and the CONTROL  and  HIGH treatments had the
largest numbers.  No discernible trends related to  fumigation are evident.

     Examination of the species data indicated that thrips and mealybugs were
the most  variable  components.   Figure 10.13 presents data without  thrips or
mealybugs.  Again  the  data  were quite variable before  fumigation,  but post-
fumigation samples  showed little variability among treatments  and  a general
decreasing trend for all treatments.

     Trends in biomass were similar for the fumigated treatments with maximum
biomass occurring on the June sample data (Figure 10.14).   In contrast to this,
peak biomass for the CONTROL treatment was observed in August.  Because of the
large  variability  associated with  arthropod  biomass, the  investigators  are
hesitant to attribute  this  difference to treatment effects based on only one
year's data.

     Table  10.6  shows  a  summary  of all  aboveground arthropods by feeding
strategy.   Of the 251 species encountered,  50% were herbivores  including tissue
and sap feeders, nectivores, and granviores; 25% were  predators and parasites;
and  10% were  scavengers  or  omnivores.  The  remaining species  were  either
nonfeeding pupae or species with unidentified feeding habits.
                                     366

-------
        500
     	Control
     	Low
     	Medium
            High
          April
                                                    Sept.
Figure  10.12.
Seasonal abundance of total aboveground  arthropods for  Taylor
Creek,  1975.

                     367

-------
      500 r
      400
      300
   O)
   n

   E
   33
   a   200
   0)
        100
                       Control

                       Low
                       Medium

                       High
                          c.
                          o

                          o
                          0>

                          e
                          3
          0
          April
       May
June
July
Aug.
Sept.
Figure 10.13.
Seasonal abundance of  total aboveground arthropods excluding

ants, Taylor  Creek, 1975.
                                 368

-------
  CM
   CP
   £
   CD
  •o
   o
   CL
   o
   C/5
   IT)
   o
   E
   o

  m
      0
     Control
     Low

     Medium
     High
        April
    May
June
July
Aug.
Sept.
Figure  10.14.
Seasonal  changes in aboveground arthropod  biomass (excluding
ants),  Taylor Creek,  1975.
                                   369

-------
     TABLE 10.6.   TROPHIC DISTRIBUTION OF ARTHROPODS, TAYLOR CREEK, 1975,

Arthropods
Tissue feeders
Sap feeders
Predators
Scavengers
Parasites
Pollen and nectar
Omnivores
Seed feeders
Others
Total


No. of
Species
63
52
47
20
18
13
9
1
27
250


% of
Total
25.10
20.72
18.73
7.97
7.17
5.18
3.58
0.40
10.75
100.00

     Belowground  arthropods were  divided  into  two  size  classes  requiring
different  sampling  techniques.  Macroarthropods  included  all   soil  insects
greater than 1 mm as well as spiders, centipedes, and millipedes,  Microarth-
ropods included the mites and wingless insects less than  1  mm.  The following
discussion  is  confined  to the macroarthropod group.  Analysis  of the micro-
arthropod data is  incomplete and does not  merit discussion  at this time.

     Total  macroarthropods  based on  time-weighted  means  for  the four post-
fumigation sample dates indicated high variability of both biomass and numbers
among treatments.  There were,  however,  a few groups that  inhibited apparent
treatment response.

     The  Coleoptera  (Figure 10.15) showed  substantially  reduced biomass and
numbers in the high  treatment.  Grasshopper eggs (Figure  10.15) were reduced in
numbers and biomass  on all of the fumigated treatments,  most likely the result
of avoidance by ovipositing females.

     Predators (Figure  10.16)  were  the  only trophic group  to  show  a strong
treatment  response  and   this  was  particularly  evident  in the  predaceous
Carabidae  (Figure  10.16)  with  a three-fold difference  in  both  biomass and
numbers between the CONTROL and HIGH concentration treatment.
                                     370

-------
100-
*l
E
x
Numbers
01
D 0
i i
P"1


n


•
•MB ',







c


                                          Coleoptero (Total)  05On
"' O.5O-
N
E
«x^
3
0.25-

o
E
0
m
n-








^







PV
ff1







T























CO
               30n
20 H
            «n
            I  10-
                 0-
                         Grasshopper  Eggs
                           BCD
                          Treatments
                                        N

                                         I
                                         O>
                                         in
                                         tn
                                         o
                                         E
                                         o
                                                              0.02H
                                                O.OH
                                                    A    B     C     D
                                                        Treatments
    Figure  10.15.
Numbers  and biomass of selected soil macroarthropod groups  (total Coleoptera and grass-
hopper eggs) for Taylor Creek, 1975.  (A =  CONTROL, B = LOW,  C = MEDIUM, D = HIGH.)

-------
                                                  Prtdotors
r°-
Numbers
3_ 5


"
I







j

J

rri
:*_

— — »
                                                           CM

                                                            E
                                                           \
                                                           3

                                                            v>
                                                            to
                                                            a
                                                            E
                                                            o
                                                           £D
                                             o.zoH
                                             Q.iCH
                                                                       A
                                                          B
CO
-^1
IN3
                
                                         0)
                                         v>
                                         a

                                         o

                                         m
                                            0.20 H
o.io-l
                                                                            B
    Figure 10.16.
 w   A    B    C     D                            A

          Treatments


Numbers and  biomass of selected soil  macroarthropod  groups  (predators and carabid
beetles) for Taylor Creek, 1975.   (A  = CONTROL,  B  =  LOW, C  = MEDIUM, D = HIGH)

-------
                                   SUMMARY

     In summary, continuous fumigation of a mixed prairie grassland during the
growing season  has  resulted in only subtle  effects  on primary producers and
arthropod  consumers.   Although  the  northern  mixed-grass  prairie may  be
particularly resistant to sulfur dioxide fumigation,  there are several reasons
why substantial changes were not recorded:  the most important of these is that
fumigation was  not  begun until 1 June which was approximately 2 months after
the beginning  of  the growing  season.  These  experiments are being  continued
this year and fumigation was started on 10 April.  These data will help explain
the 1975 results.

                                 REFERENCES

Bell, J. N.  B.  and W. S. Slough.  1973.  Depression of yield in ryegrass exposed
     to sulphur dioxide.  Nature.  241:47-49.

Brisley, H.  R.  and  W.  W. Jones.   1950.  S02 fumigation  of wheat with special
     reference to its effects on yield.   PI.  Phys.  25:666-681.

Brisley, H. R.  , C. R. Davis, and J. A.  Booth.  1959.   S02 fumigation of cotton
     with special  reference to  its effect on yield.  Agron. J.  51:77-80.

Daines, R. H.   1968.  S02:  Plant response.  J. Occ.  Med.  10(9):516-524. (Also:
     1969 Air  Quality  Monographs,  Monograph  #69-8.   16 p.  Am. Petrol. Inst. ,
     NY).

Guderian, R. and  H.  Van  Haut.   1968.   Detection of S02    Effects upon plants.
     Staub-Reinhalt. Luft.   30(1):22-35.

Guderian,  R.  and H.  Schoenbeck.   1971.   Recent  results for recognition and
     monitoring of air  pollution with the aid of plants.

Hill,  G.  R.  and M.  D.  Thomas.   1933.   Influence of leaf  destruction by sulfur
     dioxide and clipping on yield of alfalfa.  PI.  Physiol.  8:223-245.

Hill, A. C. , S. Hill, C. Lamb, and T. W.  Barrett.  1974.  Sensitivity of native
     desert vegetation  to  S02  and to  S02  and N02  combined.   APCA  J.  24:153-
     157.

National  Research  Council  of  Canada.  1939.   Effect  of  sulfur   dioxide on
     vegetation.  Ottowa, Canada.  447  p.

Thomas,  M.  D.   and  G.  R. Hill.  1935.   Absorption of  S02 by alfalfa and its
     relation to leaf injury.   P. Physiol.  10:291-307.

Thomas, M.  1956.   The  invisible injury theory of plant damage.  J.  Air Pollut.
     Contr. Assoc.  5:205-208.

Tingey,  D.  T.   1974.  Ozone induced alterations  in  the metabolite  pools and
     enzyme activities of plants.  In:  Mack Dugger (ed.) Air Pollution Effects
     on Plant Growth.

                                     373

-------
CO
                                                                            APPENDIX

                                                 PHENOLOGY OF MAJOR GRASS SPECIES, TAYLOR CREEK STUDY SITES, 1975
                                                          (A = CONTROL, B = LOW, C = MEDIUM, D = HIGH)
Date
27 May
3 Jun
10 Jun
1 6 Jun
22 Jun
30 Jun
7 Jul
13 Jul
21 Jul
30 Jul
4 Aug
11 Aug
18 Aug
26 Aug
2 Sep
10 Sep
15 Sep
Poa Stipa
secunda comata

ABCD ABCD
6666 5
8888 667
8888 7777
9988 111
9999 7777
9 10 9 10 9 9 9 10
11 11 11 11 10 10 10 10
13 13 13 13 12 13 12 12
14 14 13 14 13 13 14 13
14 13 14 14
14






Bromus
japonicus

A B

7 5
8 8
8 8
8 8
9 9
9 9
11 11
13 13
14 14







C
7
7
8
8
8
9
9
11
13
14







D

7
8
8
10
10
10
11
13
14







A
7
8
8
8
8
9
11
12
13
13
13
13
13
14



Koeleria
cristata

B C
7 7
8 8
8 8
8 8
9 9
9 9
11 11
12 12
13 13
13 13
13 13
13 13
13 13
14 14



Aristida
Bouteloua
longiseta

D
7
8
8
8
9
10
11
12
13
13
13
13
13
14



A
2
3
3
4
4
7
7
8
8
12
12
12
13
14



B
2
3
3
4
4
7
7
8
8
11
12
13
13
14



C
2
3
4
7
7
7
7
8
8
11
12
13
13
14




D
2
3
4
7
7
7
7
8
8
11
12
13
13
14




A


3
4
5
5
6
6

10
10
11
13
13
14

gracil is
B


4
4
5
5
6
6
9
10
10
11
13
13
14

C


4
4
5
5
6
6
9
10
10
11
13
13
14

D


4
4
5
5
6
6
9
10
10
11
13
13
14

Agropyron
srni thii

A
4
4
4
4
7
7
8
9
9
10
12
12
12
13
13
13
13

B
4
4
4
4
7
7
8
9
9
10
12
12
12
13
13
13
13

C D
4 4
4 4
4 4
4 4
7 7
7 7
8 8
9 9
9 9
10 10
12 12
12 12
12 12
13 13
13 13
13 13
13 13
(continued)

-------
APPENDIX (continued).
Date
27 May
3 Jun
10 Jun
1 6 Jun
22 Jun
30 Jun
7 Jul
£3 13 Jul
01 21 Jul
30 Jul
4 Aug
11 Aug
18 Aug
26 Aug
2 Sep
10 Sep
15 Sep
Taraxacum
off icinale

ABCD A
10 10 10 10 8
12 12 12 12 8
12 12 12 12 9
12121212 9
12 12 12 14 9
12 13 14 13
14 14 13
13
14








Antennaria
rosea

B
8
8
9
9
9
13
13
13
14









C
8
8
9
9
9
13
13
13
14









D
8
8
9
9
9
13
13
13
14








Tragopogon
dubius

A
5
7
7
8
8
10
12
12
12
13
14







B
5
7
8
g
9
10
12
12
12
13
14







C
5
7
7
9
9
10
12
12
13
13
14







D
5
7
8
9
9
10
12
12
13
14







Achillea
mi 1 1 ifol ium

A



9
9
10
10
10
10
11
13
14






B



g
9
10
10
10
10
11
13
14






C



9
9
10
10
10
10
11
13
14






D



9

10
10
10
10
11
13
14





Psoralea
Sphaeralcea
argophyl la

A


3
3
3
7
9
10
10
10
12
12
12
12
12
12
14

B

2
3
3
4
7
9
10
10
10
12
12
12
12
13
14


C

2
3
4
4
7
9
10
10
10
12
12
12
12
13
14


D



3
3
9
9
10
10
10
12
12
12
12
13
14


A


4
4
7
9
9
10
10
12
13
13
13
13
13
13
13
coccinea

B


4
4
4
9
9
10
10
12
13
13
13
13
13
13
13

C


7
7
7
9
9
10
10
12
13
13
13
13
13
13
13

D


4
7
7
9
9
10
10
12
13
13
13
13




A




5
5
5
6
6
6
7
7
8
8
9
9
10
Artemisia
fri
B



4
5
5
5
6
6
6



8
9
9
10
gida
C



4
5
5
5
6
6
6
6
7
8
8
9
9
10

D



5
5
5
5
6
6
6
6
7
8
8
8
9
10

-------
                                 SECTION 11

           MONITORING PLANT COMMUNITY CHANGES DUE TO S02 EXPOSURE

                                      by

                      J. E. Taylor and W. C. Leininger


                                INTRODUCTION

     This  section describes  how  the  plant  community monitoring  procedures
discussed in Section 2 have been used to monitor changes occurring on the Zonal
Air Pollution System (ZAPS) where known levels of sulfur dioxide (S02) exposure
have occurred.

                                 DISCUSSION

     Earlier studies (Taylor et aJL ,  1975,  1976) have shown that Shannon-Weaver
and  Simpson's  indices worked  best in the  Col strip  project situation.   Other
indices (Redundancy, Probability of Inter-  and Intra-Specific Encounters) were
used  in  1975,  but appeared  to be insensitive or inconsistent  in  response to
conditions in this northern Great Plains ecosystem.  The Shannon-Weaver indices
from  ZAPS  I were  tested  for  statistical  significance.  Results are  shown in
Table 11.1  and  Figure 11.1.

     The growing  season conditions  in  1975 were  reflected in the  diversity
indices.   The spring was  unusually cold, which delayed plant growth.  This is
reflected  in the  relatively  low indices observed in June and in the generally
lower values in 1975 compared with 1974.  By July, diversity indices tended to
be  higher,  reflecting  the  abundant  and  diverse  plant  growth once  growing
conditions became  favorable.   (There was good summer moisture).  This was not
the  case  in  the   control  plots.   There was  no significant  difference  among
Shannon-Weaver  or  Simpson's  Indices  for  the control  plots  among  sampling
periods.    However,  the  indices  associated  with  S02  application  rates  all
significantly increased in July and decreased in September.   At the latter date,
all were  significantly  (P g .001) lower than those of the control.

     At the July sampling period, the S02 plots showed diversity indices which
increased  with their   rates   of  gas  application.   This  may  be  due  to an
enhancement of plant growth in the period of early fumigation.  By the September
date,  however,  the  exact opposite  was observed.  All  three S02  plots were
significantly (P ^  .001)  less diverse than the  control.   Both  indices showed
this  effect.   This  appears  consistent  with the observations  reported in the
remote sensing  section  (Section 7).
                                      376

-------
     0.9 i
X
LU
Q
CC
-LI
HI
     0.8 -
0.7 -
     0.6 H
     0.5 -
                0.51
                                                  -——- JUNE

                                                  ~ - -JULY

                                                  — SEPTEMBER
                                                                    0.70
                                                   0.56
                 HIGH
                          MEDIUM
LOW
CONTROL
    Figure  11.1.
              Shannon-Weaver index of diversity values  for  four S02
              levels  over three sampling periods,  1975.
                                      377

-------
  TABLE 11.1.   STATISTICAL SIGNIFICANCE OF SHANNON-WEAVER INDICES1 AMONG ZAPS
               PLOTS,  1975.   (Values are averages of 2 lines of plots per
               observation)

                                       SAMPLING PERIODS
                       June
July
September
S02 Levels
CONTROL
LOW
MEDIUM
HIGH
0.6724
0.6059
0.6761
0.7689
MC
i I n fifinn
, ** , 0.6984
1 1
*
i i n 7.171
| | U . / 1 / \
***
i i n pcoq
| 1 U . Ou J o
NS
1 i
***
i i
I i
***
i i
1 I
***
i i
I i
0.6972
0.5560
0.5856
0.5116

Significance of t-test comparing values connected by line.

NS  = no significant difference
  * = P ^ .05
 ** = p g ,oi
*** = p g .001
  1 = Calculated with Log base 10.

     The underlying  causes  of  the  diversity differences still are unclear be-
cause of the confounding influences of pre-treatment plant community pattern and
density, coupled with treatment-induced changes in both species and numbers of
individuals.

     Diversity indices for the ZAPS sites are shown in Figures 11.2 and 11.3 for
1975 and 1976,  respectively.

     In  1975,  the  growing  season was delayed by late storms and cold tempera-
tures.   The  cool-season species  were  not  yet at peak  development  at the June
sampling date.  This  could  explain the higher diversities observed in July at
this location.  (This was  not  seen at Colstrip, where elevations and precipi-
tation are  lower).  The September decrease in diversity follows the usual pat-
tern.  There are  no  obvious  diversity differences among S02  levels,  with the
possible exception of an  enhancement on the high  rate in July.  If this is a
real difference,  it  may  reflect a transitory enhancement of plant growth with
low levels  of exposure to S02.
     A more typical growing season occurred in 1976.
in diversity followed the expected trend.
          Thus,  the  seasonal changes
     It is obvious that the ZAPS II  site is inherently less diverse than ZAPS I.
It has lower canopy cover values and fewer species.  This is  illustrated  in the
diversity values  between  the  ZAPS locations (Figure  11.3).   Also,  ZAPS  I re-
ceived more precipitation  than  ZAPS II in 1976, which could contribute to the
difference.
                                     378

-------
        3.0
        2.5 -
        2.0 -
        1.5 -
JUNE
JULY
SEPTEMBER
                 HIGH
                            MEDIUM
                                           LOW
                                                      CONTROL
                              LEVELS  OF  SO2
Figure 11.2.   Shannon-Weaver  function (HP) for ZAPS I,  three sampling
              dates, 1975.
        3-0 -i                                              	ZAPS I
                                                         	ZAPS n
        2.5 -
        2.0 -
        1.5
                 HIGH
                                           LOW
                                                      CONTROL
                            MEDIUM

                              LEVELS   OF  SO2

Figure 11.3.   Shannon-Weaver function (HP) for ZAPS I  and  II, two
              sampling dates each, 1976.
                                379

-------
     The depression of  diversity observed on the  low S02 plots on ZAPS II is
similar to the  response of ZAPS I during its first year  of fumigation (1975).
Whether the second year of ZAPS  II will show a corresponding recovery remains to
be seen.

                            ZAPS SITE SIMILARITY

     Canopy coverage data (means of 40 frames per plot) were used to calculate
Similarity Coefficients among the ZAPS plots, using the procedure of Sokal and
Sneath (1963).  This was done  for June and August, 1976.  Similarity dendrograms
are presented in Figures 11.4 and 11.5.

     In June, there was a clear separation of the two ZAPS sites.  Also, in both
cases the  vegetation  on the CONTROL and LOW treatment plots was similar (0.89
and 0.99), as was that on the MEDIUM and HIGH S02 level plots (0.90 and 0.89).
Overall similarity within sites was 0.78 in ZAPS  I and 0.71 in ZAPS II.

     In  August, the  same  trends  are  evident.   Both  CONTROL  and  LOW  plots
remained similar.  The MEDIUM and HIGH were similar on ZAPS I, and somewhat less
so on ZAPS II.  The sites were considerably more  similar in August than in June,
probably due to  the  lower overall species number  observed at the later date.

                              PHENOLOGY STUDIES

     Phenological stages  occurring  on  the ZAPS  plots are shown  graphically in
Figures 11.6 and 11.7.

     In  1975  no consistent phenological differences  were observed among ZAPS
plots for  any species.   This  may be an artifact of sampling, since the aerial
monitoring did  show a definite  tonal gradation  across the plots as the season
progressed.   This  inconsistency may be due  to  inadequate sampling frequency,
relative insensitivity of field procedures, or the avoidance of  damaged plants
in the  immediate vicinity of the gas delivery pipes in an attempt to minimize
"bias" in the  sample.  The investigators  feel  that there may in fact have been no
important phenological differences, but that the reflected differences were due
to bleaching of  the  stressed vegetation as it  matured.

                                   SUMMARY

     The Shannon-Weaver Index based on both number and on cover *s sensitive to
different S02 levels  and seasonal plant composition.  The early effect of S02
fumigation appears  to  be an enhancement of  plant  species diversity.   By late
season, 1975, this tendency was  reversed, and the diversity based on numbers for
all  three  S02  levels  were  significantly  (P ^  .001)  less diverse  than the
control.

     Cover-based diversity shows similar trends  for seasons and sites, and will
be utilized further  in future work.

     Species  similarity  coefficients  show a tendency to  group the CONTROL and
LOW versus  the MEDIUM and HIGH ZAPS treatments on both sites and both dates.
                                      380

-------
    1.0
    .9
  UJ .7
  O
  O
  5.5
  CO .4

    0
    i      i      i      IJIJIUE
CONTROL  LOW  MEDIUM   HIGH CONTROL  LOW  MEDIUM   HIGH
    12345678
                                L .9869 |
               .89O4
                      .7762
                                                .8933
                                                   .7140
                                      .4654
Figure 11.4.  Dendrogram of similarity coefficients for the ZAPS sites,
              based on canopy coverage, August 1976.
   1.0


 2.9
 I-
 •z.
 y s
 O
 £.7
 O
 O

 > .6

 E
 < .5
 CO -4

    0
             i      i      n      n     n     n     i      i
         CONTROL  LOW   HIGH CONTROL LOW  MEDIUM MEDIUM   HIGH
                    2345678
     .8169
           7924
                           .9IO4
                                .8256
                     .7851
                                                .8577
                                      .7819
Figure 11.5.
    Dendrogram of similarity coefficients for the ZAPS sites,
    based on canopy coverage, June 1976.
                                381

-------
                DEAD


     WINTER  DORMANCY


       FALL GREENUP


            MATURITY


       SEED SHATTER


        FRUIT  FORMED


      LATE FLOWERING


           FLOWERING


 FLOWER BUDS OPENING
 SHOOTING SEED  STALK

FLOWER BUDS  APPEARING
          BOOT STAGE


    VEGETATIVE  GROWTH


       EARLY GREENUP


       BASAL ROSETTE


            SEEDLING


          COTYLEDON
                 Agropyron smithii
                 Aristida longiseta
                 Koelfria ens fata
                 Poa sandbergii
                 Stipa comata
                 ANNUAL BROMES
                                MAY
                                 15
JUNE JULY
 34   12
AUG.
 28
NOV.
 3
Figure 11.6.  Phenological profile  of selected species  on Taylor  Creek
                 control  plot.
                                        382

-------
                DEAD


     WINTER  DORMANCY


       FALL GREENUP


            MATURITY


       SEED SHATTER


        FRUIT  FORMED


      LATE FLOWERING


           FLOWERING


 FLOWER BUDS OPENING
 SHOOTING SEED  STALK

FLOWER  BUDS  APPEARING
          BOOT STAGE


   VEGETATIVE  GROWTH


       EARLY GREENUP


       BASAL ROSETTE


            SEEDLING


           COTYLEDON

,„.—_ Artemisia frigida
	— Plantago patogonica
^—— Sphaeralcea coccinea
	Taraxacum officinale
	Tragopogon dubius
(NEW GROWTH,TRDU)
                                 MAY
                                           JUNE JULY
                                           24    12
             AUG.
             28
    NOV.
      3
Figure  11.7.   Phenological  profile of selected species  on Taylor Creek
                 control plot.
                                        383

-------
     The phenology scoring system has proved very satisfactory  in  a  variety  of
vegetation types.  However, phenology changes attributable to SCL have not been
quantified  because  of  the  many  confounding  influences.   The  frequency and
intensity of sampling is being increased to construct a more complete data base
so that confounding influences can be identified and separated.

                                 REFERENCES

Sokal,  R. R. and P.  H. A. Sneath.   1963.   Principles of numerical taxonomy.  W.
     H. Freeman,  San Francisco.

Taylor, J.  E. , W. C.  Leininger, and  R. J.  Fuchs.  1975.  Baseline  vegetational
     studies near Colstrip, Proc.  Ft. Union Coal Field Symp. , Mont.  Acad.  Sci.
     pp. 537-551.

Taylor,  J.  E.  , W.  C.   Leininger,  and R.  J.  Fuchs.   1976.   Monitoring  plant
     community changes due to emissions from fossil fuel power  plants  in eastern
     Montana.   Section II of the bioenvironmental impact of a coal-fired  power
     plant, second interim report.   USEPA Ecological Research  Series  EPA-600/3-
     76-013.  pp. 14-40.
                                      384

-------
                                  SECTION 12

                        EFFECTS OF LOW-LEVEL S02 STRESS
                             ON TWO LICHEN SPECIES

                                      by

                                 S. Eversman


                                 INTRODUCTION

     As part of the development of western energy resources, coal-fired power
plants  are  proliferating  in  the  Great  Plains  region.   Two  350-megawatt
generating  plants began  operation  in  autumn,  1975,  and  spring,  1976,  in
Colstrip in southeast Montana.  Two more are  in  the planning stages.

     Sulfur  dioxide  (S02),  one of the emissions  from  burning  coal, has been
repeatedly  demonstrated  to  cause  adverse  effects  in  plants,  especially
lichens.   Many  laboratory  experiments  have   shown  chlorophyll  bleaching,
decreases  in respiration  and photosynthetic rates,  and  plasmolysis with S02
exposure (LeBlanc  and Rao,  1975).   The lowest, experimental  S02 concentration
has been at 0.50 ppm for 12 hours  (Nash, 1975).

     LeBlanc   and  Rao   (1975)  cite  experimental  works  that  draw  these
conclusions.  Transplanting lichen species into areas already polluted results
in  lichens  exhibiting the  same  internal and  external pathological symptoms
demonstrated in laboratory tests.  Most of the  studies have been in  humid areas
where  S02  is  thought  to  be more  harmful than  in semi-arid climates,   They
estimate that,  in humid  climates,  long-range  average concentrations  of S02
below  0.002  ppm  would cause no damage to lichen species; acute damage occurs
above 0.03 ppm, and intermediate amounts of S02 would cause chronic damage to
lichens.

     This report  presents  results  of observations on  the effects  of chronic
low-level S02 stress  on two  lichens native to semi-arid  grassland and ponderosa
pine vegetation types.  The objectives of the  study were to:

          1) Identify and  time the anatomical  and physiological changes that
          occurred with exposure  to various low levels of  S02, and

          2)  to   make   simultaneous   observations  on  associated  vascular
          vegetation for  comparison purposes.

     Usnea hi rta is an epiphyte on trunks and branches  of ponderosa pine, the
major  tree  of this  region   It  is  a  fruticose  lichen,  i.e.,   stringy  in
appearance;  this  type of  lichen is  considered  particularly  susceptible and
sensitive to air pollutants.

                                      385

-------
     Parmelia chlorochroa is a foliose (leafy, flat) lichen that lives  on bare
soil between  grass  clumps  and shrubs on the  grasslands.  Foliose lichens are
also considered sensitive to air pollution.

     A  fumigation  system  was  constructed  on  a  grassland  site  in  southeast
Montana.  The system was composed of 4 plots,  each about 0.654  hectare.  Each
plot had  a  network  of aluminum pipes that delivered different amounts of S02
through small  holes  in  the bottoms of the pipes.  Lee and  Lewis (1976) fully
describe  the  system.  One  system  of four plots  was  in operation in 1975 and
1976, and a duplicate system operated in 1976 about 1 km from the first site.
This report refers to the system operating in both 1975 and  1976 as ZAPS I (I =
Site 1, ZAPS = Zonal  Air Pollution System).   The fumigation  or treatment plots,
within each site or system, are referred to as A (CONTROL),  B (LOW), C (MEDIUM)
and D (HIGH) (Lee and Lewis, 1976).   ZAPS  II  refers to the  second system, with
similar treatment plots A, B, C and D, operating in 1976.

     By specifically identifying the characteristics of the two  lichen species
exposed to  the S02  stress  on  the  ZAPS  sites, these species can be  used  as
biological systems to monitor increasing S02 content of the air in the Colstrip
area.  Simultaneous observations made by other researchers  in the region will
be combined with the lichen observations to identify  and interrelate ecosystem
components affected by decreasing air quality.

     Usnea  hirta and Parmelia chlorochroa  from several  ponderosa pine  and
grassland field  study sites  have  been collected  and observed  for  baseline
conditions.   They have been  compared  with  the stressed lichens  from the ZAPS
sites.   The sites are mapped  and  described in a previous  chapter (Eversman,
1976).

                                    METHODS

     Five  metal   fenceposts  were  placed  nonrandomly  in   each  of  the  ZAPS
fumigation plots.  They were  placed toward each corner and in  the  center of
each plot where there was  minimum grass cover  so  that Parmelia chlorochroa
could be  transplanted to  the soil.  There was a minimum of  2 meters between a
fencepost and any fumigation pipe.

     Transplants of both Usnea hirta and Parmelia chlorochroa were made 1  May
1975 and  23 March  1976;  fumigations began  24  May  1975  and 1£ April 1976.
Ponderosa pine branches  containing U.  hirta were wired onto the  northeast side
of each fencepost so the branches  extended about 0.25 - 1.0 meters vertically
on the post.   P. chlorochroa was moved from  the soil in a nearby field (Site P7)
to the  fumigation sites to augment the existing  populations.  A  second set of
P. chlorochroa transplants was made 17 July 1976 from Site G5 (Kluver East) to
compare vertical position  effects  of the S02  stress.  P_.  chlorochroa samples
were tied on bare ponderosa pine branches and wired to the  fenceposts on Site
1, parallel  to the branches containing U. hirta.  Duplicate  samples were placed
on the  soil at the base of  each  fencepost.

