Watershed Sensitivity Measurement
   Strategy for  Identifying  Resources  at
   Risk from Acidic Deposition
    Institute of  Ecology, Indianapolis,  IN
   Prepared for

   Environmental  Research Lab.-Duluth, MN
   Jan 84


U.S. Itepsrtment of Gemmerce
Katkmsi Techni?:2l Infc-fmallsn Ssrvke

                                                     January 1984


                            Orie L.  Loucks
                            Roland W.  Usher
                       The Institute of Ecology
                     Indianapolis, Indiana   46208
      David Rapport                                  William Swanson
  University  of  Toronto                              Miami  University
Toronto, Ontario,  Canada                           Oxford,  Ohio  45056
                           Richard W.  Miller
                       The Institute of Ecology
                         and Butler University
                     Indianapolis, Indiana   46208
                     Cooperative  Agreement CR809328
                          Project Coordinator

                             Gary E. Glass
               Environmental Research laboratory-Duluth
                       Duluth, Minnesota   55804
                        DULUTH,  MINNESOTA    55804

                                          TECHNICAL REPORT DATA
                                 (Please read Instructions on the reverse before completing)
               3. RECIPIENT'S ACCESSION NO.~
                     pap  A    1*1  209
 Watershed Sensitivity  Measurement Stragety for Identify-
 ing  Resources  at  Risk  from  Acidic Deposition
                                                                       5. REPORT DATE
                                                                          January 1984

 0.  L. Loucks
                                                                       8, PERFORMING ORGANIZATION REPORT NO.
 The  Institute  of  Ecology
 Indianapolis,  Indiana   46200
               10. PROGRAM ELEMENT NO.
               11. CONTRACT/GRANT NO.

 Environmental  Research  Laboratory
 Office of Research  and  Development
 U.S.  Environmental  Prot-ction Agency
 Duluth,  MN   55804
               14. SPONSORING AGENCY CODE

           --Several lines of research on the effects of acidic  deposition have been supported  by EPA through
 Its  Environmental Research Laboratories and through a Cooperative Agreement with North Carolina State
 University.  The study reported here  was carried out as ,a subcontract  to The Institute of  Ecology from N.C.
 State University for research supporting programs at the Environmental Research Laboratory-Duluth.  The
 objectives Included review of existing  literature on the use of  Indices for quantifying resource. *tatus and
 predicting long-term trends In relation to'acidic deposition,  review of options as to the  form of a "sensitivity
 Index" or  loading tolerance model  for use In determining resources  at  risk, and Identification  of validation .  .
 steps needed to complete testing of the measure or model, arid  to begin  Its application.

             One section of the report describes the suite of  measures  which, when taken together, best Identify
 areas potentially sensitive to acidic Inputs.  Each of the component measures, when viewed separately, has
 certain  limitations which prevent  It  from being an adequate measure of  sensitivity; when considered together,
 however, as an  Integratlve measure, the limitations are less significant.

             For non-agricultural  systems, forest site Index appears to be a wy|l-establIshed  Integratlve
 measure capable of responding to altered soil and water chemistry.  The extent to which site  Index Is related to
 changes  In cation nutrient storage (due to cation stripping by acid precipitation) or other pollutant  Impacts
 still  Is Incompletely documented,  however.  Measurements of aquatic sensitivity have been  developed more fully,
 and  a number of experimental and  field data-based approaches exist. These  Include the Calclte  Saturation Index,
 the  Henrlckson  nomograph, the Almer/Dlckson relation and an additional measure proposed here  based on pH shock
 effects  during  acid flushing events.  This  Integratlve response property appears to describe  a  complex
 environment leading to.specles and population effects associated with  periodic but physiologically Important
 exposures  to H+ and AI \.
                                      KEY WORDS AND DOCUMENT ANALYSIS
                                                       b.IDENTIFIERS/OPEN ENDED TERMS
                                   COSATi Field/Croup

19. SECURITY CLASS (This Report)
                                                                                        21. NO. OF PAGES
20. StCURITY CLASS (This page I

                                22. PRICE
 '"A Form 2220-1 {Re*. 4-77)    PREVIOUS EDITION is OB>OLTE   1


This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication.   Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.

                             TABLE OF CONTENTS


LIST OF TABLES                                                         1V

LIST OF FIGURES     .                                                    v

SECTION I - INTRODUCTION                                                 1

     Concepts of Measurements for Ecosystem Response to Acid Deposition  3
     Study Objectives                                                    4

SECTION II - SUMMARY AND FINDINGS                                        7

     Approach                                                            7
     Results                                                             8


     Water Flow Pattern                                                 11
     Nitrogen Cycling                                                   13
     Sulfur Cycling                                                     16
     Alkalinity                                                         18
     Organic Interactions                                               19
     Cation Nutrient Stripping'                                         20
     Hydrogen Ion Toxicity                                              26
     Aluminum Mobilization and Toxicity                                 35
     Heavy Metal Mobilization and Toxicity                              46
             TERRESTRIAL AND AQUATIC AREAS                              53

     Measures of Sensitivity to  Forest Productivity Response?      .     54
     Measures of Aquatic Ecosystem  Sensitivity                          56
             AND  VALIDATION                                               71

      An  Approach to Assembling Sensitivity  Measurements                  71
      Other Information/Research Needs  for Sensitivity  Measure
           Validation                                                     75
 REFERENCES                                                              77

                              LIST OF TABLES
TABLE III-l:     Mean  and  median  concentrations  of  dissolved
                ionic  materials  in  bulk precipitation,  canopy
                throughfall,  forest  floor,  and  A2  horizon per-
                colate, and springwater during the 1975 and 1976
                growing seasons  at  the  Mt.  Moosilauke subalpine
                sampling sites, New Hampshire
TABLE III-2:    Net  export  of major  ions  for calibrated water-
                sheds in Canada
TABLE III-3:    Literature survey on pH toxicity to various fish

TABLE III-4:    Summary  of  the  effect of  pH  values  on fish

TABLE  III-5:    Summary  of  literature concerning aluminum toxi-
                city to  fish
TABLE  III-6:    Times  to 50%  mortality  for fish  in trials A-F
TABLE V-l:       Listing of data needs and computation procedures
                 applying   sensitivity  measures  and  pollutant
                 loading  tolerance models  in  northeastern North

                              LIST OF FIGURES
Figure 1-1:   Flow  diagram  showing  system  linkages  for  acid
     precipitation formation, deposition and effects as a conse-
     quence of  nitrogen and sulfur oxide  emissions  from fossil
     fuel combustion.                                                    2
Figure III-l:   Detailed diagram  illustrating  acid precipitation
     formation, deposition  and effects  on terrestrial and aqua-
     ti'- ecosystems.                                                    12

Figure III-2:   Simplified  nitrogen  cycle showing  acid-forming
     and acid-consuming reactions.                                     15

Figure III-3:   Simplified sulfur cycle illustrating acid-forming
     and acid-consuming processes.                                     17
Figure III-4:  Nutrient  budgets  for  podzol-brown  earth  lysi-
     meters  treated  with  artificial  rain of pHs  3,  4,  and 5.6:
Figure III-5:  Nutrient  budgets  for  podzol  lysimeters  treated
     with simulated rain of pHs 2, 3, 4 and 6.                          22

Figure III-6:  Solubility of aluminum as affected by pH.                36

Figure III-7:  Aluminum  concentrations  found  in  Norwegian and
     Swedish clear water lakes.                                         37
 Figure  III-8:  Nutrient  release  from snowpack during 1978  spring
     melt as determined  by monitoring runoff from mini catchments
     in Norway.
 Figure  III-9:   Relationship  between  H+  loading  rate  and  observed
      aluminum  concentrations  in  lysimeter effluent,  minicatch-
      ments  or  freshwater  lakes.                          '               42
 Figure 111-10:   Effect of  altering  pH  on  cumulative  mortality of
      brook trout  exposed  to  three  concentrations of  aluminum.        47

 Figure IV-1:   Site  index  classes  determined  by  plottir.n  age
      against  average height  of dominant tree.                           55

Figure IV-2:   Theoretical   height/age  curves  for  three  soils.      57

Figure IV-3:   Results  of  a mortality  test at  various  tempera-
     tures plotted on logarithmic-probit paper..                      60
Figure IV-4:   Effects of  various sulfate loading  rates  on lake
     pH for  lakes  in very sensitive and somewhat less sensitive
     surroundings in Sweden.                                           62
Figure IV-5:  A nomograph to predict the pH of lakes.                 64
Figure IV-6:  River  pH,  rainfall  pH,  rainfall  accumulation and
     discharge rate for the Shavers Fork River, W. Va.                 67
Figure IV-7:  pH  depression  in  Little  Moose  Lake,  N.Y.,  occur-
     ring during  spring snowmelt.                                      68
Figure IV-8: . Relationship  between  sulfate  loading  rate  and
     potential  pH  changes  occurring  during  spring  melt  for
     sensitive areas and slightly less sensitive areas.                69

                                 SECTION I


             Orie L.  Loucks, David Rapport, and Richard Miller
     Atmospheric deposition of acidic pollutants appears to have increased
in both  area and  intensity  since  the middle 1950's  (Likens  et a_L  1979)
and possibly over a longer time (Davis 1980).  This increase has been cor-
related  with increasing  use of  fossil  fuel for  automobiles  as  well  as
electricity, and the  associated  releases of oxides of nitrogen and sulfur
to the  atmosphere.   Utility  emissions  increased  most  rapidly, during the
1960's (EPA  1978),  and,  increasingly over this period,  were  emitted from
elevated  stack  heights where  dispersion in the  lower  atmosphere (rather
than  local  scavenging) becomes  more effective.   Once  in  the atmosphere,
the emissions can be converted to nitric and sulfuric acids.  Processes of
rainfall   and dry deposition  then  lead  to the  accumulation of these pro-
ducts once  again on land and surface water  resources.   Despite questions
as to rt;tes of reactions and the control mechanisms, enough is  known about
the acidic deposition phenomenon, however, to describe the principal types
of impact,  the  mechanisms  by which  effects  are expressed,  and the nature
and variability in resource response times.

     Figure  1-1  is  a conceptual  model  showing  reactions,  transport path-
ways and  the principal  ecological  consequences resulting from atmospheric
deposition  of elevated  ozone,  hydrogen, nitrate  and sulfate  ions.   The
materials and processes  producing  the ultimate effects of this deposition
must be viewed as a coupled system which, for simplicity, is shown here as
a flow diagram.  The  effects of interest are  the  direct effects on vege-
tation, the alteration of groundwater and associated stream chemistry, the
alteration  of soil  chemistry and associated terrestrial productivity, and
the effects on fish and other biota  in aquatic systems.   These effects are
largely  mediated through  the watershed, alt-hough direct flux  of acidic
materials  to surface  waters can  be important in some  situations.   The
magnitude of the effects  will  vary  depending  on characteristics  of the
watershed; only those ecosystems with little or no carbonate buffering are
likely  to  show  the  listed  effects  from  acidic  inputs.    The  actual
mechanisms  of  the chemical  and  biological  transformations  implied  in
Figure 1-1 are complex and will be discussed throughout the report.

     Effects  from  oxidants   and   acid  precipitation  are  both  .partly
transient and partly  irreversible  in nature.    Transient  effects include
alteration  of  soil   solution chemistry  and  adverse  impacts  on crops,
forests,  fish and  other  biota.   Irreversible effects  include long-term
changes  in  species composition,  stripping  of  cation nutrients from soil
systems,  modification  of  groundwater.  and  relatively permanent reduction
of buffering  capacity  for entire catchment  areas  (see  later  sections for
references).  Nutrients leached from the soil column are permanently  lost,
and because  of  the rates of  natural  leaching,  geological  replacement may
be on the  order of thousands of years.  Shallow  groundwater can be des-
cribed as  a sink for sulfate and hydrogen ions, heavy metals and  aluminum
leached  from the  soil  column,  but  these ions  also reach surface waters.


                                                                   FLUX TO
                                                                SURFACE WATERS
                                                                 H*. SO, N03
SO,, NO,. 03
                       CATION DISPLACEMENT
                        METAL MOBILIZATION
                                                               EFFECTS ON 1
                                                                FISH AND
                                                              OTHER BIOTAJ
Figure 1-1.     Flow diagram showing  system linkages for acid precipitation
                formation,  deposition   and  effects  as  a  consequence  of
                nitrogen  and  sulfur  oxide emissions  from fossil  fuel com-

Once  in the  groundwater  system,  these  elements  can  'influence drinking
water.  Acid-altered  groundwter  is  released slowly as seepage to springs
and other surface waters.

     Government,  industry,  and  the  public  are all  concerned with  the
long-term  responses  to  deposition of  ozone  and acidic  pollutants.   One
possible response  is  the control  of atmospheric inputs through reductions
of S02  and  NO  emissions from fossil fuel combustion, a responsibility of
regulatory  agencies at  state and federal levels of  government.   Glass e_t
aj_. (1980)  reported on  a series of workshops held in late 1979 with state
regulatory  personnel  of  Wisconsin,  Minnesota and Michigan regarding the:r
concerns with  and  role  in emissions control and environmental protection.
These  three  states contain large areas  characteristic of  the strsitive
resources  being affected  elsewhare,  although acidification  responses in
lakes have  not yet been demonstrated.

     There  was  considerable interest among the 1979 workshop participants
in controlling the precursors of acid  deposition, while recognizing that
there must  be a clear indication of  the  prospective benefits in resource
protection  to  be  gained  from  such  action.   This  interest  resulted in
specific  suggestions  for development  of a  "damage  susceptibility index"
which could be  used to inventory susceptible resources in regions that are
not yet seriously affected.  Specific criteria suggested for such an index
included  (1) identification of buffering capacity  of  the  lake/watershed
system,  and  (2) consideration  of potential  effects  on all  biological
resources,  including  timber and fish.


     The  loading rate of  pollutants that is just below a  rate producing
demonstrable  effects  can be  defined as a  loading tolerance  for a system
with  a  defined  sensitivity.   This  tolerance  and  sensitivity must be
related  to  specific  effects observed  and  to measurable  lake/ watershed
characteristics  -  chemical, geological,  and  biological.  The pathways by
which  effects of acidification are  expressed  in ecosystems  are indicated
in Figure   1-1  as  sequential events:   Fossil  fuel  emissions,  followed by
atmospheric  reactions,   deposition,   chemical   changes   and,  finally,
biological  effects.   While this  sequence  is  an  oversimplification, given
the state  of existing knowledge,  one could  expect  that early symptoms of
acidification  need not  lie  solely in  altered biota,  nor in the chemical
properties  of the  soils  and water.   Various  biological  measures might be
responsive  solely  to che  periodic  or.3et  of acid  stress.   For example,
changes in  the  establishment and recovery patterns of algal species at the
mouth  of  a  stream receiving acidic pulses from  spring snowmelt could be a
very sensitive  measure of irregularly expressed  stresses.

     In developing  indicators of pollutant impacts on both terrestrial and
aquatic  ecosystems,  three  classes  of measures  can be  used  to  reflect
system  transformations  (Rapport  and  Regier 1980).   Two of these, termed
here  "indicators"  and "integrators," are well  :"ited to representing the
effects of  acidification.   The third, the so-called  "environmental quality
index,"  is  basically unsound  because it  attempts  to  compress  the many
dimensions  of  environmental  transformation  into  a  single dimensionless

index  number,  often without  -benefit of  a  scientifically valid  model  or
logical  basis  for  such  a  representation.    It appears  to be  unwise  to
attempt  construction of  environmental  quality  indices,  or agglomerative
.measures of environmental quality to reflect environmental transformation,
whether  due  to acid  rain or any other  factors.   Inhaber (1974) proposed
such  an  index for  Canada,  and  the  results have  led  to  a general  disen-
chantment with the environmental index approach.

     The  alternative,  namely to  suggest a suite  of direct or relatively
direct measures  which  reflect  various  dimensions  of  the impact process,
and which are based on scientifically sound models, appears mere promising
(Rapport  and Regier 1980).   These two types of measures can be defined as

               Indicators  -  measures  of ecosystem  alteration which
          are  t?latively specific or diagnostic of particular types
          of stresses,  and which  are  likely  to be useful  as early
          warning  of stressed  environments.    Some examples  in the
          acid rain context  include reduction  in  the populations of
          shcrt-lived sensitive species and specific chemical changes
          such as  the mobilization  of aluminum during acid flushing
          events.   Indicators must  be  suited to picking up transient
          events which may otherwise be difficult  to measure.

               Integrators  -  measures  of ecosystem alteration which
          are  capable  of  reflecting  a  number  of  stress  impacts
          through  time, and  which  reflect these  impacts retrospec-
          tively and  synergistically.  An example  is  the  forest site
          index  measure which  integrates a variety of environmental
          transformations as they affect  forest  responses  (growth and.
          productivity)   over  the  lifespan   of   the  trees  being

     To  a large  extent  these two types of measures complement each other.
The  "indicators"  are  diagnostic  of  particular  types  of  stresses,  and
would,  due  to  their  greater  specificity,  function  in  evaluating  both
causative  and  curative  strategies.  On   the  other hand,  the integrators,
although  retrospective  and  often non-specific,  overcome to a large degree
the  difficulty of capturing  short-term  events  and synerTistic effects in
the  ecosystem.   Tc  use  a medical  analogy,  the integrators  are somewhat
akin  to  the  "vital  signs" of an  ecosystem,  while  the indicators function
as the clinical symptoms of distress (Rapport e_t a_L 1980).


     The  initial  sections of this report have  as  their objective a review
of  the  various   measurements  and  relationships  being  used  to quantify
resource  status   in oxidant  and acid-stressed  systems,   and  to identify
integrative  measures  of  lake/watershed response to  acidic  pollutants.
Considering the voluminous  literature on  the  subject and the many possible
effects,  the  review will be  limited to  those topics most closely related
to the effects described  in Figure 1-1. '

     A  second  objective  will  be  to  examine a  group of  measures  and/or
models  selected  to  represent  ecosystem  responses,  such  as  resource
nutrient  cycling  and  biotic  composition  and  quality.   These  measures
require a compromise between generality, reality and precision because not
all of  these qualities  can be achieved simultaneously (Levins 1968).   The
emphasis  in the  report  will  ba  on  generality  and  reality for  the  time
being.  The measures used  must  enhance the  likelihood of  prediction  of
future lake/watershed responses in lightly impacted systems, given current
acid loading rates over the period of a few decades.

     Finally,  the data needed to evaluate  the  usefulness of  a  group  of
measures  or predictive models will  be identified.   The  entire  group  of
measures  should   be  tested and verified  with appropriate  studies  before
effective  regulatory use can  be  expected.   The final  test, application,
can be met best if all  others are satisfied, and if the measures relate to
obvious economic benefits from acidic pollutant abatement.

                                SECTION II

                           SUMMARY AND FINDINGS

                              Orie L.  Loucks
     In August, 1979, President Jimmy Carter established a ten-year feder-
ally funded  acid rain  assessment  program.   At  about the  same  time,  the
Environmental  Protection  Agency awarded a Cooperative  Agreement  to North
Carolina State University to conduct a program of subcontracted studies on
biological effects of  acid  precipitation.   The Institute of Ecology (TIE)
was  awarded  a subcontract  entitled "Assessment f  the Sensitivity Index
Concept  for  Evaluating  Resources  at  Risk  from  Atmospheric  Pollutant
Deposition (Acid Rain)," to be  carried out in support  of  studies  at  the
ERL-Ouluth, U.S.  Environmental  Protection Agency.

     The  study  was  concerned  principally  with  developing  sensitivity
measures  for  evaluating  terrestrial  and aquatic  resources at  risk  from
atmospheric  pollutant   deposition  (oxidants  as well  as acid  rain).   The
main objectives have been:

     (1)  To  review  existing  literature  on  the use  of indices  for quan-
     *     tifying  resource  status,  predicting long-term  trends  in  eco-
          system  and   resource  responses  to  acid   deposition,  and  for
          assessing overall  risks from atmospheric pollutant deposition in
          relation to air emissions management;

     (2)  To  consider  several  options  as  to  the  form  of  a "sensitivity
          index"  or  pollutant  loading  tolerance  model  for  use  in deter-
          mining resources  at  risk  from energy development,  and outline
          how  such a measure would function in regional inventory of risk
          from pollutant  deposition or in the  assessment  of benefit  from
          acid precursor control;

     (3)  To  identify   validation  steps needed,  data  required  (existing
          data or  new  measurements),  and  the  steps  required to complete
          testing  and   begin  application of the sensitivity  measures  or
          loading tolerance model in regional and national energy develop-
          ment decisions.