     Samples were collected about once a month through September of each year.
They were  observed for general thallus condition and photographed.  Respiration
rates were determined manometrically  (Eversman,  1975).  Samples were sent to

                                     386

-------
the Montana State University Soil Testing Laboratory for sulfur analysis.  The
details of analytical methods are presented in the Appendix.

     Plasmolysis  counts  were  made  by  preparing  3 wet-mount  slides  of  3
different  plants  and counting  100  algal  cells  on  each  slide,  recording the
number of plasmolyzed cells.

     The same set of observations was made on lichen samples  collected at about
the same times,  20-97 km from the fumigation  sites.

     Stems and  leaves of Agropyron smithi i and Koeleria  cristata, two grasses,
were collected  near  each  fencepost on Sites 1 and 2 for 3 months in 1976, to
determine respiration rates  and sulfur content.

                                    RESULTS

     Respiration  rates  and results  of  the  completed plasmolysis  counts for
Usnea  hirta  appear  in  Table 12.1.   Samples from the A (CONTROL) plot on both
Sites  1  and 2  did  not exhibit  respiration  rate and plasmolysis percentages
different  from  those of U.  hirta samples collected  simultaneously from their
natural habitats.

     The respiration rates of samples from Plot B were consistently below those
of  samples from the control site,  but  within the  range  of samples  from all
unpolluted  sites.   The plasmolysis  level  reached  about  76% of  the  control
levels in  100 days,  and leveled off throughout the 156 days of S02 treatment.
Plasmolyzed algal  cells  are  also bleached.

     The respiration rates of U. hi rta samples from  plot C decreased to 57% of
those  of control plot A samples, and remained  at  less than 50% of control rates
for  the  remainder of the test period.  Plasmolysis  was  nearly 100% in 31-33
days, and the algae  never recovered.

     Samples from plot  D  exhibited a respiration rate that was 6% of that of
control samples in 1975, but in 1976 the  rates  fell to 33-46% of control plot A
samples.  They remained 19-36% of those of control samples for the test period.
Plasmolysis and bleaching  reached 100% in 31-33 days,  with  no recovery.

     P_.  chlorochroa  samples  on  the  ground  did  not  exhibit  changes  in
respiration  rates in any  of the  treatment plots  (Table  12.2).  The results of
plasmolysis  counts  are  still  very incomplete, but some representative counts
showed a  slight increase  of plasmolysis with  S02  stress.   When samples were
elevated to the fenceposts with the L). hirta, however, plasmolysis increased to
96%  in 23  days  on plot D, and the mean respiration rate dropped to 22% of that
of control (A)  samples.  Respiration rates of elevated P. chlorochroa decreased
on plots B, C,   and D; the rates of control samples,  A,  and those on the ground
on all  plots, did not decrease.

     Grass  samples  exhibited  a generally decreasing respiration  rate  on all
plots  from June to August (Table 12.3).   This  decrease does not superficially
appear to be related  to S02 dosage.
                                      387

-------
TABLE 12.1.   RESPIRATION RATES AND PERCENTAGE OF PLASMOLYSIS FOR Usnea hirta
ON ZAPS SITES AND ON SOME COLSTRIP FIELD SITES,  1975-76.   Respiration rates
are given as the mean of 3-10 samples ± 1  standard deviation, expressed in ul
02 consumed/g dry weight/hour.   Plasmolysis is expressed as mean percentage of
3 samples ± 1 standard deviation.

Collection Site
P10, East Otter Creek
P9, transplant source
PI 5, Fort Howes natives
PI 5, Fort Howes natives
P16, Poker Jim Butte
ZAPS: IA
IIA
IB
IIB
1C
IIC
ID
IID
P6, Kluver West tr.
ZAPS: IA
IB
1C
ID
P16, Poker Jim Butte
P13, Home Creek Butte
Pll, SEAM 1
ZAPS: IA
IB
1C
ID
P10, East Otter Creek
ZAPS: IA
IIA
IB
IIB
1C
IIC
ID
IID
P9, Road
Number of
S02 Treatment
Days, Date
0
0
0
0
0
31
M
ii
H
M
M
n
ii
0
33
M
n
n
0
M
M
47
n
M
n
0
72
n
n
n
n
M
M
II
0
3-23-76
5-01-75
M
5-13-76
n
n
n
n
n
n
n
n
n
6-25-75
n
n
M
n
7-16-75
n
n
7-10-75
n
M
n
6-23-76
n
n
n
n
n
n
n
n
8-11-75
Resp.
Rate
X
709
563
511
576
694
711
800
718
755
290
438
325
264
864
821
746
472
53
630
462
545
686
555
333
0
744
704
555
359
476
236
263
115
169
508
1
s.d.
63
109
213
113
183
72
97
141
74
155
175
199
99
86
37
98
204
45
77
55
35
115
121
307
0
45
71
34
78
77
83
48
67
38
50
Plasm.
% x 1 s.d.
3




6
4
35
37
94
99
94
100

6







54
99
100
11
4
4
63
58
100
99
100
100

3




1
3
25
39
9
2
9
0

4







20
1
0
2
2
2
18
21
1
1
1
1

                                                                   (continued)
                                     388

-------
TABLE 12.1  (continued)
Number of
S02 Treatment
Collection Site Days, Date
ZAPS



P10,
P16,
Pll,
ZAPS







P10,
P8,
P3,
ZAPS



P10,
P14,
P14,
ZAPS







P10,
P15,
P15,
ZAPS







: IA
IB
1C
ID
East Otter Creek
Poker Jim Butte
SEAM 1
: IA
IIA
IB
IIB
1C
IIC
ID
IID
East Otter Creek
Morning Star View
Kluver NE1
: IA
IB
1C
ID
East Otter Creek
3-Mile top
3-Mile bottom
: IA
IIA
IB
IIB
1C
IIC
ID
IID
East Otter Creek
Fort Howes transplants
Fort Howes natives
: IA
IIA
IB
IIB
1C
IIC
ID
IID
79
11
n
ii
0
n
n
96
ii
ii
ii
ii
M
ii
n
0
n
n
110
II
II
II
0
II
II
119
n
n
M
n
n
M
n
0
II
II
156
n
M
M
n
n
n
M
n
n
M
n
7-15-76
II
II
7-17-76
n
n
M
n
n
n
n
9-25-75
M
M
9-11-75
n
n
M
8-09-75
n
M
n
n
n
n
n
n
n
n
9-15-76
n
n
9-14-76
n
n
n
n
M
n
n
Resp.
Rate
X
738
435
316
0
463
610
453
670
643
473
479
323
264
244
98
640
828
736
709
499
355
0
668
667
547
721
618
556
509
298
282
150
118
633
597
470
631
506
487
457
340
289
142
112
1
s.d.
66
87
133
0
57
31
19
103
38
39
86
35
49
92
20
40
135
72
55
128
137
0
41
76
40
18
50
84
66
73
84
29
16
53
43
50
21
61
43
37
40
85
68
46
Plasm.
% x 1 s.d.




9


5
5
69
66
99
99
100
100



15
77
98
100



5
24






10


20
4
82
71
100
99
100
100




3


2
2
17
18
2
2
0
0



3
16
3
0



2
10






1


5
2
13
18
1
1
0
0
                                     389

-------
TABLE 12.2.   RESPIRATION RATES AND PERCENTAGE OF PLASMOLYSIS FOR Parmelia
chlorochroa ON SOME COLSTRIP FIELD SITES, 1975-1976.   Respiration rates are
given as the mean of 3-10 samples ± 1  standard deviation, expressed in yl 02
consumed/g dry weight/hour.   Plasmolysis is expressed as mean percentage of
3 samples ± 1  s.d.

Collection Site
Field near ZAPS sites
G7, Pasture
Gl , Hay Coulee
G7, Pasture
ZAPS: IB
1C
G5, Kluver East
Gl , Hay Coulee
ZAPS: IA
IB
1C
ID
G4, Kluver North
G3, Kluver West
ZAPS: IA
IB
1C
ID
Field near ZAPS site
ZAPS: IA
IB
1C
ID
Gl , Hay Coulee
G3, Kluver West
G4, Kluver North
G5, Kluver East
ZAPS: IA (native)
IIA (native)
IB
IIB
IIC
IID
Number of
S02 Treatment
Days , Date
0
"
0
11
31
I!
0
n
33
II
1!
II
0
II
47
II
II
II
0
79
II
II
ii
0
n
n
n
96
"
M
11
n
M
5-01-75
II
5-14-76
n
"
"
6-26-75
11
n
11
11
11
7-07-75
n
7-10-75
II
II
II
8-11-75
II
II
II
M
7-14-76
n
n
M
7-17-76
II
"
M
II
II
Resp.
Rate
X
223
337
458
371
276
290
316
215
319
330
345
296
349
293
345
404
317
294
217
356
357
336
309
299
272
242
322
231
293
299
262
301
335
1
s.d.
39
132
43
54
50
15
40
26
44
36
36
54
21
29
80
54
50
38
28
48
36
47
37
22
25
4
26
41
36
22
31
30
22
Plasm.
% x 1 s.d.








2 2

8 2



10 7

4 4
43 34









22 3





                                                                   (continued)
                                     390

-------
TABLE 12.2.   (continued)

ZAPS



ZAPS

Collection Site
: IA
IB
1C
ID
: IA (native)
IID
After new transplants
ZAPS


ZAPS


ZAPS


ZAPS


Gl,
G3,
G4,
G5,
ZAPS



ZAPS


ZAPS


ZAPS


: IA
IIA
IA*
: IB
IIB
IB*
: 1C
IIC
1C*.
: ID
IID
ID*
Hay Coulee
Kluver West
Kluver North
Kluver East
: IA (native)
IA
IIA
I A*
: IB
IIB
IB*
: 1C
IIC
1C*
: ID
IID
ID*
Number of Resp.
S02 Treatments Rate
Days , Date x
110 9-11-75
n n
i n
M n
119 8-09-76
n M
were made 7-17-76 (* =
23 8-09-76
n n
n n
n n
n n
n n
n n
n n
n n
ii
M M
M n
0 9-14-76
n n
n n
n n
60 9-15-76
n n
n n
n n
N n
n n
n n
n n
M II
II II
II II
II II
II II
362
271
353
340
350
293
elevated
298
297
366
314
315
311
309
368
197
396
437
79
274
301
262
239
315
305
326
345
270
338
174
322
403
45
396
386
43
1 Plasm.
s.d. % x 1 s.d.
69
20
52
21
20
50
samples):
8
28
26
25
40
33
26
30
54
24 156
38
15 96 3
22
15
34
15
31
44
31
9
33
28
47
38
31
10
13
77
34
                                      391

-------
TABLE 12.3
RESPIRATION RATES
OF TWO GRASSES, Koeleria cristata AND Agropyron
smithi i, FROM THE ZAPS
SITES, 1976. Respiration rates are given
as the mean of 3 samples ± 1 standard deviation, expressed in yl
02 consumed/g dry weight/hour.

Koeleria
cristata

Col lection
Plot
IA
IIA
IB
IIB
1C
IIC
ID
IID
IA
IIA
IB
IIB
1C
IIC
ID
IID
IA
IIA
IB
IIB
1C
IIC
ID
IID
Number of
S02 Treatment
Days, Date
72 6-23-76
n M
n M
n n
n n
n n
M n
n n
96 7-17-76
n n
n M
n n
n M
n M
n n
M n
119 8-09-76
n n
M n
M
n M
M n
n n
n N

Resp.
Rate
X
702
706
706
617
893
564
1045
759
506
247
587
411
560
430
827
525
128
147
164
226
146
375
108
332


1
s.d.
51
103
102
134
104
120
42
114
4
8
130
75
72
28
113
103
35
25
19
45
29
41
31
46
Agropyron
smithi i

Resp.
Rate
X
646
579
736
699
516
551
610
349
355
335
231
328
197
366
212
249
145
312
177
107
239
394
297
530


1
s.d.
189
30
163
70
106
34
235
34
64
38
90
120
64
163
no
55
13
38
16
32
29
50
68
85
                                     392

-------
     The  data from  the  S04 content analysis  from  1975 (Table 12.4)  suggest
increasing sulfur content as S02 exposure continued, but available results are
inconclusive.  The 1976 analyses are not yet complete.

     Microscopic   examination   revealed   extensive    bacteria    populations
associated with  the lichens, especially with  ?_.  chlorochroa.   As 562 stress
increased in dosage and time, the bacteria appeared to  be increasingly  attached
to the fungal hyphae,  rather than  in the water medium (on a microscope slide).
The bacteria were associated less with algae than with  the fungal hyphae.

                                 DISCUSSION

     A reasonable question  arises  in connection with  both  the  Usnea hirta and
Parmelia  chlorochroa  transplants  on  the  control  (A)  plots.TFe IIflirta
specimens have  been transplanted from ponderosa ,pine sites that  are  shadier
than the fumigation  sites and control plot A.   However,  U.  hirta is not a plant
of deep shade; it grows on  partially exposed branches and  trunks  of ponderosa
pine.  Every effort was made to compare the measurements made on  U. hirta from
control plots A  with  those of native  samples  collected directly from their
native  habitats.   The respiration  rates,  plasmolysis  counts,   and visual
appearance of the U.  hirta exhibit no harmful  effects from the transplanting.

     P. chlorochroa is a  native of exposed bare soil and  was  transplanted in
most cases to_a similar habitat.  Some adverse effects might have been  expected
when it was tied to  pine branches and elevated  above the  soil, but this was not
the case.   A common observation when  any specimens are transplanted is  a slight
temporary  increase  in respiration  rate  over  that of  natives  or duplicate
samples,  but the  increase  is  not  permanent  and  no visual  effects can  be
detected.

     Decreasing  respiration  rates,  algal  bleaching  and  plasmolysis,  and
yellowing of the  plant  U. hirta reflect increasing S02  exposure.

     The  respj ration rates of  P.  chlorochroa samples  on  the  ground  did not
decrease  with  the  S02  treatments"nor did  the  plants   change  color.   P.
chlorochroa specimens were  then placed on the branches 0.75 m above the ground
to test  if:   1)  vertical  position  in  the grassland  stratification  had any
effects and  2)  associated  bacteria decreased  when the thalli  were  off the
ground.   The  bacterial   populations  remained  visually  the  same   but  the
respiration rates decreased dramatically in proportion  to  S02  dosage,  and the
percentage of plasmolysis  increased to nearly 97% in  23 days on plot D.

     This seems to indicate that the S02 levels on  the  sites vary  appreciably
from ground  level  to 0.5-1.0 m above the ground, although  compounding stress
effects could be a factor  in the observed reactions.   Further work  must include
using some monitoring method, possibly sulfation plates, to  detect  the  vertical
distribution of the  S02 on the treatment plots.

     Conclusions  reached  before £.  chlorochroa  samples were placed on the
branches on  the  test  site  fenceposts  were  that U.  hirta is  a  species more
sensitive than  P.  chlorochroa  to  S02  stress.   Observations  of  elevated  P.
chlorochroa sampTes  indicate that both species are sensitive to S02-   However,

                                     393

-------
TABLE 12.4.   SULFATE SULFUR CONTENTS OF Usnea hirta AND Parmelia chlorochroa
             SAMPLES COLLECTED IN 1975.   Expressed as mean percentage ± 1
             standard deviation of 3 - 5 samples.   Determined by the MSU
             Soils Testing Laboratory.
Usnea hirta
Number of
S02 Treatment %
Collection Site
P9, Road
P15, Fort Howes tr.
PI 5, Fort Howes natives
P9, Road
P3, Kluver NE1
P8, Morning Star View
P13, Home Creek Butte
P10, East Otter Creek
ii n
ZAPS: IA
IB
IB
IB
1C
1C
1C
ID
ID
Parmelia chlorochroa
Pasture near ZAPS sites
G7, transplant source
ZAPS: IA
IB
IB
IB
1C
1C
ID
ID
Days
0
n
n
0
0
0
0
II
0
47
47
79
110
47
79
no
47
79

0
n
47
47
79
110
47
no
47
79
, Date
5-01-75
n
ii
8-11-75
9-25-75
M
7-16-75
M
9-25-75
7-10-75
7-10-75
8-11-75
9-11-75
7-10-75
8-11-75
9-11-75
7-10-75
8-11-75

5-01-75
M
7-10-75
7-10-75
8-11-75
9-11-75
7-10-75
9-11-75
7-10-75
8-11-75
Sul fur
0.25
0.25
0.19
0.26
0.03
0.19
0.15
0.08
0.22
0.29
0.30
0.17
0.29
0.33
0.19
0.37
0.32
0.41

0.17
0.19
0.23
0.19
0.18
0.20
0.23
0.18
0.21
0.22
1 s.d.
0.04
0.03
0.04
0.04
0.01
0.08
0.16
0.01
0.11
0.01
0.03
0.15
0.11
0.03
0.11
0.07
0.05
0.01

0.02
0.03
0.04
0.02
0.02
0.03
0.03
0.03
0.03
0.01
                                     394

-------
U.  hirta may be more useful  as an area-wide indicator of decreasing air quality
because  of  its  more  exposed  locations on  tree trunks  and  branches.   Wind
conditions or  filtering  by  grasses and forbs may prevent a great deal of $62
from reaching  ground  level,  except in  bare  areas,  where it can demonstrably
affect P.  chlorochroa.

     Simultaneous  visual  observations  were  made on  the associated vascular
vegetation on  the  fumigation plots.  While unmistakable  bleaching was evident
in U.  hirta  on all test plots  (ZAPS  I  and II, B, C, and D plots) 31-33 days
after  the fumigation  began,  no visual  effects appeared  on the grasses during
the  test  periods.   The  height  of the  grass plants  was about equal  to the
vertical  placement of  the   transplanted  lichens.  The  generally  decreasing
respiration rates  of  Agropyron  smithii  and  Koeleria  cristata  are  probably a
function  of their  phenology.The rate  of decrease  is  similar on nearly all
plots,  both sites,  from  June to August.  Exceptions  (samples from 2D and 1C,
August, 1976), however, might reflect S02 stress, which  has been demonstrated
to raise respiration rates when  present  in very small amounts  (LeBlanc and Rao,
1975).

     Other  personal  observations  were  leaf-tip  curling  and  browning  in
Achillea lanulosa after 60 days of stress on plot ID  in 1975.   The grasses and
forbs  oh stress  plots  B,  C,  and D,  Site  1,  cured a  much  yellower  color
overwinter (1975-1976) than  did  adjacent non-stressed vegetation.
                                      395

-------
                                 REFERENCES

Eversman, Sharon.  Lichens as Predictors and Indicators of Air Pollution from
     Coal-Fired Power Plant Emissions.   The Bioenvironmental  Impact of a Coal-
     Fired Power Plant,  Second  Interim Report, Colstrip, Montana.  Corvallis
     Environmental  Research  Laboratory, Office of  Research  and Development,
     U.S. Enviromental  Protection Agency, Corvallis, 1975.

Eversman, Sharon.   Soil  and Epiphytic  Lichen  Communities of  the Colstrip,
     Montana Area.   The  Bioenvironmental  Impact  of a Coal-Fired Power Plant,
     Third  Interim  Report,  Colstrip,   Montana.    Corvallis  Environmental
     Research   Laboratory,   Office   of   Research   and   Development,   U.S.
     Environmental  Protection Agency, Corvallis, 1975.

LeBlanc,  Fabius  and  D.N.  Rao.   Effects  of  Air  Pollutants  on  Lichens  and
     Bryophytes.   Iji  Mudd,  J.B.  and T.T.  Kozlowski  (Eds).   Responses of Plants
     to Air  Pollution, Academic Press, New York,  1975.

Lee, Jeffrey,  J.  and  R.A.   Lewis.   An  Experimental  System   for  Testing  the
     Effects of Sulfur Dioxide on Grassland Systems.   In press.  1976.

Nash,  Thomas  H.   Influence  of  Effluents  from  a  Zinc  Factory  on Lichens.
     Ecological Monographs  45 (2): 183-198, 1975.

Stoltenberg, Larry E.  Availability of Sulfur to Alfalfa (Medicago sativa L.)
     and Orchardgrass (Dactyl is  glomerata L.) as Influenced by Source and Time
     of  Sulfur Fertilizer Applications.   Unpublished  MS  Thesis.   Plant  and
     Soils Department, Montana State University, 1969.
                                     396

-------
                                  APPENDIX

     The method used by the Montana State University Soils Testing Laboratory
for  analysis   of  plant  sulfur  content  is  outlined below.   The method  is
described in Stoltenberg (1969).

     This procedure  is a  modification  of one  that originally  appeared  in:
Bardsley, D.E.  and J.D. Lancaster.  Determination of Reserve Sulfur and Soluble
Sulfates in Soils.  Soil  Science Society of America Proceedings 24:  265-268.
July-August, 1960.

 1.   Grind dry  plant material in a Wiley mill, with 20-mesh screen.

 2.   Mix 0.250  g plant + 1  g ashing powder in 50 ml beaker.

 3.   Add 0.5 g  ashing powder as topdressing to minimize loss of volatiles.

     a.    Ashing powder:   27 g Na2C03
                           3 g NaHC03
                           1.8 g NaN03
                           0.06 g  KMn04

     b.    These are ground  in  power mortar and pestle until  no greater color
          intensity is  observed from KMn04 (about 5  minutes).

 4.   Beakers  are set in a muffle furnace  for 1  hour at 260°C, then 2  hours at
     480°C.

 5.   Ash is treated with 25 ml  acid solution.

     Acid solution:  a.  114.2  ml  cone HC1
                         125.6 ml  cone CH3COOH
                           8.3 g NaH2P04 -H20

                     b.  Make to 2 liters  with distilled water.

 6.   Ash and acid are shaken for 30 minutes.

 7.   5 drops of H202 are added  to  each flask for clearing.

 8.   Mixture is filtered through Whatman #40 double  acid washed filter paper.

 9.   10  ml  filtrate  is pipetted  into 125 ml Erlenmeyer  flask,  and 1  ml  acid
     seed solution is added.
                                     397

-------
     Acid seed solution:  0.5434 g K2S04  in 1 liter 6N HC1  until 20 ppm S.



10.   0.5 g BaCl2 is added to filtrate.



11.   Flasks  rest for 1 minute, then swirled until BaCl2 crystals dissolve.



12.   Solution is read in Bausch & Lomb colorimeter within 2-8 min.  at 420 nm.
                                     398

-------
                                  SECTION 13

  EFFECTS OF LOW-LEVEL S02 EXPOSURE ON SULFUR ACCUMULATION AND VARIOUS PLANT
     LIFE RESPONSES OF SOME MAJOR GRASSLAND SPECIES UNDER FIELD CONDITIONS

                                      by

                C. C. Gordon, P  M. Rice, and P. C. Tourangeau


                                INTRODUCTION

     During the  1975 growing  season, S02 fumigations  of  a cool  season-short
grass ecosystem were initiated on ZAPS I (Zonal Air Pollution System) from May
28 to September 28.  Because of the visible changes across the four plots (0, 2,
5, and 10 pphm ) on ZAPS I in October of 1975 which were not accompanied by any
significantly measurable changes in biomass or species diversity (Dodd et al.,
1976),  a  substantial  portion  of  the  previous  efforts  and  studies on  the
ponderosa pine-skunkbush sites were redirected to the ZAPS  studies during 1976.
The original 1975 hypothesis for the ZAPS fumigation studies was that most plant
species fumigated continuously with MEDIUM and HIGH S02  concentrations on these
plots during the  growing season would be severely damaged or killed outright.
This hypothesis was  based  on the S02-damaged forests studied by Linzon (1971,
1973a) and  Dreisinger  (1965,  1970) in the vicinity  of  Sudbury  and Biersdorf,
Ontario, Canada,  where  the  highest S02 annual concentration was .044 ppm (4.4
pphm).   The  investigators  also considered studies done  in  Anaconda,  Montana,
where a large stationary source of S02 emissions is located  and where severe air
pollution damage  to  vegetation occurs each year; those sites which record the
highest S02 readings average only .02 ppm or less annually.   Thus the hypothesis
that the cool  season-short grass ecosystem on the ZAPS plots would be severely
damaged or  killed outright  on the  higher S02  delivery  plots was based on air
pollution studies  in  these other two polluted  areas.

     Thus far the 1975 hypothesis has not proved to be valid, so a new hypoth-
esis was formulated  before  the 1976  growing  season  began  to test in the 1976
field studies on the ZAPS sites.   This new hypothesis was that the plant species
of cool  season-short grass ecosystems which are fumigated continuously with S02
during the  growing  season  will manifest early senescence,  loss  of viable seed
production, and   excessive  quantities of  sulfur  throughout all  plant  parts
(roots,  stems and  leaves).

     To  test  this  hypothesis,  the  investigators  established  several  study
objectives  which encompass the following:   (1) Bimonthly collections of five to
six species of plant foliage during the growing season;  (2)  collection of roots
of two grass species (one  rhizomatous and one  bunchgrass  species)  at mid and
late growing season periods; (3) analysis of all plant parts for sulfur concen-


                                      399

-------
trations; (4) data  gathering  on leaf emergence, growth,  and death of western
wheatgrass  (Agropyron  smithii)  on all plots  at both the ZAPS  I  and II sites
during  the  growing  season to  ascertain  any difference  in phenology  by  S02
concentrations,  and  (5) collection  of  mature  seeds  from various  species  of
grasses,  fumigated  with the  different  S02 concentrations,  at  the  end of  the
growing season for seed  germination tests ,under laboratory  conditions.
                            MATERIALS AND METHODS
1975 VEGETATION COLLECTIONS
     ZAPS I first was sampled on June 28.  On September 28, both ZAPS I and the
area where  ZAPS  II  was to be built  were sampled.   Each of the four treatment
plots on  ZAPS  I  was subdivided into six subplots.  A composite grab sample of
each  species  from  each  treatment was made  by collecting  material  from each
subplot.  The June  sample  was  composed predominantly of live vegetation while
the  September  sample  consisted of  cured  vegetation,  except  the Gutierrezia
sarothrae and Artemisia  tridentata samples which included both green and dead
tissue.   Samples  included  flowering parts  as well  as  vegetative  portions.
Chemical analysis was performed  for both sulfur and fluoride.

1976 VEGETATION COLLECTIONS

Species Studied for Sulfur Accumulation

     In April, ZAPS  I was surveyed visually for sampling purposes to determine
species present on  the plots and  their relative abundance  and  frequency.  This
survey was used to select the species to be  studied.  Those finally chosen were:
                   SCIENTIFIC NAME
               Agropyron smithii
               Koeleria cristata
               Aristida longiseta
               Poa sandbergii (secunda)
               Achillea millefolium
               Artemisia frigida
               Tragopogon dubius

               Stipa viridula
   COMMON NAME

Western Wheatgrass

Prairie Junegrass

Red Threeawn

Sandberg's Bluegrass

Yarrow

Fringed Sage

Yellow Salsify

Green Needlegrass
T. dubius was collected in early July.  ZAPS II was not surveyed so intensively.

A. longiseta was less frequently found so S. viridula was collected instead.

Collection Methods at  ZAPS Sites

      Pipes  delivering  the  sulfur  dioxide and  ambient  air mixture  to  the
individual treatment plots were laid out in a pattern used to form the basis for
                                      400

-------
 a  sampling grid  (Figure 13.1).  The pipe  junctions  were used as 25  sampling
 points.  Each junction was divided further into four quadrants:
                               NW             NE
                               SW
SE
 Two  digit  random  numbers  were  drawn  (00  to  99)  and  divided  by  four.  The whole
 number obtained gave the  junction  number and the remainder gave the quadrant  (1
 = NE, 2 = SE, 3 = SW, 0  = NW) to be sampled.

               random  number -f 4 = junction + quadrant

                          37 -r 4 = 9 + 1 (9-NE)

                          88 -r 4 = 22 + 0 (22-NW)

                                 etc.

     Quadrants falling  north of junctions 1, 11, and 21  were considered outside
 the  treatment plot, and their  respective south quadrants were used for sampling
 (i.e.,  1-NW,  11-NW, and 21-NW became 1-SW, 11-SW,  and  21-SW;  1-NE, 11-NE, and
 21-NE became 1-SE,  11-SE, and  21-SE, respectively).  Quadrants falling south  of
 junctions  5, 6, 15, 16, and 25, were similarly rotated to their  respective north
 quadrants to place them in the treatment plot.

     This  procedure divided each treatment plot into 86  subplots for vegetation
 sampling.  A central sampling point was established in each  "junction-quadrant"
 (subplot)  by  running  a line at a  45° angle from each junction  into the appro-
 priate quadrant.   The  line  was of a fixed length randomly  determined for each
 collection period  within  the restraints of each research team's  utilization  of
 the  treatment plots.  Appropriate  species closest to  this point  were then col-
 lected.  No samples were collected within one meter of  the pipe.

     Ten subplots  were chosen  for each treatment  plot  during each collection
 period.  Material  was collected from three separate plants  for most species  in
 each subplot, resulting in one combined sample of each  species derived from  30
 individual  plants.  Aristida  longiseta was  collected from  two plants per sub-
 plot, giving a sample derived  from 20 individual plants. Artemisia frigida was
 less frequently found,  and the sample was derived from 6  to 29 individual plants
 depending on the  sample  period.

     Samples were separated  by  subplots during early April for Agropyron smithii
 and  early  July and mid-September  for both A.  smithi i  and  K.  cristata.   This
 provided ten samples for  each type of material  (below-ground parts,  dead and
 live tops)- per species on  each  treatment plot.  Each  sample  was composed  of
material  derived  from a  minimum of three plants  in each subplot.


                                      401

-------
        un
        CO
ro
           T
            £

       ll

^^^^H
••VM




1-SE




5-NE


9-NE


7-NE



12-NE
•Bla^a
^^^
14-NE




19-NE
taHM

17-NE
25-NW

21 -SW

•••^•^




                                       MIXING  SHED
         MIXING SHED
                Figure 13.1A.   ZAPS plot sampling junction
                               numbers.
Figure 13. IB.
Agropyron smithii phenology
subplot locations.

-------
     Tops consisted of vegetative culms only for A.  smithii,  K. cristata, and S.
viridula; both vegetative and reproductive culms for £.  sandbergii and Aristida
longiseta, and leaves only for Artemisia frigida and Achillea millefolium.  T.
dubius samples were  split  into leaves,  stems, and taproots.  Top material was
classified as  live or dead based  on  color (green = live,  non-green  = dead).
Parts of partially necrotic leaf blades  were placed in  appropriate live or dead
categories.   Cocking  (1973)  in  a  field  S02  fumigation study  used  a similar
classification of live and dead.

     Below-ground parts for A.  smithii consisted of roots, rhizomes, and crowns
to a depth of 15 cm.   Weaver (1958) reported 55 percent of the root weight of A.
smithii was found  in  the top 15 cm of soil.  Jorgensen (1970) observed similar
root patterns in central Montana.  Below-ground parts of K.  cristata consisted
of roots and crowns also collected to  a depth of 15 cm.   Below-ground parts will
be referred to as roots.   Roots were washed with tap water to remove soil prior
to preparation for sulfur analyses.

     There were  ten   sample  periods  on  ZAPS  I from early  April  through mid-
October  and  nine  sample periods  on ZAPS II  from  early April through  mid-
September.  Sampling  was  done at  approximately  two-week  intervals.   Sulfur
analysis was  performed with the Leco induction  furnace.

Collection Methods at Drift Sites

     Forty marked  subplots  were  located  in the area adjacent to the treatment
plots on ZAPS I (Figure 13.2).  These will  be referred to as the drift plots.  A.
smithii was collected in mid-April, mid-July,  and mid-September.  The mid-April
A. smithii collections  included  dead  tops only.  K. cristata was collected in
mid-July and mid-September.

Phenology of Agropyron smithii

     In mid-May, 50 individual culms of A. smithii were tagged and numbered on
each treatment plot of ZAPS  I and ZAPS  II.   The culms were arranged in ten groups
of  five  on each  treatment plot.  These  groups were located in the following
subplots (Figure 13.IB):

                               1-SE      14-NE

                               5-NE      17-NE

                               7-NE      19-NE

                               9-NE      21-SW

                             12-NE      25-NW

     The culms were marked with a numbered staking flag two inches north of the
respective culms.  Each culm was encircled by a twist-tie which  led  to the base
of  the  flag.   It is felt that  neither   the  staking  flag  nor  the  twist-tie
interfered with  the  growth of the culm.    Jorgensen  (1970)  utilized a similar
                                      403

-------
                                                                         100  METERS
o
-p.
                                         Figure 13.2.   ZAPS I drift plot locati-
                                                                               ons.

-------
method of marking individual A. smithii culms and also concluded the effect was
negligible.

     The  specific  culms were  selected by measuring  five  feet north from the
appropriate junction along the pipe and then  five feet east (west on subplots 21
and  25).  The  A.  smithi i culm closest  to  this  point was #3 for that subplot.
The  culm closest to a point six inches  west  of this was #2, and 12  inches west
was  #1.  Similarly  #4  and #5 were  selected  on  the east side  of #3.  This was
repeated ten times  on each treatment plot.

     Data  collected for the  first period (mid-May)  were  the  total number of
emergent  leaves,  excluding totally  necrotic basal  leaves.   The twist-tie was
placed on  the  adaxial  side of the  highest totally necrotic blade  if any were
present.   Subsequent  inspections  involved  recording the  emergent  leaf-blade
number and  the number  of total necrotic  leaf-blades  on each  culm  at approxi-
mately two-week  intervals.  Sampling was discontinued in mid-September.  There
were eight observation periods  on ZAPS I and seven observations  on ZAPS II.

Seed Collection and Viability Studies

     Seeds of  A. smithii, K. cristata,  £. sandbergii, T.  dubius, and S. viridula
were collected from each treatment  plot on ZAPS I and II.  Seeds of A. smithii
and  T.  dubius  were  also  collected  from  a  similar vegetation community
approximately  1.6 km southwest  of the ZAPS sites.

     Each seed collection was pre-treated according to the technique utilized by
Eddleman (1977) for each species to  enhance  germination.  Tetrazolium staining
was  conducted  on the  seed collections to provide  seed viability  information
according to the  rules of the Association of Official Seed Analysts (1970).

     Germination tests were conducted to determine the percent germination per
gram weight of seeds collected.  Weights of  100  seeds were  calculated to adjust
the  gram weights in the  event that seed size is reduced as a result of the S02
treatments.  Two millimeters  of hypocotyl  growth were  considered successful
germination.   Seed  germination was  checked on a daily basis for 30  days.  From
this information, an  "energy  peak" and percent germination was determined for
each seed sample.  Analysis of variance and Duncan's (1955) multiple range test
are being conducted  on the results.

     Seed germination and viability investigations are not far enough along at
this time  to  present results  or discussion  on  this  portion of  the 1976  ZAPS
studies.

                                    RESULTS

1975 SULFUR AND FLUORIDE  ACCUMULATION AT ZAPS I

     On Table  13.1  are the sulfur and fluoride concentrations in foliage of ten
species of grasses, forbs,  and shrubs  collected from the  ZAPS I plots during
1975.  Only two grass species (i.e. , A.  smithii and B.  japonicus) were collected
during both the June and September collection periods.  Sulfur levels in these
two  grass  species  increased  four  times in  A.  smithii and three  times  in B.


                                      405

-------
TABLE 13.1.  ZAPS PPM SULFUR (S) AND PPM FLUORIDE (F) LEVELS FOR 1975.
ZAPS I
Agropyron smithii
Koeleria cristata
Achillea millefolium
Bromus japonicus
Poa pratensis

-pa
• — >
CTl
ZAPS I
Agropyron smithii
Bromus japonicus
Aristida longiseta
Artemisia tridentata
Bromus tectorum
Stipa comata
Gutierrezia sarothrae
28 June 1975 (31
A (CONTROL) B (LOW)
S F S
1200
800
600
1175
600


3.6
4.6
3.2
2.7
1.4


28 September
A (CONTROL]
S F
850
700
700
1300
400
900
1700
1.0
2.5
0.8
1.6
0.5
0.9
2.4
1500
1050
900
1200
1100


1975 (1
B (LOW)
S
1800
1600
1700
1200
1500
1800
1500
Days Fumigation)
C (MEDIUM)
F S F
2.
3.
3.
4.
2.


23
1.
1.
1.
2.
0.
1.
1.
6
9
3
2
8


Days
F"
4
0
6
4
5
0
9
1600
1200
1200
2200
1200


2.7
2.0
2.3
2.9
2.7


Fumigation)
C (MEDIUM)
S F"
3400
2400
1300
1400
2100
2800
2950
0.8
0.4
1.7
1.8
1.7
1.4
1.0
D (HIGH)
S
2000
800
2200
2550
1400


D (HIGH)
S
5250
3400
3300
2750
4150
2800
5300

1
1
3
2
1



0
1
0
1
0
0
1
F"
.8
.3
.5
.8
.9


F~
.8
.1
.1
.2
.8
.8
.3




ZAPS II Pre-Treatment
S F
900 0.8
600 1.3
800 1.2
1200 2.1
400 1.5
800 0.1
2200 1.8

-------
japonicus, while  the  fluoride levels  decreased  during  this  90-day  period.
Sulfur levels in  foliage  of all nine species collected during the latter part
of September  at the  four  fumigation concentrations  demonstrated  that sulfur
accumulation  correlated with fumigation concentration and duration of exposure.

     Because  the vegetation species on the ZAPS I plot survived remarkably well
during the 1975 growing  season after continuous  fumigation with S02,  a large
portion of the  1976  studies was redirected  from  the  ponderosa pine-skunkbush
sites to  the ZAPS I and II  sites.

1976 SULFUR ACCUMULATION AT ZAPS I  AND II

     Data analysis of  sulfur  accumulation in above-ground vegetation from the
ZAPS sites does not  include the ninth (mid-September) and tenth (mid-October)
1976 collection periods.   However,  the analysis covers the major portion of the
growing season.  This is confirmed by the 1976 phenology studies on A.  smithii
and  the   1975  phenology  studies  conducted  by personnel  of  Colorado  State
University  (Dodd   et  aj. ,  1976)  for  P_.  sandbergii,  K.  cristata,  Aristida
longiseta, A.  smithi i,  T.  dubius, and Achillea mi 1lefolium.   Dormancy is not
induced in Artemisia frigida  until later in  the  season,  as  is typical of the
sages, because  this  species is able to extract water  from greater depths.  In
the  presence  of  adequate  fall  rains,   many prairie  species  resume  growth.
Sulfur levels  in  live and dead tops  for  the six species collected throughout
the study period are  presented in Appendix A.

     The  changes  in  sulfur levels  through  time were analyzed by  fitting  a
linear regression to the data by least squares.  An initial  peak in the sulfur
concept  of live  tops from the first  collection period was  observed.   This
spring green-up peak was  noted in all species.  To allow preliminary use of a
simple linear model,  this initial  collection was excluded from the calculation
of subsequent live regression lines.   The coefficients for the regression lines
are presented  in  Table 13.2.   Of  the 80  regression  coefficients examined,  38
were found to  be significant at p ^.10  (Table  13.3).

     The  number of  significant regressions  increased with  higher  levels  of
fumigation.  Although only  two  of  the regressions on the A plots were signif-
icant, 18 of the 20 regressions showed a positive slope (p <.001), suggesting a
trend of  increasing  sulfur  levels in vegetation on the control plots from the
beginning to  the  end of the growing season.  There was a substantial increase
in the average value of the Coefficient of Determination (R2)  from the A plots
(.349) to the B  plots (.635).  The  increase from the B's (.635) to  the C's
(.683)  and  subsequently  from  the  C's  to   the  D's  (.743)  was  less.   This
degeneration  of  the   increase  in  the goodness  of fit  with  higher fumigation
levels was due to the fact that many species reached a marked  plateau of sulfur
accumulation  in late June through July.   This phenomenon will  be examined when
individual species are considered.

     The  regression   lines  for live  tops  for six  species  are presented  in
Figures  13.3  through  13.8 (regression  was  not computed  for  P.  sandbergii
because  it was collected  during  only three sample  periods).   The  trends  in
sulfur accumulation can be quickly summarized from these graphs.  Sulfur levels
in live  vegetation  increased  with exposure  time.  The  rate  of increase rose


                                      407

-------
TABLE 13.2.   REGRESSION COEFFICIENTS FOR SULFUR LEVELS THROUGH TIME.


Plot
Z1A
Z2A
Z1B
Z2B
Z1C
Z2C
Z1D
Z2D

Plot
Z1A
Z2A
Z1B
Z2B
Z1C
Z2C
Z1D
Z2D
* =
t = .
t = p

b
o
288
924
250
992
102
120
-792
1410

bo
479
660
-218
307
5
219
-19
-1110
05 < p
01 < p
' < .01

bl
3.27
-0.13
6.00
2.06
9.96
10.20
18.53
11.19

bl
1.37
0.25
8.38
4.38
10.86
10.08
16.58
22.12
< .10
< .05
Live
F
3
0
44
1
13
13
775
1
Live
F
1
0
18
3
22
9
5
7


Agropyron smithii
s
.943
.011
.500;j;
.501
.633t
.839t
.724t
.813

s
.242
.030
.022!
.927
.64?!
.562t
.394*
.257t


R2
.441
.002
.889
.231
.731
.735
.994
.266
Koel

R2
.199
.006
.783
.440
.819
.657
.519
.592


n
7
7
7
7
7
7
7
7
en" a

n
7
7
7
7
7
7
7
7


b
o
17
-15
-433
-287
-917
-1638
-1479
-2346
cristata

b
0
530
382
-288
12
27
-509
-8
-1246


bl
3.90
3.69
8.88
8.97
15.47
20.76
22.50
30.27

bl
0.31
1.10
7.73
5.56
8.17
11.19
11.32
19.66


Dead
F
3.
141.
13.
16.
11.
67.
5.
7.
Dead
F
0.
3.
136.
30.
8.
19.
20.
11.



s
891
767t
503t
006*
231t
929t
731*
909

s
109
315
236t
128t
187*
730t
310*
659*



R2 n
.565 5
.986 4
.818 5
.889 4
.789 5
.971 4
.656 5
.798 4

R2 n
.035 5
.624 4
.978 5
.938 4
.732 5
.908 4
.871 5
.854 4


(continued)
                                 408

-------
TABLE 13.2.  (Continued)

Achillea mi lief oil urn
Live
Plot
Z1A
Z2A
Z1B
Z2B
Z1C
Z2C
Z1D
Z2D

Plot
Z1A
Z2A
Z1B
Z2B
Z1C
Z2C
Z1D
Z2D
bo
339
907
542
1223
230
852
-1883
-662

bo
1188
1074
2043
832
1087
-1116
971
-2
bl
1.78
-0.75
2.26
-0.10
8.14
5.75
26.68
25.37
Li
bl
0.91
1.19
-1.70
4.13
7.63
19.41
18.29
23.01
F
1
0
1
0
16
1
33
12
ve
F
0
0
0
1
3
58
6
26
s
.496
.351
.208
.001
.306t
.468
.4901
.81 Of

5
.291
.260
.672
.841
.428
.2771
.890*
.140t

R2
.230
.066
.195
<.001
.765
.227
.870
.719
Artemi

R2
.068
.049
.199
.274
.461
.936
.633
.867

n
7
7
7
7
7
7
7
7
sia

n
6
7
7
7
6
6
6
6

bo
634
494
456
298
410
-831
-451
-2228
frigida

bo
964
680
700
758
1087
-745
1860
-1989

bl
0.13
0.75
3.65
4.91
9.28
16.25
26.64
32.57

bl
1.41
0.89
4.45
2.83
7.63
16.00
9.05
29.78
Dead
F
3.138
2.194
1.035
8.253*
4.173
4.921
6.753*
14.6191
Dead
Fs
102.084*
0.198
10.197
20.120
12.228
9. 125
8.253
4.979

R2
.007
.422
.341
.733
.582
.621'
.692
.830

R2
.990
.163
.911
.953
.461
.901
.892
.833

n
5
5
4
5
5
5
5
5

n
3
3
3
3
3
3
3
3
* = .05 < p < .10



1 = .01 < p < .05



1 = p < .01                                               (continued)




                                     409

-------
TABLE 13.2.   (Continued)
Aristida longiseta

Plot
Z1A
Z1B
Z1C
Z1D

Plot
Z2A
Z2B
Z2C
Z2D
b —
0
R2 -
H :
0

b bn
o 1
-238 4.81
-654 7.99
1473 -0.21
112 9.44

b b,
0 1
577 2.20
673 3.21
479 9.03
1669 11.33
Y-Intercept,
Live
F R2
s
0.926 .481
33.624 .971
<.001 <.001
4.928 .831
Sti
Live
Fs R2
2.379 .373
0.914 .186
3.869 .492
0.812 .169

b-| Slope, FS
Coefficient of Determination,
p = 0, F
n-2]'

n b
0
3 207
3 455
3 191
3 -41
pa viridula

n b
0
6 348
6 -920
6 -1110
6 -1938
Mean Square
Unexpl

b
2
2
5
8

b
2
9
15
24
Dead
F R2
1 s
.19 3.151 .612
.74 132.9941 .985
.82 19.1061 .905
.32 118.4711 .983
Dead
F R2
1 s
.16 4.357 .685
.87 75.000* .987
.75 66.3941 -971
.11 116.062 1 .983

n
4
4
4
4

n
4
3
4
4
due to Regression
ained
n = number of col


Error '
lection periods




* = .05 < p < .10
t =  01 < p < .05
| = p < .01

  TABLE 13.3.  SUMMATION OF NUMBER OF SIGNIFICANT REGRESSION COEFFICIENTS.

Treatment
Live
Dead
A
0
2
B
2
7
C
6
7
D
7
7
                                      410

-------
    6000  -r
    5000  -
    4000
a
oo
s:
Q.
    3000  ~
    2000  -
    1000
ZAPS I

ZAPS II
          90
     120
                                          HIGH
 T
150
 I            1
180         210
  JULIAN DATE
                                                                        I
                                                                      240
                                                                  T
                                                                 270
  I
300
        Figure  13.3.   Least  squares, Agropyron smithii, live collections 2 thru 8, ZAPS I and II.

-------
ro
            6000  -
             5000  -
            4000  -
            3000  H
          Q:
          Q
rp
oo
         g- 2000
            1000  -
                 90
                            ZAPS I

                            ZAPS II
                      I
                    120
  I
150
  I            I
180         210
   JULIAN DATE
  I
240
270
  I
300
                Figure 13.4.   Least squares, Koeleria cn'stata. live collections 2 thru 8, ZAPS I and II.

-------
oo
            6000  -I
            5000  -
            4000 -
         g 3000
         CL,
         Q.
            2000  -
            1000  H
                                                                         HIGH
                                                       X
                                                                                  X
ZAPS I

ZAPS II
                  90
        I
      120
150
 I
180
  I
210
  I
240
  I
270
 I
300
                                                         JULIAN DATE
                Figure 13.5.  Least squares, Achi1 lea millefolium, live collections 2 thru 8, ZAPS I and II.

-------
   6000 -,
   5000 -
cc
   4000 ~
   3000 -
   2000  -
   1000 -
        90
120
150
1	T
180         210
  JULIAN DATE
 r
240
 T
270
1
300
       Figure  13.6.   Least squares, Artemisia frigida,  live  collections 2 thru 8, ZAPS I and II.

-------
-pa

cn
            6000 -i
            5000 -
            4000 -
         a  3000 -|
         co
         Q-
         QL.
            2000 -
            1000 -
                 90
                       ZAPS I
120
 I
150
                                                                                      HIGH
 !            i
180         210
 JULIAN  DATE
240
 I
270
 \
300
                Figure 13.7.   Least squares, Artisti'da longiseta, live collections 6 thru 8, ZAPS I.

-------
O1
            6000 -.
            5000 -
            4000 —
         o;
            3000  -
            2000  -
           1000  -
                          ZAPS II
                                                                HIGH
                                                                           MEDIUM
                                                                                   LOW
                                                                                           CONTROL
                 90
120
150
—i	r
 180          210
  JULIAN   DATE
—r
 240
270
—I
 300
                Figure 13.8.   Least squares,  Stipa  vin'dula, live collections 3 thru 8, ZAPS II.

-------
with  increasing levels of fumigation.  Rate  of  accumulation  and total sulfur
were  approximately  similar on ZAPS  I  and ZAPS II.  The rate of accumulation on
ZAPS  II  treatment plots D and  C  tended to be higher than on their respective
ZAPS  I counterparts.  The rate of accumulation on ZAPS II treatment plots B and
A tended to be lower than on their respective ZAPS I counterparts.

Agropyron smithii

      The observed sulfur  contents and  their  regression  lines  for A.  smithii,
live  and dead,  at the various  treatments  are plotted in Figures 13.9 through
13.12.  At treatment plot A fumigation  levels  the sulfur content of dead plant
material remained below that  of live  tissue throughout most  of the growing
season.  Sulfur levels in the dead material slowly approached the  live material
levels  and can be expected to  exceed these  only in  late summer  as the grass
comes  to a  full  cure.   As  the intensity  of  fumigation  increased,  the sulfur
level  in dead tissue exceeded that  in live tissue at an earlier date.  This was
indicated  by both  the  measured  sulfur  content  and  their  fitted regression.
Initial sulfur  levels in dead tissue were  higher on ZAPS I than on ZAPS II; the
sulfur  content  of live tissue was also higher on ZAPS I, except in vegetation
from the control plots.

      A  plateau  in the sulfur accumulation by live  tissue  was  reached by late
July,  except on ZAPS I, plot D, where the rate of accumulation remained linear.
This  plateau occurred  at  a  maximum sulfur level of  approximately 4,000 ppm.
Sulfur  levels  in dead tissue peaked  as  high  as  5,675 ppm on the D fumigation
plots.   On  May  31  A.   smithi i  was  collected from  small  areas  immediately
adjacent to  the  gas outlets on  ZAPS II, plot  D.  At this time these burn areas
were  between 45 and 60 cm in diameter.  The A. smithi i collected here displayed
bifacial necrotic lesions  coalescing  to form  interveinal streaks; the streaks
were  ivory to  light tan.  Tingey et aj.  (1975) has described these symptoms in
A.  smithii treated with 150 pphm  S02  in four-hour chamber fumigations.  The A.
smithii from  these  burn areas on ZAPS  II, plot  D contained 3,350 ppm sulfur.
Similar symptoms  were  observed on ZAPS II, plot C but the sulfur level in the
tissue  was  not  determined  specifically  for A.  smithi i  because  individual
species were not separated in this sample.

Koeleria cristata

     The responses  of  K.  cristata  to  the treatments are  graphed in Figures
13.13  through  13.16.  The  relationship  between sulfur levels in live and dead
foliage was  markedly  different  from that  which occurred for A. smithii.  With
J(.  cristata,  the sulfur  content of  dead  tissue did not exceed  that of live
tissue  based on  the  regression  analysis; when  individual  data  points  were
considered, sulfur levels in live material  were exceeded only rarely by levels
in dead tissue.  Sulfur levels in dead tissue  can be expected to approach those
in  live  tissue as  K.  cristata  comes  to a full  cure.   Sulfur  levels for all
tissue  were  similar  in  K.  cristata  and  A.  smithii, with  a  maximum  of
approximately 5,000 ppm  and a plateau developing later  in July.

     A  comparison  of   initial  sulfur  levels in dead  K.  cristata  showed  a
residual effect  similar  to   that of A.   smithi i.  Sulfur  levels  in  live  K.
cristata were  higher  at ZAPS I, plots A and B than at ZAPS II, plots A and B.


                                      417

-------
a;
CL
Q.
    6000  -H
    5000  -
    4000   -
    3000   -
    2000   -
    1000   -
                                   ZAPS I
              LIVE •_

              DEAD D
           90    120    150   180    210   240     270    300
o
cc.
Q
       6000   •

       5000   .


       4000   -
£   3000   -
00
   CL.
   CL.
    2000   ~

    1000

       0
              LIVE •_

              DEAD a
                                ZAPS II

                 D
90
~T—
 120
—I"
 150
T—
 180
—I—
 210
                                              1
                                             240
                                                       1
                                                      270
—I
 300
Figure 13.9.
                             JULIAN  DATE
           Linear regressions  of sulfur levels  through  time  in
           Agropyron smithii  at CONTROL fumigation.
                             418

-------
   C£
    6000  -




    5000  -




    4000  -





    3000  -
   n_
   o_
00   2000





    1000




       0
                LIVE  •_



                DEAD  D
                                  ZAPS  I
                                                D


                                                •
             90
                   120    150    180    210     240     270    300
  Q
  Q-
  d.
    6000   -.




    5000   -




    4000   _





    3000   -





    2000   -





    1000




       0
                                  ZAPS  II
                 LIVE


                 DEAD
                 	1	1	1	1—	1	1	1


             90     120   150    180   210     240    270     300


                                JULIAN DATE

Figure 13.10.   Linear  regressions of  sulfur levels through time in

               Agropyron  smithii at LOW fumigation.
                               419

-------
    6000  -•


    5000  _


E   4000  -
  UJ
  QL
  Q

  DC
  n_
  Q-
    3000   -
      2000   -
      1000   -
         0
                                  ZAPS I
                LIVE ^_

                DEAD n
                     i      r      i      i      i       i      •
             90     120   150   180    210   240    270   300
   Q-
   Q.
    6000


    5000   _


    4000   -


    3000   -


    2000   -


    1000


       0
                LIVE

                DEAD
                                  ZAPS II
Figure 13.11
            90    120   150     180    210    240   270   300

                           JULIAN DATE
             Linear  regressions  of sulfur  levels through time in
             Aqropyron  smithii at MEDIUM  fumigation.
                                420

-------
   CtL
   CL.
   Q.
6000   __.




5000   _




4000   -



3000   -




2000   -



1000   -
                                    ZAPS  I
                 LIVE •_


                 DEAD a
                90     120    150   180   210    240    270   300
6000



5000   -



4000   —
   a   3000
   00
   a.
   Q-
2000




1000



   0
                                    ZAPS  II
                  LIVE  •


                  DEAD  n
Figure 13.12.
        90    120     150    180     210   240   270   300


                         JULIAN  DATE

        Linear regressions of sulfur  levels through time in
        Agropyron smithii at HIGH fumigation.
                                421

-------
      6000  -i





      5000  -

t—
zc          —


5    4000  —
   Q
   CO
   CX
         3000   _
         2000   _
         1000   _
                                     ZAPS I
                   LIVE



                   DEAD
                                              	•   M.
                 90    120   150   180    210   240    270   300
      6000   -,





      5000   „

\—
re            	


3     4000   _
   Q



   CCL
   O-

   Q-
      3000   _





      2000





      1000





         0
                   LIVE



                   DEAD
                                     ZAPS II
                    —r——i	1	1	1	1	1

                  90   120    150   180    210   240    270   300


                                  JULIAN DATE

Figure 13.13.   Linear regressions  of  sulfur  levels  through time  in

               Koeleria cristata  at CONTROL  fumigation.
                                422

-------
  ce.
  Q

  o:

  u_
  CO
  o_
  o_
      6000   -I
      5000   -
      4000   -
3000  -
      2000  -
      1000  -
                                    ZAPS I
                LIVE •
                     •••



                DEAD D
                                   D
              90
 I


120
                     I

                     150
180
 T


210
  i


240
270   300
      6000  -i
      5000
  CD

  Hi   4000
  a:
  a
  CsL
  GO
  D_

  Q_
      3000
      2000
      1000   -
           LIVE •_



           DEAD a
                                    ZAPS II
Figure 13.14.
         90    120    150    180   210   240    270    300



                       JULIAN DATE

         Linear  regressions  of  sulfur  levels  through  time  in

         Koeleria cristata at  LOW fumigation.
                                 423

-------
   CtL
   n.
   o.
       6000   -,
       5000   _
       4000   -
       3000   -
       2000   -
       1000  -
   0







6000




5000
                            ZAPS  I
          LIVE •_


          DEAD D
              90    120    150   180    210   240   270    300
   CD
   uj   4000  -
   cc.
   Q
   o:
   oo
   0.

   CL.
3000   -





2000




1000




   0
                                   ZAPS  II
                 LIVE •_


                 DEAD n
Figure 13.15.
        90    120    150     180   210    240   270   300


                         JULIAN DATE


        Linear regressions of sulfur levels through time in

        Koeleria cristata at MEDIUM fumigation.
                                424

-------
   ex.
   D_
6000  -,



5000  -




4000  -



3000  -



2000  -



1000  -
          0
                                   ZAPS I
                 LIVE  •_


                 DEAD  a
                 	1	1	1	r	1	1	1

               90     120   150    180   210   240    270    300
   CD
   CXL
   Q.
   D-
6000  -H



5000



4000  -



3000  -



2000



1000



   0
                                     ZAPS  II
                  LIVE •_


                  DEAD a
Figure 13.16.
       90     120    150    180   210   240    270   300

                        JULIAN DATE

        Linear regressions  of  sulfur  levels through time in
        Koeleria cristata at HIGH fumigation.
                                425

-------
Levels in K.  cristata were lower initially at ZAPS I, plots C and D than at the
same ZAPS ~II  plots,  and the green-up sulfur levels at these ZAPS I sites were
among the lowest observed in K.  cristata.

Achillea millefolium

     Initial   sulfur  levels  for A.  millefolium, live and  dead,  were  higher on
ZAPS  I,  plots B,  C,  and D  than  on the pre-treatment ZAPS  II  plots (Figures
13.17 through  13.20).   On ZAPS I,  plots B, C,  and D,  the initial dead levels
exceeded the  live.   This  relationship  continued throughout the sample period.
On  ZAPS  II the  sulfur  levels  in  dead  leaves  did not exceed  the live leaves
until  later   in  the  season.   A July plateau  was observed,  but the  rate  of
accumulation  on the D plots again began to increase.   Maximum values  in excess
of 6,000 ppm were observed in both  live and dead tissues.

Artemisia frigida

     The  observed   levels  and  regressions  for A.  frigida  are  presented  in
Figures  13.21  through  13.24.   Sulfur levels  in the live leaf material of this
species  exceeded  that  of the  dead material,  and  this  relationship  continued
throughout the period covered.   A  residual  S02  effect was  evident on ZAPS I and
very  pronounced  on the C and  D plots.   The  plateau  was  less  distinct  for  A.
frigida;  the   process  of accumulation  continued  until  later in the growing
season on the  higher fumigation plots.

     Baseline  levels for A.  frigida were the highest of any of the species that
have  been monitored continuously  on  the  ZAPS   plots.   The baseline  level
(treatment A)  was  in excess of 1,000 ppm  and  slowly increased throughout the
season  with   peaks  of  as  high  as  1,550  ppm.   This  high  baseline made  a
substantial contribution  to the  higher  levels observed  throughout  the  year.
Maximum  peaks  for  live  and  dead material appeared to be in the 5,000 to 6,000
ppm range and were reached by early July on the D treatment plots.  The rate of
accumulation was  higher  on the ZAPS II plots.