     Early  in  the  study,  attempts were made to compress various .acid rain
indices (i.e.,  the  McFee. soil sensitivity measure, the Calcite Saturation
Index, etc.) into one sensitivity index.   As this concept was examined, it
was  recognized  as  an unsound approach because  it  attempts  to express too
many  dimensions of environmental transformation  into  a single dimension-
less  index  number.   Instead, a  suite  of  integrative  sensitivity measures
was  considered  which,  when viewed as  a wh^le,  would  be more likely to be
suitable for quantifying sensitive systems and their response to pollutant
inputs.   To provide a  technical basis for  considering such measures, an
                                              Preceding page blank

 extensive review is presented on  the mechanisms  by which  pollutant inputs
 alter  the  biological  productivity  of  terrestrial  and  aquatic  systems.


      One major  section of  the  report, entitled  Sub-Components of  Lake/
 Watershed  Sensitivity,  contains  the  background  material  on  pertinent
 characteristics  of acid  precipitation  effects  on soils,  lakes  and water-
sheds,  and  lays  a  foundation for both the  recommended  sensitivity measures
 and  the outline  of  research  and  data  needs  in later  sections.   This
 section also provides  an  understanding  of the  various processes occurring
 within  the system, information that  is  required  for developing and evalu-
 ating measures of watershed/lake  sensitivity.  Topics of  interest include
 hydrologic  flows,  the  nitrogen  and sulfur  cycles,   alkalinity  relation-
 ships,  interactions of acids with  organic  material, nutrient stripping,  H+
 toxicity, the mobilization  and toxicity of aluminum and heavy metals,  and
 synergisms  between H+, aluminum and  heavy  metals.

      The section  principally focused on new results,  entitled Methodolo-
 gies for Identifying  Sensitive  Terrestrial  and  Aquatic  Areas,  describes
 the various options for  measures  that  best identify potentially sensitive
 areas.   Three separate measures  are employed  for the  terrestrial  compo-
 nent:    McFee's  (1980)  soil sensitivity  classification  based on  cation
 exchange capacity;  the  soil  sensitivity classification  based  on  base
 saturation  (Coote  et  al.  1980); and  the  forest  site index.   The site index
 (SI) concept  has  been accepted as  a measure  of forest  productivity  for
 many decades and  is  examined here as  a methodology for  measuring changes
 in potential forest growth  du,: to long-term  acidic inputs.   The magnitude
 of site  index changes  due  to a combination of  oxidants,  changes in cation
 nutrient storage  (resulting  from  cation  stripping  by acidic  precipita-
 tion),   and  aluminum  toxicity  effects   is  still   incompletely  quantified,
 however.  Studies  will be  required  using  available  data  bases  on  oxidant
 exposures,   apparent   changes  in   total   nutrient stocks,   and  aluminum
 mobilization in  relation to acidic inputs.

      Integrative  measures  for expressing  aquatic sensitivity  have  been
 developed more  fully,  and  a number of experimental   and  field data-based
 methods exist.   These include the  Calcite  Saturation  Index,  the Henricksen
 nomograph and the  Almer/Oickson relation.   An additional  measure, based on
 pH  shock  effects  during  acid   flushing  events,  attempts  to  identify
 species/population  impacts   associated  with  short-term,   physiologically
 important exposures of critical life stages to  H+ and  A13+.   This approach
 can be  quantified  for the  pH depression levels already being observed in
 various sensitive  regions.   As with  the  other models,  its  applicability is
 relatively untested  for regions  where  acidic  inputs  are moderate and pH
 shock effects are  intermediate in  significance.

      The final  section,  entitled  Concluding Comments  and  Research Needs,
 is a brief statement on the data  needed to achieve an  effective validation
 of  the sensitivity measurement approaches.   Insufficient  data presently
 exist  for  fully quantifying hydrogen  ion  or  sulfate  fluxes through a wide
 variety of watersheds,  or  aluminum  mobilization  during peak H+ concentra-
 tions.    Nutrient   stripping  effects on  forest  productivity,   effects  on

mammals and predatory birds from metal mobilization and food chain altera-
tions, and the role of organic matter in mediating acid deposition effects
are  all  too  poorly known  to have  a  fully reliable,  locally applicable
sensitivity measure at this time.

                                SECTION III


       Richard Miller, Roland Usher, Orie Loucks and William Swanson
     Devising  an  adequate  measure of forest, watershed  and lake sensiti-
vity to  combinations  of pollutants requires an  understanding  not only of
the  response  mechanisms,  but  also of various  processes occurring within
the  system which  might mediate  or  interact with  the  effects  of acidic
deposition.   Within  the  lake/watershed  system  these  processes  include
water  flow,  nitrogen  and   sulfur  cycling,  alkalinity production,  inter-
actions with organic material, nutrient stripping (leaching), H+ toxicity,
mobilization  and  toxicity  of aluminum and  heavy metals,  and synergisms
between H+, aluminum and heavy metals.

     Figure III-l is an expanded version of Figure 1-1 and is meant to put
all  the  sub-components  of   forest,  lake  and  watershed  sensitivity  into
perspective.    The  scheme  is obviously complex  and  highly linked; changes
in one place  have  effacts  elsewhere.  Systems of differing overall sensi-
tivity are indicated, and fluxes from the different systems will vary from
zero  in  the  highly buffered  areas  to generally high  fluxes  in systems
undergoing alteration  (Overrein  1980).   The  events  occurring following
watershed deposition are discussed below.

     The pathways by which watar flows through a watershed are implicit in
Figure III-l and are of crucial importance to acidification by determining
the  contact  time  between acid  inputs  and p-tt  .ially  neutralizing soil
constituents.   A  considerable amount  of surface  runoff  will  greatly in-
crease the through-put  of acid directly to the  aquatic components of the
system.  The exact flow patterns depend on a variety of factors, including
topography and slope,  iioil   type,  rate  of input  and vegetative cover.

     Two  types of water  flow exist:  surface  flow and sub-surfaca flow.
Surface flow  is water move.-?'/., across the top of the soil to an arbitrary
depth  of  2  cm.   This flow  .an  be interpreted  in  two  ways:   dynamic and
continuous.  Dooge  (1973)  haj developed mathematical equations describing
surface  flow  that  take  into account  flow velocity,  slope,  friction of
slope, depth  of flow, flow velocity/unit area,  length of slope, length of
flow, outflow,  outflow equilibrium, time, characteristic time dependent on
flow intensity, and duration  of uniform flow.   Predicting the flow rate of
water  through   the surface  soil  is  important  because  this  layer has the
most  biological  activity  and  is  therefore one of  the  most influential
components of  the system.

     Sub-surface  flow is the movement of  water through  the soil profile
below the surface.  This aspect of hydrology is  perhaps the most difficult
to  accurately   simulate  in  mathematical  form.    The .movement of water can
proceed  in  two directions:   (1)  percolation   (downward  movement  to  the
                                              Preceding page blank

                         NATURAL SOURCES
                         FLUX TO WATERSHEDS
              SO,   )    f   SOI'
              NO.      _    NO',
              H,0   	*"    H*
              H.C.O.         O,
           HIGHLY BUFFERED  i
         (     SOILS       j
\ SOILS   /
        ! SOILS !
                 SOLUBLE NUTRIENT
                 IONS Ca^Mg7tNa*,K:
         IONS HtAI^PbT
          ELEVATED H*. METALS,
           AND NO', LEVELS IN
            TO FISH AND
           AOUATiC BIOTA
 ..I..     ,-J-_
,' ACID N|    ,' POORLY ")
1   LAKES  ;
                  i HIGHLY BUFFERED
                  yCARBONATEl LAKES*
               DISRUPTION OF
              ORGANIC MATTER A
               NITROGEN CYCLE
                N01,NH4,ORG. N
                            SOLUBLE TOXIC
                            IONS H'.AI ,Pb ,
           IONS Ca2*,Ma2*. NO;
    SOILS   / /
                           EFFECTS ON CROPS
                             AND FORESTS

Figure III-l.  Flow  diagram showing system linkages for acid precipitation  formation, deposition and effects as a
              consequence of  nitrogen  and sulfur  oxide emissions from fossil  fuel  combustion.

groundwater),  and (2) capillary  n' ~e  (upward  movement from  the ground-
water).   Implicit in both  direction,  of flow  is  equilibrium  between the
water  and  the soil.  Mathematical  equations describing  percolation have
been developed  that take  into  account such factors  as  soil  suction; hy-
draulic  conductivity,  hydraulic  diffusivity,   soil  moisture  content and
elevation above water table.

     The above discussion  has  dealt with the liquid form of precipitation
only.  The  solid  form,  snow, does not contribute to v.aterflow until snow-
melt occurs.   Light  (1941)  has  proposed a  complex  bjt realistic approach
for  predicting snowmelt  that takes into account many  of the major micro-
climatic variables that can  influence snowmelt.

     Once the snow is in liquid form  it can be placed  into the surface and
sub-surface  flow  equations.  Wright and Dovland (1977) have discussed the
problems  associated  with snowmelt,  snowpacking and  major  ion concentra-
tions  in Norway.   Fahey  (1979) and Wright and  Dovland  (1977) imply that
slow  input  of  acid, metals and  ions from the snow into  the  watershed
occurs as  a result of snowmelt.  One question  that has not been properly
addressed in the  literature  is  the importance of snowmelt and soil contact
with the  water before it enters  the  lake  and/or river, as it pertains to
acid and heavy metal inputs  to  the watershed.

     Finally,  Glass  e_t a_L  (1981) note that because, in northern soils,
the  sulfate  ion  is  a  relatively  conservative substance (Harvey  et al_.
1981),  high  rates  of  evaporation  can leave  the  precipitation sulfate
concentrated  in the  soil  solution (and lake water) by a factor controlled
by  evaporative losses.   The  equations for  lake  sulfate  concentrations
developed by Henriksen  (1980,  see later sections)  show this factor, plus
the  dry  deposition,  to be  1.9  for  central  Norway.  Regions  of propor-
tionately  higher  evaporative losses  have  higher observed sulfate concen-
trations  in  lake watar  than   are  predicted  by the  Henriksen equations.
These  processes  vary with  precipitation and  temperature patterns between
regions  and,  in  areas of  strong topography,  from one watershed  to the

     Becaus*  large quantities  of  amironium and  nitrate  can be present  in
precipitation,  it  is  of interest to  Jiscuss  their relationship to acidi-
fication.   First,  a brief review of  ".he natural reactions  of  the  nitrogen
cycle will  help in understanding the possible fates of added  ammonium and
nitric acid.

     Atmospheric nitrogen  (N2)  is  reduced by the  soil microflora, collec-
tively termed  nitrogen fixerr>,  to ammonia (NH3).  The NH3  is  hydrogenated
in  the  soil system to  the ammonium  ion,  NH4-f-,  resulting in a  loss of 1H+
from the  soil  system.   The NH4+ either  may  be assimilated by  plants (re-
sulting  in  an  input of 1H+,  thus balancing previous output),  or it may  be
oxidized  to nitrate (N03~) forming 2H+.   The ammonium ion assimilated  by
plants  will eventually be  returned  to the soil system  as  NH3.   The fol-
lowing  reactions   simplify   the  oxidation   of  NH4+,  a   process termed

     NH4 + r-sOj, * 2H  + H20 + N02                                (eq.  1)

     N02" + 5s02 -> N03~                                           (eq.  2)

Nitrate  is  generally taken up by  plants,  resulting in an  input  of  1 OH-
ion to  the system,  but  under anaerobic  conditions it may  be  reduced to
nitrite (N02-), summarized in equation 3:

     N03" + 2H+ + 2e- -> N02" + H20                               (eq.  3)

The nitrite (under anaerobic conditions) may be further reduced to nitrous
oxide (N20) or atmospheric nitrogen (N2).   Both of these reactions consume
H+  ions from  the  soil  system.  .The  magnitude of  H+ consumption by the
process  of  nitrate conversion to N2  (denitrification)  is  largely unknown
for  most  terrestrial  ecosystems.   These  reactions  probably  occur  more
readily  in aquatic  systems,   where  anaerobic conditions  more  frequently

     The molecular  forms  of nitrogen  deposited in acid rain can result in
either  acidification or  neutralization of  surface waters.   Nitrogen as
nitrate  ions   (N04-) can  be  incorporated  directly by  plants,  and the
resulting  reactions  with water  release hydroxyl  ions  (OH-)  into the en-
vironment (Figure  III-2).   The hydroxyl ions can raise the pH of the soil
and water  or   neutralize  the  hydrogen  ions  of the  nitric acid.   Natural
decomposition  of nitrogenous  plant materials also releases hydrogen  ions,
but net  accumulation of  plant tissue dominates in  most ecosystems;   hence
net production of  neutralizing capacity from  nitrate  addition  is usually

     Several  scenarios  operating  in  terrestrial  and/or  aquatic systems
involving  ammonium  and  nitrate  inputs and  acidification  are possible:

     (1)  If all nitrogen  in  precipitation is in the  form of HN03,   or if
          the concentration of HN03 is much greater than the concentration
          of NH4+,  and  if all nitrogen is assimilated, then a decrease in
          the  acidity of  the  system  is possible.   (The H+ associated with
          N03-  may  be  neutralized by  the  OH-  released  when  nitrate is

     (2)  If  the  HN03 concentration  equals  the NH4+ concentration  and if
          there is no leaching, then  there may be a slight increase in the
          acidity  of the system.   (The H+  from  HN03 may be neutralized
          when  N03-  is  assimilated;  the  NH4+ adds  1H+  to  the  system).

     (3)  If  ell  the nitrogen in  rain  is  in the  form of  NH4+,  or  if the
          NH4+ concentration  is  much  greater than the N03- concentration,
          and  if  there  is no  leaching, then an increase in the acidity of
          the  system will occur.

     (4)  If  HN03  predominates in precipitation, and  if  leaching occurs,
          then an  input of H+  to the  system will occur.

     (5)  If  NH4+  predominates in precipitation  and if assimilation does
          not  occur,  three routes  are possible:


r\  v
_,  pj -^a 	 i\j n, " 	 ^p- 	
1 organic nitrogen H
i ii
RtBm. M1 ' ^fel>- l~i.
\ft~- | VJ ( | > J-

4 ^ +

1 1 "*" -^ &- r
"4 ^ f

                R  organic  nitrogen
Figure  III-2.  Simplified nitrogen cycle showing  acid-forming and acid-consuming reactions.  Assumes  all N
              utilized as N03-.   From Reuss 1976.

          (a)  Leaching of NH4+ intact (acidifies receiving water);

          (b)  conversion  to  N03-  and  then leaching,  adding 2H+  to  the
               jystem; or

          (c)  exchange  reactions  on soil  colloids between  NH4+  and soil
               cations,   resulting   in   cation  stripping   aM   loss   of
               nutrients from the soil system.

     What  route  incoming nitrate  or  ammonium will  take  depends on  a
variety  of  factors,  including:   soil type  (base  saturation, cation  ex-
change  capacity and  mineralogy);   vegetation  type,  density  and  nutrient
condition; and water flow patterns.

     One  further  impact  of  acid  deposition  on the  nitrogen cycle is  a
possible disruption of  the rate (and quantity)  of nitrogen  cycled.  Many
steps  of the  nitrogen  cycle  (i.e.,  nitrogen  fixation,  ammonific-'-ion,
nitrification and denitrification) are mediated by microorganisms known to
grow only in  conditions above  a pH of approximately  5.0.   Acid conditions
are  known  to  inhibit modulation  of legumes  (Andrew  1978;   Evans  et  a_T.
1980),  resulting  in a  decreased nitrogen  fixation  rate.   Besides direct
injury  to  organism? involved,  acid deposition  could  lead  to a  decreased
availability  of  molybdenum,  which is  an  essential  constituent  of  the
enzynes  involved  in nitrogen  fixation,  nitrification and denitrification.
Because  nitrogen  is such a critical plant nutrient,  any disruption of the
nitrogen  cycle   could  have   severe  consequences  in   terms   of  plant

     Large quantities of sulfur are being added to terrestrial and aquatic
ecosystems via  precipitation.   It  is  important to determine  what is the
fate of  this added  sulfur  in  terms of acidification.  A  brief review of
the sulfur cycle is in order.

     The predominant  form of  sulfur in the soil system is organic sulfur,
generally in  the  S2- state  or as  sulfide,  elemental  sulfur,  thiosulfate,
tetratliionate  and  sulfite  (Reuss  1975).   This  sulfur   is   oxidized  to
sulfate  by  autotrophic and   heterotrophic  microorganisms.   Autotrophic
bacteria of  the genus  Thiobacillus are recognized as  the most important
group  of  sulfur-oxidizing  microorganisms.  Five  species  predominate:   T.
thioparus,  T.  denitrificans,   T.   thiooxidans,  T.   ferroxidans,  and  T.
novellus.   T.  ferrodoxins  and  T.  novi?1lus are found  in  acid soils,  even
dovn to pH 2.0.   Regardless of the pathway or microorganism involved, the
oxidation of sulfides or sulfur results in the formation of H+:

     S > 3/2 02  H20 - S0|"   2H*                                 (q.  4)

     H2S + 202 -" 2H* + SO^"                                       (eq.  5)

Under  anaerobic conditions  ^'primarily  deltas  and flooded systems), sulfur
may  be reduced  to  sulfides by members  of  the genera Desulfovibrio and
Desulfotomaculum.   Iron sulfides often accumulate in such areas.

Figure  III-3.  Simplified sulfur cycle illustrating acid-forming and acid-consuming processes.  (From Reuss

      A simplified sulfur cycle  in  Figure  III-3  shows  the acid-forming and
 acid-consuming  reactions  (Reuss 1976).   As  shown,  the  sulfur cycle  is
 balanced and no  net change in  acidity  sho-jld occur.   (The 2H+  ions  pro-
 duced when  organic  S is  oxidized to S0|- are neutralized  by the 2 OH-  ions
 released when the  plant incorporates the SO^-.   The  2 OH- ions  lost are
 balanced when 2H+ ions  are  used to reduce  S0|- to  organic S  in the plant
 system.)  However,  because of  lags between  sulfate  formation  and  plant
 uptake,  the  H+  ions produced  will displace  basic  cations from  exchange
 sites, and the cations  plus  S0|- will  leach  through  the system.   Sulfate.
 reduction by anaerobic  heterotrophs  occurring in  flooded areas results in
 neutral  or basic soils  as  H+  is consumed.   Such soils  will become acid if
 they become aerobic .due  to  the oxidation  of sulfides  to  sulfates.   Sulfur
 entering the soil system as  S0,-S02, H2S03  or H2S04 will lead to the  same
 net increase in  acidity (2H+)  only  if  the  sulfate produced  is  not imme-
 diately  absorbed  by the  plant  system  (i.e.,  if  sulfur is   in  abundant
 supply).    If sulfur  is  in  short  supply,  no acidification will  occur
 because the 2 OH- ions  released will  balance the 2H+ ions proJuced.

      In spite of  many possible reactions,  much of  the  sulfate  deposited in
 acid  precipitation  is not  retained in  the  Precambrian Shield watersheds,
 thus  functioning  freely as  an  anion  balancing  the   transport of H+  in
 surface water and  shallow groundwater.    The amount of  sulfate  in  runoff
 from  the  Shield areas  is  very close  to  the  amount  deposited.   At the
 Experimental   Lakes  Area  in   Ontario,  Schindler  et  al_.  (1976)   found
 virtually 100% of the atmospheric S0|-  input in  the runoff.  Likens  e_t a_L
 (1977)  found  67%  of the  total input  in  runoff at  Hubbard  Brook,  New
 Hampshire,  and Harvey et a 1.  (1981) found 25% more sulfate in runoff from
 four watersheds  than was measured  in the  bulk deposition.  This "excess"
 sulfate may  be  dua  to  unmeasured inputs  by dry  and  gaseous deposition.

      Ultimately,  excess  sulfate can be  leached through the poorly-buffered
-soils of these regions  in  association with  cations released from exchange
 sites.  Sulfur taken up  by  plants  (or  S02  absorbed by foliage) is  subse-
 quently  reduced  to  the  sulfide state  (primarily  as  amino acids)  during
 decomposition,  thus  returning  to  an  initial state  of  the  sulfur cycle
 (Figure III-3).

      One  of  the  most important  factors  influencing  the  sensitivity  of
 lakes  to   acid  input  is  alkalinity,  or  the capacity  of  the water  to
 neutralize acid.   It is normally measured by titrating a water sample with
 acid to a  fixed  end-point,  depending on total inorganic carbon concentra-
 tion.  Carbonate  and  bicarbonate  are the major contributors to alkalinity
 in  most  natural  waters,  but  other  anions of weak  acids  (e.g.,  organic
 anions) can also accept protons.

      Lake  alkalinity   is  largely  a  function of  the  type  of rock  and
 biological activity (soil  layr C02 production) in  the drainage basin.   If
 carbonates are  present,  weathfring by acid will release  bicarbonate into
 the  water  as  the major buffeiing  agent.   In this  case alkalinity  can  be
 high, and  the lake  will  be insensitive  to  acid  input.   In the absence of
 carbonate  minerals,  bicarbonate  can be released  by  weathering of  other


minerals,  but the  quantities are  slight.   On granitic  substrates,  very
little  alkalinity will  be  made  available  by weathering,  and  the  main
sources are from carbonic acid, as controlled by C02 concentrations in the
soil (e.g.,  see  Johnson et aj. 1977).   Lakes  in  such areas will often be
very poorly  buffered  (Wright and Henriksen  1978,  Glass  and Louc.ks 1980).

     There may be some alkalinity produced in the anaerobic hyoolimnion of
small  lakes.   Cook  and Schindler (1980) report incrt'asintj bicarbonate and
ferrous  iron  during  summer stratification due to the oxidation of organic
matter to C02 and the reduction of ferric iron.  Hutchinson (1957) note* a
similar  phenomenon  in  a  number  of lakes and explains  that  ferrous and
manganous  bicarbonat.es  may diffuse from the sediments  and add noticeably
to alkalinity in softwater lakes.

     Kramer (1976) has reviewed the relationship between  pH and alkalinity
in  lakes.   Water in equilibrium with CaC03  has  pH 8 and about 2 meq/ of
alkalinity.   The  pH  and alkalinity  levels  drop  as   this   solution  .is
diluted,  but  between  pH  4  to 6,  alkalinity remains  nearly  constant at
0.1-0.2  meq/.    Kramer  attributes  this buffering  to  alumino-silicates,
FeOOH  and  soluble organic acids.   He concludes  that low alkalinity  lakes
in  non-calcareous  terrain are likely to be  altered by acid rain, but that
detailed  analysis  of  soil   minerology  at  all   depths  is necessary  to
accurately assess the risk.


     The role  of organic materials in soils and  water  as  they affect the
impact  of  acid  deposition has been  the subject  of  some speculation but
little  quantitative  study.   This is a major oversight  since organics may
influence  almost all  the reactions relating to acidification.  Of course,
the  formation of hurmc  acid substances  is  one of  the  factors leading to
natural  soil acidification.   Acid deposition can be seen  as an audition to
this process, so the role  of  organics deserves consideration.

     One  of  the major  characteristics  of  humic  substances  in  soils is
their  behavior  in cation exchange.   McFee e_t  a_[.  (1976) list the average
cation  exchange  capacity  of  humus  as 200 meq/lOOg.  This  can serve as a
significant  buffer  for added H+ in precipitation.   If  base saturation is
high,  the  added H+  will  exchange  for  cations  absorbed  to  the organic
matter,  such  as  Ca2+ and  A13+.  Organic soils with lower  base saturation
are. already acid,  and  H+ in  rainfall will  have  little additional effect
(Petersen  1980).

     Another  characteristic  of  organic materials   is  their  ability to
chelate  various  ions.   These  ions rr.ay then  either be retained in the soil
or  leached  out,  depending   on  the  behavior  of  the   organic  molecula.
Schnitzer  (1980) notes that moderate acidity will decrease the solubility
of  humic acids but  increase  the solubility of  fulvic  acids, which may then
be  lost from the soil  system with their absorbed  ions.    Organic  chelation
and  mobilization could  be a  significant  factor  in ion mobilization from
watershed  soils,  but  the  quantitative effects  are  unknown.

     Once  into  the aquatic  system,  dissociated organic  acids  may add--feb
the alkalinity of  the  s- ..tern by their  ability  to  take up H+ (Chen et ai.

     R-COOH = RCOO" + H+

This  reaction  may  be   especially  important  in waters  where bicarbonate
buffering  is low,  and  organic concentrations are  high.   The buffering by
this reaction will  probably occur at pi! less than 5.

     One  final  effect   of  organics  relative  to acidification problems is
the chelation of toxic metals.   Baker  and  Schofield  (1980)  have observed
that  aluminum   chelated with organics  is  not toxic  to fish.   Similar
detoxification is possible with other metals mobilized by acid deposition,
e.g., Hg, Cd, Cu, etc.

     Quantitative determination  of  the role of organics  in  lake/watershed
sensitivity  to  acid  deposition will  depend  on  the  amount of organics
present,  and the  quantitative  effect  on  the  processes  discussed above.
Various methods  are  available to measure organic content depending on the
medium:  soil or water.  Total organic matter in soil can be determined by
combustion  of dried  samples.   Various  organic  fractions  can be extracted
by  treatment with  acid or alkali as  described  by  Schnitzer (1980).  Both
humic and  fulvic acids  dissolve in alkali, then humics can be precipitated
by  subsequent acid treatment.  This  allows  separate  determination of the
quantity of organic material  in each fraction.  Specific.organic molecules
or  classes  of molecules (e.g.,  emino  acids)  may be separated arid quanti-
fied  by  various  techniques,  including  GC,  GLC  and spectrophotometry.

     Organic matter  dissolved in water can  be  determined by oxidation to
C02,  followed by  infrared  absorption measurement of C02.  This  quantifies
total  organic  carbon  (TOC).  Natural waters can  also  be  analyzed by UV
absorption  as  an  indication  of relative organic  content,  but  this would
not be  as accurate as   TOC analysis.  Other methods include  color compari-
sons with  standards and methods for specific compounds as indicated above.
Each  method has advantages  and disadvantages:   TOC  analysis is accurate
but time-consuming and  non-specific, whil'i color comparisons are quick but
difficult  to relate  to actual organic  content.  The  choice  of  a suitable
method  depends  on  the   relationship between quantity and  effect, e.g., how
much  organic matter and what type  is necessary  to  cKlate and detoxify
aluminum at various concentrdtions.

     Nutrient stripping is one  of  the most  serious  consequences  of acid-
deposition  because  loss of critical nutrients  from the soil system may be
expressed  in lowered growth rates and  reduced productivity.  Abrahamsen et
aj_.  (1976), working  with podsol and  podsol-brown  earth  lysimeters, docu-
mented  the  effects  of hydrogen  ic  concentration on  nutrient leaching.
For  the.  podzol-brown   earth  lysimeters,   net  losses  of Ca,  Mg,   Al  and
sulfate  occurred;  whereas  in the podsol  lysimeters,  net losses occurred
for Ca, Mg, Al,  and K,  (Figures III-4  and III-5).   For all elements except
Al  in tiie podsol-brown earth lysi.neters,  increasing  hydrogen ion concen-
tration resulted in  increased leaching  of nutrients.

 pH 3
C* Tiq/m'/vear Mg mq/m'/ysaf
500 1000 1500 2000 '00 200 300 400
111.! 1111


960 j


50 100 150
1 1 1


NH4 +N0i) - N
100 200 300
1 1 1




100 j


10 20 30
1 1 1

17 )
" 21 1
SO,) (1975 onil
1000 2000 3000 9500
i 1 1 , 1 K t ' I
550 N

2440 j
'335  |
pH 5.6
  pH 4
  pH 3

 Figure III-4.   Nutrient  budgets  for pedzol-brown earth  lysimaters treated
                 with artificial  rain of  pHs 3, 4, and  5.6.  (From  Abrahamsen
                 et al. 1976.)

                II	1	1	i_J	1	L_/s/-J
                       - N
                                               J0i I
                                                 Reproduced from
                                                 best  available copy.
Figure III-5.   Nutrient  budgets  for  podzol  lysimeters treated with  simu-
                lated  rain  of  pHs  2,  3,  4 and  6.    N.W.  =  not watered;
                received incident'rain of Norway, pH  4.4.   (From Abrahaiisen
                et a_L  1976.)

     At Mt. Moosilauke,  New Hampshire,  Cronan (1980) found  sulfate  to be
the major  anion  in  precipitation,  throughfall,  percolate and springwater;
aluminum the major cation in springwater;  and hydrogen the major cation in
precipitation,  throughfall  and percolate.   Potassium is  readily  leached
from the  forest canopy  (as is calcium)  but is  rapidly taken  up  by the
vegetation.  Harvey et  aj_.  (1981)  also note  potassium retention by vege-
tation  in  Canadian  watersheds.   Calcium  is also  retained  to  a  certain
extent  in  the  soil  system,  although  more  than  half of  the  throughfall
concentration  is  found  in  percolate  and  springwater  (Table  III-l).
Magnesium  concentrations were highest  in the  springwater and  lowest in

     Harvey  et ah   (1981)  have summarized  nutrient budgets  for  several
Canadian watersheds  (Table  III-2).   All  watersheds studied  have  a net
output of  major  cations (ICa + Mg + Na + K) except  for Clear Lake.   The
ELA and Harp Lake results are of comparative interest (in terms of effects
of H ion  input)  as  they both have similar bedrock geology and soil  types.
The  increased  H inputs  at  Harp  Lake have resulted  in  increased leaching
rates  of  Ca +  Mg.   Harvey et aj. note  that on the  basis of  the  limited
evidence available,  it  appears that  the  increased  H+   input  to southern
Ontario has  resulted  in a 2- to 4-fold increase in .net  output of cations.

     There  are  numerous problems  associated  with  quantifying  nutrient
leaching from  the soil.  These  include  (1)  organic  composition,  content
and percentage in the soil (see Goring ?nd Jamaker 1972); (2) pH-dependent
CEC  sites  (Clark et  aK 1956);  (3)  the buffering  capacity  of ions other
than Ca2+  and Mg2+  (DeVilliers  and  Jackson 1967,  Turner  and  Clark 1965,
Clark  and  Turner  1965,  Clark 1966);  and (4) the toxicities of ions (Clark
1966,  McCorroick  and Steiner  1978,  Cronon and  Schofield 1979).  The two
areas  that  have  the  most influence with regards to acid precipitation are
(1) organics and (2) pH-dependent CEC sites.

     Currently there are only a  few general  equations that  deal with the
complexity  of  nutrient  leaching.   The parameters  for these  equations are
molar  ion  concentrations denoted  by  "[  ]" and ion  activities denoted by
"( )".   Addiscott (1977) and Reuss (1978) have proposed simplified models
describing the  dynamics  of  nutrient  leaching.  Reuss1 model will  be out-
lined and discussed briefly.

     Reuss  utilizes  (1) sulfate  (SO?,-) absorption,  (2)  H+ and HC03 equi-
libria,  (3)  Ca2+  and H+ equilibria,  (4)  A13+ equilibrium,  and (5) elec-
trical  neutrality.  This proposed  model  assumes that there  is a low Ca2+
base  saturation and  a  low exchangeable  Ca2+/unit  soil.   The level  of
SO2.-is an  indicator of H+ input and can be expressed as:

          SOl- (sorbed)  = Km [SO;-] / K^ + [SO2,-]                 (eq. 6)

where  K  is the number of moles of S042-/unit soil at saturation and K, is
equal  tS h K .  The total S042- present can be expressed as:           "*

          SOf- (total) = SO2.- (sorbed) + S01--9-D                (eq. 7)

where 6 is the volumetric moisture content and D is soil  depth.  Equations
(6)  and  (7)  express  the balance  of  SO2,- between  the   soil  and the soil

IABIE lll'l.  Mean and median concentrations of dissolved ionic aaterials In bulk precipitation,  canopy throtighfal I,  forest  floor  anil  A2  torizon  perco-
             late, and springwater during  the 1975 and 1976  growing  seasons  at the Ht. Moosilauke  subalpfne  sampling  sites.  New Hampshire.  Springs
             also include 19/7 data.   (From Cronan 1980.)

Sample pit
Bulk precipitation 4.08
Ihroughfal 1
mean 4.02
summed equivalents
ned i an
Tiean 4.04
sunned equivalents
mean 4.66
summed equivalents
C o n c e n

II* Ca'*
83 9

95 36


91 25

22 26
t r a





t i 0





n ( o i c r

Na NH^
4 13

3 6
0.2 0.5

3 S

7 5
0.3 0.7
5 3

13 3
0.7 0.7
13 3
o - e





q u i v





a 1





e n t s per
* Cations SOj'
118 75

201 143


219 137

163 132





r )
Total Aniun
NO, 110)3 Anions Deficit*
21 0 103 15

12 0 168 33
2 -

8 -

8 " 0 161 5S
1 - -
5 -

15 0 154 , 9
3 -
14 - - -
*  The anion deficit is presumed to represent organic ligands;  this has been confirmed on the basis  of GLC and  ultra-violet  irradiation  experiaents.

 TABLE  III-2. Net  export  of major ions (meq nr2 yr-1) for calibrated watersheds in Canada (From Harvey et
              al.  1981).
                                               Net Export (meq tn-2 yr-1)
      	    ^~r"- " v"""i '"	^^	  Input of H+
       Mg++  Na*    K+   ZM     NH4+-N  N03"-N  HC03"    S04-S  (meq m-2 yr-1)
 Carnation Creek,
   Vancouver  Is.

 Jamieson Creek, B.C.

 Haney, S.W.  B.C.
  12 watersheds, ELA,
   Western Ontario

  Rawson  Lake Watershed,
   Western Ontario
269    61.2  118.5  5.8  454.5   -7.7    -4.3   270.0   131.4
171.6  54.5   53.9  4.3  284
-0.2    -2.2    48.7
 70.4  22.2   29.4  2.0  124     -8.8   -13.4   approx.   18.0        -v
                                                  200   (range 3.1-47)
*36.6  24.0   18.0  3.4   82.0   -0.9   -0.8
217.7  16.4   11.1  0.5   45.7

 11.2  12.1    9.0  0.4   32.7  Total N reported
                        38.8   est.  7-10
 4  subwatersheds of Harp      52.7  33.0    6.6  2.3   94.6  -30.2   -32.3    31.7    43.9        67
    Lake; Hdi iburton-Muskoka,

* Clear  Lake, Haliburton      -36.8 -25.3    1.3 -0.4  -61.2  NO
                                                              retention = 62. 1
  1   Gross output

  2   Estimated  net output using  input in precipitation from Rawson Lake watershed studies.

solution.   This  balance can  be  interpreted as SO2,-  retention or holding
capacity of the  soil  system which has obvious  implications  for buffering
capacity related to the sulfur cycle.

     Distilled water  in  equilibrium with atmospheric C02 should have a pH
of 5.6._ Therefore,  it is necessary to  express the  equilibria between H+
and HC03.   This can be expressed as:

          (H+) + (HCOa) = (C02)g  10-7-81                      (eq. 8)

     The balance between  Ca2+ and H+  is an  indicator of  the buffering
capacity.   Reuss assumes  that the input of Mg2+ is insignificant in rela-
tion to the Ca2+ input.  Therefore, the equilibria is expressed as

          pH - \ pCa = KL                                        (eq. 9)

which  is  simplified  from  Schofield  and  Taylor's  (1955) "Lime Potential"
{ K.  =  pH  -  ^(Ca  +  Mg)  }.   Equation  (9) can be  modified  to include the
exchangeable Ca2+,  Ca2+ in solution and the CEC.

     The A13+ equilibrium can be expressed as

          (A13+)   (10-14 /(H+})3 = KA1                           (eq. 10)

                                             and is pH dependent

(DeVilliers and Jackson 1967).

     There must  also  be  an electrical  balance between  the  soil  and soil
solution.   This  balance  is a function of molar ion  concentration and ion
activity.   The electrical balance can  be expressed as

          [H+] + 2[Ca+2]  3[A1+3] = 2[SO;2]  + [HCOg]  + [Cl-].    (Eq. 11)
     It  appears  from the  literature  that Reuss'  simulation  model  may be
applicable to acid precipitation-sensitive systems only; however, as Reuss
states,  the  model  needs verification  in the  field and in the laboratory.
This model  deals  superficially with (1) the  input/output and neutraliza-
tion of H+ and (2) the variability of the organics.

General Reactions of H+

     Exchanges of H+  throuah the watershed depend largely on soil charac-
teristics, the balancing anions, and water flow patterns.  So:Is with high
pH  usually are  well  buffered  by  carbonates  or  have  Ir'gh  base exchange
capacity to balance anions, and have little flux of H+ or mobile anions to
surface  waters.    As   soil   pH,  base  saturation  and/or  cation exchange
capacity  decrease,  the  likelihood  that H+  and  a mobile  anion will pass
through  in the  soil  water greatly  increases  (Wiklander  1980).   In very
acid soils, almost  no H+ will  be  retained  by  the soil.   In general, pod-
zolic  soils should  be most susceptible to much alteration by acid inputs,

(Petersen 1980),  but exchangeable  cations  and water  flow  can  modify the
results (Johnsen and Freedman 1980).  Some podzols may contain appreciable
cations to exchange  for  H+,  depending on inputs.   In addition,  water flow
through the system may occur too quickly at times for complete exchange to
occur.   Bache (1980)  points  out that soil texture  and intensity of rain-
fall are  important factors  determining  the opportunity  for  H+ uptake by
soils.   One  further factor affecting H+ movement  through watershed soils
is  sulfate  retention (Johnson 1980).  Cation movement requires the move-
ment of balancing  anions,  so that if sulfate is retained ind other anions
are  not  mobilized,  then  H+ cannot  be  leached.   Soils  with a  high  per-
centage  of  sesquioxides   are   likely  to  be resistant  to  suIfuric  acid

     The toxic  effects of elevated H+ ion concentrations have  been known
over a considerable period,  for both terrestrial  and aquatic ecosystems.
Acid  soils  are  known to  be infertile  (Hewitt 1952), mainly  because of
adverse concentrations  of nutrients  and elements  (Andrew  1978),  and the
known  lower production rates of acid lakes (EIFAC 1969).

     Because, at  low pH,  the H+ ions also may displace aluminum and heavy
metals from  exchange  sites,  it is reasonable to assume that all three can
occur  simultaneously,  a*,  least  during  peak mobilization events.   Syner-
gistic effects  of these  three pollutants are  thus  a potential result of
acid   deposition.   Some   research  has  been  conducted  in  the area  of
synergisms,  but more  is  needed,  especially with  terrestrial  organisms.

Terrestrial Effects.

     Hewitt  (1952)  has  summarized the factors affecting  crop productivity
in acid soils:

     (1)  Direct injury to below-ground organs by hydrogen  ions.

     (2)  Indirect effects of low pH:

          (a)  Physiologically  impaired  absorption of calcium,  magnesium,
               and phosphorous.

          (b)  Increased   solubility,  to  a  toxic  extent,  of aluminum,
               manganese,  heavy metals and possibly  iron.

          (c)  Reduced  availability  of  phosphorus  partly  by  interaction
               with aluminum or iron, possibly after  absorption.

          (d)  Reduced availability of molybdenum.

     (3)  Effects from the low  base  status:

          (a)  Calcium deficiency.

          (b)  Deficiencies  of magnesium,  potassium  or  possibly  sodium.

     (4)  Induced abnormal biotic factors:

          (a)  Impaired nitrogen cycle.

          (b)  Impaired mycorrhizal activity.

          (c)  Increased  attack by  certain  soil  pathogens,  e.g.,  "club

     (5)  Accumulation  of soil  organic  acids  or other  toxic compounds.

These same  factors  could  affect forest and lake productivity, and many of
them could  be induced  by acid precipitation,  depending  on the buffering
capacity of the system.