Aristida longiseta

     A.   longiseta  was  available in sufficient quantities for sampling only on
ZAPS  I.   These   data   are  portrayed   in  Figures  13.25  and  13.26.    The
relationship   between  live and  dead  was  typical   of   the  other  perennial
bunchgrasses studied (K. cristata  and S.  viridula).  Live  levels exceeded those
of the dead throughout the study period.   A residual  S02  effect was seen in the
dead samples  collected  before  the 1976 fumigation was started.  Live material
could not  be  collected  until  early July because A. longiseta is a warm season
grass.

Stipa viridula

     S.   viridula  was  collected only  on  ZAPS II.   Live sulfur  levels  were
generally in excess of those in the dead tissue (Figures 13.27 and 13.28), but
the dead approached the  live  at   an  increasing rate as  the  fumigation level
increased.    Live   tissue  reached  a  mid-July  plateau  on  the  plots  under
intentional fumigation.  This  was  followed by  a decline  in  sulfur  until  mid-


                                     426

-------
      6000  _
      5000  -
  :n
  CD
  LU  4000  -
  C£.
  Q
  CO
   Q_
   CL.
      3000  -
      2000  ~
      1000  -
LIVE •_

DEAD n
                    i       i      »i      i      i      i
             90   120     150     180   210    240   270   300
      6000  -,
      5000  -
      4000  -
      3000  -
      2000  -
      1000  __
                   ZAPS  II
LIVE •_

DEAD n
                    i      i       i      i	1	n	1
             90    120    150     180    210    240   270   300

                              JULIAN  DATE

Figure 13.17.   Linear regressions of sulfur  levels  through time in
               Achillea  millefolium at CONTROL  fumigation.
                                427

-------
    6000  -i


    5000
i—
CD
§   4000  -I
>-
Q
^   3000  J
   CO
   o_
   (X
    2000  _


    1000


       0
                 LIVE •_

                 DEAD a
                                   ZAPS  I
              90    120   150    180   210   240    270    300
       6000  -,
       5000  -
   oz
   CD
    4000  ~


o:  3000  -
   o;
   Q
   Q-
   D.
    2000  _


    1000  „


       0
                 LIVE ^_

                 DEAD a
                                ZAPS II
              90    120    150   180    210     240    270   300

                              JULIAN DATE
Figure 13.18.   Linear regressions  of sulfur  levels  through  time  in
               Achillea millefolium at  LOW fumigation.
                                428

-------
     CD
     D_
     Q.
6000  _
5000  _.
        4000  _
        3000  -
         2000  -
         1000  -
                  LIVE •
                  DEAD a
                                    ZAPS I
         D
         •
                90     120    150     180   210    240   270   300
         6000  _
         5000 _
     ~   4000  J
     a:
     O
     o.
3000  -


2000  -


1000  -
                           ZAPS II
          LIVE

          DEAD
               90    120    150   180    210   240    270   300

                                JULIAN DATE
Figure 13.19.   Linear  regressions of  sulfur  levels through time in
               Achillea  millefpliurn at MEDIUM fumigation.
                                429

-------
   Q.
   Q-
6000  -,


5000  —


4000


3000  •


2000


1000
         0
         D
             90    120   150    180   210    240   270   300
   ID
   OO
   D.
   CL.
6000  -,


5000  -


4000  -


3000  -


2000  -


1000  _
                               ZAPS  II
                LIVE
                DEAD D
                    i      i       i      i      i       i      I
             90   120    150   180    210   240    270   300

                             JULIAN DATE
Figure 13.20.   Linear  regressions of sulfur  levels through time in
               Achillea mi liefoil urn at HIGH  fumigation.
                                430

-------
     CD
     Qi
     Q-
     CL.
    6000 —


    5000 _


    4000 -


    3000 _


    2000 -


    1000 -
            0
                   LIVE _•_

                   DEAD  n
                                     ZAPS  I
                                                  •
                                                  "a
                 a-
               90    120    150   180    210   240    270   300
    6000  -,


    5000  .
i—
m
S   4000  _
     o:
     Q_
     Q-
         3000 -
         2000 -
         1000 -
                                   ZAPS  II
                   LIVE •_

                   DEAD D
Figure 13.21.
          90    120    150   180    210    240   270    300

                          JULIAN DATE

          Linear regressions  of sulfur  levels through time in
          Artemisia frigida  at  CONTROL  fumigation.
                                431

-------
6000  —

5000  -
    £2   4000 _
    UJ
    o:
    CSL
    D.
    Q_
3000  —

2000  _

1000
           0
                                    ZAPS  I
                  LIVE
                  DEAD
      90     120   150
                                 —I	T~
                                  180   210
-T	1	1
 240   270   300
    LU
    OL
    Q
    Q_
    Q.
6000  -

5000  -

4000  ~

3000  -

2000  -

1000  —
                  LIVE
                  DEAD
                                   ZAPS II
                                             •	 a
                 a-
Figure 13.22.
          	1	1	1	1	1	1	1
       90      120   150   180    210   240    270   300
                         JULIAN DATE
       Linear regressions of sulfur levels through time in
       Artemisia frigida at LOW fumigation.
                                432

-------
6000  -r



5000
    o

    5  4000
    C£.
    Q
    00
    D-
    D-
3000



2000



1000



   0
                  LIVE •_

                  DEAD D
                                    ZAPS I
                       I       I      i       I      i       i      i

               90    120     150     180    210     240    270    300
6000   ~



5000   -
    o
    £  4000   _
    Q

    DS  -3000
    oo
    CL.
    Q.
2000



1000



   0
                                   ZAPS II
                  LIVE
                  DEAD D
Figure 13.23.
        90    120   150     180    210     240   270    300

                          JULIAN  DATE

       Linear  regressions  of sulfur  levels through time in
       Artemisia  frigida at  MEDIUM fumigation.
                                433

-------
   CZ3
   o:
   C/O
   D-
   Q.
       6000  -i
       5000  -
       4000 -
       3000 -
       2000 _
       1000  -
                            ZAPS I
              90   120    150    180   210    240   270   300
   IS)
   D-
   Q-
6000



5000  _



4000  -



3000  -



2000



1000
          0
                                  ZAPS II
                 LIVE

                 DEAD
                                                 D
Figure 13.24.
          	1	1	r-	1	1	—T	r
       90    120    150   180    210    240    270   300

                         JULIAN DATE

        Linear regressions  of sulfur levels through time in
        Artemisia frigida at HIGH  fumigation.
                                434

-------
 Q


 C£
 OO
 D-

 Q.
6000   -,





5000   _





4000   -





3000   -





2000





1000




    0
                        ZAPS I — CONTROL FUMIGATION
                 LIVE •
                 DEAD a
90    120   150    180    210
                                               240
                                               270
                                              300
  o;
  Q
  oo


  a.
  a.
       6000   -i
       5000   -
       4000   -
       3000    -
 2000   -
        1000    -
           0
                     ZAPS  II —LOW  FUMIGATION
            LIVE ^_



            DEAD n
                                                      270
                                                       300
Figure 13.25.
 90    120   150    180   210   240



                 JULIAN DATE


Linear regressions of sulfur levels  through  time  in

Aristida longiseta at CONTROL and  LOW fumigations.
                                435

-------
    6000  _






    5000  _



i—         ~
:DC

2   4000  -
UJ
3

>-


°   3000  -

a:
^>
u_
_i

^   2000






    1000





       0
   D-

   O.
                            ZAPS  I—MEDIUM  FUMIGATION
                 LIVE •_


                 DEAD a
                                                  n
               90     120    150   180    210   240
                                                   270   300
        6000
                          ZAPS II--HIGH FUMIGATION
    a:
    a
    Q-

    Q-
        5000   _
        4000   -
        3000   -
        2000   -
        1000   -
           0
                  LIVE •
                  DEAD a
                                         --a
                          i
                                    1
               90    120    150   180     210    240    270   300


                                JULIAN  DATE


Figure 13.26.   Linear regressions  of  sulfur  levels  through time  in

               Aristida longiseta  at  MEDIUM  and  HIGH  fumigations.
                                436

-------
   O-
   o.
6000  .



5000



4000  _




3000  -




2000  -




1000  -
                         ZAPS II--CONTROL  FUMIGATION
                 LIVE


                 DEAD
              90    120    150   180    210   240   270    300
   OL
   Q
       6000   -,
       5000   -
       4000   -
       3000   -
       2000   -
       1000   -
                  ZAPS  II —LOW FUMIGATION
          LIVE •_


          DEAD a
              90    120    150   180    210    240     270    300

                               JULIAN DATE

Figure 13.27.   Linear  regressions of sulfur levels through time in
               Stipa viridula  at CONTROL and LOW fumigations.
                                437

-------
01
Q
ZD
in
       6000   -r


       5000   _


       4000   -


       3000   -


       2000


       1000    -


           0
                      ZAPS II—MEDIUM FUMIGATION
             LIVE

             DEAD
            90    120   150    180   210    240   270    300
    6000   -T
        5000
    i—

    E  4000
oo
    3000
    2000
    1000
                          ZAPS  II —HIGH FUMIGATION
               LIVE •
               DEAD n
               	T~
            90    120
                            —i	r—
                             150   180
-i	1—
 210   240
~I	1
 270   300
Figure 13.28.
                             JULIAN DATE
           Linear  regressions of sulfur levels through time in
           Stipa viridula at MEDIUM and HIGH fumigations.
                            438

-------
August when  the levels  rose a  second  time.   Maximum  observed  level  in live
foliage was 4,850 ppm at the D fumigation levels.

Poa sandbergii

     The measured sulfur levels for P_.  sandbergii are graphed in Figures 13.29
through 13.32.   Observed levels in both live and dead tissue increased with the
fumigation level and  tended to increase with time.  The maximum level  in both
live  and  dead tissue  was  approximately 2,400 ppm.  A  limited  number  of data
points precluded regression computation  and interpretation of trends.

Tragopogon dubius

     T. dubius was collected only in early July.  The measured sulfur levels in
the  leaves, taproots,  and  stems are presented in Table  13.4.  The small sample
size  precluded  testing for  significance.  T. dubius  showed increasing sulfur
levels  in  all  plant  parts with  higher levels  of  fumigation.   Sulfur levels
decreased  in the following order:  leaves, stems, taproots.  The sulfur levels
in  this plant  exceeded those  in any  others  that  have  been studied  by  the
investigators.

Roots                         r

     A. smithii  roots  collected in /early April  were improperly stored prior to
preparation  for sulfur   analysis.   Molding  caused  a  large  reduction  in
measurable sulfur  levels.   These  data are not presented.  Root collections of
A. smithii and K. cristata made in early July were first tested for homogeneity
of variances  (Table  13.5).   As  the variances could be assumed equal, an anova
was  performed  (Table  13.6).  A  significant difference  in  root  sulfur levels
across  the treatment  plots  could not  be  detected for  either  species during
early July.

     A. smithii  and  K. cristata roots  (ZAPS I and ZAPS II) from mid-September
(September 17,  18,  19; Collection Period 9) displayed  equal  variances (Table
13.5).  Anova  indicated  significant  (p S.05)  differences between  means  for
individual species subjected to  the different fumigation levels (Table 13.7).
The  observed differences in mean sulfur levels were inspected for significance
(p ^ .05)  by the Least Significant Difference  (LSD)  method (Table 13.8).   A.
smithii roots from ZAPS I, plot D had significantly more sulfur than roots from
the  other  three treatments.  On  ZAPS  II,  plot D, sulfur  in  A.  smithii  roots
also  exceeded  levels  in roots from the other three treatments.  Additionally,
the  sulfur content of roots from plot  C  was  significantly greater than those
from  plot  A.   In  the  case  of  K.  cristata, roots  from ZAPS I,  plot  D,  were
significantly  higher  in   sulfur  content  than  roots  from  the  other  three
treatment  plots.   Roots from treatment C also  had  significantly more sulfur
than those from  treatment plot B.  K.  cristata roots from ZAPS II, plot D, also
had significantly higher sulfur content  than those from plots A, B, and C.

Drift Areas

     The   results   of  the   drift   plot   collections   (ZAPS  I)  (including
mid-September)  are  presented in  Table  13.9.   Several  significant differences


                                      439

-------

>_
CD
i— i
UJ
3
>-
OL
Q
ZD
U-
s:




fe
CD
H-4
li 1
3
o;
Q
Qi
§
00
s:
O-
Q_


6000 — j
5000 _
_
4000 _
3000 -
2000 _
1000 _
0 -
9
ZAPS I
LIVE •
DEAD n




• ' • D
! 1 1 1 1 II
0 120 150 180 210 240 270 300
JULIAN DATE
6000 -n
5000 —

4000 -

3000 —
_
2000 _
1000 _
0
c
LIVE • ZAPS II
DEAD n







a
I 1 I 1 	 1 	 1 	 1
10 120 150 180 210 240 270 300
Figure 13.29.
                  JULIAN  DATE
Sulfur levels through time in Poa sandbergii  at CONTROL
fumigation.
                                440

-------
o
UJ


OL

C£


00

Q_
6000  -,

5000  —


4000  —


3000  -
    g: 2000  -


       1000  —

          0
                 LIVE •

                 DEAD D
90
                     1      1
                    120   150
                              1
                             180
                                  1
                                 210
 1      1
240   270
                                             300
6000 -
" ^
| 5000 -
i__<
UJ —
3
£ 4000 -
Q
D;
— i
j 3000 -
oo
SI
2: 2000 -
^F—
1000 -
0

ZAPS II
LIVE •
DEAD D







' . • D

i r i ' i i i
             90     120   150    180   210   240    270    300
                                JULIAN DATE
Figure 13.30.   Sulfur  levels through time in Poa sandbergi i at  LOW
               fumigation.
                               441

-------
UJ
3


a:


a:
rs
LJ_
_i
ZD
00
       6000  —i
       5000  —
       4000  -
       3000   -
       2000   -
       1000   _
                                   ZAPS  I
               LIVE*


               DEADD
                      I       i      i       i      i      i       '
              90     120   150   180     210    240    270    300
        6000   —,
    H-   5000   _
    CD
    t—i
    UJ
    Q

    a:
        4000   _
        3000
        2000   -
        1000
                               ZAPS  II
                LIVE •


                DEAD D
Figure 13.31.
                   i	1	1	1	,	,	
            90    120    150   180     210    240   270

                               JULIAN  DATE

           Sulfur levels through time in Poa sandbergii  at MEDIUM
           fumigation.
                               442

-------
      6000  —i
      5000  —
   CJ
   I—I
   LU
   4000  -



£  3000  _
   OL
   Q
   £  2000
       1000 -
          0
              LIVE •

              DEAD a
                                  ZAPS  I
                    T      I       I      I      I      I      I

             90    120    150    180   210    240   270   300
       6000  —r
       5000 —
       4000  _
       3000
       2000  —I
       1000  -
                              ZAPS II
              LIVE •

              DEAD a
              90    120   150    180   210    240   270    300
                               JULIAN DATE

Figure 13.32.  Sulfur levels through time  in  Poa  sandbergi i at HIGH
              fumigation.
                                443

-------
TABLE 13.4.   SULFUR (ppm)  IN Tragopogon dubius  FROM EARLY JULY

Treatment
Z1A


X
Z1B


X
Z1C


X
Z1D

(col lection
Leaves
4775
4450
3100
4108
6675
5750
7850
7125
8625
5750
7850
7408
5700
9450
period 5)
Taproot
1950
1550
1150
1550
1550
1900
1400
1617
1550
1450
1750
1583
1150
2450

Stem
-
-
850
-
-
-
1825

-
-
2000

-
3150
                       7575
1800
                              444

-------
TABLE 13.5.  MEAN SULFUR CONTENT AND BARTLETT'S TEST OF HOMOGENEITY OF
             VARIANCES FOR ROOTS FROM A. Smithii AND K. Cristata.


Collection Period 5, Early July.
A. smithii
Treatment
Z1A
Z1B
Z1C
Z1D

x ppm
780
720
790
811


df=n-l
9
9
9
8
X2 = 5

Collection Period
A. smithii
Treatment
Z1A
Z1B
Z1C
Z1D
Z2A
Z2B
Z2C
Z2D

x ppm
730
738
720
830

720
775
790
870

df=n-l
9
9
9
9
X2 - 2
9
9
9
9
X2 = 4
S2
9,565
7,327
6,006
1 ,739
.28

x ppm
792
789
805
725


9, Mid-September,
S2 x ppm
3,481
6,889
2,916
6,241
.27
1 ,764
3,969
6,561
7,921
.92
765
725
835
970

738
775
825
935

1976
K. cristata
df=n-l
9
8
9
9
X2 =

1976
K. cristata
df=n-l
9
9
9
9
x2 -
8
8
8
9
X2 =
S2
6,115
11,109
6,368
7,921
0.94

S2
14,400
6,241
13,917
7,921
2. 15
14,884
5,041
6,400
13,924
3.13
                   Critical Chi-square x2 -05 [3] = 7.82
                                   445

-------
TABLE 13.6.   ANOVA OF SULFUR CONTENT OF ROOTS  FOR Agropyron smithii  AND
Koeleria cristata FROM ZAPS I IN EARLY JULY.
(collection period 5)


Agropyron smithii
Source of Variation Sum of Squares Degrees of Freedom
Treatments 44,470 3
Residual 219,889 35
Total 264,359 38
Critical F Q5[3>35] = 2.875
Koeleria cristata
Source of Variation Sum of Squares Degrees of Freedom
Treatments 38,542 3
Residual 272,451 35
Total 310,993 38
Critical F_05[3)35]= 2.875
Mean Square
14,823
6,283
FS = 2.359
Mean Square
12,847
7,784
F =1.650
                                  446

-------
TABLE 13.7.   ANOVA OF SULFUR CONTENT OF ROOTS FOR A.  smithi1 AND K.  cristata
             FROM ZAPS I AND ZAPS II IN MID-SEPTEMBER.
                             (collection period 9)

Source of Variation
Treatments
Residual
Total
Critical F Q5[3
Source of Variation
Treatments
Residual
Total-.
Critical F;05[3
Source of- Variation
Treatments
Residual
Total
Critical F_05[3
Source of Variation
Treatments
Residual :
Total
Critical F Q5[3
Agropyron smithii ZAPS
Sum of Squares Degrees
77,797
174,562
252,359
,36] - 2-88
Koeleria cristata ZAPS
Sum of Squares Degrees
382,750
347,188
729,938
.36]. = 2'88 :
Agropyron smithii ZAPS
Sum of Squares Degrees
115,188
182,250
297,438
,36] = 2'88
Koeleria cristata ZAPS
Sum of Squares Degrees
202,250
309,000
511,250
= 2 92
,30] ^'^
I
of Freedom
3
36
39

I
of Freedom
3
36
39

II
of Freedom
3
36
39

II
of Freedom
3
30
33

Mean Square
25,932
4,849
F = 5.348
Mean Square
127,583
9,644
F = 10.885
s
Mean Square
38,396
5,063
F = 7.584
Mean Square
67,417
10,300
F = 6.545
                                      447

-------
TABLE 13.8.   DIFFERENCES IN SULFUR CONTENT OF ROOTS COLLECTED IN MID-
             SEPTEMBER COMPARED BY LEAST SIGNIFICANT DIFFERENCES.
       Agropyron smithii  ZAPS I
     LSD
      63
.05
     720
      C
     730
      A
     738
      B
     830
      D
720
 C
      10
      18
     no
730
 A
     100
738
 B
      92
830
 D
                                     Koeleria  cristata ZAPS  I
       Agropyron smithii ZAPS II
     LSD
      65
.05
     720
      A
     775
      B
     790
      C
     870
      D
720
 A
      55
      70
     150
775
 B
      15
      95
790
 C
      *
      80
870
 D
                                     Koeleria  cristata  ZAPS  II
LSD
104
.05
                                   738
                                    A
                              775
                               B
                         825
                          C
                    935
                     D
738
 A
                                   47
                              87
                         *
                        197
775
 B
                              50
              160
825
 C
               110
935
 D
 ''Signif icant at p£. 05
              TABLE 13.9.   SULFUR LEVELS IN DRIFT PLOT COLLECTION.
Col lection
   Date

 13 April
 16 July
    Julian
     Date

     104
     198
          Species   Material

        A.  smithii
      	    Dead
                 Live
                 Dead
                 Roots
   K.  cristata   Live
                 Dead
        "        Roots
                        X

                       472
                      1004
                       528
                       600
                       802
                       612
                       752
                        Sx

                        12
                        36
                        32
                        18
                        44
                        31
                        22
              40
              34
              18
              18
              20
              19
              20
                                                     95%
                                                 Con/idence
                                                   Limits
                                                  L1      Lo
               449
               933
               459
               562
               711
               546
               705
                 496
                1075
                 596
                 638
                 894
                 678
                 800
                                     448

-------
(p ^  .05)  in  partitioning of sulfur within  the  plants and between A. smithii
and  K.  cristata were  observed in  the  mid-July collection.   A.  smithii live
tissue  had  more  sulfur than either the dead tissue or the  roots.  K.  cristata
had  more  sulfur  in the live tissue than  in the dead.  The level of sulfur in
the  dead  tissue was  less than that  in the roots.   Finally,  live  A.  smithii
contained more  sulfur  than K.  cristata, while the dead tissue and roots of A.
smithii contained less  than did K.  cristata.

     A  cluster  pattern of differing sulfur  levels within  the  drift areas was
not  established by the sampling grid used (Figure 13.5, Materials and Methods).
The  levels  of sulfur  by  species  in live culms  for the  entire mid-July drift
collection  were  compared  to the levels predicted by the A and B treatments on
ZAPS I.  Regression for both species at the LOW treatment level was significant
and  used  to establish  95% prediction belts for  sample sizes of 34 in the case
of A.  smithi  and 20 in the  case  of K.  cristata.  The linear  model for the A
treatment (0  pphm)  was rejected,  and a prediction belt was constructed on 7 =
Y.   The 95% confidence interval  for the grasses from  the  drift plot area was
then  compared  to  the prediction belts for the appropriate point in time (July
16,  i.e.,  day 198).  This  is portrayed graphically in Figure 13.33.   The drift
collections showed  significantly  lower  levels  in both species than in grasses
fumigated on  the  B treatment plot.  Although the means were higher than those
from the  plot A collections, they were  not  significantly  so.   No  clusters on
the  drift  areas  were  recognized  as  significantly  higher than  the plot  A
controls.

PHENOLOGY

     The  results  of  the  phenological  observations of the vegetative stages of
A.  smithii  are summarized in  Tables 13.10  through  13.12.  Total  leaf number
increased rapidly  through  early July and leveled off  by the  next observation
period  in early August (Tables 13.10 and 13.11).   No significant differences (p
^  .05)  in total  leaf number were  observed  between  the treatment plots at any
observation date (Table 13.12).

     The  ratio  of necrotic  leaves to  live  leaves  was  examined  to determine
differences  between treatments on  each  ZAPS   plot  during each  observation
period.  The  last  row  of  Table 13.10 places the treatment  plots  on ZAPS I in
order  from  that with  the highest  ratio  of necrotic leaves to that  with the
lowest  ratio  of  necrotic  leaves  for  each  observation   period.   The  only
significant (p  i  .05)  difference  observed was  the September  2  collection in
which  vegetation  from  ZAPS I,  plot  D  (HIGH)  displayed   a  higher  ratio  of
necrotic  leaves than vegetation from ZAPS I, plot B (LOW).   However, the trend
from early July through mid-September was consistent,  treatment plot D showing
the  highest proportion of necrotic leaves, followed  by A (CONTROL),  then C
(MEDIUM),  and  finally  B  showing  the smallest proportion  of  necrotic leaves.
This  relationship  is  portrayed graphically in  Figures  13.34  through 13.36,
where necrotic leaves (expressed as a percent of the total  leaf number) of plot
D are compared to plots A, B, and C.  The exact probabilities  (as opposed to a
^ .05) of obtaining the observed differences on each date are  presented in the
next to the last row.  The low probabilities from July on are indicative of the
strength  of the  trend  DACB.   Only the HIGH treatment showed a higher level of
necrosis than  A (CONTROL).


                                      449

-------
    2000  -r
CD
Q
co
Q_
O-
    1500  4-
    1000
     500
                  LIVE AGROPYRON SMITHII
                  ZAPS
        130
    2000
    1500 --
    1000
Q-
O-
     500
        130
Figure 13.33.
                                               CONTROL
150      170
                                     190
210
230     250
                  LIVE KOELERIA CRISTATA

                  ZAPS I      ( K=20 )
                  150      170      190      210

                                    JULIAN DATE
         230     250
              95% prediction belts for treatment plots and
              95% confidence interval (  J ) for drift plot
              col lection.
                               450

-------
     TOO -,
     80 -
oo
LU
     60 .
O
LU
LU
O
LU
Q.
     40 -
     20 -
    ZAPS I  D (HIGH)

 ZAPS I A (CONTROL)
                       INDICATES  OBSERVATIONS  WHEN  THE
                        DIFFERENCE WAS  SIGNIFICANT AT  P<.05
                                                                              X
                                                                            X
        120
 i
140
160
  I

180
        200

JULIAN DATE
                                                                    220
                                                            240
                                                            260
                         Figure 13.34.   Percent necrotic  leaves  of  Agropyron  smithii.

-------
en
ro
      oo
      LU
      o
      o
      o;
      o
      UJ

      a.
            100 -,
             80 •
             60 -
            40 •
            20-
     ZAPS  I  D  (HIGH)


      ZAPS  I B  (LOW)
                              INDICATES OBSERVATIONS WHEN THE

                               DIFFERENCE WAS SIGNIFICANT AT P<.05
              120
140
  •
 i

160
180        200



    JULIAN DATE
220
240
260
                               Figure  13.35.   Percent  necrotic  leaves of Agropyron smithi i.

-------
           100-t
            80
      oo
      LU
                   ZAPS  I D  (HIGH)  -

                   ZAPS  I C  (MEDIUM)-
                              INDICATES OBSERVATIONS WHEN THE
                               DIFFERENCE WAS SIGNIFICANT AT P<.05
      o
      ^—t
      I—
      O
            60
on
CO
      O

      LU
      Q_
40 -
            20-
              120
               140
  i
160
  i            i
180          200

     JULIAN DATE
  1
220
240
260
                               Figure 13.36.  Percent necrotic leaves of Agropyron smithii.

-------
  TABLE  13.10.   Agropyron  smithii  ZAPS  I  PHENOLOGY  STUDIES AND TEST FOR  INDEPENDENCE OF LEAF NECROSIS.

Observation #
Date 17,

7TA



ZIB




ZIC




/1U



Total
Live
Necrotic
% Necrotic
Total
Live
Necrotic
% Necrotic
Total
Live

Necrotic
% Necrotic
Total
Live
Necrotic
%Necrotic
RXC Test of
Independence*
1
18 May
147
147
0
0
157
157
0
0
154
154

0
0
151
151
0
0
G
I-P(G)
Maximal Nonsignificant


ae us u v . u j



2
1 Jun
193
161
32
16.6
201
161
40
19.9
198
159

39
19.7
200
162
38
19.0
0.908
.823
B C D A


3
16, 18 Jun
216
175
41
19.0
246
186
60
24.4
234
175

59
25.2
222
169
53
23.9
3.031
.387
C B D A


4
4, 5, 6 Jul
288
196
92
31.9
292
213
79
27.1
259
178

81
31.3
237
146
91
38.4
7.757
.051
DACB


5
2 Aug
310
180
130
41.9
328
204
124
37.8
322
197

125
38.8
301
172
129
42.9
2.305
.512
DACB


6
16 Aug
311
137
174
55.9
331
171
160
48.3
327
167

160
48.9
312
136
176
56.4
7.352
.061
DACB


7
2 Sept
312
112
200
64.1
332
148
184
55.4
327
128

199
60.9
313
100
213
68.1
11.736
.008
DACB



8
17 Sept
312
83
229
73.4
333
112
221
66.4
329
89

240
72.9
313
82
231
73.8
5.894
.117
DACB


*Source:  Sokal and Rohlf, 1969.

-------
 TABLE 13.11.  Agropyron smithi i ZAPS II PHENOLOGY STUDIES AND TEST FOR INDEPENDENCE OF LEAF NECROSIS.

Observation.*
Date


Z2A










Z2D




23,
Total
Live
Necrotic
% Necrotic
Total
Live
Necrotic
% Necrotic
Total
Live
Necrotic
% Necrotic
Total
Live
Necrotic
% Necrotic
RXC Test of
Independence*
1 2
24 May No Sample
159
159
0
0
158
158
0
0
164
164
0
0
145
145
0
0
G
l-P(G)
Maximal Nonsignificant
Sets a < .05



3
18, 21 Jun
201
153
48
23.9
230
184
46
20.0
230
174
56
24.3
220
150
76
31.8
8.578
.035
DCAB


4
6, 7 Jul
275
201
74
26.9
280
213
67
23.9
294
201
93
31.6
289
170
119
41.2
22.444
.001
DCAB


5
3 Aug 1
281
151
130
46.3
292
158
134
45.9
304
161
143
47.0
316
127
189
59.8
16.855
.001
DCAB

6
5 Aug 2
281
137
144
51.2
296
143
153
51.7
307
148
159
51.8
319
124
195
61.1
8.686
.034
D C B A


7
Sept
285
90
195
68.4
302
104
198
65.6
307
102
205
66.8
321
88
233
72.6
4.149
.246
D A C

8
19 Sept
286
45
241
84.3
303
64
239
78.9
308
68
240
11. B
323
57
266
82.4
5.130
.163
B A D B C

^Source:   Sokal and Rohlf, 1969.