Aquaeic Effects

     The  elevated H   ion concentration  in  lakes where  bicarbonate  buf-
fering  is negligible,  or  has been eliminated by acid inputs, is now being
linked with adverse effects on numerous fish species, and on phytoplankton
and zooplankton  species.   Hall  and Likens (1980) have presented data sug-
gesting that the emergence of adult mayflies, stoneflies, caddis flies and
some true flies decrease  in part due to lower pH, although elevated levels
of A13+,  Ca2+, Mg2+,  K,  Mn2+,  Fe, ard Ca2+  were  present.   Although  many
parts of  aquatic ecosystems  are detrimentally  affected  by acidic condi-
tions, our discussion will focus on the effects on fish.

     Laboratory  experiments  and field data  are  i.n  general  agreement  con-
cerning  effects  of  pH on fish  (Table  III-3),  but only  in  cases  where
toxicity  is  not  complicated  by the presence of ferric salts (EIFAC 1969).
Although  factors ether  than  hydrogen  ions  often  make  it  difficult  to
identify the cause of the observed response, the development of a pH below
5 (or above 10)  is considered to be unsafe for fish (Table III-4).  Fromm
(1980)  notes  that  a pH around 6.5  is  the 'no effect1 level of pH effects
on fish  reproduction.   A  pH between 5  and  9 is generally agreed as being
safe for fish.  However,  some species of  fish can tolerate pH values lower
than  those  reported in  the  laboratory to be  lethal,  which suggests  that
long-term acclimation is  possible (EIFAC  1969).

     A generally agreed observation in the literature is thnt early stages
of  fish  development  are  much  more sensitive  to  acid stress  than  adult
stages.    Thus,  acid  deposition  in sensitive  lakes  characteristically
results in a failure to replace the older age  rlasses; eventually the fish
population  is eliminated.

     Most  researchers  also  agree that  sodium imbalance is  the dominant
cause of  death,  although Packer and Ounson  (1970)  contend that death is
attributable  to  lowered  blood pH.  The ameliorating effect of high levels
of  Ca  on toxicity at  low pH is probably a  result  of reduced sodium loss
from  fish  (Wright  and   Snekvik  1978).   Fromm  (1980)  has  reviewed  the
physiological  and  toxicological  responses   of freshwater  fish  to  acid
stress.    1.v'er& is  some  question whether H+ ions alone  are toxic at the
observed  concentrations,  but the development of  low pH is  a major factor
in lake sensitivity.

fish species

Salmon and Trout

Salmon and I rout




S.i I mon  

Atldiillc Salmon

Atlantic Salmon

Atlantic Salmon

Atlantic Salmon

Atlantic Salmon





 (brook x Take trout)

Lake trout
 (SalveIinus namaycush)
pll level


5.3 & lower















Effect and Comments

"critical to young"

"highly lethal"

Lethal to SOX of eyed eggs

Critical range for hatching

LOf,-, in 12 days to yolk sac fry

96% Hatching success
4BX Hatching IUCCOK*

lUfto r embryos and alevlns

5G~ rortalHy of yolk sac fry in i9.0 days

Lethal pll In 2-day test; possible
  CO, interaction

Delayed hatching

Matching prevented

High and probably total lethjlHy

Critical range for hatching

80* yolk sac died in 20 days

10% yolk sac died in 20 days

50% mortality of yolk sac in 16.0 days

Approximate pll when reproduction ceased

Sunde 1926 as cited llagen & Langeland 19/3

Sur.de 1926 as cited llagen & Ljngeland 1973

Johansson et aj. 1977

Bua & Snekvik 1972 as cited Hagen & Langeland 1973

Oahl 1927 as cited EIFAC 1969

M. Grande personal comm. as cltod EIFAC 1%9

Oaye and Cars(de 19/9

Grande et aj. 1976

Bishai 1960 as cited EIFAC 1969

Peterson et a_l.  1900

Peterson et a].  1980

Dahl 1926 ar cited llagen & Langeland 1973

Oua and Snekvik 1972 as cited llagen & Langeland 1973

Oahl 1927 as cited EIFAC 1969

Dahl 1927 as cited EIFAC 1969

Grande et aj. 1978

Beamish 1976

FABLE 111-3. (continued)
Fish species
Lake trout
Rainbow trout
Rainbow trout
Rainbow trout
Ralnb'-i. \rout
Rainbow t -out
Rainbow trout
Rainbow trout
Brown trout
(Salrco trutla)
Brown trout
Brown trout
Brown trout
Brown trout
Brown trout
Brown trout
(Avaa strain)
Brown trout
(Avaa strain)
Brown trout
(R. Dalai ven strain)
Brown trout

pll level
1 age x
1 age x
1* age x
1 age x
1 age x
1 age x

= 6.10
= 5.59
= 4.98

= 6.20
= 5.50
= 4.77

(R.  QJ la I veil strain)
                                                  Effect and Comments

                                                  50% mortality of yolk sac In ->>1.3 days

                                                  50% mortality of yolk sac fry In >]..3 days

                                                  3.5 no.  exposure; 3% mortality

                                                  3.5 00.  exposure; 4% mortal Ity

                                                  3.5 ma.  exposure; 7% mortality

                                                  Approx.  lower Halts of tolerance

                                                  LDSO to finger)Ings in 15 days

                                                  8 day LD60

                                                  100% mortality of yolk sac fry due to ron-
                                                    bined low pll and low salt concentration

                                                  Approx.  lower Molts of tclerance

                                                  50% mortality of yolk sac fry In MO.O days

                                                  3.5 no.  exposure; 5% mortality

                                                  3.5 mo.  exposure; 2% mortality

                                                  3.5 mo.  exposure; 6% mortality

                                                  66% of eyed eggs survived yolk-sac stage

                                                  Lethal to 90% of eyed eggs

                                                  Lethal to 100% of eyed eggs

                                                  Lethal to 20% of eyed eggs

Grande et al. i<"'

Grande et al. 1978

Edwards A fljeldnes 1977

Edwards & Hjeldnes 1977

Edwards & Hjeldnes 1977

Berlins I960 as cited Johansson et aj. 1977

Lloyd & Jordan 1964 as cited EIFAC 1969

Lloyd & Jordan 1964 as cited EITAC 1969

Grande & Anderson 1979

Berlins 1960 as cited Johansson et a_l. 1977

Grande et al. 1978

Edwards & Hjeldnes 1977

Edwards & Hjeldnes 1977

Edwards & Hjeldnes 1977

Johansson et aj.  1977

Johansson et aj.  1977

Johansson et al.  1977

Johansson et al.  1977

1ADIE II1-3.   (continued)

f ish species

Brown trout

Brown trout                   4.77

Brown truut                   <5.0

Adult Brook Trout             4.5

Adult Brook (rout             S.08

3rook Trout                   5.09

Brook Trout                   5.57

Brook Trout                   6.13

Brook Trout                   6.55

Brook Trout (Control)         7.04

Brook trout                   4.8

Brook trout                   4.5

Brook trout                   4.49

Sea trout                     5.8-6.2

Perch                         4.5 and 5.0

Perch                         4.4-4.9

Yellow perch                  4.7-4.5
  (Pcrea fiavescens)

IABIE 111-3.   (coullmied)
F ish species
Umbra llmi
(Slizostedion vitreum)
(Percopsis oroiscomaycus)
Arctic char
Arctic char
Arctic char
to Arctic ctidr
Arctic char
pjl leyej
1 age x -
\ age x =
1 ay
FABLE 1II-3.   (continued)

fish species
pll level
Smal 1 mouth bass
(Hicrupterus dolumieui)
Ruck bass
(Ainhloplf tes r.upestrls)
lake Chub
(Coues Ins plumbeus)
'Ljt.a lota)
Bur-bo t
Lake herring
(Coregonus arredli)
Brown bul Ihead
(Ictalurus nebulostis)
6.0 -5.5
5.2-4. 7
effect and Comments

Approx. pll when reproduction ceased

Approx. pll when reproduction ceased
                                                    Healthy populations found in Wisconsin
                                                      Lake Survey

                                                    Approx. pll when reproduction ceased
                                                    Only found when pll 5.0

                                                    Approx. pH when reproduction ceased

                                                    Critical lower level for embryo

                                                    Critical lower level for post-embryo

                                                    Approximate pll when reproduction ceased

                                                    Approx. pll when reproduction ceased

                                                    8 day LDSO

                                                    Critical pll for reproduction

                                                    Critical pll for reproduction

                                                    Critical pll for reproduction

                                                    Critical pll for reproduction

Beamish 1976

Beamish ?976

Rahel A Haijnuson 1980

Beamish J976

Rahel & Haymtson 1980

Beamish 1976

Volodin 1960 as cited E1FAC 1969

Votodln 1960 as cited EIFAC 1969

Beanish '"-6

Beamish 1975

Lloyd & Jordan 1964 as cited flfAC 1969

Alner et aj. 1978

Aimer et al. 1978

Aimer et aj. 1978

Alner el al. 1978


Range          	Iff feet	

3.0 - 3.5      Unlikely that  any  fish can survive for more than a few hours in this range
               although some plants and invertebrates can be found at pH values lower than

3.5 - 4.0      This  range  is lethal  to salmonids.   There  is  evidence  that roach, tench,
               perch  and  pike  can  survive  in  this  range,  presumably  after  a  period of
               acclimation  to slightly higher,  non-lat-hal  levels,  but  the lower end of
               this range may still be  lethal for roach.

4.0 - 4.5      Likely to be harmful to  salmonids, tench, bream, roach, goldfish and common
               carp which  have  not previously been  acclimated  to  low pH values,  although
               the  resistance to this  pH range  increases  with the  size and  age of the
               fish.   Fish  can become  acclimated to  these levels,  -but  of perch, bream,
               roach and pike, only pike may bs able to breed.

4.5 - 5.0      Likely  to  be  harmful  to the  eggs  and  fry  of  salmonids  and,  in  the lor.g
               term,  persistence  of  these  values will be  detrimental  to such fisheries.
               Can be harmful to common carp.

5.0 - 6.0      Unlikely to  be  harmful  to any species  unless  either the concentration of
               free carbon dioxide is greater than 20 ppm or the water contains iron salts
               which  are  precipitated  as  ferric hydroxide, the toxicity  of which is not

6.0 - 6.5      Unlikely to  be  harmful  to fish  unless  free carbon  dioxide is present in
               excess of 100 ppm.

6.5 - 9.0      Harmless to fish, although the toxicity of other poisons may  be affected by
               changes within this range.

9.0 - 9.5      Likely to be  harmful  to salmonids and  perch if present for  a  considerable
               length of time.

9.5 - 10.0     Lethal to salmonids  over a prolonged period of time, but can  be withstood
               for  short periods.   May be harmful to  development  stages of some  species.

10.0 - 10.5    Can be withstood by roach and salmonids for  short periods but lethal ove-  a
               prolonged period.

10.5 - 11.0    Rapidly lethal to salmonids.  Prolonged exposure to the upper limit of this
               range is lethal  .:o carp, tench, goldfish and pike.

11.0 - 11.5    Rapidly lethal to all  species of fish.

     Reference is  made  to  different  species on  the  basis of  information known to us; the
absence of a reference indicates only that insufficient data exist.


General Reactions of Aluminum

     One of  the indirect  effects  of watershed  acidification may  be the
mobilization  of  aluminum  in  soils,  causing  toxic  effects on terrestrial
organisms and  the  biota of streams and  lakes.   Aluminum solubility  is pH
dependent,  and  increase;'  with increasing acidity (Figure III-6).  Several
reports have documented elevated aluminum concentrations in acid lakes and
streams in areas  known to be impacted by acid inputs (Figure II1-7,  Davis
1980;  Cronan  and Schofield  1979;  Schofield  1980),  and  in effluents from
lysimeters   treated with  acid solution  (Dickson 1978;   Abrahamsen  et a1.
1976).  While aluminum is typically leached from the upper soil horizon of
podsol  soils  by  carbonic  acid  and  organic  chelation,   it is  usually
deposited  in  lower  horizons.  Under  the  influence of -'strong  acids  in
precipitation, .however,  the  aluminum may be  transported through the soil
and  leached  into  lakes  and   streams (Hall  and  Likens  1980;  Herrmann and
Baron  1980).   Peak aluminum  concentrations generally  occur during  spring
melt  of the  sr.owpack  (Figure  III-8)  when large  quantities of  H+  ions,
accumulated  over  winter,  are released  and flush  aluminum from  the soil
system  (Schofield  1980).   As  illustrated   in  Figure  III-8,   the   large
majority of  impurities are released during the  first  stages  of snowmelt.
Seip  et a_L  1380 note that interactions  of meltwater  with soil and  vege-
tation  are  important  in  determining the amounts  of aluminum  leached to
receiving waters.

     Current  precipitation in  the  northeast United States,  Norway, and
Sweden  has  pH  levels  between 3.0  -  4:0 and lower (Likens  et  aj.   1979,
Wright  et aj_.  1980,  Wright and Dovland 1977).   Magistad's (1925) data and
more  recent  data  (Baker and  Schofield 1980,  Clark  1966, Cronan and  Scho-
field  1979,   Dalai  1975,  Driscoll  1980)  show  that the  pH  of incoming
precipitation  is well  within  the range of aluminum mobilization resulting
in toxic levels (as identified in the laboratory).

     The  mechanism  supplying  A13+  is  the  decomposition  of  alumino-
silicates and gibbsite  (Norton 1976, Reuss 1976)

          A12 Si2 05 (OH)4)<1  + 6H* = 2A13+ + 2H4Si02+ H20

          A1(OH)3 + H+ = A1(OH)2'1' + H20

          AKOH),"1" + H* = Al(OH)2"1" + H20

          Al(OH)2* + H+ = Al3* + H20

This  is likely  to  occur  in  watersheds  where  there are no carbonates to
consume H+,  and the above reactions become a primary  buffering mechanism
(Johnson iD79,  Kramer  1976).   The pH  at which  this  buffering occurs is
around  4.5-5.0  as  indicated   by the Al solubility diagram  of  Dalai (1975),
which  shows  that A1(OH)3 begins to decline rapidly at pH  5 as Al3-*-  begins
to   increase.   The  other  soluble  Al   species,  A1(OH)2+,  -.A1(OH)2+ and
Al6(OH)153-t-,  never exceed  10% of total aluminum.

            2.0   3.0  4.0  5.0  6.0   7.0   8.0   9.0 10.0  11.0 12.0

  Figure III-6.   Solubility of aluminum as affected by pH.   Note increasing
                 solubility  with  increasing  acidity.   (From Andren et  al.

              _  500

              g  100

              i   so

Figura III-7.  Aluminum  concentrations   found  in  Norwegian  and  Swedish

              clear water lakes.   (From  Davis  1980.)

                   5 so
                                                   Mini-catchment  5
5 (

10 IS
.'i i
; \ ,
.  '
' i
i i
i \^
; N> '<

/ u
20 25 X 5 X) 15 20


                                     20   25
JO   5
D    IS
Figure III-8.  Nutrient  release  from snowpack  during  1978 spring  melt as
               determined   by  monitoring  runoff  from  mini catchments  in
               Norway.   Note  majority  of  contaminants  released  simul-
               taneously   at  initial  melt.   Shaded area  in  lower  graph
               corresponds to 30% of the runoff during  the  period.   Dashed
               and  solid  lines to  right of  top  figure  represent  average
               concentrations  of  S04 and  H+,  respectively, during summer
               and  autumn  1973.  (From Seip e_t  al_.  1980.)           .  .-.-*.

     In aquatic systems, aluminum forms a variety of complexes with water,
hydroxide, fluoride,  silicate,  organic  matter  and sulfate  (Everhart and
Freeman  1973,  Oriscoll  1980,  Baker  and Schofield  1980).   It  is  rarely
found  as  the free  aluminum  ion.   According  to  Johannessen  (1980),  these
complexes act  as  a  buffer  in the  pH range 4..5-5.0, but  above  and  below
these  levels buffering  cannot  be  ascribed to aluminum complexes.  Henrik-
sen (1980) shows  that lakes  with pH 4.6 - 4.8 are less acid than expected
from a theoretical  "titration"  curve based on  bicarbonate  buffering, and
that the extra buffering can be explained by aluminum.  He postulates that
mineral acid addition to lakes  will cause  them  to change from a carbonic
acid-bicarbonate buffered system  to one buffered by strong acid-aluminum.

     In  surface  waters  of  the  Adirondack Region  of New York,  Driscoll
(1980)  found aluminum-organic complexes  as  the predominant monomeric form
(Avg.   = 44%), which increased  linearly with total organic carbon content.
Aluminum-flouride complexes were  the  most abundant inorganic forn. 'avg. =
29% of the total  monomeric  Al),  with their concentration  increasing with
decreasing pH, although  their  formation was generally limited by fluoride
concentration.   Aluminum sulfate species  were of  relative  unimportance,
although  they  did   increase  with  decreasing pH.   Everhart and  Freeman
(1973)  note  that  the solubility  of aluminum is a direct function cf pH in
the  vase  majority  of  natural  waters  (i.e.,  under  conditions  where hy-
droxide complexes dominate).   In  acid environments the  soluble  forms are
cationic  and polymeric.  Overall,  pH controls  complexation,  polymeriza-
tion,  hydrolysis and solubility.

     Aluminum  also   indirectly  affects  availability  of heavy metals and
phosphorus.   Heavy  metal  availability increases because humic substances,
to which heavy metals are normally bound, are instead complexed and preci-
pitated by aluminum (Davis  1980).   Aluminum also precipitates phosphorus,
as aluminum  phosphate,  thus  decreasing its availability to aquatic biota.

Terrestrial  Effects

     The aluminum mobilized  by  strong acid deposition  is  of most concern
due  to its  toxic effect on  terrestrial and  aquatic  organisms.   Much re-
search on aluminum  toxicity has been conducted on crops and legume species
(Black  1968,  Andrew  1978,  Hewitt  1952,  Foy et  aj_. 1978).  From  these
reports the following statements can be made:

          (11  Excess   aluminum   affects   cell   division  in  roots,
               causing  inhibition of  root  growth  leading  to  stubby
               and  brittle  roots.   The  best indicator  of  aluminum
               toxicity  in  the field  is abnormal  root development.
               The  tips  and lateral  roots  become  thickened  and  turn
               brown,  and no  fine branching roots  develop.  Al  is
               often  found  in  root  cortex cells,  being especially
               concentrated  in  nuclei.    It  generally accumulates in
               roots  and  is only found in above-ground plant parts of
               a  few plant  species.    Germinating plants  and  young
               seedlings  are  generally  more susceptible  than  older
              , plants.

          (2)  Excess  aluminum  fixes  phosphorous  in  less  available
               forms in  the  soil  and  in or on plant  roots.   Conse-
               quently,  phosphorous  deficiency  is  often  associated
               with  toxic  aluminum  concentrations  and  is  generally.
               the  form  of  symptom expression  in above-ground plant
               parts.   Al  has  also  been  shown *.o   interfere  with
               uptake,  transport  and use of other nutrients  (Ca,  Mg
               and K) and water.   Thus, Al  toxicity  can appear as Ca,
               Mg or  K deficiency.   In  cotton, excess aluminum also
               decreased uptake of  Mn,  Fe,  Ni and B,  some  of which
               must be due to a  decreased absorptive  surface.

          (3)  Excess Al also decreases  root respiration,  interferes
               with  certain   enzymes  governing  the  deposition  of
               polysaccharides in  cell  walls, and increases cell-wall
               rigidity  (by   cross-linking  pectins).    All  of  these
               effects of aluminum lead to decreased growth and occur
               below a pH of  5.0,  although effects  have been noted at
               pH 5.5.   Andrew  (1978)  also  notes  that  excess  Al  is
           1    detrimental   to  nodule  initiation   in  legumes,  the
               efficiency of  the  nitrogen-fixing symbiosis  and plant
               growth.    He also notes  that  low Al  concentrations for
               short  periods of  time  can  increase  growth of  some
               plant species.

     Various plant  species  have  quite different tolerance  levels  of Al.
Many  factors affect tolerance/susceptibility,  including:   pH  around  the
root  zone  (tolerant plants  can  produce a  higher  pH  around  root zones);
NH4+  vs. N03-  nutrition; aluminum exclusion processes; calcium nutrition;
phosphorous  nutrition;  and   organic-aluminum  complexes.   All  of  these
factors are discussed in detail  by Foy et a_L  1978.

     Species  known  to  be  sensitive to  aluminum include:    barley,  sugar
beet, corn  and  alfalfa.   Decreased  growth of  barley  occurred  at < 1 ppm;
for corn,  4-17  ppm  Al  was toxic;  and for alfalfa 2.7 ppm was toxic (Black
1968).   In  experiments  with alfalfa  grown  in nutrient solutions,  Munns
(1965)  found decreases  (from controls  containing  0.008 ppm Al)  in shoot
dry weight of 70.3% and 83% at aluminum concentrations of 1.08 ppm and 5.4
ppm,  respectively.   The  first  observed  symptom was  inhibition  of root
elongation  and  lateral  formation,  and plants  with high aluminum concen-
tration showed characteristic phosphate deficiency.   Tolerant crop species
include  oats,   most brassicas  (i.e.,  cauliflower,   narrowstem  Kale,  and
swede), rye, rice, soybeans,  azaleas, cranberry, and triticale.

     For forestry  applications,  McCormick  and Steiner  (1978)  tested the
effect cf aluminum on root elongation in six tree genera.  A hybrid poplar
was  most  sensitive,  with complete  inhibition  at  less  than  10  ppm  Al.
Autumn olive was  sensitive at 10-40 ppm, while the  other genera (Quercus,
Pinus,  AljTus,  Betula)  did not  show effects  until  the  concentration  was
80-100  p;m.   However,  concentrations  greater  than  4  ppm  Al  are  rare in
soil  solutions (Andrew 1978).

     Manganese  toxicity  also is  associated  with aluminum  toxicity (both
occur simultaneously,  i.e.,  under acid conditions).   Low concentrations of

1-4 rng/2 affect  sensitive  plants such as lespedeza,  soybeans  and barley,
although corn  tolerates >15  mg/  and Oeschampsia  flexuosa-' tolerates >60
mg/. (Black 1968).  Tolerance is usually attributed to reduced absorption,
less translocation of excess Mn to plant tops, and/or greater tolerance to
high Mn levels within plant tissues.

Aquatic Effects

     Graphs  such as  those  in  Figures  III-6 to  III-8 are  important in
establishing the relationship between pH and aluminum solubility, but they
do not  adequately  describe peak aluminum mobilization events, nor do they
describe the effects  of continual  input of H+  ions.   Peak concentrations
in lakes are  of  utmost importance biologically because  they may occur in
conjunction with  spawning  and hatching and are thus  most  likely to cause
adverse effects on fish recruitment (Gjessing et al. 1976).  Peak aluminum
concentrations  also  will   occur  in  conjunccion  with  peak  hydrogen  ion
concentrations  (Figure III-9)  which may increase  the overall  effect on
fish  and  other  biota  (see  Schofielc! 1980;  and  pH and  synergisms,  this

     Figure  III-9 graphs  the  relationship  between hydrogen  ion loading
rate and aluminum concentration, and  indicates the effacts large  inputs of
hydrogen ions  would  have  on aluminum concentrations in lakes.  The points
labeled N  are from  lysimeter studies in which  complexation with organic
matter  increased  aluminum  solubility.  Caution should be used when inter-
preting the  data point 0,  as the  full  data set was not available for its
calculation.   It  was  assumed that 100 kg-H2S04/hectare'year resulted in 1
mg Al/ effluent (see Oickson 1978).  Data used  in Figure III-9 are from
Abrahamsen  and Stuanes (unpublished  data),  Crisman  and  Brezonik (1980),
Dickson (1978),  Cronan and  Schofield (1979),  Likens  et aJL  (1977),  and
Wright  et aj_.  (1978).

     The scatter  in  Figure III-9 indicates that other variables  influence
Al  mobilization.   Factors  such  as  soil  type  (clays  retain  aluminum),
amount  of  organic  matter  present,  base  saturation,  cation  exchange
capacity  and  absolute  aluminum concentration  will affect  a watershed's
response to  acid inputs.   These factors emphasize  the  need for  more data
from various watershed types before  conclusions can be made regarding the
hydiogen  ion   loading  rate  that will  result  in  toxic aluminum concentra-

     Graphs such  as  Figure III-9 could be  used  to determine the hydrogen
ion  input  resulting in potentially  toxic aluminum concentrations to  lake
and  stream biota,  but aluminum toxicity  is not  that  simple,  as  it is
affected  by a variety  of  chemical  and  biological  factors.   For example,
the  points  labeled  N  would suggest  that elevated (and hence potentially
toxic)  aluminum  concentrations  can occur with  relatively  small   inputs of
hydrogen ions.  However, these points are a result of increased r.olubility
due  to  complexation  with  organic matter, and  aluminum  is  less  toxic when
complexed to organic matter.  Other  substances act as aluminum ligands and
have the same  effect  (see previous discussion).

     Wright  and  Snekvik (1978) observed that  aluminum  caused no deleter-
ious  effects  at  concentrations  ranging from 0.05 - 0.3 mg/2, found in 90%

- 30

I 25


i 20


                      50              100

                            H+Loading meq/m2/yr
    Figure III-9.   Relationship between H+ loading rate and observed aluminum

                   concentrations  in  lysimeter  effluent,  minicatchments  or

                   freshwater  lakes.  Data from a variety of sources cited  in

                   the text.

of  the  700 lakes  which  they surveyed.   Their results  indicate  that fish
st-^'is was correlated most with. Ca and pH,  with barren lakes having low pH
levels and  low  Ca  concentrations.   They note  that  acid  stress to fish is
apparently more acute  in waters with very  low ionic  strength.   They also
found that Ca ameliorates the effects of low pH, with more fish surviving
low pH with high  Ca-concentration than low pH with low.Ca concentrations.
Nutrient stripping  of  soils  leads to increases  in  Ca2+  concentrations in
lakes which could ameliorate the toxic effects of high aluminum and hydro-
gen ion concentrations.

     Some  research  has  been completed  on  the  physiological effects of
elevated  aluminum  concentrations  on  fisn.   Muniz  and  Leivestad. (1980)
noted a  rapid  loss of plasma Na and Cl at toxic  Al levels, and also large
amounts  of  mucus  clogging  gills,  resulting  in  hyperventilation  and-
coughing.  Venous oxygen tension was decreased in fish with clogged gills.
The  toxic effect  was thus  a  combination  of  impaired  ion  exchange  and
respiratory distress brought on  by mucus clogging of  gills.   Cronan  and
Schofield  (1979)  observed that mortality from acute  exposure to aluminum
appeared to result from severe necrosis of the gill epithelium.

     Estimating  peak  aluminum  concentrations and  toxicity  to  fish with
sufficient  accuracy is not possible from current  data.   The relationship
between  H+ ion  input  and aluminum concentration  graphed in Figure III-9
requires  further   data  for validation.   Data  on aluminum  toxicity (both
chronic  and shock  effects)  to  various fish  species,  -*nd  under various
physical  and  chemical situations,  also  are  needed.  Numerous factors --
pH,  ionic strength,  calcium concentration,  phosphates,  organic matter,
nitrate,  fluoride,  sulfate and  silicates -- all  affect aluminum toxicity.
Relationships  between  these factors  and  aluminum toxicity  need  to be
identified before "safe" aluminum concentrations  can be set.  Based on the
data  presented  in  Tables  III-4  and  III-5,  highly tentative  ranges  are
proposed  as  follows:  non-toxic concentrations would be  below 0.05 mg/,
potentially toxic  concentrations  would  be between 0.05 and  0.3  mg/,  and
toxic concentrations  would be greater than 0.3 mg/.   These ranges should
only  be  applied to  species already tested, trout  and  salmon,  and should
not be applied to other fish species.

H+ and Synergisms with A13+

     Thtre  has  been  some  work  on responses of  fish  to  pH in conjunrtion
with  other variables.  McLeay  et al. (1979)  have  investigated  the toxic
effects  on salmonid fish  of paper  pulp  effluent as  a  function  of  pH.
Their  results   suggest that as pH  decreases, mortality  increases.   They
found 50% to  67% mortality at a  pH below  7.0, which was the normal pH of
the  receiving water.   Baird e_t a_L  (1979)  have  investigated the toxicity
of ammonia as a function of pH.

     Aluminum has  been  found  to  lower the toxicity of pH to fish, i.e.,
low pH and some aluminum are more toxic than low pH and no aluminum (Davis
1980,  Dickson  1978,  and  Baker and  Schofield  1980).   Davis  (1980) notes
that  for brook trout,  pH 4.9 was  not  lethal  but that pH  4.9 and 1.0 mg
Al/2 were  toxic to 50% of  the population.  At  pH 4.0, 1.0 mg Al/2 was  less
toxic, although  pH 4.0 and 4.4 were  toxic  by themselves  (Figure 111-10).
Schofield  (1980)  also found less aluminum  toxicity at  pH 4.0 than at 4.4

Species p_H
Sa 1 mo salar 5.0
0.07 mg/
0.2 mg/S.
Commt its
Toxic within
6 days
Dickson 1978
Dickson 1978
Brook trout
0.2 mg/S.       Toxic response     Cronan & Schofiel
Brook trout
                  0.1-0.3 mg/S.   Growth reduction   Cronan & Schofiel
Brook trout        5.2
  (Salvelinus fontinalis M.)
                  0.42 mg/S.
               28% survival
                 after 2 weeks
                 (mean of 4
                   Driscol 1 et a_L
Brook trout
0.48 mg/je
42% survival
  after 2 weeks
  (mean of 4
Driscol 1 et aj_.
Brook trout
Brown trout        4.3, 4.5,
  (Salmo trutta)     5.0 & 5.3
1.0 mg/
0.2 mg/2
                                  Davis 1980 (work
                                    of Schofield?)
                                                    Muniz &  Leivestad
Brown trout        4.0 & 6.0
  (Salmo trutta)
                  0.2 mg/i.
               No effects
                   Muniz & Leivestad
near 5.0
0.2 mg/2
Very toxic
Grahn 1980
Ciscoe             5.0-5.5
  (Coregonus albula)
                  max 0.5 mg/S,   Large-scale  fish   Grahn  1980
                                   kills  in 2  lakes
                   (majority <5. 0)
                   ,05-. 3 mg/2.
               Not deleterious
                 in 700 Norwe-
                 gian  lakes
                   Wright &
                     Snekvik  1978

                                  I    I    I   I    I    I   I    I   I
                              4       6      8      10
                                DAYS OF EXPOSURE
Figure  111-10. Effect  of altering  pH  on cumulative  mortality  of brook
              trout  exposed  to  three  concentrations  of  aluminum.  (From
              Davis 1980.)

and  higher.   His  results  (Table  III-6)  conclusively show  that  pH  and
aluminum  do  interact  in producing  fish  toxicity.   Cronan  and Schofield
(1979) found that  brook trout showed a toxic response to 0.2 mg Al/2. (and
above)  in the  pH  range of  4.4-5.9.    Everhart and  Freeman  (1973) .-and
Oriscoll  et a_K  (1980)  also note that aluminum  toxicity  is pH dependent.
Grahn (1980) reports  that 0.2 mg Al/2 in  water  near pH 5.0 is very toxic
to salmon.  He also notes that two fish kills of Ciscoe, Coregonus albula,
were associated with  low pH (5.0-5.5)  and maximum aluminum concentrations
of 0.5 mg  Al/2.   Oickson (1978) relates that  0.2  to 0.6 mg Al/2 in asso-
ciation with  a low  pH  (4-5)  is  toxic to  many  fish  species.   Muniz and
Leivestad  (1980),  working  with brown  trout,  Salmo trutta,  and aluminum
concentrations of 0.2-0.8 mg/2, recorded mortality at pH 4.3, 4.5, 5.0 and
5.3, but  no  effects  at pH 4.0 and  6.0.   Maximum toxicity was  at  pH 5.0
where mortality occurred in 18 hours.

     No work has  been done en the combined effects of acid pH's and heavy
metals, or of aluminum  and heavy metals,  on  terrestrial  and aquatic or-
ganisms.    Synergisms between zinc and copper on fish have been reported by
Bandt  (1946)  and  Doudoroff  (1952)  (both  cited  by Lloyd   1961),  and
synergism  between nickel  and  copper  on  decomposing  organisms  has been
reported  by   Freedman  and  Hutchinson  (1980)-   Lloyd  noted  a possible
synergistic effect of zinc  and copper sulfates  on  rainbow trout in soft-
water.  Czuba and  Ormrod (1974) noted synergistic response in lettuce and
cress when ozone  (03) was added in conjunction with Cd or Zn, and similar
interactions were noted with S02 and Ni or Cu.


General Soil  Reactions

     The toxicity of heavy metals to biota was", at one time, assumed to be
determined by the total metal concentration in the system (Mancy and Allen
1977).   It is now realized that this  assumption does  not  hold, and that
toxicity  of   heavy  metals  is  determined  by  their  particular chemical
species.    Research must center on determining what species of heavy metals
are  toxic  and  at  what  concentrations;   what  physical,  chemical  and
biological factors determine  the speciation of heavy metals; and the flux
rates of heavy metals from the atmosphere to soil to ground water.

     Sources of  heavy metals  fall into two  categories:   natural and man-
made.  Natural  sources  include weathering  of  mineral  bedrock,  deposition
of  marine salts,  erosion, volcanism,  and  forest fires.   Man-made sources
include  industrial  and  mining wastes,  fossil  fuel   combustion,  atomic
testing,  fertilizers,  pesticides, manures, and  sewage-'s-tudge.   To deter-
mine'transport  rates between  ecosystem  compartments;  accurate  determina-
tions  of  the  quantity  of  heavy  metals   originating  from  these  sources
should be  calculated,  something which  has not beer, done for source inputs
to the atmosphere (Goldberg 1975).  Goldberg recommends that more research
be  conducted on  the  vapor phase  of  heavy metals,  and on  vertical  and
horizontal  concentrations of  heavy  metals, before  calculating transport
rates from the atmosphere.

             CENTRATIONS ON MORTALITY AT pH 4.0.  (From Schofield 1980.)
Time (days)






Nominal Aluminum
pH (mq/2)
4.0 0.0
4.4 0.0
4.9 0.0
4.9 0.25
5.2 0.0
4.9 0.0
Repl icate
90% (80%)*
Repl icate
90% (70%)*
90% (75%)*
90% (75%)*
 *  days survival

**  Brook water

     Our knowledge of chemistry of metals in soils is incomplete and often
speculative  (Goldberg  1S75).   However,  numerous  factors  are  known  to
affect availability, toxicity, ar* mobility of heavy netals (Ratsch 1974).
These  factors  include   (1)  relative  abundance,  (2) form,  (3)  soil  pH,
(4) interactions with other element1;, (5) physical conditions of the soil,
(6) temperature,  and  (7) soil  moisture.   Generally,  heavy metals  do not
react to the  same conditions in the  same  manner, so generalizations from
one metal to another are not recommenced.  However, an increase in H+ ions
in precipitation  may decrease  the ab'iity  of  the -i"il  system to retain
heavy metal ions  (Tyler 1972,  Abrahamsen an^ Dollard 1979).   In the soil
system, heavy metals are found as (1) minerals of soil structures, (2) in-
organic precipitates, (3)  soluble  and insoluble organic complexes or che-
lates,  and  (6) absorbed ions  on  charged  surfaces  of clays,  precipitates
and organic matter (Goldberg 1975).

     Heavy  metals tend   to  accumulate in soils  (VanHook  et al.  1977, and
Tyler  1972)  and  generally  do  so  by  exchange reactions,  whereby organic
matter  of  the  soil  complex binds  heavy metals  forming  stable complexes
(Abrahamsen and Dollard 1979, Tyler 1972).   Accumulation occurs in/on dead
organic matter,  litter  and humus,  with concentrations increasing with age
and extent of decomposition (Tyler 1972, Van Hook et  aj_. 1977).  Humic and
fulvic  acids  are believed  to  play  a prominent  role in accumulation and
cycling  of  trace metals  in soils  and  sediments  (Nriagu  and  Coker 1980).

     In  aquatic systems,  heavy-metal  speciation is  subject  to change by
biological, physical  and  chemical  factors  (Wollast  et  a_[.  1975,  Elder
1978, Chakoumakos et  aj_.  1979).   These  factors  include:   pH, water hard-
ness, alkalinity, organic matter contant, nutrient content, and oxidation-
reduction  potential   (Goldberg  1975,  Gambrell  et  a_L  1980,  Elder 1978,
Chakoumakos et  a_L  1979).   In freshwater systems'"treavy  metals can exist
in three forms:   readily available (biologically), potentially available,
and  essentially  unavailable.   Processes  affecting  the  availability  of
toxic  metals   include:    (1) precipitation  as   insoluble   sulfides  under
highly reduced  conditions; (2) formation of metal cxiaes and  hydroxides of
less  solubility;  (3) absorption to  colloidal  hydrc-js  oxides of iron and
manganese  (primarily  under aerobic,  neutral  or  alkaline conditions); and
(4) complex formation with soluble and  insoluble organic matter under all
conditions of  pH  and oxidation potential (Gambrell et a_L  1980).  Accord-
ing to  Allen  et al_.  (1980), most studies suggest that free metal ions are
the  most toxic  form,  and  that  the  stronger the complex, tne  lower the

     The following  summaries  illustrate how various  factors  interact with
specific heavy  metals.

        Mercury:   Increased levels  of  soluble  mercurv were found by
   Gambrel   et  a_l_.  (1980)   in  moderately acid,   reduced  conditions (pH
   5.0,  -15CmV)  and weakly   alkaline,  oxidized  conditions  (pH  8.0,
   +500mV).  Wollast  et aj.  (1975) note that mercury is highly soluble
   only  in  well-oxygenated water,  and  that  under moderately oxidizing
   conditions   the .predominant species of  mercury  is  undissociated.
   They  report  that  under more reducing conditions extremely insoluble
   cinnabar (HgS) precipitates, but that under very reducing  conditions
   mercury  may increase in solubility.  Dissolved  mineral mercury can

   be present as Hg(OH)2  in  the aerobic zone and in elemental  form (or
   as organically complexed oxidized  species)  in oxidizing conditions.
   Mercury also exists in  several  organic forms, e.g.,  methyl  mercury,
   which are products  of  bacterial  activity.

        Zinc:    In  freshwaters  zinc  has  been found  to occur  in  both
   ionic and  colloidal  inorganic  forms  (Allen  et  a]_.  1980).   pH  was
   found to affect  the levels  of total  dissolved  and exchangeable  zinc,
   with highest concentrations  found  at the most acid pH employed,  5.0
   (Gambrell et a_L  1980).   At that pH it is most likely to be  found in
   the  free  cationic  form.   Soluble  zinc  decreases  under  reducing

        Lead and Cadmium:   Florence  (1977),  as  cited  in Allen et  a_l_.
   (1980), has  shown  that  in  freshwaters, lead  is  present  to  an  equal
   degree in both inorganic and organic forms, while cadmium is present
   as the  free metal   ion.   Gambrell  et  a_L  (1980)  found  little  dis-
   solved lead at any  pH-redox combination (pH 5.0 to 8.0; redox -150mV
   to  +500 mV).   They  also  found  that  exchangeable  lead  (readily
   available)  was  more influenced by  pH than redox  potential  (higher
   concentrations found at pH  5.0  than 6.5 and 8.0),  whereas reducible
   lead (metal  oxides  and  hydroxides)  was more influenced by oxidation
   strength than pH (higher  levels found at +500 mV  than -150, +50 or

        Copper:  In  fresh  waters  copper  is  predominantly associated
   with  organic colloidal  matter  (Allen  et al_.  1980),  and  is  often
   bound to humic  acids  (Nriagu and Coker 1980).  Elder (1978) studied
   copper speciation in two alkaline freshwater lakes in California, pH
   7.7 -  9.0.   Based  on  the copper concentration and  pH of a lake, he
   constructed a model of  expected speciation of copper  in an aqueous
   solution.   Below  pH 5,  all  of  the  copper is  free Cu2+.   He  also
   discusses  the  fact that all  copper in the lake eventual'iy reaches
   the sediments.   Under  anoxic conditions, there is a low redox poten-
   tial and a release  of hydrogen sulfide from sediments.  Copper com-
   plexes with  sulfur  to  form CuS, which permanently sequesters copper
   in sediments.