-------
   TABLE  13.12.   CHI-SQUARE TEST OF TOTAL  LEAF  NUMBER  AT  EACH  OBSERVATION.
Col lection
Number
1
2
3
4
5
6
7
8
ZAPS
X2
0.360
0.192
2.314
7.487
1.392
0.982
0.941
1.090
I
E
152
198
230
269
315
320
321
322
ZAPS II
X2
1.259
No
2.546
0.777
2.296
2.609
2.182
2.289
E
156
Sample
220
285
298
301
304
305

X2.05[3] - 7.815;  See Tables  13.10 and 13.11  for Observed Total  Leaf Values.


     On ZAPS II (Table  13.10)  the D treatment  showed  significantly (p S .05)
higher necrosis than the A treatment from early July through mid-August.  Plot
D was also significantly higher in necrosis  than plot B from late June through
early August.   Finally,  the D treatment  showed significantly more necrosis than
the  C  treatment in  early July.   The  trend  was  DCAB until mid-August  when  it
became less consistent.   These  data are  portrayed graphically in Figures 13.37
through 13.39,  where  plot D is compared to plots A, B,  and  C.

                           DISCUSSION AND SUMMARY

SULFUR

     Sulfur levels in live plant foliage of the seven  species of grasses, forbs,
and semi-shrubs collected at  ZAPS  plots  during 1976 adequately demonstrate that
elevated concentrations  are  directly related to  the  duration of fumigation and
the S02 concentrations delivered at the four treatment levels.  The foliage of
the  four  species  (Agropyron smithii,  K.  cristata,  Achillea  millefolium,  and
Artemisia frigida) which started to grow before  fumigations  began at ZAPS sites
I and  II  in  April  of 1976 showed  higher sulfur levels than are normally found
during the rest of  the  growing season (i.e., average 1,370 vs.  950 ± 75 ppm).
Evidence of residual  sulfur concentrations  in live  tissue  resulting  from the
1975 fumigation on ZAPS  I was  demonstrated clearly  only in one of the species
(A.  millefolium) on the  B, C,  and  D treatment plots,  while residual sulfur was
evident in live foliage  of A. frigida from the C and  D treatment  plots.  Neither
of the two early season  grasses (A.  smithii  and K.  cristata) collected in 1976
demonstrated evidence of  elevated residual  sulfur concentrations  due  to 1975
                                      456

-------
          100 „
           80 -
           60 -
-p.
en
      O
      i—i
      o
      o
      Di
      UJ
           40 -
           20
              120
ZAPS ii D (HIGH;)
ZAPS II A (CONTROL)
                             INDICATES OBSERVATIONS WHEN THE
                              DIFFERENCE WAS SIGNIFICANT AT P<.05
 i
140
 i
160
                                               220
 i
240
                   180         200

                        JULIAN DATE
Figure 13.37.   Percent necrotic  leaves  of Agropyron smithii
260

-------
          TOO
           80
           60
      §
-fs.
en
00
           40
          20
ZAPS II  D (HIGH)

ZAPS II  B (LOW)
                           INDICATES OBSERVATIONS WHEN THE
                            DIFFERENCE WAS SIGNIFICANT AT P<.05
                                                                                                  x
                                                                                                   X
                                                                 X
                                                                                       /•
                                                                                     s
                                                                                   X
                                                                                 X
                                      x
                                                                 X
                                          x
                                    X
                                      X
120         140
                                       I
                                      160
                      I
                     180
 I
200
 r
220
 I
240
                                                      JULIAN DATE

                               Figure  13.38.   Percent  necrotic  leaves of Agropyron smithii.
 I
260

-------
cn
         100^
          80-
      S  60-
      §
      o
      o
      LU
      Q,
          40-
          20-
           o-.
            IfO
  ZAPS II D (HIGH)

  ZAPS II C (MEDIUM)
                             INDICATES OBSERVATION  WHEN THE
                              DIFFERENCE  WAS  SIGNIFICANT AT P<.05
                                           /
                /
                                      /
 I
140
  I
160
 I
180
                                               JULIAN DATE
  I
200
                                                                        220
  I
240
 r
260
                               Figure  13.39.   Percent  necrotic leaves of Agropyron smithii

-------
fumigation.   However,  because of the  elevated sulfur levels  in  both  of these
grasses during the early flush,  any residual  sulfur in the roots  which could be
translocated up to the foliage could be masked (see Table 13.8, Differences in
Sulfur Content of  Roots  Collected  in Mid-September Compared  by  Least Signif-
icant  Differences).   More  detailed  studies  of  sulfur levels  in the  root
tissues of these two species will  be undertaken during the 1977 growing season
because it  is  necessary  to  understand the role  of residual  sulfur concentra-
tions if one is to understand and correlate sulfur levels in the foliage being
fumigated and leaf  necrosis and/or early senescence.

     The need for more detailed studies on the role of residual sulfur also was
made apparent  by  the  sulfur levels found in  remnant populations of A.  smithii
collected in  the winter of  1976  at two air  monitoring  stations in Anaconda,
Montana.   These concentrations are  presented  in  Appendix B, Table B-l, as are
the sulfur contents of A. smithii  foliage collected in mid-February (1977) from
the  D  treatment plot (ZAPS  I) one  week after collections were  made  from the
Anaconda sites (Appendix  B, Table B-2).

     Figure B (Appendix B)  relates the sulfur data in this appendix to ambient
S02 concentrations and sulfur content of A.  smithi i foliage.   The total monthly
dosage of S02  delivered  during  a 14-hour day length period (6 a.m.  to 8 p.m.)
from April  1 to September  30, 1976,  at the  two  Anaconda  monitoring  sites is
compared to  the  total  monthly dosage for similar day length periods on ZAPS I
at  the  four  treatment levels  (basic  data  for  ZAPS   sites   is  from  1975
fumigation,  unpublished  data,  Lewis and Lee,  EPA).   Dosage was  computed from
hourly geometric means of the ambient concentrations.   Figure B also shows the
95%  confidence  intervals for the  sulfur content of A. smithii foliage for the
February (1977) collections  from ZAPS I, plot  D, and the Anaconda sites.

     The  data on  Figure B  suggest three  explanations  for the  discrepancies
between S02 air concentrations and sulfur accumulation in the foliage from the
Anaconda  sites  and the  ZAPS I,  D  treatment plot:   (1) The  sulfur  levels in
foliage  from the Anaconda sites are  due  to  residual  sulfur  in  the  roots and
rhizosphere  of  these  plants;  (2)  the remnant  population  of  A.  smithii at the
Anaconda sites consists of ecotypes which are S02 tolerant and have adapted to
accumulating excessive levels of sulfur, or (3) the ambient S02 concentrations
measured at either the Anaconda or ZAPS monitoring sites are not revealing the
actual  amounts  of  S02  impacting  the  foliage  of  A.   smithii.   Until  the
investigators  finish  the 1977 ZAPS studies   and  grow A.  smithTi  in Anaconda
soils (from Highway Monitoring Site) in the University of Montana Department of
Botany  gardens this  year,   it  will  not  be  known whether the  forementioned
factors  are  collectively or  singly responsible  for  the  difference  in sulfur
accumulation between the  ZAPS I, D treatment plot  and the Anaconda sites.

PHENOLOGY

     The 1976 phenology studies  on A.  smithii across the ZAPS  I treatment plots
demonstrate that this grass  species from the  control plot (plot A) is possibly
influenced  by  abiotic factors other  than S02.  The  rate  of  leaf  necrosis on
plot A  is  more similar  to  that  which occurred  on  the  most heavily-fumigated
plot (plot D) than to either the  B or C treatment plots.  This similarity could
be caused or  partially  caused by the fact that the A treatment plot has lower


                                      460

-------
soil moisture  (1976  soil  moisture data  from  Dodd and Fort Collins, Colorado,
study group)  in  the  top 30  cm  of  soil than the  D treatment plot.  Differences
between soil  moisture  in  the top  30  cm  of soil   on  plots  A  and D occurred in
late June  (June  24)  and  continued  on  through  the  remaining growing season.
While the  overall  difference was  only 19 percent  (i.e., 4.39 g/cm2) from June
24  to  September  6  (74 days),  this measured reduction of  soil  moisture on
treatment  plot A  might explain why  leaf necrosis  at this  location  is most
similar to the leaf necrosis on treatment plot D  (see Table 13.10,  Phenology of
ZAPS I).

     The soil moisture  profile  of  all four treatment plots on ZAPS I should be
examined thoroughly  during  the 1977  growing  season to ascertain whether there
are significant  differences  between the plots which  could be influencing the
results of phenology, biomass,  and species diversity studies being carried out
on ZAPS I.

     Phenology studies  of A. smithii  on ZAPS  II (Table 13.11)  during the period
of  mid-June  (Julian  Day 170) to mid-August (Julian Day 228) demonstrated that
S02 concentrations on that D plot  caused significantly more leaf necrosis than
that  occurring   on  the  other three   treatment plots.   However,  the  S02
concentrations  delivered  to  the  B   and  C treatment  plots  did  not  cause  a
significant  difference  in  leaf necrosis from that  occurring on  the  control
plot.   This  would indicate  that the  gradient for these  concentrations,  while
causing increases in sulfur  accumulation in  foliage  over baseline,  are  not
adequate to cause differences in visible damage to the foliage of A.  smithi i.

     Phenology studies  of A.  smithii and one other  grass  species during the
1977 growing  season  (possibly  K.   cristata) will   be carried out to compare and
verify the results of the 1976 phenology studies.

                                 REFERENCES

Association of Official Seed Analysts.   1970.  Rules for Testing Seeds,  1970.
     Proc.  Assoc.  Official Seed  Analysts.  60(2):116.

Cocking, W.  D.   1973.   Plant  Community Damage  and  Repairabi1ity  Following
     Sulfur  Dioxide  Stress  on  an  Old-Field  Ecosystem.   Ph.D.  Thesis,  Rutgers
     University.   133 pp.

Dodd,  J. L.  et. aj..  1976.   Effects of S02 and Other Coal-Fired Plant Emissions
     on  Producers,  Invertebrate  Consumers,  and  Decomposer  Structure  and
     Function  in  an  Eastern Montana Grassland.  Quarterly  Progress  Report,
     Grant  No.  A-803176-01.  Natural  Resources   Ecology  Laboratory,  Colorado
     State University,  Fort  Collins.

Dreisinger,  B.  R.   1965.   Sulfur Dioxide  Levels and Effects  of the Gas  on
     Vegetation Near Sudbury, Ontario.  Presented at 58th Annual Meeting of the
     Air Pollution Control Association.  Paper No. 65-121.

Dreisinger,  B.  R. and  P.  C. McGovern.   1970.   Monitoring atmospheric sulfur
     dioxide and  correlating  its  effects on  crops and forests  in the Sudbury
                                      461

-------
     area.   Proceedings of  the  Impact of Air  Pollution  on Vegetation Confer-
     ence,  (S.N.  Linzon, ed.), Toronto, Ontario.

Duncan, D.  B.  1955.  Multiple  range  and multiple  F  tests.  Biometrics.   11:1-
     41.

Eddleman,  L.  E.   1977.  Personal Communication.   (Rangeland Ecologist,  School of
     Forestry, University  of Montana, Missoula, Montana).

Jorgensen,  H.  E.   1970.   A  Life  History Study of Agropyron  smithii  Ryalb in
     Central   Montana with Related Effects of  Selected Herbicide Treatment of
     Rangeland.    Ecological  Effects   of  Chemical   and  Mechanical   Sagebrush
     Control.   Progress Report for period ending 6/30/70.  Montana Department of
     Fish and  Game, Game Management Research Bureau, Bozeman, Montana.  VM05-R-
     3,4,5.

Lee, J. J.  and R.  A.  Lewis.  1976.   Unpublished  Report.   U.  S.  Environmental
     Protection Agency, Corvallis  Environmental Research Laboratory.

Linzon, S.  N.  1971.  Economic effects of sulfur dioxide on forest growth.  J.
     Air Poll. Control Assoc.  21:81-86.

Tingey, D.   T. ,  R.  W.  Field,  and L.  Bard.   1975.  Physiological responses of
     Vegetation to coal-fired power plant  emissions.  In The Bioenvironmental
     Impact of  a  Coal-Fired  Power  Plant,  Second  Interim  Report,   Col strip,
     Montana—June 1975  (R. A.  Lewis, N.  R.  Glass,  and A.  S.  Lefohn, eds.).
     Environmental   Research Laboratory,  Office  of  Research  and Development,
     Corvallis, Oregon.  EPA-600/3-76-013.

Weaver, J.  E.   1958.  Underground  development  in natural grassland communities.
     Ecological Monograph.   28:55-77.
                                      462

-------
                                                    APPENDIX A
en
                                     TABLE A-l.  ZAPS I - TREATMENT A (CONTROL)
                                                  1976 COLLECTION

                                                          Sulfur Levels (ppm)
Collection
Date
6 April
13 May
1 June
18 June
4 July
2 August
16 August
1 September
Jul ian
Date
97
134
153
170
186
215
229
245
A. smithii
Live
1300
700
850
1000
700
1000
850
1275
Dead
500
NS
NS
NS
500
900
750
1225
K. cristata
Live
1675
700
550
900
650
800
700
875
Dead
600
NS
NS
NS
550
500
550
750
P. sandbergii
Live
NS
500
550
550
NS
NS
NS
NS
Dead
NS
NS
NS
NS
400
NS
NS
NS
A. longiseta
Live
NS
NS
NS
NS
1000
750
950
900
Dead
400
NS
NS
NS
NS
700
850
600
A. mill
Live
800
600
550
850
550
550
750
900
efolium
Dead
700
NS
NS
NS
550
650
600
800
A. frigida
Live
1100
1250
-
1550
1300
1250
1300
1550
Dead
1100
NS
NS
NS
NS
NS
1300
1300
     NS^not sampled

     -=missing sample

-------
                                                     APPENDIX A
en
                                     TABLE A-2.   ZAPS II - TREATMENT A (CONTROL)
                                                   1976 COLLECTION

                                                           Sulfur Levels (ppm)
Col lection
Date
10 April
8 May
31 May
18 June
6 July
3 August
15 August
1 September
Julian
Date
101
128
152
170
188
216
228
245
A. smithii
Live
1716
800
850
1050
1050
800
950
800
Dead
350
NS
NS
NS
NS
800
850
850
K. cristata
Live
1575
650
750
900
500
650
650
850
Dead
500
NS
NS
NS
NS
550
700
650
P. sandbergii
Live
NS
650
600
700
NS
NS
NS
NS
Dead
NS
NS
NS
NS
450
NS
NS
NS
S. viridula A. mill
Live
NS
800
800
1100
950
1150
1000
1100
Dead
550
NS
NS
NS
NS
950
750
850
Live
1175
850
700
950
650
600
850
750
efolium
Dead
550
NS
NS
NS
650
700
700
600
A. frigida
Live
1600
1550
913
1350
1133
1250
1350
1550
Dead
750
NS
NS
NS
NS
NS
1050
750

-------
               APPENDIX A
TABLE A-3.   ZAPS I - TREATMENT B (LOW)
             1976 COLLECTION

                     Sulfur Levels (ppm)
Col lection
Date
7 April
13 May
1 June
16 June
4=.
171 5 July
2 August
16 August
2 September
Jul ian
Date
98
134
153
168
187
215
229
246
A. smithii
Live
1400
1050
1050
1350
1450
1500
1700
1650
Dead
600
NS
NS
NS
850
1400
1650
2000
K. cristata
Live
1575
900
900
1425
1200
1850
1550
1800
Dead
500
NS
NS
NS
1050
1400
1550
1600
P. sandbergii
Live
NS
950
783
850
NS
NS
NS
NS
Dead
NS
NS
NS
NS
500
NS
NS
NS
A. longiseta
Live
NS
NS
NS
NS
850
1050
1200
1300
Dead
725
NS
NS
NS
NS
1050
1050
1150
A. mil lefol ium
Live
1275
800
800
1200
700
1200
1150
950
Dead
-
NS
NS
NS
1050
1350
1400
1225
A. frigida
Live
1700
1950
1450
1850
1750
1933
1600
1500
Dead
1150
NS
NS
NS
NS
NS
1600
1900

-------
                                                      APPENDIX A
                                       TABLE  A-4.   ZAPS  II  -  TREATMENT B (LOW)

                                                    1976 COLLECTION
                                                            Sulfur Levels (ppm)
en
en
Col lection
Date
10 April
11 May
1 June
21 June
7 July
3 August
15 August
2 September
Julian
Date
101
132
153
173
189
216
228
246
A. smith ii
Live
1250
1100
1300
1450
1600
1550
1250
1450
Dead
600
NS
NS
NS
NS
1900
1500
1950
K. cristata
Live
1050
800
1100
850
1300
1550
1050
1350
Dead
550
NS
NS
NS
NS
1350
1250
1300
P. sandbergii
Live
NS
975
800
950
NS
NS
NS
NS
Dead
NS
NS
NS
NS
850
NS
NS
NS
S. viridula
Live
NS
1275
950
1350
1600
1300
1100
1600
Dead
-
NS
NS
NS
NS
1200
1350
1500
A. mi lief oli urn
Live
850
1050
1050
1800
1100
1300
775
1350
Dead
700
NS
NS
NS
1400
1550
1300
1350
A. frigida
Live
2125
1350
1050
1900
1950
1750
1550
1800
Dead
1050
NS
NS
NS
NS
NS
1350
1500

-------
               APPENDIX A
TABLE A-5.   ZAPS I - TREATMENT C (MEDIUM)
             1976 COLLECTION

                     Sulfur Levels (ppm)
Col lection
Date
7 April
13 May
1 June
16 June
5 July
2 August
16 August
2 September
Jul ian
Date
98
134
153
168
187
215
229
246
A. smithii
Live
1300
1350
1500
2050
1750
2550
2550
2225
Dead
800
NS
NS
NS
1200
2750
2950
2800
K. cristata
Live
1200
1200
1600
2250
2100
2350
2400
2600
Dead
850
NS
NS
NS
1250
2200
2000
1800
P. sandbergii
Live
NS
1150
1450
988
NS
NS
NS
NS
Dead
NS
NS
NS
NS
900
NS
NS
NS
A. longiseta
Live
NS
NS
NS
NS
1450
1250
1750
1275
Dead
750
NS
NS
NS
NS
1400
1700
1500
A. mill
Live
900
1150
1450
1650
2000
2200
1850
2150
efol ium
Dead
1050
NS
NS
NS
2700
2850
2150
2350
A. frigida
Live
2700
2050
2050
-
3100
2700
2350
3150
Dead
1100
NS
NS
NS
NS
NS
2325
2100

-------
               APPENDIX A
TABLE A-6.   ZAPS II - TREATMENT C (MEDIUM)
             1976 COLLECTION

                     Sulfur Levels (ppm)




-P-
en
co



Col lection
Date
9 April
11 May
31 May
21 June
7 July
3 August
15 August
2 September
Jul ian
Date
100
132
152
173
189
216
228
246
A. smithi i
Live
1100
1150
1700
2200
2150
2616
2200
2450
Dead
400
NS
NS
NS
NS
2850
3400
3200
K. cristata
Live
1675
1550
1300
2250
2450
2650
2250
2550
Dead
550
NS
NS
NS
NS
2250
1950
2050
P. sandbergii
Live
NS
1450
1250
1250
NS
NS
NS
NS
Dead
NS
NS
NS
NS
1100
NS
NS
NS
S. viridula
Live
NS
2500
1600
2250
2600
2100
2250
2950
Dead
450
NS
NS
NS
NS
2500
2250
2800
A. millefolium
Live
550
1400
1200
2550
2200
2350
2150
1800
Dead
500
NS
NS
NS
2450
3750
2800
2250
A. frigida
Live
1250
1650
1750
-
2325
3150
3050
3950
Dead
800
NS
NS
NS
NS
NS
3350
2800

-------
                                                     APPENDIX A
-P-
en
                                      TABLE A-7.  ZAPS I - TREATMENT D (HIGH)

                                                   1976 COLLECTION



                                                           Sulfur Levels (ppm)
Collection
Date
8 April
13 May
1 June
16 June
4 July
2 August
16 August
2 September
Jul ian
Date
99
134
153
168
186
215
229
246
A. smithii
Live
1500
1750
1950
2400
2600
3175
3450
3800
Dead
1000
NS
NS
NS
1519
4725
3200
4100
K. cristata
Live
1200
1650
2017
3350
3850
4275
3400
3400
Dead
1050
NS
NS
NS
2100
2850
2350
2650
P. sandbergii
Live
NS
1550
1500
1900
NS
NS
NS
NS
Dead
NS
NS
NS
NS
2350
NS
NS
NS
A. longiseta
Live
NS
NS
NS
NS
1850
2100
2350
2400
Dead
800
NS
NS
NS
NS
1700
1800
2100
A. mill
Live
1350
1300
2750
3600
3725
3900
4400
5325
efol ium
Dead
1600
NS
NS
NS
5750
5250
6367
4750
A. frigida
Live
2650
3300
3400
-
5150
5450
4250
5550
Dead
2725
NS
NS
NS
NS
NS
4200
3850

-------
                APPENDIX A
TABLE A-8.   ZAPS II - TREATMENT D (HIGH)
              1976 COLLECTION

                      Sulfur Levels (ppm)
Col lection
Date
9 April
13 May
31 May
21 June
7 July
3 August
15 August
2 September
Jul ian
Date
100
134
152
173
189
216
228
246
A. smithii
Live
1400
1850
3250
4350
4200
4150
3100
3950
Dead
450
NS
NS
NS
NS
5675
3950
4450
K. cristata
Live
1625
1300
1550
3750
4183
3550
3550
3950
Dead
600
NS
NS
NS
NS
3250
3800
2900
P. sandbergii
Live
NS
1400
1900
2400
NS
NS
NS
NS
Dead
NS
NS
NS
NS
1800
NS
NS
NS
S. viridula
Live
NS
3250
2150
4650
4850
3600
4150
4250
Dead
500
NS
NS
NS
NS
3000
3800
4000
A. mill
Live
500
2600
2800
4983
3875
4100
4950
6000
e f o 1 i urn
Dead
550
NS
NS
NS
4750
5400
5500
4550
A. frigida
Live
1450
2700
3550
-
4850
5150
5450
5100
Dead
850
NS
NS
NS
NS
NS
5925
4350

-------
                                 APPENDIX B

TABLE B-l.   COLLECTION OF Agropyron smithii AT TWO ANACONDA MONITORING
            SITES, 1976.
     Site
                  Sulfur Concentrations (ppm)
     Highway Junction
      Monitoring Site
        3500
        3550
        4025
        3750
        3700
        3000
Average 3587
     Mill Creek
      Monitoring Site
        2400
        2150
        1950
        1950
        2050
        1950
        2650
        2950
        3000
        1950
Average 2300
 Arithmetic Average 0.02
 Maximum 1  hr.  Ave.  1.52
 Maximum 3  hr.  Ave.  1.25
 Maximum 24 hr.  Ave.  0.48
                                        Number of Monitoring Days  329

                                        Total 1 hr. Readings Taken 7665
 Arithmetic Average 0.014
 Maximum 1  hr.  Ave.  0.79
 Maximum 3  hr.  Ave.  0.48
 Maximum 24 hr.  Ave.  0.13
                                        Number of Monitoring Days  317

                                        Total 1 hr.Readings Taken 7202
      TABLE B-2.  PPM SULFUR IN Agropyron smithii FROM ZAPS I, TREATMENT D
                  collected February 16, 1977.
                                     1450
                                     1550
                                     1300
                                     1350
                                     1100
                                     1500
                                     1250
                                     1200
                                     1450
                                     1500
                x = 1365
           = 149
s- = 47
 x
n = 10
                                     471

-------
                                    APPENDIX B
UJ
CO

 CM
O
CO

s
a.
a.
UJ
to
o
h-
o
  10-
   0
          AT 0,10 PPM
          FORiHR. READINGS
          CONTINUOUSLY
   AT 0.05 PPM
   CONTINUOUSLY
   AT 0.02 PPM
   CONTINUOUSLY

     HWY JCT(I976)
           MlLLCR.(l976
  AT 0.005 PPM CONTINUOUSLY
                                                       HWY JCT
                                                       MILL CR
ZAP ID (1975)


      ZAP  I D

ZAP 1C (1975)


ZAP IB (1975)

ZAP 1 A (1975)
                                                                    _j
                                                                    UJ
                                                               3600 >
      o:

      u_ —
      -J x
-27OOS S|

      co10
      -I 2
      < O

      &>
-1800.H- o
      2 Q;
      o Q
      2 <
      Ul UJ

-900 ?Q
                                                                    L
                                                                           u.
                                                                           o
                                                                           o
                                                                    10
 I      I      I       •       •

APR.   MAY   JUNE   JULY   AUG.
                                         SEPT.  OCT.
   Figure B.  Dosage and accumulation of sulfur in Agropyron smithli.

-------
                                 SECTION  14

            INVESTIGATION OF THE IMPACT OF COAL-FIRED  POWER  PLANT
            EMISSIONS UPON  INSECTS:   ENTOMOLOGICAL STUDIES AT THE
                         ZONAL AIR  POLLUTION SYSTEM

                                      by

                              J.J.  Bromenshenk


                                INTRODUCTION

     This  portion  of the report concentrates  on  the entomological investiga-
tions  at  the  EPA  sulfur  dioxide  fumigation  plots  (Zonal  Air  Pollution
System-ZAPS)  at  Taylor  Creek,  near  Fort  Howes,  Montana, for  the periods of
June  15  through   October   20,  1975,  and May  1  through  November  15,  1976.
During  1975,  attention  was  given  primarily  to  collecting baseline  data in
the vicinity  of  Colstrip,  Montana,  since baseline conditions  would  no longer
prevail  there once  the  Colstrip   power  plants  began operation.    The  first
plant  began  to burn  coal  in early September  1975.  With  baseline data banks
established  (see  sections  on  animal  and  vegetation studies in the vicinity
of  Colstrip,  Montana, by  Gordon,  Tourangeau,  and Rice  and  the entomological
investigations by Bromenshenk), attention  was redirected to the ZAPS.

     Investigations  were initiated  in the  late summer  of 1975  and  efforts
were intensified  in  1976.   The  reason for the  intensification  of  the  work at
ZAPS was  twoTold:   (1)  The baseline data collections were  completed  and the
study  had entered  a monitoring  phase;  and  (2) we were convinced that the
Zonal  Air Pollution System  could  be useful  in testing  the  effects  of sulfur
dioxide  (S02)  upon   plant  and  animal  (arthropod)   species or  populations.
This belief  was  based  on  first season evaluations  of field experiments, the
system  design,  and testing  of  the gas delivery.  These  indicated that meas-
urable  changes  in response  to  the  S02  had occurred  (unpublished reports to
EPA).

     Experiments  were conducted  to  study the  behavioral   and  physiological
responses  of  honeybees  to  jjn  situ  S02 fumigation  in 1975,  and  in  1976,
studies were  initiated  of  the  inter-  and  intra-plot distributions of  other
insect  species:    Ground-dwelling  beetles, airborne  or  flying insects,  and
native pollinators.

     As  noted  previously,  this  research  is  one   aspect  of  the  Colstrip
project  to  develop  methods  to  predict  the  bioenvironmental  impacts  of
emissions  from  coal-fired  power plants  before  any  damage actually  occurs.
The most  prevalent  toxic,   gaseous  materials  released  by  the  combustion of

                                     473

-------
coal  are sulfur oxides  (DHES,  1975;  WES, 1973).  Because  of  their abundance,
consequences  to  human  and  vegetative   health,  and  deteriorative  effects,
oxides  of   sulfur  are  significant  air  pollutants  (Smith,  1975).  They  may
impact  ecosystems   as   gaseous  oxides,  as   acids   in  precipitation  (Likens,
1974), or as synergists with other pollutants.

     Known   interactions of  insects  and  sulfur compounds include:   (1)  The
use of  S02  as  a fumigant against stored  product and  household pests (Kanaga,
1956; Swisher,  1944;  Negherbon,  1966);  (2) changes in  the  population dynam-
ics  of  forest  insects  induced  by S02  air  pollution stress  (Linzon,  1966;
Bosener, 1969;  Donaubauer,  1968; Scheffer  and Hedgcock,  1955;  Kudela,  1962;
Sierpinski,  1966,   1967;  Templin,  1962);  (3)  decreases   in  populations  of
social  bees  and   parasitic  wasps  and  increases   in  aphids  linked  to  S02
stress  from a  615 MW  coal-fired power plant  in  Pennsylvania  (Hillmann  and
Benton,  1972);  (4)  decreases  in  populations  of  ground-dwelling  beetles,
mostly  Carabids, near  a Kraft mill  in Canada  (Frietag et  al_. , 1973),  and (5)
reductions   in  brood   rearing,  reduced  pollen  collection,  depressed  flight
activity,  and  uncertain mortality effects  produced by  fumigation for 9-  and
14-  week periods  using 0 to 5  ppm S02 on 20  honeybee colonies  (Hillmann  and
Benton,  1972).   They  noted  that   reduced  flight  activity  contributed  to
decreased honey production.