Terrestrial Effects

     As indicated  above,  the  soil  system is capable of accumulating heavy
metals  in potentially  toxic  concentrations,  depending  on the  level  of
inputs  and mobilization  reactions.   The  most  common  symptoms  of heavy
metal  toxicity on  crops  and  other  plants  are  chlorosis  and  stunting.
Chlorosis  is  generally caused  by  direct  or  indirect  heavy metal  inter-
actions with  foliar Fe (Foy  et  a_L  1978).   Oats have  often been  used as
indicator  plants  for  Ni   toxicity  because of  their unique  chlorotic  and
necrotic  reaction.   Stunting  can  be caused  by  specific  metal  toxicity,
antagonism  with  other  nutrients,  or  inhibition  of  root  penetration.
Toxicity  is first expressed in root tips, and lateral root development can
be  severely  restricted.    If  root  development is altered,  this will  also
affect normal nutrient uptake, which could lead  to  decreased productivity.

Aquatic,. Effects

     Once  in  the aquatic environment, heavy-metal  chemistry  becomes very
complex, as many factors interact to determine the chemical species found.
Because  of  this,  extrapolating  experimental  results  from  one  aquatic
system  to  another can  lead  to error; i.e.,  "safe"  concentrations in one
system may  not  be "safe" in another.  Even so, water quality criteria aro
set, with  a single  standard for all  freshwater systems.  The  following
brief  review  of  the  literature  shows  the  variability  in  heavy  metal
toxicity, but also gives an idea of the concentrations which are "safe" to
freshwater  fish.   Unfortunately,  most  of the experiments  were conducted
under  neutral  to  alkaline conditions and thus  do  not reflect  toxicities
under acid conditions.

        Copper:   Factors known to  affect copper toxicity  include pH,
   hardness,  alkalinity, and  inorganic  and  organic  complexes.   Cha-
   koumakos  et  aj.  (1979)  studied  the  effects of  alkalinity,  water
   hardness, and  pH  on copper toxicity to Cutthroat  trout.   There was
   an  inverse  relationship between  acute toxicity  and  water hardness
   and  alkalinity.   Copper  was more  toxic  in  soft  water  than  hard
   .water.   Toxic  species,  of  copper  were Cu2+,  CuOH+,  Cu(HO)2 and
   Cu(OH2)2+ while CuHC03,  CuC03  and Cu(C03)|- were  not toxic.   Lett
   et ajL (1976) studied copper toxicity on rainbow trout.  They deter-
   mined  a 96-h  LCSO   of  0.25-0.68  mg   Cu/  for  hard  water  (365  mg
   CaC03/) and  pH 7.8-8.2.   Beamish (1976)  notes  that  17 ug Cu/ did
   not  affect  survival,  growth,  or reproduction of adult rainbow  trout
   in soft water.  Sauter et al_. (1976) recommend that the safe concen-
   tration proposed in 1973 by the National Academy of Sciences (not to
   exceed 0.1  of the 96 h-LC50 for  the  species  of interest) should be
   changed, based on the results of their study.

        Zinc:  Spehar  (1976)  determined  a 96-h LCSO of 1500 pg Zn/ to
   juvenile  (4  to 5-week-old) flagfish and an  estimated MATC (maximum
   acceptable  toxicant concentration) of  26-51  ug/  in Lake Superior
   water.   He  found  survival  of larvae and growth of  females to be the
   most  sensitive  measures  of  toxicity.   Lloyd   (1960)  notes  that
   rainbow trout are capable of acclimation to lethal  concentrations of
   zinc  if  they  are  first exposed  to  sublethal  concentrations.   Ball
   (1967a)  studied zinc toxicity to  four  freshwater  species  and  found
   five-day LCso's for each:   rainbow trout 4.6.mg/; bream 14.3  mg/;
   perch  16.0  mg/;  and  roach  (two  strains)  17.3 mg/.   Cairns and
   Scheier  (1957)  as .cited  by Mount (1966)  found  that  bluegills were
   killed  in  soft water by 1.93-3.78 ppm Zn, whereas  in hard water tne
   -Values were 10.13-12.15 ppm.

        Cadmium:   For   juvenile (4   to  5-week-old)   flagfish  in  Lake
   Superior water, a concentration  of 2500 ug Cu/2 was  toxic to 50% of
   the population  in 96  hours (Spehar 1976).   Spehar determined an MATC
   of  4.1-8.1  ug Cu/  with  the  latter  concentration  inhibiting repro-
   .duction on a chronic  basis.  Spawning and embryo production were the
   most sensitive measures of cadmium effect.   Ball (1967b) estimated a
   7-day   LC50   of  0.4  mg Cd/  for  rainbow  trout,  but  he   cites
   Schweiger's  (1957)  4 mg  Cd/  as   lethal to  rainbow  trout  in  seven
   days, and 3 mg/2 as a safe dose.  Sauter et al. (1976) conclude that

   the National Academy of  Sciences'  recommended safe concentration of
   0.004  mg  Cd/   for  water  with  hardness  100  mg  CaC03/  is  barely
   adequate;  and that  for water  with hardness 100 mg CaC03/,  a recom-
   mended safe  conce.itration  of  0.03 mg Cd/ is  totally  inadequate to
   protect aquatic life.

        Lead:    Davies  and  Everhart (1973) found  the  following toxici-
   ties  for  rainbow trout:    in  hard water (243  |jg  CaC03/)  96-h LC50
   for  total  lead was  471  mg/,  but  for dissolved  lead it  was  1.32
   mg/;  in  soft  water (26.4  mg CaC03/) the 96-h  LC50  for dissolved
   lead was 140 ug/.   They estimated MATC's  for lead:   hard water and
   total lead (0.12-0.36 mg/); hard  water and dissolved lead (0.018 to
   0.032  mg/);  and for  softwater dissolved  or total  lead  (6.0-11.9
   mg/).   Pickering  and   Henderson  (1965)  as  cited   by  Davies  and
   Everhart  (1973)  working with hard water  (300-360 mg CaC03/)  and
   s,oft  water  (18-20 mg/),  determined  the  following  96-h  LC50's  for
   lead:  fathead minnows in  soft  water -  5.58  mg/;  in hard water -
   482 mg/, for  bluegills  in soft water - 25.8 mg/,  in hard water -
   442 mg/.   Sauter et  aj. (1976) conclude  that  the National Academy
   of  Sciences'  recommended  maximum allowable  concentration  of  0.03
   (jg Pb/ appears adequate  to protect most fish species.

     Heavy-metal  toxicity   in  soils  and  freshwater  ecosystems is  very
complex, with  numerous  factors  (physical,  chemical and biological) inter-
acting  in  complex ways to  determine the  chemical  speciation,  and hence
mobility and toxicity,  of heavy  metals.   All  heavy metals do not react in
the same manner to  the same set of  conditions,  so extrapolation from one
heavy metal to another is not reliable.   More research is needed to deter-
mine the  residence  times  of heavy  metals in various ecosystem components,
and to  find  the extent to  which heavy metals  are recycled in terrestrial
ecosystems.    Finally,   *he  chemical  species  of heavy  metals  which  are
toxic,  and  those  that -re not, need to  be  determined.  In  determining
toxicities, experiment should be  conducted using soil  and organisms from
the terrestrial site in question.

                                SECTION IV


       Orie Loucks, Richard Miller, Roland Usher and William Swanson
     The exposure of  forests  to  oxidants and the addition  of  acidic pol-
lutants to a lake/watershed ecosystem both alter the producing  system, but
in very  different (although  potentially  interactive) ways.   The  oxidant
effects (mainly 03) on  forest growth have been reviewed by  the authors in
the recently  completed  ORBES study  (Loucks  et  al_.  1980) and  in a recent
symposium (Miller  1980).   No additional  review will be  added  here.   The
aggregate effect  of  the  acidic  inputs  to the ecosystem, however,  may be
thought  of   as  a  large-scale titration  (Henriksen  1980),  but  it  is  a
complex, uneven process only  superficially similar  to a laboratory titra-
tion.    Gorham  and McFee (1980)  and  Last  et  a_l_.  (1980)  discuss  the
variables involved and note that the hydrogen ions deposited with precipi-
tation  (or   generated  by  other  components  of  precipitation)  may  have
several  fates  in  the lake/watershed system, as  reviewed in the  previous

     Various consequences  follow.   First,  the mobilization  of  elements by
cation displacement can  cause a long-term loss of soil fertility and lower
productivity.  Even if there  is  an  initial  fertilization effect  from the
added nitrogen in precipitation,  continued acid input may result in future
nutrient limitation through  the  loss of essential cations.   Secondly, the
mobilization  of  aluminum and  heavy metals by  cation  displacement  and
weathering  can  result in  toxic  effects for both the aquatic  and  terres-
trial  components  of the  system.   Finally, unneutralized  H+ may  itself be
toxic and a stress to the system.

     The  combined  effects  of   these  above-ground   and   below-ground
processes,  as  they   influence  the  more  highly  integrative  measures  of
response, are  discussed   in  the  following sections from  the viewpoint of
data needs in measuring  forest,  lake or watershed sensitivity.


     The  yields   of  forests  usually are  expressed  as  wood volumes,  and
growth  rates  are expressed  as  cubic feet (or meters)  per  unit  area, per
year.   These volumes are affected by the annual diameter growth rate, but,
since  total  volume is greatly  infuenced by the height  of  the  cone (and
diameter  increment  is intercorrelated with  annual  height growth), height
growth  is  a very  frequent measure  of  forest  productivity  (Forbes 1955).

     Oxidants and  acid  rain,  however,  are relatively recent incursions on
forest processes  (Miller 1980), and, therefore, most studies have reported
changes in foliate condition and Jiameter growth of mature trees -- not in
young,  developing  stands.    Many  other  factors,  including  insects  and
disease problems,  interact with  the mature  tree, however, suggesting that
the height response of younger stands would  have more general application.

                                  53              Preceding  oaee blank

     Various measures  of  soil  texture and soil  nutrient  status  have been
correlated  with  forest  growth  rates,  and  (as  reviewed in  the  previous
section)  these suggest various measurements linking a  forest response to
the  agents  of  chemical  alteration  in the  soil,  i.e.,  acidic deposition.
However,  height  and diameter  growth  of  the  fores't represents  an  "inte-
grator"  of  soil  conditions,  including,  potentially,  the  alteration  of
nutrient  status   due   to  cation  stripping  or  aluminum  toxicity.   These
growth  measures,  however,  are also the "integrative"  measures  that would
respond to  toxic  gaseous  pollutant inputs to  the  forest canopy   S02 as
well as 03.

     Of  the height and  diameter  measures  used   in  forestry,   the  most
generalizable  is  Site  Index,  a basic measure  expressing rate  of  height
growth.   The  Site  Index  (SI)  concept  has  been  accepted  routinely  in
forestry  as  the  best measure of current  or  potential  forest  productivity
because rates  of  height  growth are readily  converted  to cubic volumes of
wood.  The SI is most commonly expressed as the average height of dominant
and  co-dominant  trees  at the age of  50 years.   It is used in forestry to
express the rzce of productivity for a single species in a specified site,
or mapped aggregate of sites.   Although SI usually is specified by height
and  age measurements,  it  can  be estimated  indirectly by  using  soil  and
topographical  factors  (Hannah  1968,  Mader  1963),  height and age  (Spurr
1952),  available  nutrients  (Ellerbe and  Smith  1963,  Gordon  1964, Heilnian
1968)  and available soil  water and  nutrients  (Fralish  and  Loucks 1975,
Mogren and Dolph 1972).

     A  review  of the  literature since 1960 shov~.  that  the  primary vari-
ables  associated with  SI are  the  available  soil  water and  topography
(Mogren  and Dolph  1972,  Fralish  and  Loucks   1975,  Hannah   1968).   The
importance  of  soil  type  and structure has  been discussed (Hannah 1968,
Richards  and  Stone 1964,  Linnartz  1963)  and,  although the  literature
appears  to be  divided  (Broadfoot  1969,  Hannah 1968),  one  of  the most
important  underlying   factors   affected  by  site quality  is   the  rooting
pattern of  the trees  under  investigation.   This  directly correlates with
the  available  soil  water and nutrients in solution, which,  in turn, are a
function  of precipitation and water use.

     To determine site index directly the height  and  age of  the dominant
and  co-dominant trees must be known (Curtis t a_[.   1974).  Plotting height
versus  age  (Figure  IV-1)  of dominant  stems  in  a series of stands results
in  a  collection  of points  which  can be  broken  down  into  a  family of
curves, each representing a separate site class.  Thus, simply stated, the
site  index  is  the height  of  the  stand  projected through  time,  either
forward or backward.  This projection can only be accomplished through the
series  of  harmonic curves  illustrated   in  Figure IV-1.   In the  United
States,  50 and  100 years have  been  chosen  as  reference points  for  the
expression of site  index.

     There  have  been  a  number of  investigations relating  the  role  of
nutrients  to the  SI (Gordon 1964, Heilman 1968).   Mogren and Dolph (1972)
observed  the influence of  pH on  the  SI and concluded  that  there  was no
effect.   It  should  be  noted that they did not report any pH  values in the
report.   A more recent  preliminary literature  search  (1965  to  date) has
shown  that chronic and  acute pH  changes as they  affect the SI  have not
been investigated.





 30       40

Figure IV-1.    Site   index   classes  determined  by  plotting  age  against
               average  height of dominant tree.

     Use of  the forest  site  index as a measure  of  terrestrial  ecosystem
stress  from  acidic <.1eposition  is  an attempt to  integrate  the  effects of
nutrient additions  (such as N  and  S),  nutrient losses such as  Ca,  Mg,  K
and Na,  and the toxic  effects  of  aluminum, H+ and  heavy metals (Drablos
and Toll an 1980).

     There are several liabilities  to this approach.   The first relates to
age classes.   The  site  index  can  only  be applied  if  the average site
quality  is  the same  for  each  age  class.   This liability is  important in
managed  forest  sites  (Spurr 1956)  due to many  practices  that  are used to
increase yields.   A  second  liability stems from the  assumption that the
shape of the height/age curve is consistent between sites.  This generali-
zation  is capable  of  reliable  results; however,  it  does  not hold for all
soils  (Figure  IV-2).   A  third  liability,  related to  the second,  is that
site differences are  apparent  at an  early  age.   In  other words, the har-
monized  curves  that  predict a  taller tree  at 50 years  assume  thac this
tree  has always, been  taller.   To  alleviate the  influences  of  these lia-
bilities, the  standard  site index  curve should be based  on specific soil
types and field measurements of species growth on that site.

     The site  index  is  typically  used in  forest management  to determine
harvest  schedules  for even-aged,  monocultural  stands.   This  relationship
severely limits the  uses of the site  index  from  an  ecological  viewpoint.
A  mixed-species  and uneven-aged stand presents many fundamental problems
in  determining  a  site  index  curve.  To alleviate these  problems,  Curtis
and Post (1964) have  proposed  a  "Composite  Site Index"  to  express site
quality  in  mixed  hardwood  stands.   Their concept uses  the- measured site-
index  of each  species,  each of which is converted  by means  of equations
(Curtis  and  Post  1962)  to  an  estimated maple site  index.  All  the esti-
mated  maple  site  indices are  added to the measured maple  site  index and
then averaged, which results in the composite site index.   This provides a
more  reliable  measure of  site  quality for an  unmanaged, even-aged site.
Curtis  t  a_L  (1974)  have presented  a  detailed  review article  on the
importance  of  the  site  index  and  how  best  to  interpret the  series of
curves obtained.

     The effects of acid precipitation on a watershed/lake system are many
and  intricately balanced,  especially  in  a mature  system.   As  has been
outlined above, the  site index may be an  appropriate  tool  to measure the
effects  on,  or changes in, site quality due to acid precipitation.  Use of
the site index  as  an integrator of  acid  deposition  effects (via nutrient
stripping and  Al  toxicity)  has the advantage of being readily convertible
to economic terms  for cost-benefit analysis.  At this time, however, there
is not enough data to validate its use.


     At  the  present  time there is  no  single,  fully  validated  methodology
for  estimating the  sensitivity of  aquatic  systems  to  acid  deposition.
Thus,  measures  of  aquatic sensitivity to acid inputs  should be  thought of
as hypotheses constructed with a reasonable amount of quantitative inputs,
but evaluation  should not be based on any one model  or measurement proce-
dure.    All  of  the methodologies given belr.w  are  limited in  scope and/or


                      Age,  years
Figure IV-2.   Theoretical  height/age  curves  for'three soils:   (1)  homo-
              geneous  soil,  (b)' shallow  soil,  (c) poor surface soil with
              richer lower horizons (redrawn  from Spurr 1964).

availability of data, but taken together they present a comprehensive view
of our  ability to  quantify current resource status  and,  therefore, some
prediction of prospective future change.

Predicting Toxicity to Aquatic Organ'!sms

     Determining toxic effects  of  acid precipitation on aquatic organisms
requires that standard methods be identified.  Such standard methods exist
for determining general  toxicity to aquatic organisms, and can be applied
to the  toxicity of  acid precipitation.   Bioassay  tests,  conducted under
natural conditions  and  employing  standard  procedures, are  a recommended

     Toy/!city  is  determined by  dosage -  the amount  (in  relation to body
weight  or  growth  medium)  of toxicant and  the  frequency  and duration of
exposure  (Gough et  al.   1979).   The term  is  used by  a  wide  variety of
professionals,  including doctors,  pharmacists,  microbiologists, and regu-
latory personnel.

     Although  chemical  analysis  and  standardized  tables  would  be  the
easiest way  to  determine toxicities, they have not  proven to be feasible
for many  reason:-:    (1)  the specific  organism  in question  often has  not
been tested;  (2) different  bodies  of wat^r have different characteristics
which greatly  affect toxicity of many substances, and  (3) there are dif-
ferences  in temperature,  fish  size,  seasonal  lolerance  of  various fish
species,  testing   procedures,  physiology,  previous  history,  nutritional
state,  and  certain  laboratory  conditions  which may  have  affec;ed  the
results (Katz 1971; American Public Health Association 1965).

     It is  now  recognized that bioassay with organisms of the indigenous
population  under  natural  conditions  is  the most  satisfactory  means of
determining  toxicities of substances (Peltier 1978,  Geckler  et al_. 1976,
American Public Health Association 1965,  Katz 1971).   Ideally,  the tests
should  employ  a range of concentrations of  toxicants,  and should be con-
ducted  at  different  seasons  of  the  year to   reflect  differing  water
quality, which has  been  shown  to  greatly  affect  the toxicity  of sub-
stances, especially heavy metals (Geckler et al.  1976, U.S. EPA 1976, Katz

     The earliest  recommendations  for standard methods were published by
Hart  et al_.  (1945)  and   Doudoroff  et  al_.  (1951),  and have  largely been
adopted by the American  Public Health Association (1965).   Other documents
are available  which outline standard  procedures  for  determining toxicity
of elements  to aquatic  organisms:   U.S.  EPA (1972);  Peltier (1978);  and
The Committee on Methods  for Toxicity Tests with Aquatic Organisms (1975).
Numerous review articles have been published on the measurement of pollu-
tant toxicity  to  fish (Sprague 1969, 1970 and 1971; Herbert 1965; Edwards
and Brown 1967; Alderdice 1967; Burdick 1967; and Warner 1967).

     Acute .toxicity  to fish is most frequently measured by the concentra-
tion of toxicant  that is lethal to  50% of  the  population within a  speci-
fied time period (usually 96 hours).  Various terms are used, e.g., Median
Lethal  Dose  (LD50), Median  Lethal  Concentration  (LC50),  Median Tolerance
Limit (TLm), Median  Effective Dose (ED50), and Median Effective Concer.tra-

tion  (EC5o),  although LC50  and  EC50  are preferred.  The  latter also can
refer to  lethal  or  sublethal  responses.   When used to determine sublethal
responses,  EC5(j  is   defined  as  the  concentration  causing  an "adverse
effect"  in  50 percent of  the population within a  specified  time period,
whert "adverse effect" is generally defined as immobilization.

     Three procedures  for  determining LC5o's and EC50's  are  discussed by
Sprague  (1969)  from  which  the following is largely taken.  For all three,
tests are conducted using a series of concentrations of the toxicant, with
mortality being  recorded at  various  times.   The  results are  plotted on
log-probability paper, as  shown  in Figure IV-3 (from Sprague 1969).   From
such  a graph,  one can determine median survival times (when acute mortal-
ity  ceases),  and the  presence  of differing modes  of  action.   Changes in
slope  or grouping of  lines  are  clues to differing modes  of  action.   The
next  step in  determining the LC50 is  to  plot all  the median lethal  times
(obtained  from graphs similar  to Figure  IV-1) against  concentration to
find a lethal threshold concentration.

     A second  procedure  for  determining LCso's is  described in Standard
Methods  (American Public  Health  Association 1965)  and by  the Doudoroff
Committee (1951).  It yields LCSo's for specified time periods, usually 1,
2  or 4  days.   For  each fixed time,  percentage mortalities  are plotted
against  test  concentrations, and  the concentration lethal to  50% of the
population  is  then  estimated by  interpolation.  Sprague  (1969) suggests
that  more frequent  observations   be  made and  that  investigators should
construct time-concentration toxicity curves as described above.

     No  reports  specifically outline  standard  procedures  for determining
chronic  toxicity.   This  underscores  the fact that  chronic toxicity tests
have  only recently  gained  the  interest of  toxicologists (Cough  et al.
1979).   Long-term studies using a wide range of concentrations of a parti-
cular  toxicant  on large  numbers  of aquatic organisms, which determine the
overall  community response,  are  needed  by  regulatory  agencies (U.S.  EPA
1976).   Geckler  et  aJL  (1976)  suggest  that  tests  other than  those on
survival, growth, and reproduction be conducted; specifically, that tests
based  on behavioral  responses may better predict effects  of  toxicants on
natural  systems.   They concluded from their  tests  with copper, conducted
over  a 33-month  period,  that organisms avoided the highest" concentrations
of  copper.   Results  of chronic  tests  can be  expressed either  in terms of
the  lifetime of one organism or the time span of more than one generation.
Chronic  effects  often  influence  the  species  population  and  not  just

      Sprague  (1970,   1971)  and  Mount  and  Stephan   (1967)  discuss various
methods  of  determining sublethal or  "safe"  concentrations,  including the
use  of toxic units and application factors.   The latter procedure  involves
taking the  LC50  and applying some factor (generally  .1)  to determine the
"safe" concentration.

      In  conclusion,   the  following recommendations  can be made for mea-
suring effects of acid precipitation on lake/watershed ecosystems:

      (1)  That  additional  standard bioassay  tests  determining  LC5o's
          and  EC50's' be  used  when   toxicity  tests  are  conducted;


       3  70

       ~  50
                 10         50    100         500  1000

                    Time  Of  Observation,  min.
Figure  IV-3.   Results of a mortality test at various  temperatures plotted
              on  logarithmic-probit  paper.   The  results  indicate  two
              modes  of  'lethal  action  at  27.5C.   (From  Sprague 1959.)

      (2)  That the results of  these  toxicity  tests  be presented as a
           range,   indicating  the variable effect  that water  quality
           can have on  the toxicity of many substances,  particularly
           heavy  metals;

      (3)  That.extrapolation   f results  of toxicity tests  from one
           body of water  to another,  and  from  one  species to another,
           be avoided  except where absolutely necessary.   It should be
           noted  that  extreme variability can  exist  from one aquatic
           system  to  another,  from one species  to another,  and from
           one season  of the year to another.

 The Calcite Saturation Index  (CSI)

      Kramer  (1976) has  summarized the relationship between  pH and alka-
 linity  in  the calcite  saturation index (CSI)  which  expresses the relative
 saturation of water with CaC03.

      CSI = p(Ca2+) +  p(Alk) - p(H+) * pK,

 where p(X) =  -log10X,  pK = 2,  (Ca2+) is  given as  mol/ and (Alk) and (H+)
 are given  as  eq/2.   When CSI < 1, the lake  is nearly saturated with CaC03
 and not susceptible to acidification.  Increasing  CSI indicates increasing
 susceptibility,  and,  at CSI > 4, the lake is very  likely to be affected by
 acid (Glass and  Loucks 1980).

      The CSI  has  been  used  to classify  lakes sensitive to acidification
 (Glass  and Loucks 1980,  and others)  and  is  a relatively simple  way  to
Account  for  lake  buffering  capacity.    However,  it is  not comprehensive
'enough to be used as  a sensitivity index  as  defined  earlier for watersheds
 because  buffering   capacity   is  only   one   aspect   of  lake/watershed
 sensitivity.   It  is  necessary  to account explicitly for  other possible
 consequences of  the acid input, including the  mobilization if aluminum and
 the  loss  of soil  fertility  by  cation stripping.   Moreover,  the  CSI
 reflects  only  present   conditions  and  does  not  relate  past  or  future
 changes  to  loading rates.   There is no  indication in  CSI  of  how a lake
 came to  have  a  particular value, or of  any explicit relationship between
 the CSI and acidic inputs.

 The Dickson Relation

      One way to  summarize the relationship between lake pH and atmospheric
 loading  of  acidic pollutants  is  to  plot lake pH in  relation to measured
 local  sulfate  loadings,  as  attributed  to  Dickson  (Aimer e_t  a_[.  1978),
 Figure  IV-4.   Various  combinations  of  watershed  characteristics  are
 reflected  in  the family  of  curves connecting the different  sets  of lake
 data.   A  larger  number  of  curves  may  be  envisioned,  each representing
 lake/watsrshed  systems  of  comparable   sensitivities  as  they  apparently
 respond  to  increases in  atmospheric acid inputs.   The  curves  are super-
 ficially similar  to titration curves, as  might be  expected, and each shows
 buffering  at  high pH,  then a rapid  drop (as  this-buffering apparently is
 depleted)  to  lakes with somewhat lower  pH and minimal  buffering.   The pH
 in the more sensitive lakes drops sooner  and more  rapidly, relative  co the
 loading of SO^-.


            0                 30               60               90
                Sulfate  Loading  to  Lake  Water   [Kg/ha-yrl
Figure  IV-4.   Effects  of various sulfate loading  rates  on lake pH for
              lakes  in very sensitive (1) and somewhat less sensitive (2)
              surroundings in Sweden.  From Glass and Loucks 1980.   Added
              points  are for:    Florida (Crisman and  Brezonik  1980);
              8 Como  Creek  (Lewis  and  Grant 1979);  A Hubbard   Brook
              (Likens  e_t al.  1977);  and X Norway (Wright and Snekvik

     A graph of sulfate deposition versus lake pH could be used to express
lake/watershed  sensitivity,  and  therefore  acid  loading  tolerance,  by
evaluating  the  position  of  the  lake/watershed  system  on the  response
curves.    For  example,  the  lakes  of  southern  Norway  are underlain  by
granites and  felsic gneisses  (Wright and  Snekvik  1978),  and  almost all
prove to be  somewhat  sensitive.   From a knowledge of the response of very
sensitive geochemical/biological  systems in  moderately  impacted systems,
it  appears  possible  to anticipate  similar  responses  in  the  future  on
similarly sensitive  systems if  acidic  inputs increase  or  are sustained.

     Examination of Figure  IV-4 suggests that annual  sulfate  loadings  of
less  than 15 tc 17 kg/ha would  be unlikely to degrade  lakes  of the type
represented  in curve  (1).   However,  if the  lower  envelope of  the data
distribution is viewed as a potential "family" of the most sensitive lakes
and streams, these  appear  to be just barely  free  of potential acid load-
ings  effects  at an annual  rate of  9 to 12  kg/hayr.   Thus,  two "toler-
ances" can be defined, one associated with a possible protection of nearly
all sensitive  aquatic resources,  and the  other with  protection of some-
thing less  than all  sensitive  lakes,  e.g., only the  half  of  the "sensi-
tive" resources that lie above curve (1) in Figure IV-4.

     Since these curves were developed on  the basis  of  Swedish data, and
represent a  north-south  geographic gradient in S0|- loadings,  and are not
actual observations of  acidification responses over time, these estimates
of  acid  loading tolerances  must be viewed very cautiously for application
in a North American context.

The Henriksen Nomograph

     Henriksen  (1980) presents  a model  based on the  concepts  implicit in
titration of  a  bicarbonate-buffered  lake  with  strong  acid (principally
H2S04) from  the atmosphere.   In the process,  bicarbonate  is depleted and
lake  pH  can  fall  below 5 with consequent effects on aluminum mobilization
and fish.  These  relations  are basically the same as those in the Dickson
work, and  are  summarized  in  the  nomograph  (Figure  IV-5)   using  two key

     (i)  ambient  concentrations  of  in-lake  calcium  (or  Ca+Mg)  as  an
          estimator of the pre-acidification alkalinity;  and

     (ii) lake  sulfate  concentrations  (in  excess  of  marine  input) as  an
          estimator of H+ added to the system.

     The  resulting   semi-predictive  nomograph   is  divided  into  three

     (i)   bicarbonate  lakes,  where  original  alkalinity was  high and/or
           added H+ is low, uo that  the  lakes remain bicarbonate buffered;

      (ii)  acid  lakes,  where original  alkalinity was low relative to acid
           irputs,  and  all   bicarbonate  appears  to  have  been  depleted by
           the acid addition; and

 5 200
             250 -
                      25   50        100        150       200

                     Excess Sulphate  in Lakewater, ueq/l
4	j	1_

 4.4  4.3              4.1

 pH  of  Precipitation
Figure IV-5.    A  nomograph to predict  the  pH of  lakes  given the-sum of
               non-marine  calcium and  magnesium  concentrations  (or  non-
               marine  calcium  concentration  alone)  and  the non-marine
               sal fate  concentration  in  lake water  (or  the weighted-
               average  hydrogen   ion   concentration   in   precipitation).
               Nomograph  also can  be  used  to  predict  future changes in
               lake   pH   when  precipitation  acidity   increases.    (From
               Henriksen  1980. )

     (iii) transition lakes, in which bicarbonate appears to be undergoing
           reduction  (or  is  almost depleted)  end large  pH  fluctuations
           occur during runoff events.

     The  transition  phase,   in   which  the  lake  is  shifting  from  a
bicarbonate-buffered  equilibrium  at moderate  pH to  an  aluminum-buffered
equilibrium at  low pH, represents  the  key  process  requiring prediction.
This  shift  apparently  is  forced  by  H+ and  S05j- inputs, and  Henriksen
presents  regression  equations based on Norwegian lake  and precipitation
data for  representing  sulfate in  lake water (S04*) in terms of sulfate in
precipitation [S04(p)] and H+ in precipitation [H-t-(p)]:

          S04* = -19 f 1.9 S04(p)

          S04(p) = -2.7 + 1.37 H+(p)

All  concentrations  are ueq/,  and  the  sulfate  is excess  over  that from
marine  origin.   These  equations  have  not  yet  been  validated  for  North-

     The  methodological   component   that  can  be applied  to   determine  a
threshold  loadings  tolerance'  must  u^e assumptions  as  to the  processes
maintaining steady  state  lake  pH  in various  types  of aquatic resources,
ranging  from  very  sensitive to only moderately  sensitive.  Maintaining a
pH of  5.3 would,  presumably, prevent lakes  from entering  the "transition
phase".  This method., however, does not consider the depression of lake pH
due  to  spring  snowmelt,  although  the  results (based  on  summer  data)
suggest the spring depression is being  taken into consideration.

Episode Receptor Dose/Response Relation During Shock Events

     A  possible alternative model  for describing and projecting aquatic
resource  response to acid loading emphasizes the non-equilibrium nature of
the  acidification  process,   equivalent  to  Henriksen's "transition phase".
During  this phase,  aquatic  ecosystems   can show  some buffering and a rela-
tively  unaltered pH  during  much of the year, but they also may experience
sudden  pH drops during hydrologic  flushing  events,  i.e., at  times when
accumulated  acidic  material  in  the  watershed  is  flushed  relatively
suddenly  into streams and lakes.  The rapid  increases  in H+ concentrations
can  also  mobilize  A13+ for  a brief period,  and  the shock event can become
a  multi-agent  toxic  stress affecting the  survival  of  fish  and  other
organisms at various life stages,  depending  on the timing and magnitude of
the pH  depression.

     Research on brook  trout and  Atlantic salmon  by  Daye  (1980) and Daye
and  Garside  (1975,  1976,  1977,  1980a,  1980b), and  related  research by
Beamish (1974,  1976) and Harvey (1975,   1979, 1980a, 1980b) have provided a
broad  understanding  of the  response of several  pH-sensitive  fish species
to  both  long-term and  short-term  elevated  H+ exposures.   Mortalities of
fish  eggs,  sac fry  and  adult  fish have  been  viewed  as a  response to
continuing  chronic  pH  depression.   At  the seme  time,  effects on egg via-
bility,  hatching1  success,  and adult   survival  are   known  to occur  as a
response  to short-interval  acute  H+ and A13+  exposures.  The experimental
data b-,se supporting these relationships can be  generalized to support the
following statements (Andrews e_t aj_. 1980; see previous section).


     (1)  The short-term acute exposure,  or  shock effect,  can be expected
          when pH drops  by  a range of  0.5 to 1.5 units of  the pH scale in
          a background of pH 5.5.  to 6.5;  and

     (2)  These shock exposures may be  significant at a pH  above the level
          at which chronic effects ordinarily would be produced.

     Taken  together,  these  data  suggest  that  for  waters  normally  in  a
range of  pH 5.5  to  6.5, a pH depression  (A pH) of 0.5 to 1.0  can cause
substantial, and  physiologically  significant,  acid-induced alteration of
water chemistry.   This relatively  large increase in H+ during  shock events
may  be  occurring  only  in  regions where  acidic  deposition  already  has
produced  a  downward  time-trend in  stream pH and alkalinity, as  shown in
several  monitoring programs.   Data  from  very sensitive regions  relatively
free of long-term acid  inputs (or in the  first to second decade of moder-
ate deposition) indicate it is unlikely that  a short-term pH depression of
these magnitudes can  occur with an unusual watershed stimulus.

     Given  this  dose/response relationship.,   a  procedure for defining an
acid-loading tolerance,  or  loading  threshold, also can be  suggested.   The
annual  S0|- loading  which,  when  subjected   to  a defined  flushing event
(e.g., snowmelt or first m-jor rainfall  following drought),  leads  to the
minimal   biologically-significant   short-term  H+  and  A13+  exposure,  as
determined by controlled environment studies.

     The data available on pH depression during flushing events  indicate a
range in responses from very  little to  as  much as 1.0 unit  of  the pH scale
during snowmelt in northern MinnecJta,  a region of recent  and only moder-
ate acid deposition (Glass et a_L  1981a).   Depression of pH is shown to be
in  the  order  of 1.0  unit on the  Shaver's   Fork  River in West  Virginia
(Dunshie  1979,  Figure IV-6),  to  more  than  2.0 units  in  the Adirondacks
(Schofield  1980,  Figure IV-7).   Values  in this range also have  been re-
ported  for  Hubbard  Brook  (Likens  et aj.  1977)  and  Plastic Lake,  Ontario
(Zimmerman  and Harvey  1979).  Jeffries  et   aj_.  (1979)  report  spring pH
depression  in several Ontario lakes and streams of 0.3-1.2.
     The data presently available on the general  relationship between
deposition  ana  pH depression  (A pH) are  presented in  Figure  IV-8.   The
dashed  lines indicate  areas  of  differing geological  sensitivity  (i.e.,
buffering capacity).   It  is  also important to recognize that two types of
streams will  show little  effect:   "acidified" streams, those  with  pH in
the  range   of 4.7 to  5.0  during much  of  the year,  and which  have  gone
through the "transition phase"; and well-buffered  streams.   Those  likely
to  show  appreciable pH  depression  are those  believed  to  be experiencing
the "transition phase" of Henriksen.

     Pending  further  testing,  a  "significant" shock event  response, de-
fined as a A pH of 0.5 to 1.0 unit,  mav be a useful  estimate of the annual
S0|-  threshold  loading required  to  produce  an  unacceptable level  of pH
depression  (this  level  to  be based  on  the  controlled  environment effects
observed  on  fish populations).   These  estimates   must  be defined for
watersheds within a specified sensitivity range,  a limited range of hydro-
logic dilution (i.e., stream size),  and with a defined return interval for
the episode.  Recurrence  an  average of once  a year during critical   life-



                        II ICtll



rl,;.n;irA:i DAILY PII ret us sii/v/rns FORX nivrn AF DUIIS. i/.v.
PI![Ciri!A(IG.-| LVtiir rl! All!) ACCUIIULAIIttl At AliBOT/AlC. I
                                                          A RIVER HI

                                                          O BAIill-AU Pll
                                                          j] nAi:ir/\iL ACCIESJLAI 101

                                                                                     10    o    
                                              ilnIl_Ln	t'l.nnll'illL
                                                  10     15     70    25
                                                                                     Reproduced Irom   ||pl
                                                                                     bes! available copy. %yj
                                                                                                                           1.0  -;
         Figure  IV-6.    River pH,  rainfall  pH,  rainfall  accumulation and  discharge rate  for the Shavers Fork River,
                          West  Virginia,   illustrating  river  pH  depressions  during snowmelt,  when  river  acidity  is
                          generally  greatest.    Precipitation  data were collected  at Arborvale,  West Virginia.   (From
                          Dunshie 1979.)

             7 -
                                          7 I
Figure IV-7.    pH  depression  in  Little Moose  Lake,  New  York,  occurring
               during spring snowmelt as measured in lake outlet and water
               entering  the  laboratory from  a  3-m intake  from  the lake.
               Stream pH was  more  severely affected than lake pH.   Little
               Moose Lake aluminum  concentrations  increased from pre-thaw
               leve'is  of less  than 20  ueq/  to  320 ueq/  during early
               thaw.  (From Schofield 1980.)

          3.0 r
pH    1.5




                    10     20     30     40     50     60

                      SULFATE  LOADING   [Kg/ha/yrl
Figure IV-8.    Relationship  between  sulfate loading rate and potential pH
               changes  occurring  during  spring melt  for  sensitive areas
               (top dashed line) and  slightly  less  sensitive areas  (bottom
               dashed  line).   Rapid  pK  changes  of  0.5 to 1.0  unit are
               considered  detrimental to  most fis;.  species.   Data shown
               are from:   (1) Minnesota (Glass 1980); (2) Ontario  (Harvey
cycle stages  would be  consistent  with the  physiological  data  base  des-
cribed above,  and implicit in the model.   Flushing events,  however, appear
to  be possible  at almost  any  time  of  the year  and, therefore,  could
threaten other life stages and other species.  From Figure  IV-8, a sulfate
loading of 5  to  7 kg/ha-yr is seen to produce a critical episode response
(A pH) in the range of 0.5 to 1.0 for the most sensitive streams.

     The  major  advantage of  the  above  model   is  that  it relates  acid
loading directly  to biological  effects of interest, namely, a  pH depres-
sion  sufficient  to cause  observable  mortality  of  organisms at sensitive
life  stages.    It  also  emphasizes  non-equilibrium  transition  events,
thereby predicting when resource impacts will tend to begin, not when they
will  be  past  and  the  lake or  stream already  acidified.   The  model  also
appears likely to apply equally to lakes as well as streams.

     Although  the  episode  receptor/dose  relation  incorporates  physio-
logical data  on  organism  sensitivity to  H+, further work  is  required to
investigate  fully  the assumptions   and  relationships  implicit  in  this
approach  to  measuring water quality  alteration  and. defining acid-loading
tolerances.  The model needs to be evaluated fully in relation  to types of
geological materials in the watershed, the pa-v, duration of acid loadings,
the  range in  flow rates  of  the  streams or  seasonal   rivulets,  and the
interval  for  deposition accumulation.   It also  needs to be evaluated for
applicability  in  southern watersheds  where  snow does  not  accumulate and
where  flushing  events  are  related  to   significant   rainfall  following
drought (Ounshie  1979).   Finally,  the role  for fish   mortality  of  short
periods   of   elevated  aluminum   concentrations   must   be   investigated,
particularly  the  apparent  interactions  with  H+  (Schofield,  personal

                                 SECTION V

     This  report  has considered various measures  of ecosystem alteration
with respect  to  the  effects of acid  rain  on  terrestrial  and aquatic eco-
systems.   The  details  of  the  measurement proposals,  however,  beg  the
question of  what constitutes an appropriate or  minimally  sufficient des-
cription of ecosystem response.  One should be particularly concerned that
the  "indicator"  measures  not  be  selected  from a  restrictive  set  that
documents only a limited aspect of total ecosystem function.  For example,
concerns over eutrophication seldom take  into account  the  larger changes
in fauna and  flora that accompany such processesbut more often restrict
documentation  to  a  few  "water  quality"  parameters,  reflecting  the
state-of-the-water chemistry.