     These  literature   reports   suggest:   (1)  Trees   or  plants  which  are
physiologically weakened by air  pollutants  such as S02  become  more  suscep-
tible  to  attack by those insects which  seek out weak plants  for feeding  and
reproduction;  (2)  host-predator  and  host-parasite   imbalances  may  occur,
probably as a result  of a  concentration of toxic  substances  in  food-chains,
and  (3)  pollinators,  namely  honeybees  and social bees,  demonstrate a  variety
of  physiological   and  behavioral  responses  to  S02.    In  his  review  of  air*
pollution  and  apiculture (1972), Debackere  discussed  the chemical reactions
in the  atmosphere which  convert  S02   to  sulfurous  acid  (H2S03)  and sulfuric
acid  (H2S04).  He  reported  that it  is well   known that these sulfur compounds
have  an  irritating effect on the gastrointestinal  tract of mammals.  Although
almost  nothing  is  known about the effects on  bees  as  indicated by literature
reports,  it is  assumed  that  these   acids  have a  deleterious effect  on  the
intestinal   tract   of   bees.   The  ground beetle  studies  of  Frietag  et  al.
(1973)  demonstrated  significant decreases in the numbers  of Carabids  associ-
ated with  increased  fall-out of  S042   in the form of sodium sulfate (Na2S04);
these  beetles  were  all  predatory   species.    Kulman  (1974)  Deported  that
Carabids  are  sensitive  to  most  insecticides  and pollutants   and  that  it
appeared  that  seed-eating   beetles  contain  fewer  ppm  of  insecticides  than
predatory  beetles.  A bibliography  of literature pertaining  to  air pollution
and  insects is included as  Appendix  A in section 5;  a  more in-depth  litera-
ture  review appears  in the  section  on  the  entomological studies  near Col-
strip.

     In  the introduction  to  a  problem  analysis  of  air pollution  in  urban
areas  which  divides   air  pollution   influences  on  temperate forest  systems
into three  classes based on pollution load, Smith  (1975)  presented the known
responses of trees to these loads and indicated associated ecosystem impacts.
This   breakdown  provided a generalized conceptual model  for  other ecosystems;
his model is presented  in Table 14.1.

                                      474

-------
                    TABLE 14.1.  INFLUENCE OF AIR POLLUTION ON TEMPERATE FOREST ECOSYSTEMS1
Designation
                     Air Pollution
                         Load
Response of Trees
                                                                         Impact on Ecosystems
Class I
                         Low
    Class  II
                 Intermediate
en
    Class III
                    High
                                    1
                                    2.


                                    1 .
Act as a sink
contaminants
                  for
    No or minimal  physiological
    alteration

    Reduced growth
    a.   detruded nutrient
        avail abi 1 ity
        1)  depressed litter
            decomposition
        2)  acid rain leaching
    b.   suppressed photosynthesis
        enhanced respiration

    Reduced reproduction
    a.   pollinator interference
    b.   abnormal pollen, flower,
        seed,  or seedling
        development
3.   Increased morbidity
    a.   predisposition to
        entomological  or
        microbial  stress
    b.   direct disease induction

1 .   Acute morbidity
                                        2.  Mortality
1.   Pollutants shifted from
    atmospheric to organic or
    available nutrient com-
    partment
                                                                             2.
                                     Undetectable influence or
                                     fertilizing effect

                                     Reduced productivity,
                                     lessened biomass
                                                                                 Alteration of
                                                                                 composition
                                                                                           species
                                                                                 Increased insect outbreaks,
                                                                                 microbial epidemics
                                                                         1.   Simplification:   increased
                                                                             erodibility,  nutrient
                                                                             attrition,  altered micro-
                                                                             climate  and hydrology

                                                                         2.   Reduced  stability
          Smith, 1975

-------
     Because  the  Colorado  investigative  team  (CNREL)   conducted  tests  to
determine the  effects  of  sulfur  dioxide fumigation  on  the density, biomass,
and  trophic composition  of  above-ground and  below-ground  arthropods,  this
research concentrated on  a few insect species which,  according to  literature
reports, were sensitive or susceptible to perturbations,  easy  to monitor and
measure,  and  reliable  predictors  of  bionomic  and  economic  impacts.   In
addition, these  species could be  related to the  studies  in the Colstrip area
and  to  the  needs  of  the   overall  project.   Honeybees were used as  detectors
and  accumulators  of  air  pollutants  at apiary  locations  throughout  south-
eastern  Montana.   These  bees  have  an estimated  economic  value  of  $6  to $33
million  as   pollinators   and  $300  to $725,000  as  honey   producers annually
(Montana Agricultural  Reports,  1977; McGregor, 1976),  have bionomic value as
pollinators   of   rangelands  and croplands,  and  have  proved  to  be  extremely
sensitive and  useful  indicators  of a variety of  atmospheric pollutants (see
section  on  Colstrip  vicinity  studies).   Smith  (1975)   listed  pollinator
interference  and  reduced  reproduction,   which  alters plant  species composi-
tion,  as an  important effect  of intermediate air  pollution  stress  on eco-
systems.  Therefore, high  priority  was placed  on studying  the  effects  of S02
fumigation on honeybees.

     Data  collected  by  the  Colorado  team  in   1975  (unpublished  reports)
indicated that   during  the first  year of fumigation  no order  of arthropods
showed  any   treatment  response  except  the  Lepidopterans,  which  are  plant
tissue  and  nectar  feeders.   They  were  lowest  in  the  highest  fumigation (10
pphm)  plot;  however,  low population  numbers  made the  results inconclusive.
The  Colorado investigators  noted  that  scavengers,  represented  primarily by
Collembolan  species,  decreased in  numbers  in  the high plot  and that numbers
of  Carabid  beetles also  may  have  changed  in response  to the  highest  fumi-
gation  levels.    Again,   the  population   numbers   were  too  low  to  make  any
definitive  statement.   Finally,  more  grasshopper  eggs  were   found  on  the
control  plot,  an  area that  from  observations  had  less   cover  and somewhat
different  soil   characteristics   than  the  other plots.   The  Colorado  group
added  investigations   of   grasshoppers  in  1976;  the  investigators of this
research added  studies of ground-dwelling beetles  (predators and scavengers)
and  of  native  nectar  feeders  (Lepidopterans,  and pollinating  Dipterans  and
Hymenopterans, namely  native bees).

     The main  objectives   of  the  entomological studies at  ZAPS were:   (1) To
establish honeybee  colonies  for  behavioral  and physiological exgerimentation;
(2)  to  conduct  preliminary tests  of experimental procedures  and equipment to
maximize  the potential  usefulness  of the  ZAPS  system;  (3) to  evaluate  the
initial  responses  of  honeybees and other insects  to j_n  situ  S02  fumigation
so  that further  research directions  (e.g.,  process and  mechanism studies)
could  be determined,  and  (4) to evaluate intra-  and inter-plot distribution
of the selected species of insects via grid trapping and gradient analysis.

     We  were interested  in   characterizing  species  of insects  and temporal
(season)  differences   in   susceptabi1ity to  S02  stress   in  the  cool  season
grassland.    The   hypotheses  were  that:   (1)  Insects of  the  higher  trophic
levels  such  as  predators, scavengers, and pollinators would be more suscep-
tible  to damage  than  foliage feeders;  (2)  the  effects  of  S02   fumigation
would  be  reflected by behavioral  and physiological  responses  of the  target

                                      476

-------
organism;  (3)  these behavioral and physiological  responses  could be observed
and   measured   (e.g.,   distributions,   immigration,   emigration,  mortality,
shortened  life  span,  changes in fecundity  and  fertility,  etc.);  and (4) that
long-term  chronic  exposure  to  S02 would be  reflected  by increased suscepti-
bility, increased resistance, elimination, or  no discernible effects.


                             MATERIALS AND METHODS

     To  reduce  physiological  and/or  behavioral   responses  due to  genetic
variability, in  May 1975, the USDA Bee  Stock Center  (Baton Rouge, Louisiana)
provided  eight  dihybrid  sister  queens  which  had  been  artificially insemina-
ted  with  sperm  from drones  of  a  single  inbred  line  (see  Appendix B  for a
description).  These  queens were  introduced  into  eight  colonies and  used to
convert  the colonies to the "same"  genetic  composition.   Four colonies were
moved  to  the Taylor Creek site; the other  four were left at Broadus, Montana,
to serve  as controls and to provide replacements.  By  late August  1975,  all
of the original  queens  had died  or disappeared,  although  daughter  queens
continued  three  colonies  at Taylor  Creek and one  colony  at  Broadus.   Dr.
Harbo  of the Bee  Stock  Center  believed that  the  problem  of  the supercedure
of the queens was  due   to  inadequate sperm  reserves and/or to injury caused
during  artificial  insemination.    The Bee  Stock Center  provided  replacements
in  May  1976.   This  line  of  bees  has  continued  in use  because  they  have
proved  to be docile  in temperament and as such caused  a  minimum of disturb-
ance  to investigators  at the ZAPS  sites.   Due to the  lag time  of reestab-
lishing  colonies  from   introduced queens,   four  colonies  of "local"  bees,
provided by a commercial  keeper, were utilized  at ZAPS I in 1976,

     The glass observation  hives  described in the  Second  Interim Report were
not colonized in  1975 because of the supercedure  problem.

     During  the  1975 and 1976 growing  seasons,  a  beehive was  set 70  m  above
(north)  each of the  ZAPS I  treatment  plots.  The hives were  not set  on  the
plots  because  of potential  encounters  between bees and  the crews conducting
the  biological  sampling.   It was  requested  that  the  gas  delivery  line  be
extended  from  each  of  the  treatment plots to  the  respective hive,  but  the
EPA  design  engineers   concluded   that  this  might alter  the  delivery  flow
patterns  of the  entire  system.   Thus,  we could  perform only those  experi-
ments  which  involved the  foraging of the  bees on the  plots, but  could  not
control the exposure,  if any, of the bees  to  the S02.

     Because of  uncertainty about  the  dosages  of S02 to which the  honeybees
were   being  exposed and  because  we were  interested in the  effects  of  S02  on
native  pollinators,  a  trapping  series  was  begun in 1976 designed to  capture
samples  of  several  species  of  solitary  and social   bees   (Hymenoptera)  and
flies  (Dipterans)  on each  of the treatment  plots.  Drop-traps were  based  on
a  design  by Olsen,  Elliot,  and  Associates  (1976)  which  was  economical,
practical,  and well-suited  for this purpose.   They reported that these  traps
appear  to  act  as  artificial  flowers.  Each  trap  was composed of a 14-ounce
disposable  white,   plastic   beverage  container.   A white,  plastic,  conical
disposable  coffee  cup  from which  the  bottom  had  been  removed  was  inserted
into   the  mouth  of  each beverage  cup.   The coffee cup  was  held  in  place  by


                                     477

-------
a piece  of strapping tape.   Thus,  each trap was  a  simple funnel-trap.  Each
cup was  set into the ground  so  that the upper two-thirds  extended above the
surface  of the  ground.   Approximately one  inch  of water  was placed  in the
bottom  of  each  beverage  cup to  drown any  captured  insects.  The  cups  were
inspected  and  insects  removed at approximately ten-day  intervals.   Twenty of
these  traps  were installed on each  of the  eight treatment plots  for a total
of 160 traps  (Figure 14.1).

     The data  from  the  CNREL  group  suggested  that  ground-dwelling beetles
were  affected  by the  S02; literature  reports  suggested  the same.   Therefore,
a  network  (Figure 14.1)  of  pit-fall  traps  was set  on  each  treatment plot.
The traps  consisted of  the nested cup  design  utilized to obtain pollinators.
Olsen,  Elliott,  and  Associates  (1976) found that  sinking these  cups  deeper
into  the  ground  decreased their  attractiveness  to  pollinators  but greatly
increased  the  catch  of  beetles  (personal   communication,  D.  Schmidt,  ento-
mologist).    In  each  treatment plot,  24 of  these  traps were  arranged  into a
grid  pattern  and sunk  into  the ground so  that the mouth  was  level  with the
surface  of  the  soil.   In  addition,  ten extra  cups  were  placed   on ZAPS II
below  the  S02  gas delivery  line  of the high (10  pphm)  concentration fumiga-
tion  plot  in  areas  of severe vegetation burn,  and  ten extra cups  were placed
on  the control  plot  in the  same  relative  position  as  the  cups  on the  high
concentration  plot.   Thus,  each  network consisted  of 212 collection  points
(Figure  14.1)  utilized  for  approximately  ten-day intervals  during the first
half  of each  month  from May  through September.   An inch  of water  in the
bottom  of  each trap  drowned  the beetles and  prevented the predacious species
from devouring each other  and the other captured insects.

      The  number  of  beetles   obtained   in this  manner declined  rapidly after
mid-June  and  introduced  a problem  of  random-baiting  effect because  a  trap
containing  a  few beetles became much more  attractive  than  one  without any
beetles.   To  minimize  this  effect  and to increase  trapping success,  all
traps  were baited with  decayed meat.  This appeared  to  resolve the immediate
problem  and greatly  improved  the  trapping yield  of  Scarabaeids and Silphids
(Figure 14.2).

      We  recognize the  limitations  of pit-fall trapping  and  of the possible
sources  of error resultant  from  baiting.   However, we agree with the state-
ment  of Kulman   in  his  1974  review  of  the  ecology of  the  North American
Carabidae  in which  he  remarked that  the pit-fall method  persists in studies
of  ground-dwelling  beetles because  "...other sampling methods (mark capture,
total  population over  unit  area,  etc.)  also  have problems  and  are  more
difficult to use."

      Finally,  to  study  the distribution of pollinators and of other airborne
or  flying  insects over the  plots,  which the D-Vac  sampling  systems utilized
by  the Colorado  entomological  investigative team did not adequately sample,
a series of  cylindrical,  sticky traps  were  employed.  Each trap consisted of
a  14-ounce  disposable  plastic  beverage  cup,  coated  with   Tack  Trap®  and
taped  above  each junction   of  the   delivery  pipeline  (Figure  14.2).   The
sticky traps were replaced every two  weeks.  Since color  may affect capture,
both  white and  yellow  cups  were  used.   Reportedly, native  pollinators are
attracted  to  white  (Olson,  Elliott,  and  Associates,   1976),  while  aphids

                                      478

-------
                                   73m
5 •




4 •


T 	 /
0 3 •



	 L
2



•
• 15 •
•-— A " '•
o


•
•
7
*

— z
8 •

\ 	



c

14 •
• (

i3~~"'
• 24"
• ft
1~A
D

'
' 0-S
17
^_
o
>.— °
(
«MH«B^
12
^

o — /:
18"
D


/
i


2~«
\ —

22?

^—

— ^
21
V.
19 *•
o

m •
\ 	 — i
•



\ *«HB
•






1 	




"
                                10
20
                                        Mixing shed
              A  Sticky-traps
              •  Pit-traps
              •  Water-traps
              o  Pit-traps in vegetation burn  areas
Figure 14.1.   Distribution  of  insect traps on ZAPS plots - Summer 1976
                                479

-------

Figure 14.2.
A.  Sticky-trap^tapBd to S02 delivery line.  B.  Coating
sticky-trap cup.  C.  Pit-fall trap and captured beetles.
Cup was raised slightly for photographic purposes.
                                      480

-------
are  attacted  to yellow  (Broadbent,  1948).  Hi 11 man  reported increased aphid
numbers  in  areas subjected  to elevated levels of  sulfur  from a Pennsylvania
coal-fired power plant.

     Presumably, color  has  the greatest effect at  low wind speeds (Southwood,
1975).    Taylor  (1962) demonstrated  that a  white  sticky trap  caught insects
almost  as  if they  were inert  particles,  and  he  constructed tables relating
the  efficiency  of  cylindrical  traps for different insects  at different wind
speeds.  The  relation of catch to  size  of sticky  traps  was investigated by
Heathcote (1957) and  by Staples and Allington  (1959).   They  found  that catch
usually  increased  with trap  size, but this increase was  not proportional to
trap size;  the  smallest  trap  caught  the  largest  number  of  insects per unit
area.   Based  on  these findings,  we  decided  to  test  the effects  of  color
During  each  six-week trap  period, ten  white  and  ten  yellow stock-cups were
used  on each  treatment  plot  for two  weeks,  then  20 yellow  cups for  two
weeks,   followed  by  20 white cups  for  two  weeks,  at the end of which time the
cycle  was  repeated.  When  two colors were  used,   the  colors  were  alternated
in each row of traps.

     Captured insects  were  frozen  until  they could be identified and counted.
The  number  of   insects  captured  per cup  per  trap day  was  calculated  for
purpose  of  analyses.   Destroyed  (hail,  mice,  wind) or upset cups  were  not
included in  the analyses.   All traps (sticky, drop, and  pit-fall)  were laid
out  in a grid  system based  on the pattern  established by  the  gas delivery
line.   The  pipe junctions  were used  as sampling  points.   Each  cross-shaped
junction visually  marked out  four quadrants.   Quadrants  falling outside  the
north  and  south  perimeter  of the  delivery system  were considered to fall
outside  of  the treatment  plot.  The  traps  were arranged  in  a  systematic
manner  so  that  each  junction and quadrant was  utilized,  each quadrant being
replicated  an  equal   number  of  times within  a  plot.   Trap  locations  from
treatment plot  to  plot  were  identical.   Within each  "junction quadrant"  the
pit-fall or  drop trap  location was  determined by measuring  125 cm from  the
horizontal  gas  pipeline diagonally  to the  ground  at the  first  orifice away
from the junction.  This  placed each trap just  over  one meter from the line.
A  systematic  grid  trapping  network  was  used to study  dispersion patterns as
well as changes  in population size.

     During  1975,  dead-bee  traps,  pollen traps, magnetic  labels, and feeding
stations were  used in  tests  on honeybee  responses to  the  fumigations.  The
dead-bee traps  consisted of  plastic traps  below  the entrance  of  each hive,
enclosed by a  2-cm  mesh to  the inch  screen.   Living bees could  easily fly or
crawl  through  the  screen,   but housecleaning  bees  could  not carry  the dead
bees out.   Dead bees  in the  traps  were  counted  daily during  those periods
when investigators  were at  the ZAPS  plots.   Numbers  of  bees  that  died away
from the hive were not included in the analyses.

     The original  design for  a pollen  trap,  which is  part  of  the  dead  bee
trap (EPA,  1976),   was  based  on a design by  Hillmann  (1972).   However, this
design   restricted   the  entrance to  the  hive,  and the  bees  piled  up  at  the
entrance o'n days when they  were actively  foraging.  Therefore,  a pollen trap
based  on the design  by  Henderson  (1962) was used.  This  consisted  of  a deep
honey  super  (box)   with  a  2-cm  mesh  screen over  its  top, which then  became

                                     481

-------
the new entrance.  The  pollen  trap was set  under  the rest of the hive struc-
ture.   Bees  entered  via a  normal  entrance,  then  crawled  through  the screen
which formed  a false bottom  in  the  hive.   The screen  brushed  pollen off the
bees'  legs.   A pull-out drawer  lined with  plastic  sheeting allowed for the
removal  of the pollen.

     We  used  a  ferrous metal  tag  capture-recapture  technique  (Gary,  1971)
when  appropriate  for distribution,  movement,  and flight  tests.   Coded metal
tags  were  glued to  the abdomen of  bees  captured  in the  field.   These  tags
were  retrieved at the  entrance  to  the  hives by a  strip  of ceramic magnets.
The  magnets   either  pulled the  tags off  the  bees  as  they entered,  or  else
both  tags  and bees  stuck to the magnet but the bees were able to wiggle free
within  a  short period  leaving the  tag  behind.   The ferrous  metal technique
demonstrated  a 78  percent retrieval  success, which is consistent  with the
success ratio  reported  by  Gary (1971).  Gary  discussed  reasons  for less  than
100 percent retrieval  in his paper.

      For  some  experiments,  bees  were  marked  with  dots of  enamel  paint
(color-coded  to each hive).   Direct counts  of  bee   visitations  and foraging
on  each plot  or  at  feeding  stations were  made  by   walking  transects across
the plots.

      Behavioral  tests  using  feeding  stations (glass-watch glasses  on which
jars  were  inverted  and  filled with  a  one  molar sucrose solution as described
by  von  Frisch,  1967) were conducted  in  late  August and  early  September of
1975.  At  that time  very few flowering plants  were  in  blossom, and honeybees
were  easily  and  rapidly trained to  feeding  stations.   During  this same 1976
period,  honeybees  were attracted  to  the  water-containing  drop   traps  and
captured,  although  at  other  times  honeybees  were  seen only  infrequently in
or  around these  traps.  Apparently, the  shortage  of water  in  nearby reser-
voirs forced the bees  to the traps.

      Chi  square  analyses  of  variance were  used  to  test whether  number of
honeybees  or  other  insects on the fumigation  plots  fit an equal distribution
ratio or  whether numbers  fit  some other ratio such  as dependency  on fumiga-
tion  level.    Regression  and  co-variance  analyses  were applied  where appro-
priate,  but  for  the  1976  period  the tests relied  on  the theoretical levels
of  mean  S02  concentration  because the actual  data were not available.  These
analyses  are  useful  approximations   but  the  actual data  are  needed before
more  intensive and  accurate   mathematical  methods  can be  applied,  such as
within plot distribution.
                                   RESULTS

     Chi  square analyses  of variance  were used  to test  whether numbers  of
honeybees on  the  fumigation plots fit  an  equal  distribution ratio or  whether
numbers  of  bees were  dependent on  the concentration  of  S02-   The  following
observations were made  at the ZAPS sites during 1975:
                                      482

-------
     (1)  Honeybees  fed  at all  feeding stations,  even  when a feeding station
was placed directly  under a point source emitting concentrated S02.
     (2)  The number of  honeybees  at each feeding  station  increased from the
time each experiment  was   initiated,  indicating  either recruitment  or dis-
covery by other bees.
     (3)  The  number of  bees at a  feeder  was  greatest just after the sucrose
solution  was  depleted  and  then decreased  rapidly if  the  feeder was not re-
filled.
     (4)  The  bees  generally  were  not  successful   in  locating   a  feeding
station  that  was  hidden  from  sight,  even  when fragrance (an  essential oil
such as  peppermint)  was added.
     (5)   Marked  and  tagged bees  from each  of  the  colonies  (located  75  m
north of  the  fumigation areas) were  observed  on  the  plots;  no preferential
response  to  S02  as indicated  by  the number  of  bees  visiting  the feeding
stations or different plots  could be discerned (Table 14.2).
        TABLE 14.2.  VISITS BY ADULT WORKER HONEYBEES TO FEEDING DISHES
                     (ONE MOLAR SUCROSE), ZAPS I, AUGUST-SEPTEMBER, 1975.
      Experiments  1-6.  Total Number of Bees/Plot Versus Fumigation Level
                             (Theoretical Levels)

Number of Bees
CONTROL LOW MEDIUM HIGH X2 P
91 118 127 109 6.38 >0.05
 Experiments 3 and 4.  Total Number of Bees/Plot at Dishes a) Hidden in Grass
 with  No  Fragrance;  b) Hidden with Essential Oil Fragrance; c) Not Hidden,
 e.g., white 20 cm squares  under dishes.   (30 minute tests).
                        CONTROL
LOW
MEDIUM
HIGH
Number of Bees


a)
b)
c)
0
1
172
0
3
124
0
0
169
0
0
120

      During  both  1975  and  1976,  mortality  of the  bees  at  the  ZAPS site
 appeared  to  be  normal;  no  obvious  lethal  effects  could  be  discerned.  It
 should  be  noted that  the  hives  were  not  placed  directly  on  the fumigated
 plots  primarily  because of  concern  expressed about possible  problems  asso-
 ciated with honeybees and researchers  in  close proximity.

      The  bees  did  not  appear to  avoid different  concentrations  of  S02 gas
 nor  did  they  exhibit  obvious signs of intoxication,  either  lethal  or sub-
 lethal,  such as  interference with foraging  activity.   But  in  1975, the bees
 did  not  successfully  locate  feeding stations  hidden   from view,  even if a
 fragrance  was added.   Odor may be more important than  vision as a guidance
 system  to  bees,  and  the  addition  of  small  amounts  of an  essential  oil  should

                                      483

-------
be  a  strong  attractant (von  Frisch,  1967).   It  is  possible  that  the  S02
interfered with  the bees'   olfactory  sensory systems or  that  other odors are
more attractive than peppermint.

     The long-term  effects  of  continued exposure to S02  are  still  not known,
but it  appeared  that  during the summers of  1975 and 1976, a large portion of
the bees  from each colony  did not  forage on the  control  or  fumigated plots
and as such, may have had little contact  with the S02.

     Acetylcholinesterase  (Ache)   enzyme   analyses  on  heads  or brains  from
bees  (both .foraging and dying  bees),  which  were presented  in  the  portion of
this  report  describing  the  chemical   analyses  of  bee  tissue   from  sites
throughout  southeastern Montana,  were  performed by  R.   Barker,  Bee Research
Laboratory, Tucson, Arizona.   He  found that the rates of  enzyme  activity of
bees  from  the ZAPS I  site  and controls from apiaries  near Broadus, Montana,
were  as high  or  higher than  the normal  activity  rates  of  bees  from  the
Tucson  area and  considerably  higher  than  the  rates  of  enzyme-activity  in
bees  poisoned by  organo-phosphate  pesticides.    Failure to  detect  an  S02-
induced decrease  in the Ache  activity of bees  at the ZAPS sites  may be due
to  avoidance   of  the   S0?  since the  hives were  not  placed on  the treatment
plots  and  the bees were free  to  forage   elsewhere.  Foraging  outside of the
ZAPS  plots  may have  been  related to  a lack of  good forage materials on the
plots  rather  than avoidance  of  the  sulfur, which  from  the  feeding  tests
appeared to be the case.

     During  1976,  the  queens  from each  hive,  except the  one  above  the  LOW
plot,  superceded,  leaving  daughter   queens  to  take  their  places.  However,
the  hive  above the MEDIUM  plot was  left in a  queenless state.  This colony
was  replaced  in   late  June  with a colony  from the Broadus  area and an extra
colony  was  set above  the  control  plot to use  in  case  of  additional super-
cedures.   Because  of   the  supercedures,   in  the   fall   the  strength  of  the
colonies and  the  amount of stored honey   ranked  from the largest  to smallest
was:  LOW,  CONTROL,  MEDIUM, HIGH.

     In late  September  of  1976,  honeybees began  to seek  out the drop-traps
for  water,  as  indicated by  their  lapping  of  water from  the sides  of  the
trap.  The  result  was  that  bees were captured in  these  traps  at this time of
the year  when they had not been captured before.  For example,  by walking a
transect across  the plots,  the numbers of bees  on  the plots near  these traps
(September  21, 1976)  was 31, 25,  2,  and  2 for the CONTROL,  LOW,  MEDIUM,  and
HIGH  treatments,  respectively.   However,   because  of the  supercedures,  these
numbers approximated the difference  in the strength and  size  of the colonies
nearest each plot.

     Bees  from the ZAPS  I site in 1975  had slightly lower sulfur content (x
 = 4,000,   SD  = 436)  than   those obtained  in the previous  year near Col strip
(x  =  4,392,  SD   = 286).   In  1976,  bees   from ZAPS I had  lower sulfur  (x =
2,508, SD =  504)  in late June  than in  August (x =  3,990, SD = 942).  However,
in  August,  the  bees   near  Colstrip  contained more sulfur  (x  = 4,933,  SD =
506) than bees from ZAPS.   Hillmann  (1972) found  lower  sulfur in  larval bees
than  in adult bees.   In June,  bee  colonies would have contained a greater
proportion of young bees than the same  colonies  in August.

                                     484

-------
     Currently,  the drop-trap  captures  are  being  processed, and  those  data
will be  presented  in  the next  report.   The traps worked well  and were  very
specific to  the  collection of bees and  flies.   The  pollen  they were carrying
verified them as  pollinators.

     We  also are in  the process  of  completing the  sticky  trap evaluations.
So  far,  we  have not  found  any  demonstrable  trends  in  numbers  across  the
plots.    Yellow  sulfur  butterflies (Colias   eurytheme  Bdv.)  were attracted to
yellow   sticky   traps  over  white  traps.   The  numbers  were  significantly
different  for  some plots   (Table  14.3)  but did  not correlate  to the  S02
treatment levels.

     The deposition of  small  insects on  the sticky  traps  may  be  useful  for
the  interpretation  of prevailing  wind  patterns on the plots  at the level  of
the  pipeline,  assuming  that  the insects are deposited more  or  less as inert
particles  as suggested  by  Taylor (1962).  Wind roses, based on insect depo-
sition  for  each  plot,  are  presented  in Appendix  C.  This  information  has
been  included because  the  instrumentated wind  data are  not  yet  available
from  EPA,  and  because there  is only a  single wind  monitor at each  of  the
ZAPS sites,  and  it  is located approximately  three  meters  above the surround-
ing  terrain  and may  not produce readings representative of wind  patterns  at
the  level  of  the pipes.   The insects  counted  on the  traps for this purpose
were small,  weak fliers  such as  aphids  and thrips,  which  would  most  likely
be active during  the daylight hours.

     To  date,  the  identifications   and   counts  of  beetles  in the pit-fall
traps  have  been  completed for  about 80  percent  of the traps.   Figure  14.3
presents  linear  regression  lines  (based on  theoretical   fumigation  means)
which  depict the population  changes  of beetles of  the  genus Canthon (mostly
Canthon  laevis),  a  decomposer  dependent  on  fecal  and dead  organic materials
(Ritcher,  1958).   The trapping  data (Appendix  D)  for May  at  ZAPS  I  versus
ZAPS II  shows that more beetles  were  present  on each of the  ZAPS II  plots.
The  number of beetles  captured  on ZAPS  I  in May was  low compared to ZAPS II,
but  the population  numbers  increased  until they  were  only  slightly  lower
than those  at ZAPS  II  in  August.   Also,  the  number of beetles  captured  on
the  LOW  plot at  ZAPS   I  in  May  was unusually low (21) compared to other plots
and  dates, which produced a positive regression  slope;  but  the number of the
beetles  on  the  CONTROL  plot exceeded  those at any  other ZAPS  I  plot.  Com-
puting  a regression  line for the  number of beetles on the  CONTROL,  MEDIUM
and  HIGH plots  gave  a  negative  slope,  which  is  still  less  steep  than  that
for ZAPS II in May.

     The relatively low, beetle  populations  on  ZAPS I in May  could  be  due  to
a  residual  effect  from  the previous year's  fumigations, to S02 levels above
the  desired  delivery   concentrations  on  some of the  plots  during  the  spring
equipment adjustments, to grazing influences (ZAPS  II was fenced  in 1976 and
ZAPS I  in  1975)  or  to attraction  of  beetles to rodents caught  in traps which
had inadvertantly been  left on ZAPS I during the  first part of  May.
                                      485

-------
   TABLE 14.3.   STICKY-TRAP CAPTURE OF COLZAS  EURYTHEME (SULFUR BUTTERFLIES)
                FOR 1520 TRAP DAYS, FIVE SAMPLE PERIODS, N = 985.
Period Ending
Mean number
 Total Capture

     SD

     SE

Goodness of Fit
 Total Capture

Plot by Plot
                             ZAPS I
                              S02
                                         ZAPS II
                                           S02
CONTROL  LOW   MEDIUM   HIGH   CONTROL  LOW   MEDIUM   HIGH
July
Aug.
Aug.
Sept.
Sept.
19, 1976
5, 1976
16, 1976
2, 1976
19, 1976
35
50
18
10
14
18
55
26
9
3
28
27
29
7
12
24
35
22
19
10
26
26
43
19
27
28
25
44
13
13
21
41
29
10
8
54
36
38
13
20
25.4    22.2   20.6    20.0

16.7    20.3   10.3     9.0

 7.5     9.1    4.6     4.0


X2  = 2.7, df = 3, P >0.4
                                 28.2    24.6   21.8    32.2

                                  8.9    12.8   13.7    16.1

                                  4.0     5.7    6.1     7.2
                                  X2n = 9.6, df = 3, P <0.05
  X2n = 109.9, df = 1, P <0.001  X2n = 5.0, df = 1, P <0.05
                    for HIGH vs MEDIUM
                    X2n = 5.1, df = 1, P <0.05

                    for MEDIUM vs CONTROL
                                  for HIGH vs LOW

                                  X2n = 1.3, df = 1, P >0.3

                                  for CONTROL vs HIGH
     By  August,  the Canthon  beetle  populations on the  control  plots of both
ZAPS  I  and  II  were very  similar,  and the  regression  slopes  describing  the
population decreases  across  the plots  at both  sites  were  almost  identical,
indicating that  the  populations  of  this  beetle were  influenced by  the  in-
creasing  S02  concentrations.   We  have  not presented significance  levels  for
regression because  at  the time  of this  report data concerning the measured
levels of S02  in the ambient air were not available.

     The  results  of chi  square analyses  to  test significance  of the differ-
ent total  catches  for  each  plot  were highly  significant  (Appendix D).   This
conclusion is  consistent with that of  Frietag et  al_.  (1973),  who  reported a
negative  correlation between  the  numbers  of ground beetles  (primarily preda-
tory Carabids  and  one   carrion-feeding  Silphid) and  sulfate fallout from a
Kraft mill.
                                     486

-------
           700 -r
           600 --
           500 ..
        oo
        tt  400  --
        55
        CO
           200
           TOO
                         MAY  1976  (NOT  BAITED)

                         AUGUST  1976  (BAITED)
                    ZAP I
             CONTROL       LOW            MEDIUM

                                    SULFUR DIOXIDE
                                                       HIGH
Figure 14. 3.
Number of beetles (Canthon sp.) in pit-fall  traps/S02 treat-
ment plot,  plotted against theoretical fumigation levels.
     In May,  significant differences were documented  (X2  analysis)  (Appendix
E)  at  both  sites  in  the  numbers  of  five  species  of beetles  representing
three  families  (Carabids,  Silphids,  and  Scarabaeids).   In  most cases,  the
number  of  beetles  tended  to decline sharply  (Figure 14.4)  as  the S02  con-
centration  increased.    Later  the  capture  numbers   of  all  but  the  Canthon
beetles and  of  the  Silphids  (Nicrophorus  spp.) decreased to very  low levels
on  all  plots,  including the  controls,  and the  numbers of  Silphids  no longer
showed any  demonstrable relation to S02 level  (X2 analyses).
                                     487

-------
00
00
       oo
       >-

       Q
       Q-
       T

                                                        <
                                                        Q
                                                                  I  30-
                                                       o
                                                       o
                                                       ro

                                                       to
                                                        !-  20-
                                                        111  t-u
                                                        LiJ
                                                        CO

                                                        U_
                                                        O
                                                                  CQ
                                                                  s:
                                                                  ID
                                                           10-
                                                                        •Trox sonorae


                                                                        v Pasimachus elongatus
                                                              CONTROL
MEDIUM
                                                                      S02  CONCENTRATIONS
HIGH
               Figure 14.4A.   Pit-fall  captures  of  beetles  on ZAPS I treatment plots, May 1976.

-------
-P=>
00
         540 -
      oo

      
-------
                                 DISCUSSION

     The honeybee  work at ZAPS  I has  been  limited by the  need  to  place the
hives off  the  treatment plots.   They  may receive some drift  from the plots,
but  this  has  not  been  measured.  Extension  of  the  delivery  lines to  the
hives appears  to  be  impractical.  Moving of  the  hives onto  the plots  would
insure  exposure but  could result in  interference with the  sampling by  other
investigators,  at  least one of  whom   is  known to be allergic  to bee stings.
Bees forage on the  plots;  one  colony  swarmed  onto the  HIGH plot in June 1976
and  settled  under  a  small  shelter  from which  they   had  to  be removed.   A
greater number  of  bees  in  the drop traps  in  September  1976  on the CONTROL
and  LOW pphm  plots  more  likely  was   related  to  the  fact  that there  were
smaller, weaker colonies above the higher  plots than  to an  avoidance of the
sulfur,  since in 1975 the bees showed no differential avoidance.

     The  supercedure  rate   seems  rather  high  and  exceeds  that of similar
colonies in  Broadus.   In 1975, we thought  this problem was  a consequence  of
artificial   insemination.   In  1976,  queens  were  used  that had  been insemi-
nated during a nuptial  flight,  and were selected to insure healthy colonies
and  queens.  Since  the  queen  may live for several years,  while  workers only
live  for  a  few weeks  or months at  the longest, the  queen  is  usually the
oldest  member  of  the  colony  and as   such  would  be exposed to  any environ-
mental  contaminants  for a longer period than  the  workers.   The  inability  to
detect  significant  differences   in  total  sulfur  between  bees  at  the  ZAPS
plots and  bees in  other areas may be  due  to  a masking effect from the high
natural   levels of   sulfur  of  animal   tissues  bound  in proteins.   Hillmann
(1972)  reasoned that  this  explained  the  lack  of  significant increases in the
sulfur  content of  adult worker honeybees fumigated with as much as 5 pphm of
S02  for 9  and  14  weeks as compared to non-fumigated  controls.  He found that
S02  produced significant  reductions in brood rearing which  led to correlated
declines  in pollen   collection.   Both  brood   rearing  and   pollen  collection
showed  an  inverse  linear  relationship to   increasing  S02  concentrations.
Hillmann thought  that  reduced  brood production may reflect  an effect of S02
on  the  queens.  Also,  he found that control colonies  exhibited  significantly
more  flight  activity than fumigated  colonies   but  saw  no  significant differ-
ences in  the  degree  of flight suppression among  the three fumigation treat-
ments.  Reduced  flight activity  contributed   to  lessened  honey and colony
weight.   The toxic  effect of  S02 in  Hillmann's studies was  unclear:  fumi-
gated bees  showed  significant  mortality increases in  the   firs£ year of the
study but  not  in  the second year.   He  did not observe behavioral abnormali-
ties  induced by S02  fumigation based   on  his observations  of bees in a glass-
observation hive.   Acute dosages of  approximately 1,000  ppm  S02   indicated
that  adults were  more  sensitive  than   larvae,   eggs, and  pupae,  respectively.
However, we  believe  that it  is  unlikely  that these  extreme  levels  of S02
would be encountered  in  the ambient air

     Available  and  preferred  forage   is  often better  off  of  the  ZAPS  plots
than on because there  are nearby alfalfa and  clover  hay fields.  Thus,  it is
likely,  as  observations  suggested,  that the   majority of  the  bees in each
colony  did not  forage  the  ZAPS treatment  areas,  including  the   control,
unless  something  attracted  them  onto  the plots such as the feeding  stations
or the water in the drop-traps.

                                     490

-------
     Since significant  increases in  fluorides  in  bees  were found  near Col-
strip and  a  probable  toxic  effect was  identified at one  site  (see Colstrip
vicinity studies,  this report),  it is  believed  that honeybees  are particu-
larily  efficient  and  sensitive  bioindicators  of  some  major and  trace ele-
ments.   Sulfur and  sulfur  compound  impacts may be  more  difficult to identify
and  measure,  but  Hillmann's  data  (1972)  indicate that sulfur  dioxide does
have significant  impacts  on  honeybees.   In 1977, we intend  to  move the bees
into areas where  the  bees  may  have a  greater probability of exposure to S02,
such as  in the buffer zones,  and the  levels to  which they  are  exposed using
sulfation  plates  will  be  monitored at  each  hive.   Behavioral  responses such
as  flight  activity and physiological  responses  such  as mortality  and brood
rearing  influenced  by S02 will  be examined.   For  comparison purposes, tech-
niques will be  similar to those used  by  Hillmann.

     Hillmann  (1972)   found   that  tissues  of  larval  or immature  bees  had
significantly  less  sulfur than  those of  adult  bees.   The  bees at  the ZAPS
plots  contained   less  sulfur  in  June than in August.   June  is a  period  of
rapid population  increase; whereas  August  is a period of a relatively stable
or  declining  population size.   Thus,  the  bee populations in  June  would have
had  a  greater proportion  of young adult  bees,  which probably  accounted for
the  lower  sulfur content.    Bees  at the ZAPS plots consistently demonstrated
lower  sulfur  levels than  bees at the other  sites.  Also, there  seemed to  be
a  tendency towards lower sulfur  in both bees and  beetles on  the HIGH treat-
ment  plot.   Lower  sulfur  content  in insects  on  plots  fumigated with  S02
could  occur  as a  consequence of rapid population  turnover  or  shortened life
span.   This  would  increase   the  proportion  of  young to old members  of the
population.   Assuming  that   sulfur  increases  in  older  insects,  this  could
explain the observed trend.

     Data  from the  beetle  trapping   are  important as  regards   soil  cycles.
The  Scarabaeidae  and  the Silphidae   gather,   bury,  and  break   down  organic
material.  This  in turn releases nutrients  to  the soil systems.   There are
several  literature  reports  that  support the  hypothesis  that air pollutants,
particularly  heavy  metals,  accumulate in   litter and the upper  soil horizons
and  interfere  with  nutrient  cycling.   In a review  on  depressed  litter decom-
position  and  reduced  nutrient availability,  Smith (1975)  stressed  the need
for  studies  on   the  effects  of contaminants  on  these  processes   and  their
ultimate   effect  on  plant  growth.    Also,  many  contaminants   are  toxic  to
microorganisms,  particularly fungi, which  are primarily responsible  for the
decomposition  of organic  materials in  the soil  (Horsfall,  1956).   A number
of  insects,  including  the dung  beetles  such as Canthon,  disperse spores  of
saprophagic  or coprophilous   fungi  (Lodha,  1974).  These insects act  both  as
agents  of  mechanical   breakdown and  movement  of  organic  materials  and  as
disseminators  of  primary microbial decomposers.   Thus,   any  reduction in the
populations  of these  beetles  could   have  effects  on  the   decomposition  and
nutrient cycles.

     Based  on  the  data,  the  numbers  of Canthon beetles  on ZAPS  I  built  up
through  the  summer to  the  level of   the beetles on  the ZAPS II plots.  This
indicates  that what may  be   occurring is  a  depressed  immigration  or popula-
tion  growth  on the  fumigated  plots  rather than an  elimination or migration
                                      491

-------
of the beetles  from  these plots.   Waldow (1973) discovered  an  antennal  cell,
a  coelosphaerica,   in   carrion  beetles   (Nicrophorus   vespilloides)  which
responds  to  the  odor of  carrion,   H2S,  and  a few  cyclic  compounds.   This
receptor may be  responsible  for a coding of  the  odor of rotten meat.  Sulfur
dioxide could  act in a  somewhat  similar manner,  interacting on  the insects'
ability to  orient to food  material  by  olfaction.   This can  be  tested  in a
simple y-tube olfactometer patterned on the design by Burkholder (1970).

     Data from  the other  sampling procedures at ZAPS as well  as within-plot
distributions of insects,  will be presented in  the next interim report.


                           SUMMARY AND CONCLUSIONS

     We  have focused  our studies  on ground-dwelling  beetles  and  on honey-
bees—insects  which   our   studies  and  those   of   the  Colorado  investigators
indicate are most susceptible to S02.

     Both  the   types  of  experiments  and  the  results  of   experiments  using
honeybees at the ZAPS plots  were limited by  the  placement  of the beehives 75
m  above  each  of  the  treatment  plots.   This  averted  possible  problems  for
investigators  from bees  in  close  proximity.  However,  it introduced  a  new
problem since  only bees  foraging  on  the plots  would  be  expected to receive a
full  strength  exposure  of short  duration  to  the  S02 gas.  The  bees did not
demonstrate  measurable  avoidance  of the  sulfur gas  on any  of  the  plots,  nor
did  they  exhibit  obvious  signs of sublethal  effects such  as  intoxication or
of  lethal effects as  indicated by mortality.   However, bees  did  not success-
fully  locate feeding stations  hidden  from view,  even  if  an  attractive fra-
grance  was   added to the sugar  solution.   Odor  may  be more  important than
vision  as a guidance  system to  bees,  and the addition  of  small  amounts of
essential  oils  should  be  a  strong attractant  (von Frisch,  1967).   It  is
possible  that  the S02 interfered with the  bees' olfactory  sensory  systems or
that odors other than  those tested are more attractive.

     In  May,  several  species  of  ground-dwelling  beetles,  mostly Scarabaeids
and  Silphids  (carrion  beetles),  demonstrated  significant  inverse relation-
ships  in  numbers  compared to  S02 fumigation  concentration.  By  late summer,
tumblebugs  (Canthon   spp.)  continued to  demonstrate this  relationship.  The
response  of carrion  feeders was  variable, and the  capture of  other beetle
species  was too low for  meaningful  comparisons.   The  inverse 'relationships
of  numbers  to S02 levels  appeared to  be related  to movement of  beetles onto
the plots,  since the  beetle  numbers  on  the ZAPS  I plots increased during the
season.   The increase was  greatest  on the  control  and the  low  fumigation
treatments.   Hydrogen  sulfide  apparently is an attractant  to carrion beetles
(Waldow,  1973).    Since   hydrogen  sulfide  attracts  carrion  beetles,  it  was
hypothesized  that other  sulfur  compounds  may act  in  a similar manner,  and
the release  of  S02 gas  over large areas  could mask the ability of beetles to
detect and to orient towards food sources of  decaying organic materials.

     The  sulfur  content  in  honeybees  reared at  the  ZAPS  plots during 1975
and  1976  tended to   be  somewhat  lower than  the sulfur content  of  bees at
sites  in  other  areas of  southeastern Montana.   Also,  the  sulfur  content of

                                      492

-------
bees  at  ZAPS was  relatively low  in  June  and  increased by  August  1976.   In
both  bees  and beetles,  there appeared to  be  a trend towards decreased sulfur
in  insects  on the HIGH treatment  plots.   Since  Hillmann (1972) reported that
immature bees  had significantly less sulfur than older  bees, it was suggested
that  decreased  sulfur  in  insects  exposed  to  S02 could be  explained by high
turnovers  or  shortened life  spans  of  the  members of a  population,  which
would result in a  higher proportion of younger  to  older individuals.

     The data  processing  and analyses of  insects  other  than beetles captured
by  the  trapping  systems  (pit-fall,  drop-trap,  and  sticky-trap)  in  1976 were
not  complete at  the time  of  this   report,  but  this   information  should  be
available for the  next interim report.

      Emphasis  on  sampling  programs  (population  estimates  and  population
dispersal)  will  be  continued  but  will  be  restricted to selected species  and
refined  to maximize trapping or capture  precision  and  success, applicability
to  statistical  analyses,   and  efficiency in terms  of  man-hours.   Trapping
regimens will  be  limited  to three  or  four periods  coinciding with seasonal
population  cycles,  replacing more  or less  continuous sampling at considerable
effort  and expense.  The  intent   is  to  de-emphasize repetitive  sampling  and
redirect   efforts   toward   process-   and   mechanism-oriented  experiments  to
determine the basis for  observed responses to S02.

     At  present,  beetles  seem to   offer the greatest  potential  for studies at
the  ZAPS plots.  Not only  do they appear to respond to  S02  but  they include
important  predators,  saprophages, and  necrophages.   The  latter two  groups
undoubtedly  play  a  role  in  litter  decomposition  and  nutrient  cycling.
Rigorous  investigations of  one  or two  major species of beetles and  of  the
major ecological  events and processes interacting with  these populations  are
needed.   Also,  the  role of these   species  in  the grassland ecosystem and soil
ecosystem  must  be appraised.   This requires both quantitative and qualitative
information  regarding  behavior,   seasonal  cycles,   and population  dynamics
relative  to -changes  in biotic and  abiotic  components  of the  ecosystems  in
order  to  evaluate  the  chronic   effects   of  S02  stress.    This  work  would
contribute  both to  the  goals of the  overall  project as   regards the impact of
S02  and  would  provide  valuable information  concerning  soil  systems  and  the
ecology  of  soil   animals.   It  is  particularly well-suited  to   the  current
needs of the  project in view of a lack  of information on soil  systems and of
the  cessation  of  above-ground arthropod sampling  and  re-directed emphasis  by
the Colorado investigators  on soil  arthropods.


                                 REFERENCES

Bb'sener,   R.   1969.   Zum Vorkommen ridenbriitender Schadinsecten  in  rauchges-
     chadigten  Kiefern-und  Fichtenbestaenden.   Arch.  Forstwes.  Bd.  18:1021-
     26.

Broadbent,   L.   1948.   Aphis Migration  and  the Efficiency  of  the  Trapping
     Method.  Ann. Appl. Biol. 35:379-94.
                                      493

-------
Burkholder, W.  E.   1970.   Pheromone  Research  with Stored-Product Coleoptera.
     In:   Control  of Insect  Behavior  by Natural Products.   D.  L.  Wood, R.  M.
     Silverstein,  and  M.  Nakajima  (Eds).   Academic  Press.   New York-London.
     p.  1-20.

Debackere, M.   1972.   Industriele  luchtvervuiling  en bijenteelt.  (Industrial
     pollution and apiculture).  Vlaams Imkersblad.   2(6):  145-155.

DHES  (Montana Department  of Health  and  Environmental   Sciences),  Bureau  of
     Air  Quality.   1971.   Implementation  Plan  for  Control  of  Air Pollution
     in Montana,   (revised January 24,  1972).

Donaubauer,  E.   1968.    Sekundarschaden  in  Osterreichischen   Rauchschadens-
     gebieten.  Schwierigkeiten  Der diagnose  und  Bewertung.  Miedzynarodowej
     Konf. Wplyw  Zanieczyszczen Powietrza na  Lasy,  6th  Katowice, Poland,   pp
     277-284.

EPA  (U.   S.   Environmental   Protection  Agency).   1976.   The Bioenvironmental
     Impact  of a Coal-Fired  Power  Plant.   (2nd Interim  Report)  EPA-600/3-76-
     013.

Frietag,  K. ,   L.  Hastings,  W.   R. Mercer,  and  A. Smith.   1973.   Ground  Beetle
     Populations near a Kraft Mill.  Can. Entomol.  105:299-310.

Gary,  N.  E.    1971.   Magnetic  Retrieval   of   Ferrous  Labels   in  a  Capture-
     Recapture System  for  Honeybees  and  Other Insects.   J.   Econ.  Entomol.
     64:961-965.

Heathcote,  G.  D.   1957.   The  Optimum Size   of  Sticky  Aphid  Traps.   Plant
     Path.  6:104-107.

Henderson, K. D.   1962.  Pollen Trap.  Can.  Bee J.  73(10):3-4.

Hillmann,  R.  C.   1972.   Biological   Effects   of  Air  Pollution  on   Insects
     Emphasizing  the  Reactions  of the  Honeybee (Apis mellifera L.) to  Sulfur
     Dioxide.   Ph.D.  thesis.   The  Perm.  State University, University  Park,
     PA.

Hillmann,  R.  C.  and A. W. Benton.  1972.  Biological Effects of $ir Pollution
     on  Insects,  Emphasizing  the  Reactions  of the  Honeybee (Apis mellifera
     L.) to Sulfur Dioxide.   Elisha Mitchell Sci. Soc.  88:195.

Horsfall,  J.   G.   1956.  Principles  of  Fungicidal   Action.   Chronica  Botan.
     Com. Waltham, Mass.   280 pp.

Kanaga,  E.  E.   1956.   An  Evaluation of  the  Use of Sulfur  Dioxide  in  Fumigant
     Mixtures  for Grain Treatment.   J.  Econ. Entomol.   49:723-9.

Kudela,  M. ,  and  E.   Novakova.   1962.   Lesni  skiidci a skody ve  zveri  v  lesich
     Poskozovanych Kourem.  Lesnictvi.  6:493-502.
                                      494

-------
Kulman,   H.  M.   1974.   Comparative Ecology  of North  American  Carabidae  with
     Special  Reference  to Biological  Control.   Entomophaga,  Mem.  H. S.  7:61-
     70.

Likens,   G.  E.  and  F-  H.  Bormainn.   1974.   Acid Rain:   A  Serious Regional
     Environmental Problem.   Science.  184:1176.

Linzon,   S.  N.  1966.   Damage  to  Eastern  White Pine  by Sulfur Dioxide  Semi-
     mature-Tissue  Needle Blight,  and Ozone.   J.  Air Pollut.  Control Assoc.
     16:  140-144.

Lodha,  B.   C.   1974.   Decomposition  of  Digested Litter.   In:   Biology  of
     Plant  Litter  Decomposition.   C.  H.  Dickinson  and G.  F.  Pugh (Eds).   Vol.
     1,  Academic Press,  New York-London,  p. 213-238.

McGregor,  S.  E.  1976.    Insect  Pollination  of Cultivated Crop  Plants.   USDA,
     ARS.  Agric.  Handbook No. 496.

Montana Agric. Reports.  1977.  Unpublished Draft.   February, 1977.

Negherbon,  W.  0.   1966.  Sulfur Dioxide,  Sulfur  Trioxide,  Sulfuric Acid and
     Fly  Ash:  Their  Nature and Role  in Air  Pollution.    Prep,  for Edison
     Electric  Inst. ,  EEI  Publ.   66-900.   Hazelton  Lab.,  Falls  Church, VA.
     1213 pp.

Olson-Elliott  and  Associates.   1976.   Effects  of Spruce  Budworm Control  on
     Pollinating   Insects.    Report   to:    USDA  Forest  Service,   Region   I,
     Missoula, Montana.

Ritcher,  P.  0.    1958.   Biology  of Scarabaeidae.   Ann.   Rev.   of Entomol.
     3:311-334.

Scheffer,  T.   C.  and  G.  G.  Hedgcock.  1955.   Injury to Northwestern Forest
     Trees  by  Sulfur Dioxide from Smelters.   USDA Tech.  Bull.  No. 1117.  49
     pp.

Sierpinski,   Z.   1966.    Znaczenie   gospodarcze  skosnika  tuzinka  (Exoteleia
     dodecella  L.)  na terenach uprzemyskowionych.   Sylwan.  110:23-31.

Sierpinski,   Z.   1967.    Einfluss von  industriellen  1 ufterunreinigungen ant
     die populations dynamik  einiger primarer Kiefernschadlinge.   Proc.  14th
     Congr. Int. Union Forest.   Res. Organ.  (IUFRO Congress) 5:518-31.

Smith,  W.   H.    1975.  Depressed Litter Decomposition.   In:   Air Pollution and
     Metropolitan  Woody  Vegetation.   W.  H.  Smith  and L. S.  Dochinger (Eds).
     Yale Univ. Printing Service, New  Haven, CT.  p. 33-34.

Smith,   W.  H.   1975.   Introduction.   In:   Air  Pollution  and  Metropolitan
     Woody  Vegetation.   W.   H.  Smith  and  L.  S.  Dochinger  (Eds).   Yale  Univ.
     Printing Service, New Haven, CT.
                                      495

-------
Southwood, T.  R.  E.   1975.   Ecological  Methods  with Particular Reference  to
     the  Study  of  Insect  Populations.   Chapman   and   Hall   Ltd.,   London,
     England.  391 pp.

Staples,  R.  and W.   B.  Allington.   1959.   The  Efficiency  of Sticky Traps  in
     Sampling  Epidemic  Populations  of the  Eriophyid  mite Aceria tulipae  (K.)
     Vector of Wheat Streak Mosaic Virus.   Ann.  Ent.  Soc. Amer. 52:159-164.

Swisher,  E.  M.   1944.   Sulfur Dioxide-Acetone  as a  Household Fumigant.   J.
     Econ. Entomol.   37:694-7.

Taylor,  L.  R.   1962.  The  Efficiency of Cylindrical  Sticky Insect Traps and
     Suspended Nets. Ann. Appl. Biol.  50:681-685.

Templin,  E.    1962.   On  the  population  dynamics  of  several  pine  pests  in
     smoke-damaged   forest   stands.   Wiss.   Z.   Tech.  Universitaet  Dresden.
     113:631-637.

von  Frisch,  Karl.    1967.    The   Dance   Language  and  Orientation  of  Bees.
     Bel knap Univ. Press.  Cambridge, Mass.  566 pp.

Waldow,  U.   1973.    Electrophysiologie   eines   neuen  Aasgeruchrezeptors und
     seine  Bedeutung  fur  das Vernal ten  des  Totengrabers  (Necrophorus)  J.
     Comp. Physiol.   83:415-424.

WES  (Westinghouse  Environmental  Systems).   1973.   Colstrip  Generation and
     Transmission   Project.    Applications  Environmental   Analysis.   Westing-
     house Electric  Corporation.   Pittsburgh, PA.
                                     496

-------
                                  APPENDIX A



See Appendix A in the section 5 of this report by Bromenshenk entitled:






             INVESTIGATION OF THE IMPACT OF COAL-FIRED POWER PLANT



             EMISSIONS UPON INSECTS:   ENTOMOLOGICAL STUDIES IN THE



                         VICINITY OF COLSTRIP, MONTANA
                                     497

-------
                                 APPENDIX B

                 DESCRIPTIONS OF THE INBRED LINES OF HONEYBEES
                  UTILIZED IN THE ARTIFICIALLY INSEMINATED
               QUEENS OBTAINED FROM THE USDA BEE STOCK CENTER.


Queens were a dihybrid YDGk.


Drones were a Pa line.


YD  is  a yellow  line.   Queens  are  stocky.   Brood  rearing  is  initiated early
in  the season and  continues  late  into  the  fall.   They winter  poorly,  con-
suming lots of honey and often producing inadequate stores.


Gk  is  an   intermediate  color  between  black and  yellow.   The workers  are
small.  The  queens   do  not  lay eggs  well.   Honey  production  is  low.   Win-
tering is poor.  Workers are marked  by 1 or 2 yellow bands.


YDGk hybrids produce  a good line.


^a  is  a yellow  line.   The queens are  large  and  are the best layers of the 18
lines  at  the  Bee Stock  Center.  Brood rearing  begins early and  ends  late in
the  season.   Population  size  is  good  during the  foraging  season.  The  bees
store little honey and are very susceptible to American Foul Brood.


YDGkxPa makes a good cross.
                                     498

-------
                                 APPENDIX C

     Wind  roses  derived  from  the  deposition  of  wind-carried  insects  on
sticky traps.   The  length of  each  line  indicates  the relative  frequency  of
insects captured and as  such  the prevailing wind pattern  during  the  daylight
hours.   The  angle  indicates  wind  direction  relative  to  the  layout of  the
plots, thus a  horizontal  line  indicates wind current  perpendicular to  the  N-
S axis  of the  plots.  In  all  cases the lines indicate wind  directed  towards
the point of intersection.
                                      499

-------
                                         ZAPS I WIND ROSES, JULY 15, 1976
          Control Plot
Low Plot
  \
  Medium Plot
                                                                                                     High  Plot
en
O
O
                                                     Figure C-l
        Control  Plot
                                          ZAPS I WIND ROSES, AUGUST 5,  1976
  Low  Plot
Medium Plot
                                                                                                       High  Plot
                                                     Figure C-2.

-------
                                        ZAPS I WIND ROSES, AUGUST 16, 1976
       Control Plot
 Low  Plot
  Medium Plot
 Hiqh  Plot
en
o
                                                    Figure C-3.
                                        ZAPS I WIND ROSES, SEPTEMBER 2, 1976
     Control Plot
Low Plot
Medium Plot
High Plot
                                                    Figure C-4.

-------
                                       ZAPS I  WIND  ROSES,  SEPTEMBER  19,  1976
     Control Plot
Low Plot                    Medium Plot




                Figure C-5.
High Plot
en
o
ro

-------
                                        ZAPS II WIND ROSES, JULY 15, 1976
en
O
CO
       Control Plot
         Control  Plot
  Low  Plot                Medium Plot



            Figure C-6
                                        ZAPS II WIND ROSES, AUGUST   5,  1976
Low  Plot                     Medium  Plot



             Figure  C-7.
High Plot
  High  Plot

-------
        Control  Plot
en
O
                                       ZAPS II  WIND ROSES,  AUGUST  16,  1976
    Low Plot              Medium Plot



              Figure C-8.






ZAPS II WIND ROSES, SEPTEMBER 2, 1976
       Control  Plot
   Low Plot
Medium Plot
                                High  Plot
High Plot
                                                    Figure C-9.

-------
       Control  Plot
                                      ZAPS II  WIND ROSES, SEPTEMBER 19, 1976
Low Plot             Medium Plot




        Figure C-10.
High Plot
en
O
en

-------
                                  APPENDIX D

     NUMBER OF BEETLES (CANTHON SP=)  IN PIT-FALL TRAPS/S02 TREATMENT PLOT.

                   May 1976--300 trap days/piot--not baited

Site	Number of Beetles	Computations	

ZAPS I  CONTROL            240                         Y = 141.908 + 1.198X
        LOW                 21                         r = .0565
        MEDIUM             149                         without LOW data
        HIGH               148                         Y = 220.000 - 16.200X
        All plots          558                         r = .667
        X2 = 174.2, df=3,  P<0.005

ZAPS II CONTROL            589
        LOW                374                         Y = 527.216 - 35.426X
        MEDIUM             358                         r = .932
        HIGH               185
        All plots         1506
        X2 = 218.3, df=3,  P<0.005


                  August 1976 - 120 trap days/plot - Baited

Site	Number of Beetles	Computations	

ZAPS I  CONTROL            767
        LOW                476                         Y = 658.405 - 44.154
        MEDIUM             371                         r = .894
        HIGH               269
        All plots         1883
        X2 = 294.1, df=3,  P<0.005

ZAPS II CONTROL            864
        LOW                443                         Y = 683.934 - 43.573X
        MEDIUM             352                         r = .764
        HIGH               336
        All plots         1995
        X2 = 370.0, df=3,  P<0.005
                                                 	9	

                  August 1976 - 50 trap days/plot - Baited
                          (Vegetation Burn Areas)

        CONTROL            426
        HIGH                66
        X2 = 263.4, df=3,  P<0.005


^Regression computations are based on theoretical rather than measured S02
 delivery concentrations.
                                      506

-------
                     APPENDIX E

X2 EVALUATION (EXPECTED VALUES EQUAL) BEETLE CAPTURES
    ON ZAPS I AND II TREATMENT PLOTS IN MAY 1976.


Species

Canthon sp.
Nicrophorus sp.
Qnthophagus sp.
Trox sp.
Pasimachus sp.


Canthon sp.
Nicrophorus sp.
Qnthophagus sp.
Trox sp.
Pasimachus sp.



CONTROL

589
153
136
43
39


240
38
315
12
8



LOW

374
100
131
19
13


21
9
22
0
5


Pic
MEDIUM
ZAPS I
358
26
92
12
20

ZAPS II
149
32
189
17
7


)t
HIGH

185
21
49
6
11


178
20
109
1
8



X2

218.3
160.4
48.1
39.5
23.6


173.4
20.15
292.9
27.9
.86



P

P<0.005
P<0.005
P<0.005
P<0.005
P<0.005


P<0.005
P<0.005
P<0.005
P<0.005
P<0.5

                        507

-------
                                  SECTION  15

              THE RELATIVE  SENSITIVITY OF  SELECTED  PLANT SPECIES
                TO SEVERAL POLLUTANTS SINGLY AND  IN COMBINATION

                                     by

                  D. T. Tingey,  L.  Bard,  and R. W.  Field


                                 INTRODUCTION

     With  the  introduction of  coal-fired power plants  into grassland  eco-
systems it becomes important to recognize and determine the possible effects of
power plant emissions  on the  ecosystems.   There  have been only limited studies
on native prairie plant species  either describing injury symptoms or providing
relative sensitivity data (Davis  et al_. ,  1966; Hill et ah ,  1974).

     The objectives of  this  study were to describe the  air pollution-induced
symptoms and to determine the relative sensitivity of selected plant species to
common air pollutants, singly and in combination.

                             METHODS AND MATERIALS

PLANT GROWTH

     Plant growth and  exposures  to air pollutants were conducted in greenhouse
facilities  in Corvallis, Oregon.  The plant material used in the study (Table
15.1) was  collected in  the Otter Creek Valley, Montana, near  the Fort Howes
Ranger Station  during August,  1974.   Since then,  the  individual species were
propagated  vegetatively;  the  grasses  by  division   and  Fringed  Sage  Wort by
cuttings.  The  plants  were grown  in  225  ml  styrofoam cups  containing  a 2:1
(v:v) mixture  of perlite:Jiffy  Mix.   Plants  for  all  studies  were  grown in
greenhouses;  watered  daily  with North  Carolina  State  University  phytotron
nutrient  solution  (Downs  and Hellmers,  1975)  and  periodically leached with
water.   The  plants  were grown  at day/night  temperatures  of 26-32/18^0; the
sunlight was supplemented and  the light period extended to 16  hr/day with light
from  HID  sodium  vapor  lamps.   Maximum  light   intensities   were  450-650
microeinstenins M-2  sec-1.

PLANT EXPOSURES

     Pollutant  exposures  were  conducted  in  single  pass  exposure  chambers
located in a greenhouse  (Heck, Dunning, and Johnson, 1968).  Sulfur dioxide and
nitrogen dioxide diluted in nitrogen were metered into  the exposure chambers at
a rate  sufficient  to  maintain the  desired gas  phase  concentrations.  Sulfur

                                     508

-------
TABLE 15.1.  PERCENT FOLIAR INJURY OF PLANTS EXPOSED TO SULFUR DIOXIDE FOR 4
             HOURS1
Plant Species
0.5
Agropyron smith ii Rydb
WESTERN WHEAT GRASS 0
Bouteloua gracilis Lag.
BLUE GRAMMA 0
Koeleria cristata (L.) Pers.
PRAIRIE JUNE GRASS 0
Stipa comata Trin. and Rupr.
NEEDLE & THREAD GRASS 0
Artemisia frigida Willd.
FRINGED SAGE WORT 0
Triticum aestivium L. em Thell.
WHEAT, HYSLOP 0
S02 Cone.
1.0

0

3

0

0

0
!
0
(ppm)
1.5 2.0

0 3

3 6

1 6

0 2

6 13

4 5

  1  Plants were exposed for 4 hr and foliar injury was assessed 96 hr after
     exposure.   The injury was measured as the percentage of the leaf area
     showing S02 injury.  Each mean was the average of 9 observations:

     Sx=1-

TABLE 15.2.  PERCENT FOLIAR INJURY OF PLANTS EXPOSED TO NITROGEN DIOXIDE FOR
             4 HR1
„
Plant Species

WESTERN WHEAT GRASS
BLUE GRAMMA
PRAIRIE JUNE GRASS
NEEDLE & THREAD GRASS
FRINGED SAGE WORT
WHEAT, HYSLOP

0.5
0
0
0
0
0
0
N02 Cone.
1.0
0
3
0
0
0
0
(ppm)
2.0
1
4
1
1
0
0

4.0
1
8
4
1
0
1
  1   Plants were exposed for 4 hr and foliar injury was assessed 96 hr after
     exposure.   The injury was measured as the percentage of the leaf area
     showing N02 injury.   Each mean was the average of 9 observations:
                                     509

-------
dioxide was measured with a flame photometric sulfur dioxide analyzer; nitrogen
dioxide and ozone were  monitored with chemiluminescence analyzers.  Prior to
each exposure the analyzers were calibrated using  known concentrations of each
pollutant.   The  plants  were  exposed  to  the  pollutant 4  to  6 weeks  after
vegetative propogation.  Plant injury was visually  assessed  in 5% increments 96
hours following  exposure  as  the percentage of the plant leaf area exhibiting
visual pollutant effects.

                                    RESULTS

PLANT INJURY  DESCRIPTIONS

     Plant injury resulting from sulfur dioxide, nitrogen  dioxide and ozone was
generally similar to previous reports (Jacobson and Hill, 1970).   Injury from
sulfur  dioxide  and  nitrogen dioxide  was  similar on the monocots.   On  young
leaves,  injury   developed  at the  leaf  tip,  on older  leaves  injury  usually
occurred at  the  bend  of the leaf.  Injury usually occurred as  small bifacial
lesions, and  as  severity  increased,  the lesions  coalesced  and spread up and
down the  leaf.   Interveinal streaks of necrotic tissue were frequent.   Lesion
color ranged from light tan to ivory.  On Fringed Sage Wort the  injury occurred
on  the  middleaged  leaf  tissue  as a bifacial  collapse  of the tissue,  killing
both  veins and  intervenial areas.  The ozone-induced injury on  Fringed Sage
Wort was  similar to  that  caused by sulfur dioxide.   Ozone  injury to monocots
appeared as tip  necrosis and small necrotic lesions  between the veins; as the
interveinal  lesions coalesced,  necrotic streaking resulted.  Injury tended to
be  concentrated between  the leaf  tip and  bend in  the  leaf.   Injury that
resulted  from the mixtures  of  sulfur dioxide  +  nitrogen  dioxide or sulfur
dioxide + ozone  were similar to  sulfur dioxide  injury.

RELATIVE SENSITIVITY

     To determine the relative  sensitivity of each of the plant species, they
were  exposed  to  either  single  pollutants  or mixtures of pollutants and their
injury  rated.   Wheat  was  included in  all  studies as an example  of an  agri-
cultural crop.   All species  were injured  following a four-hour exposure to 2
ppm S02, with Fringe Sage  Wort exhibiting the most injury and Needle and Thread
Grass the least  (Table 15.1).  Only Blue Gramma was  injured at  a concentration
of  less than  1.5  ppm.

     In exposures to nitrogen dioxide (Table 15.2) only Prairie June Grass and
Blue Gramma developed  significant  injury with Blue  Gramma  being  more sensitive.

     All plants  except Needle and Thread Grass were injured  by the lowest ozone
concentration (0.4  ppm)  (Table  15.3).   Blue Gramma and wheat were the most
sensitive, Western Wheat and Needle and Thread Grass  the least sensitive.

     Plants  were exposed  to either  mixtures  of sulfur dioxide  +  nitrogen
dioxide (Table 15.4) or sulfur dioxide + ozone (Table 15.5)  to determine  if the
gas mixtures interacted to alter the  injury response of the  plants.   Injury was
very  similar  on plants exposed to  sulfur  dioxide  with or without  nitrogen
dioxide.  When plants were exposed to sulfur dioxide with  or  without  ozone,
                                     510

-------
   TABLE 15.3.   PERCENT FOLIAR INJURY OF PLANTS EXPOSED TO OZONE FOR 4 HR1-
Plant Species

WESTERN WHEAT GRASS
BLUE GRAMMA
PRAIRIE JUNE GRASS
NEEDLE & THREAD GRASS
FRINGED SAGE WORT
WHEAT, HYSLOP

0.4
1
6
5
0
2
9
03
0.6
2
12
5
0
4
11
Cone, (ppm)
0.8
5
15
11
2
11
17

0.95
7
25
12
2
12
23

  1  Plants  were exposed for 4 hr and foliar injury was assessed 96 hr after
     exposure.   The injury was measured as the percentage of the leaf area
     showing 03 injury.   Each mean was the average of 9 observations, S- = 2.

  TABLE 15.4.   PERCENT FOLIAR INJURY OR PLANTS EXPOSED TO SULFUR DIOXIDE OR
               SULFUR DIOXIDE PLUS NITROGEN DIOXIDE1.


Plant Species               S02 Cone,  (ppm)    S02 Cone,  (ppm) + 0.1 ppm N02

                      073    O    O    T72    O    076O    T772~~
WESTERN WHEAT GRASS
BLUE GRAMMA
PRAIRIE JUNE GRASS
NEEDLE & THREAD GRASS
FRINGED SAGE WORT
WHEAT, HYSLOP
0
0
0
0
0
0
0
1
0
0
0
0
2
3
1
0
1
1
4
4
3
0
4
1
0
0
0
0
0
0
3
1
1
0
0
0
4
3
1
0
0
1
5
4
2
0
0
1

     Plants were exposed for 4 hr and foliar injury was assessed 96 hr after
     exposure.   The injury was measured as the percentage of the leaf area
     showing injury.   Each mean was the average of 8 observations:   S- = 2.
                                     511

-------
injury was  similar  except in Fringed Sage Wort  where the gas mixture caused
less injury than the sulfur dioxide alone.

  TABLE 15.5.  PERCENT FOLIAR INJURY OF PLANTS EXPOSED TO SULFUR DIOXIDE OR
               SULFUR DIOXIDE + OZONE1.
Plant Species
                            S02 Cone.  (PPM)

                       075O    ^
SO
                                     1.5    2.0
I2 Cone.  (PPM) + 0.1 ppm 03
7T5    O    T75    27o~
WESTERN WHEAT GRASS
BLUE GRAMMA
PRAIRIE JUNE GRASS
NEEDLE & THREAD GRASS
FRINGED SAGE WORT
WHEAT, HYSLOP
0
1
0
0
0
0
1
2
1
0
9
4
3
2
3
1
24
4
3
2
4
2
33
6
2
1
0
0
0
1
2
3
0
0
0
1
3
2
2
0
4
2
4
4
3
0
6
3

     Plants were exposed for 4 hr and foliar injury was assessed 96 hr after
     exposure.  The  injury was measured  as the percentage  of  the leaf area
     showing injury.   Each mean was the average  of 8 observations:

                                  DISCUSSION
                                                                  Sx=1'
     The injury observed on the native species was similar to that previously
described  (Jacobson  and Hill,  1970; Hill   et  al_. ,  1974).   Injury thresholds
suggested for the species tested in these experiments do not  take  into account
the effect of soil water potential  on plant sensitivity.  Low  soil  water poten-
tial could  reduce  plant sensitivity.   The levels of pollutants  used in this
study are higher than would usually be expected in the  field.  They were used
only to provide concentrations  that would  readily cause visual injury to aid in
injury description and relative sensitivity ranking.

     Davis  et ah (1966) reported that levels of sulfur dioxide  that might be
expected around  a  Phelps Dodge smelter in Arizona,  and which would defoliate
cocklebur,  did  not  injure  Blue  Gramma.   Hill   et  al_.  (1974),  working with
established native plants  in  the field,   showed that Agropyron caniun and A.
desertorum  required  between 6 and  10  ppm S02 for two  hours to  cause visual
injury.   This  suggests that these two species of Wheat  Grass  are more tolerant
of S02 than Western Wheat Grass.   Hill  watered the plants for a mtinth prior to
exposure to ensure that they  were in a stage of rapid growth and thus highly
sensitive.  He also  reported that  Stipa   occidental is required 10 ppm S02 for
              induce  visual  injury.  This suggests that Stipa occidental is is
               than  Stipa  comata which  showed  injury  at 2  ppm  S02 for four
two hours to
more tolerant
hours.
     Previous reports  (Tingey   et al_. ,  1971; Tingey  et al_. , 1973) of green-
house studies suggested that mixtures of sulfur dioxide + nitrogen dioxide or
sulfur dioxide + ozone interacted to cause more foliar injury than the effects
of the single gases.   These observations were not substantiated  in this study.
The gas mixtures did not increase the injury over that caused by  sulfur dioxide
alone.   The  lack of  synergism  from sulfur dioxide  + nitrogen has previously
                                     512

-------
been  reported  for  native  plants  (Hill  et al.,  1974)  and in growth chamber
studies (Bennet  et al.,  1975).

                                  REFERENCES

Bennet, J. H.,  A.  C.  Hill, A. Soleimani and W.  H.  Edwards.  1975.   Acute effects
     of  combination  of  sulphur  dioxide  and   nitrogen  dioxide  on plants.
     Environmental Pollution 9:127-132.

Davis, C.  R. , D. R. Howell and G. W.  Morgan.  1966.  Sulphur dioxide  fumigations
     of  range   grasses  native to  southeastern Arizona.  Journal  of   Range
     Management 19:60-64.

Downs, R.  J. and  H.  Hellmers.  1975.   Environment and Experimental Control  of
     Plant Growth.  Academic Press.  New York,  New York.

Heck, W.  W. , J.  A. Dunning  and H.  Johnson.   1968.   Design of a simple  plant
     exposure  chamber.   DHEW, National  Center  for  Air  Pollution  Control,
     Publication APTD-68-6.   Cincinnati,  OH.  24pp.

Hill, A.  C. , S. Hill, C. Lamb  and T. W.  Barrett.  1974.   Sensitivity  of native
     desert  vegetation  to  S02 and to  S02   and N02  combined.   Journal Air
     Pollution Control Association 24:153-157.

Jacobson, J.  S. and A. C. Hill.  1970.   Recognition of Air Pollution  Injury  to
     Vegetation.  A  Pictorial  Atlas.    Air   Pollution   Control  Association.
     Pittsburg,  PA.

Tingey; D. T. ,  R.  A. Reinert,  J.  A.  Dunning and  W. W. Heck.  1971.  Vegetation
     injury  from   the  interaction  of  nitrogen  dioxide  and  sulfur dioxide.
     Phytopathology 61:1506-1511.

Tingey, D.. T. , R.  A.  Reinert, J.  A.  Dunning and W.  W.  Heck.   1973.  Foliar
     injury  responses   of   eleven  plant  species  to  ozone/sulphur dioxide
     mixtures.   Atmospheric Environment 7:201-208.
                                     513

-------
                                 SECTION 16

             PROGRESS IN MODELING THE EFFECTS OF S02 FUMIGATION
                       ON AN EASTERN MONTANA GRASSLAND

                                     by

                 J. L. Dodd, M.  Coughenour,  and W. K. Lauenroth


                                INTRODUCTION

     The  modeling  effort  is charged  with (1)  adapting ELM  (the  grassland
ecosystem simulation model)  to  the  Montana sites;  (2) developing and implem-
enting simulated direct  effects of atmospheric  S02 on ecosystem components; and
(3)  developing  a  sulfur  submodel  to  examine  the   sulfur  cycle.   Work  is
proceeding ahead of schedule in the cases of (2) and (3)  and is slightly behind
schedule in the case of (1).

                      ADAPTATION OF ELM TO  LOCAL  SITES

     Adaptation  of ELM  to the mixed-grass prairie  ecosystem  has  proceeded
through completion of the application  and validation of the abiotic submodel.
Initial  conditions,  parameter  values,  and  validation  information for  the
producer, decomposer,  nitrogen, and phosphorus submodels are completed.

     The   investigations   are   currently  examining   procedures   for   model
adaptation.  Information developed  by  the Natural Resources Ecology Laboratory
(NREL) modeling  team  and supported  by new findings of  Dave  Tingey (personal
communication)  suggest  that modification  of the  structure  of  the producer
submodel will  lead  to considerable conceptual  strengthening of the whole model,
especially  in  regard to  the effects  of S02 on  the  ecosystem.   The dilemma,
which is influenced by limited funding, is  whether to proceed wit|j adaptation
of  the currently  implemented  model  and then  make the  new  changes  in that
version, or wait a few months until the new and more conceptually sound version
of ELM is available and then modify that version.  The best representation of
the effects of S02 will  be incorporated into the new version. In either case,
the data-based information needed for  adaptation will  be essentially the same
as that now being completed.

DEVELOPMENT AND IMPLEMENTATION OF SIMULATED DIRECT EFFECTS OF  ATMOSPHERIC S02

     Currently the NREL  Grassland Biome staff is modifying the structure  of the
primary producer submodel to more accurately represent the biomass dynamics of
grassland ecosystems.  A current draft of the  new structure is  shown in  Figure
16.1.    The  most  important changes  to  be  incorporated will  be (a)  the

                                      514

-------
                                                                                      Old
                                                                                 Standing Dead
                                                                                    Recent
                                                                                 Standing Dead
en
en
Photosynthetic
     Pool
                                                           Mature
                                                           Leaves
                                                          Juvenile
                                                           Leaves
                                                        Annual  Stems
                                                         and  Flowers
                                                       Perennial  Stems
                                                         Live  Roots
    Figure 16.1,   Structure of the new version of the primary production submodel now being implemented by the
                   NREL  modeling  staff.   The  significant  changes  in  the structure  are the  addition of an
                   explicit photosynthetic pool compartment and the splitting of the existing leaf compartment
                   into two compartments,  mature and juvenile leaves.

-------
photosynthetic production of a pool of translocatable carbohydrates  and (b) the
partitioning of this  photosynthate  among various portions of the  plant.  The
investigators feel that the new treatment of these two processes  in plant pro-
duction will be especially well suited for showing the direct effects on plants
of  inorganic  sulfur taken up  as sulfur dioxide.   It  is  hypothesized  that a
close  interrelationship  exists  between  the  process  of photosynthesis  and
partitioning of photosynthate.   Field experiments for summer of 1976 are being
formulated to test these  hypotheses.

     A new feature of  modeling mature leaves separate from juvenile  leaves will
also be amenable for showing the effects of sulfur taken up as sulfur.dioxide,
since these two leaf categories are  expected to exhibit differential  effects.
Related to this concept is the phenology of the plant and the  varying effects
with the  age  of the plant.  The plant  phenology  model  will  also be improved
before the final coupling to the sulfur cycling mode.

     Other  effects  on  primary  producers  are  expected  to be  shown on  the
processes of (a) transpiration, as reflected by changes  in  stomatal resistance,
(b)  uptake  of sulfate  from  the  soil  as reflected by  changes in  the  sulfur
status  of  the  plant,  and (c)  plant death,  as  reflected  by chlorosis  and
necrosis.

     If significant differences are  shown to exist among major species  of the
mixed-grass  prairie  with  regard  to  sulfur dioxide  uptake rates  and  related
effects on  the  above-mentioned processes,  the primary producer model  will be
able  to  reflect  these  changes  as   altered community  composition.   Modeled
functional groups would be selected  with regard to similarities of fundamental
production processes,  and similarity of responses to  sulfur dioxide.

            DEVELOPMENT OF SULFUR SUBMODEL TO EXAMINE SULFUR CYCLE

     The direct effects of sulfur dioxide pollution  on  primary production will
depend not only on the amount of sulfur dioxide present in the atmosphere, but
also on its distribution throughout  the plant system, and the  closely related
soil system.  A sulfur  cycling model  is ideally  suited to  demonstrate  these
relationships, and  a  current  draft  of its structure is shown  in Figure 16.2.
The state variables and processes are selected to maintain a level  of resolu-
tion comparible with ELM, and similarity exists with the existing nitrogen and
phosphorus models.  Not only  will the effects  of pollution be shown on primary
producers, but also on the entire sulfur cycling system, a capability set  in a
more entire ecology  perspective.

     Atmospheric sulfur inputs will  be in the form of sulfur dioxide,  and the
acid-rain  sulfate.   Rainfall  inputs  will  be  simply  a function  of rainfall,
allowing  coupling  to  atmospheric   models which  facilitate  prediction  of
concentrations in rainfall. Gaseous  inputs  to plants might be described by the
fol lowing scheme:
                    V  -  1     1      1      F
                                     516

-------
01
      Primary Producer



 Af    Abiotic

 '-"\
(BIO)  Soil Blotlc
    Figure  16.2.   Structure of  the  sulfur submodel  for ELM now being  developed to represent  the dynamics of
                   sulfur cycling  in grassland ecosystems.

-------
Where  V~    = deposition velocity
        (j
       r     = stomatal resistance
       r     = atmospheric resistance
        a
       r. .   = internal plant resistance
        int.            p
       F     = flux into plant
       C(Z)  = gas concentration at height Z
Where  F     = atmospheric flux
        cl
       K(Z)  = eddy diffusivity of air at height Z

       K(Z) is a function of wind speed and plant height;

       r  will be a function of light intensity, humidity, soil water, plant
        s species, plant condition, temperature;

       r. .   will be a function of metabolic activity of plant, sulfur
        in  ' concentration in plant, and plant species.

     Gaseous  uptake  or  deposition to the soil  and  to  surface litter will be
treated  in  analogous manner,  with an  atmospheric  resistance and  a surface
resistance.   The surface resistance of these two components will be functions
of water content, relative humidity, temperature, and amount of sulfur already
deposited.

     Such processes as decomposition of  litter may then  be altered according to
sulfur concentrations resulting from fluxes.  Although  the soil pH is expected
to  be  well  buffered in this geographic area, the pH of surface litter may be
altered.  This, in conjunction with the altered C:N:S ratio of the litter, may
then affect decomposition.  The complexities of the soil  sulfur  transformations
are  demonstrated in  Figures  16.3 and  16.4.  Although  the resolution of the
model  will  not permit a detailed treatment of these relationships, the major
processes could be represented and subsequently be reflected in the  cycling as
shown  in Figure  16.1.   The  great  predominance of  aerobic  versus* anaerobic
processes  in  this particular  system will  eliminate much  of  the  complexity
associated with Figure 16.3.   A detailed soil chemistry model is not  within the
scope  of  this effort,  however,  the major processes of Figure 16.4 might be
represented in a  rather coarse way.

     Another major reason for modeling  the soil components, as well as plant,
is  that  the  relationship  between plants and  soils  is  an established factor
governing plant distribution and  abundance.  Any alteration in soil  properties
or mineral  distributions  in  the  soil may  then be expected to play a role in
plant competitive relationships and the  related community structure.   No effort
is expected to model direct toxic effects of sulfur  dioxide upon consumers in
this model,  only the  consumer relationships  to  the system  by  way of plant
consumption.

                                      518

-------
              Microbe
               Death
                                Aeroebic
                              Heterotrophic
                           Aeroebic   \
                           Reduction   J
                                                     Anaeroebic
                                                     Oxidation
        Aeroebic
        Oxidation
                                  f  Anaeroebic
                                  (   Reduction
                                        Aeroebic
                                        Oxidation
Figure  16.3.
Conceptual  high resolution of sulfur  transformations  that poten-
tially  occur in soils.


                         519

-------
so2s


Adsorbed

Bases

Metal

co2
°2

N, P

Mineral

i
r
—
—

Organic

S03

•A

$2°*,

H2S

s


Figure 16.4.
Schematic to represent the complexity of chemical reactions that
can occur in soils.
                                      520

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
  EPA-600/3-78-021
                              2.
                                                           3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
  The  Bioenvironmental  Impact of a Coal-fired Power  Plant
  Third Interim Report,  Col strip, Montana
  December 1977	
                                                            5. REPORT DATE
                                                              February 1978
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

  Edited by Eric M, Preston  and  Robert A. Lewis
             8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS

  U.  S.  EPA
  Con/all is Environmental  Research Laboratory
  Corvallis, Oregon 97330
             10. PROGRAM ELEMENT NO.

                EHA541/E-AP-77ACV
             11. CONTRACT/GRANT NO.
 12. SPONSORING AGENCY NAME AND ADDRESS
                                                            13. TYPE OF REPORT AND PERIOD COVERED

                                                             Interim April  - Nnvpmhpr 1975
                                                            14. SPONSORING AGENCY CODE
  Same
                                                              EPA/600/02
 15. SUPPLEMENTARY NOTES
     In this  series of reports, the  First  Interim Report is EPA number
     and the  Second Interim Report is  EPA  number  EPA-600/3-76-013.
                            EPA-600/3-76-002
 16. ABSTRACT
       The EPA has recongnized  the  need for a rational approach  to  the incorporation of
  ecological  impact information  into power facility sitinq decisions  in the northern
  great plains.   Research funded by the Col strip, Coal-fired  Power  Plant Pro.iect is a
  first attempt to generate methods to predict the bioenvironmental effects of air
  pollution before damage is  sustained.  Pre-construction documentation of the en-
  vironmental  characteristics of the grassland ecosystem in the  vicinity of Colstrip,
  Montana  began in the summer of 1974.   Since then, key characteristics of the eco-
  system have  been monitored  regularly to detect possible pollution impacts upon plant
  and  animal community structure.
       In  the  summer of 1975, field stressing experiments were begun  to provide the
  data  necessary to develop dose-response models for SO^ stress  on a  grassland eco-
  system.   These experiments  involve continuous stressing of  one acre grassland plots
  with  measured  doses of SO^ during the growing season (usually  Aoril  through  October),
       Results of the 1975 field season's investigations are  summarized in this
  publication.   The six-year project will  terminate in 1980 and  a final  report will
  be published after data analyses  are complete.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS  C. COSATI Field/Group
 plant and animal  response to pollution
 coal-fired  power  plant
 air pollutants
 grassland ecosystems
 mathematical  modeling
 remote sensing
 micrometeorological  investigation
  coal-fired power plant
            emissions
  air  quality monitorina
  aerosol  characterization
51
 8. DISTRIBUTION STATEMENT
 release to public
19. SECURITY CLASS (This Report)
  Unclassified
                                                                         21. NO. OF PAGES
                                                                           561
20. SECURITY CLASS (This page)

  Unclassified
                                                                         22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE


   H U S. GOVERNMENT PRINTING OFFICE' 1978-799-936/85 REGION 10

-------