     Recognizing  the need  for  a minimally sufficient  set  of well  chosen
measures to  reflect  the  various dimensions of  ecosystem transformation,
Statistics Canada  has developed a  framework  for environmental statistics
which  seeks  to establish  taxonomies,  both  for  human  stress  factors  and
environmental  responses,  that  are  reflective of the  major  aspects  of the
environmental transformation process (Rapport and Friend 1979; Rapport and
Regier  1980;  Rapport  1981).   With  respect to  ecosystem  response,  the
provisional   taxonomy   for   response   measures   includes  indicators  and
integrators  pertaining  to  the  following  ecosystem  attributes:   produc-
tivity,  nutrient  concentration and  cycling,  composition  of  the  biota,
quality of biota, and quality of the environment.

     The measures  discussed in this  report cover a  number of aspects of
the  overall   response  taxonomy.  The  forest  site  index  reflects  forest
ecosystem productivity; the calcite saturation index deals with aspects of
nutrient cycling;  the  toxicity measures refer to  the quality of biota 
both with  respect  to mortality and morbidity; the composition of biota is
best documented  in the sequential  decline  in  reproduction  and ultimately
extinction of  various  fish  species in acidifying lakes.  This latter work
could be extended to identifying the more sensitive "function:;1! groups" of
species  that are  displaced as well  as  new groups  of  dominants  that may
become  established.   With respect  to quality of biota, referred to above,
measures  of  changes   in   size  distributions   (particularly  stunting  of
trees), disease incidence on trees  in acid-stressed areas, and failures to
reproduce  might  also  be  included  in  addition  to  the  toxic aspects
already mentioned.


     A  preliminary  listing  of  the  data needed to document  lake and water-
shed  sensitivity  is  given  in Table  V-l,  effectively  summarizing  the
various  methodologies  discussed previously.   The  table  is meant only to
suggest  the  range of factors  involved  in  lake  and watershed  sensitivity,
not  to  be  comprehensive or  complete  at  this  time.   However,  one need not
have all  the listed  data to estimate some  of the  integrative  measures.   A


Area (ha):
% Urban 	
% Forest
Mean Drainage Basin:  Slope
Water Flow Volumes:   Input
Water Flow Rates:     Input
Retention Time:
Avg. Precipitation (mm) 	
% Agricultural
% Other
                                              Mean Annual pH
Regional Atmospheric Bulk Loading (kg/ha-yr):  S02 	  NO  	
SO^- 	  NH4 	   Pb 	  Hg 	  C.3 	
Regional Cumulative Growing Season Exposure (ug/m3-yr above background):  0;
Major Soil Types Parent material
Horizon Depth (cm)
CEC (meq/100g)
Base Saturation (%)
Soil pH (water)
Soil texture
Extractable (ug/2)





Soil Sensitivity:
    Coote et al.
             Non-sensitive     Slightly sensitive

Major Plant Communities:  	
Dominant/Codominant Tree spp.  	
Age of Dominants 	
Height of Dominants 	
Understory Spp.
Field Measurement of SI (ft.) 	
Regression Estimate of SI (ft.) 	
Calculated Nutrient Loss (Gain) Due to Acid Deposition (kg/ha-yr):
     Ca 	 +Mg 	  K 	  Other 	
Calculated peak AL3>  (ug/2) 	   +
Estimated SI with Nutrient Change  and Al3  (ft.) 	
Estimated Change in Annual Fiber Production (kg/ha) 	


TABLE V-l.   (continued)


pH 	  Alkalinity (meg/2)
Specific Conductance (|jnihos/cm)
Secchi Disk Transparence (m) +
Concentrations (mq/2): DOC POC Caz
K+ . Na+ NO,- NH^-f
P02~ HCO.,- Al Mn
ZN Cu Hg Cd
Calcite Saturation Index

   CSI = p(Ca2+) + p(Alk) - p(H+) + 2

1.   Ca2+ (mol/2) 	;  p (Ca2*) 	
2.   Alk (eq/2) 	;  p (Alk) 	
3.   H+ (eq/) 	;  p (H+) 	
4.   CSI = 	

(<1  . . Non-sensitive; 1-3  . . Potentially Sensitive; >3  .  . Sensitive)

Henriksen Nomograph (See Figure IV-5)

1.   Ca2+ cone. 	  (ueq/2)          2.   Mg2  cone. 	  (ueq/2)
3.   Excess S0|-	 (peq/2)
4.   Lake Status:  Acidified 	

     (Based on Henrikson, 198C)

Dickson Relation

SO2,  loading (kg/ha-yr) 	
Lake pH 	
Sensitivity category (see Figure IV-4) 	


Chronic effects:  Al (mg/2) 	  pH 	
(See Tables III-3 and III-4)                                    '

Shock effects:  SOj  loading
  Sensitivity  category  (from  Dickson)
  Predicted A  pH 	
  Predicted peak Al 	

viable sensitivity assessment  strategy would fill in  theso  data  needs as
studies or  existing  resource  surveys  (e.g., soi'ls  are  assembled for as-
sessment purposes).

          General Watershed Data

     Coarse estimates of  lake/watershed sensitivity could be made through
evaluation  of the  data collected  under this  heading.   For  instance,  a
watershed with a  large percentage  of  urban  and agricultural  lanJ (and
associated  fine  particulate  runoff)  is unlikely  to  be as  sensitive as an
area with a large proportion of forest and water.  Knowledge of water flow
rates, volumes and soil  retention time also is important in assessing the
ability of soils to neutralize acidic inputs.

          General Airshed Data

     Application of  the Site  Index  measure requires  the use  of several
dose/response  relationships  for sensitive tree  species.   Thus, ozone and
other gaseous pollutants should be available from regional monitoring.  In
addition, use  of Almer/Dickson relation, Henriksen Nomograph and pH shock
effects models  requires measuring the atmospheric  loading  of acidifying

          Soil Sensitivity Data

     The data needs listed here are derived from McFee (1980) and Coote et
aj. (1980).   Extractable  aluminum should be measured  to allow preliminary
estimation of potential  toxicity problems, should sulfates accumulation in
soils  reach  that  point.   Together,  these  data  provide  a  broad-scale
assessment of sensitivity.

          Site Index Sensitivity Data

     Forest site index  is  proposed as  an integrative  measure  for terres-
trial sensitivity because  oxidants,  aluminum toxicity and nutrient  losses
can all  affect  the  rate of height growth,  a term that translates readily
into  terrestrial  productivity.   The  magnitude  of  site  index Decreases,
therefore,  represents a  measure of the  resulting  loss in wood fibre pro-
duction due to 03 and acid deposition.

          Aquatic Sensitivity Data

     The  methodologies  for  evaluating  aquatic  sensitivity  have  been
developed to  a greater  extent than terrestrial  methodologies, but field
validation  is still required.  The data needs listed here are based  or. the
methodologies of Kramer  (1976), Henrikson (1980) and  Aimer et a]_. (1978),
as discussed  in  previous  sections.   Monitoring heavy-metal concentrations
would help  to define potential  toxicity problems.

          Fish Sensitivity Data and Watershed Tolerance

     The chronic effects of pH  and aluminum on fish  reproduction  have been
described as  fully as  possible,  although more  work on aluminum toxicity
will  be  important.   The  pH  and  aluminum  shock effect  on  various  life-

stages of fish  is  a dose/response relationship now being developed and in
need of validation.  Development of this methodology is critical, as it is
the only  measure  that  bears  directly on  the biological  effects  of acid


     A considerable amount  of research  will be necessary to show that the
measures  given  above  are adequate and appropriate  to  determining forest,
lake/watershed  sensitivity.   While  a  variety of  data needs  is implicit
throughout  the  previous  discussion, this  section attempts  to  summarize
necessary research according to major subject areas.

          Effects on Forest Productivity

     Since  forest  site  index  is a highly integrative measure, it reflects
a  variety  of  effects  mechanisms,  including  oxidant  damage  to foliage,
nutrient addition and depletion, and the toxic effects of H+, aluminum and
heavy metals.  The magnitude of the effect on height growth in relation to
the combined dose must  ultimately be measured  in.the  field,  not simply
calculated  from a variety of separate studies.  In general, much more data
are needed  on  the  response of  various forest  species  and entire communi-
ties to the various toxic agents.

          Sulfate and H+ Transpo. t

     The  movement  of H+ within the  sytem is, along with  movement of the
anion  (SO^-),   fundamental  to predicting all  other effects.  Water flow
patterns  influence the pathways  of  S0|- and H+ transport  and  the inter-
action  with other system  components.   Simple  quantitative methods  are
necessary to predict  water flow patterns in relation to soil type, topog-
raphy,  rate of flow  and other relevant variables.  Of  particular impor-
tance  is  study of flushing event processes  as they  affect S0|-  and  H+
contact with the  soil, and  the  regulation  of  the nitrogen and sulfur
cycles  in soil.   The theoretical  considerations of these  questions given
in this paper need to be supplemented with experimental work and watershed
mass balance studies.

          Nutrient Cation Stripping

     While  theory and some field studies indicate acid precipitation leads
to  Teaching of  cations  such as calcium and magnesium,  other field studies
indicate  such  losses  may  be  small.   Quantitative  studies of nutrient
stripping and the consequential effects on forest productivity are needed,
therefore,  for  a  variety of soils with  different  ratis  of acidic inputs.
This  is  especially important  in  soils  which  are   not now  acid, but have
relatively  low base saturation.

          Aluminum Mobilization

     The  relationship between  SOij   and  H+ input  to a  watershed and the
subsequent  output  of  soluble  aluminum must be studied on the watershed as
well as the soil  column level.  Of critical importance are the  effects of

variation in slopes,  soil  type and water,  flow  (flushing event)! patterns.
In order to predict the occurrence of spring or .fall aluminum flushes, one
must  understand  the   dynamics  of  water  flow  during  snowmelt or other
flushing  events  as   water  penetrates  various  soil  horizons.   Research
requirements include  comparative watershed studies  as  well  as regionally
placed lysimeter studies.

          Role of Organic Substances in Water

     Superimposed on  many  of  the H+ and metal  mobilization  processes -and
effects  are the modifying  influences  of organic  molecules  within  the
system.   These  range  from buffering  of  H+  input, to  detoxification of
aluminum,  to  mobilization  and  transport  in  chelated  form  of   various
elements and compounds.   At  the present time, many of these processes are
poorly understood, particularly in aqueous media.   Further research snould
identify sources  of organics  in a variety  of  soils and determine  quanti-
tatively the impact of the amounts and  kinds  of  organics being observed.

          Other Aquatic Effects

     Most of  the questions  concerning'aquatic effects  centc^ around the
dose/response  relationships of  H+,  aluminum and heavy metals for fish and
other  components of  the aquatic  community.    Of  particular  concern are
potential synergisms  between the various  toxicants and overall community
and  ecosystem  responses.  Effects  on  fish-eating  birds  and  mammals  as  a
lc:e/stream fishery declines,  and  certain species are c-1 inrinated,  must be
investigated  fully.    It  is  possible  that  research  in this area  may
identify  additional   broadly  based measures  of aquatic  ecosystem stress
comparable to the site index for forests.

Abrahamsen, G.,  !(.  Bjor,  R.  Horntvedt and  B.  Treite.   1976.  Effects  of
     acid precipitation on coniferous forest,   pp.  36-63  In:   F.H.  Braekke
     (ed.)  Impact ojf  Acid  Precipitation on  Forest  and Freshwater  Eco-
     systems i_n Norway.  SNSF Project, NISK, Oslo, .Norway.
Abrahamsen, G. , and  G.  J.  Dollard.   1979.   Effects of acid  precipitation
     on  forest  vegetat"'  n  and soil.   In.    Ecological  Effects  of  Acid
     Precipitation.    Report  of a  Workshop held  at  Galloway,  Scotland,
     September 1978.   EPRI SOA  77-403.  17 pages.
Abrahamsen, G. ,  and A.  0.  Stuanes.   Unpublished Manuscript.  Effects  of
                rain  on  the effluent  from  lysimeters  with  acid,  shallow
                in organic matter.  9 pp.
                1977.   Simple  computer  model   for  leaching  in structured
                Soil Sci. 28:554-563.
                  1967.  The detection and measurement  of water pollution-
                assays.   Canada Dept. Fisheries:   Can.  Fish Rept.  No.  9,
     soil, rich
Addiscott,  T.
     soils.  J.
Alderdice, D. F
Allen, H. W., R
     on aquatic
                 H. Hall, T. D. Brisbin.   1980.
                toxic-'ty.  Environ. Sci. Techno!
Aimer, B. ,  W.  Dickson,  C. Ekstrom, E.  Hornstrom
     and the aquatic ecosystem,  pp.  273-311  In:
     j_n  the Environment:   Part J_I  Ecological
     Sons,  Inc.
American  Public  Health  Association.   1965.   	
     Examination  of Water and Wastewater  Including
 Metal speciation.  Effects
   1978.   Sulfur pollution
  J.O. Nriagu (ed.) Sulfur
 Impacts.   John  Wiley and

Standard  Methods  for the
      Bottom  Sedime~nts and
     Sludges.  12th  Edition.   New York:  American  Public Health  Associa-
     tion.  769 pp.
Andren, A.,  M.  Anderson,  N. Loux, R. Talbot  and G.  Glass.   1980.   Element
     flow  in aquatic  systems  surrounding  coal-fired  power plants:   Wis-
     consin  power  plant   impact study.   EPA-600/3-80-076.    U.S.   EPA,
     Duluth, Minnesota.  84 pp.
Andrew, C.  S.   1978.   Legumes and  acid  soils.   In:  J. Dobereiner,  R.K-.
     Burris,  and  A.   Hollaender  (eds.)   Limitations  and  	
     Biological Nitrogen  Fixation j_n the  Tropics.   Plenum  Publishing,
                                                            Potentials  for
     York.  398 pp.
Andrews,  A.K.,  R.A.  Ellison, D.8. Hamilton,  J.E.  Roelle,  and  D.  Marmorek.
     1980.  Results  of a modeling worksnop concerning  acid  precipitation.
     Western Energy and Land Use  Team. Office of  Biological  Services,  U.S.
     Fish and Wildlife Service,
               1980.   The sereitivity of  soils to acidification,   pp.  569-
               T.C.  Hutchinson and M. Havas,  (eds.)  Effects  f Acid  Preci-
                 TerrestriaJ Ecosystems.   Plenum  Press,  New  York.
                           and H.  Taitz.   1979.   Ammonia toxicity  and  pH
             in fish toxicity bioassays of tr-'^tec1 watersheds.   Water  Res.
Bache, B.  W.
     572 Jn:
     pi tat ion p_n 	
Baird, R. ,  J.  Bottomley,
Baker,  J.  P. ,   and  C.  L.  Schofield.   1980.   Aluminum toxicity to  fish  as
     related to acid precipitation  and Adirondack surface water  quality.
     pp. 292-293 In:  D. Drablds  and  A.  To!Ian  (eds.) Ecological  Impact  of
     Acid  Precipitation.   Proceedings  of  an  international   conference,
     Sandefjord, Norway, March  11-14,  1980.   Sf.'SF  Project,  Norway.
Ball,  I.R.   1967a.    The  relative  susceptibilities  of some  species  of
     freshwater fish to poisons.   II. '  Zinc.  Water Res.  1:777-783.

Ball,  I.R.    1967b.   The  toxicity  of  cadmium  to  rainbow  trout  ( Salmo
     gairdnerii Richardson).  Water Res. ].:805-805.
Beamish, R.J.  1974.   The loss of fish populations from unexploited remote
     lakes in  Ontario,  Canada,  as a consequence of atmospheric  fallout of
    . acid.  Water Res. 8:85-95.
Beanr'sh,  R.  J.    1976.   Acidification of lakes in Canada by acid precipi-
     tation  and  the  resulting effects on fishes.  Water  Air Soil Pollut.
Black,  C.A.    1968.   Soil-Plant  Relationships.   John  Wiley &  Sons,  New
     York.  792 pps.
Broadfoot, W.M.   1969.   Problems relating soil to site index for southern
     hardwoods.  Forest Sci. 15:354-364.
Burdick,  G.  E.  1967.   Use of  bioassays  in  determining  levels of toxic
     wastes  to aquatic  organisms.  Trans. Amer.  Fish.  Soc.  96(1) Suppl.,
     Spec. Pub. No. 4:7-12.
Chakoumakos,  C. ,   R.   C. Russo,  and R.  F.  Thurston.   1979.   Toxicity of
     copper  to cutthroat trout  (Salmo  clarki) under different  conditio'.s
     of alkalinity, pH and hardness. Environ.  Sci. Tech.iol. 13(2):213-219.
Chen,  C.W. ,  S. Gherini  and R.   A.  Goldstein.  1978.   Spelling the  lake
     acidification process.   In:  Ecological Effects of Acid Precipitation.
     Report  of a Workshop,  Galloway,  Scotland, September 1978.   EPRI SOA
Clark, J. ,  and R.  Turner.    1965.   Extraction of  exchangeable cations and
     distribution  constants  for ion exchange.  Soil Sci. Soc. Amer. Proc.
Clark, J.   1965.   The relation  between pH and  soluble and exchangeable Al
     in some acid soils.  Can. J. Soil Sci. 46(l):94-96.
Clark,  J.,   J.  McKeague and W.   Nichol.   1966.   The  use  of pH-dependent
     cation  exchange  capacity for characterizing  the  B horizons of brun-
     solic and podzolic soils.   Can. J. Soil  Sci. 46(2):161-166.
Committee  on  Methods  for  Toxicity  Tests  with Aquatic  Organisms.   1975.
     Methods  for  acute toxicity  tests  with  fish, macroinvertebrates, and
     amphibians.   U.S.   EPA 660/3-75-009.   U.S.  EPA,  Con/all is.   62 pp.
Cook,  R.  B. ,  and D.  W.  Schindler.   1980.   Alkalinity production  in an
     anoxic  hypolimnion.   Paper  presented  at  43rd  Annual  Meeting, Amer.
     Soc. Limnol. and Oceanog.,  Knoxville, TN.
Coote, D.R. ,  D.  Siminovitch, S.S. Singh and  C. Wang.  1980.  The signifi-
     cance of  the  acid  rain problem  to  agriculture  in eastern  Canada.   A
     draft report prepared for the Canada-U.S.  Research Consultation group
     on the  Long Range Transport  of Air Pollutants (LRTAP).
Crisman,  T.  L.,  and  P.  L.   Brezonik.   1980.   Acid Rain:  Threat to sensi-
     tive  aquatic e  .osystems.    Paper  for   presentation  to  73rd  Annual
     Meeting Air Pollut. Contr.   Assoc., Montreal, Quebec, Canada.
Cronan,  C.S.   1980.    Solution  chemistry  of  a  'New  Hampshire   subalpine
     ecosystem:  a Mogeochemical analysis.   Oikos 34:272-281.
Cronan,  C. ,  and C.  Schofield.   1979.   Aluminum leaching  response to  acid
     precipitation:   Effects on  high-elevation watersheds  in   the north-
     east. Science 204:304-306.                                .    . 	
Curtis, R.O. ,  and B.W.  Post.  1962.  Comparative  site  indices for riorthern
     hardwoods  in  the  Gresn Mountains of  Vermont.   U.S.  Forest Service,
     Northeastern  Forest Expt. Sta.  Paper #171.   6  pp.
Curtis,  R.O. ,  and  B.W. Post.    1964.   Basal area,  volume,  and diameter
     related  to  site  index   and  age  in  unmanaged  even-aged northern hard-
     woods  in  the Green Mountains.  J.  Forestry 62:864-870.

Curtis, R.O. , D.J.  DeMars and F.R. Herman.   1974.   Which dependent  vari-
     able  in site  index-height-age  regressions?   Forest  Sci.  20:74/87.
Czuba, M., and D. P. Ormrod.  1974.  Effects of cadmium and zinc on ozone-
     induced phytotoxicity  in cress and  lettuce.  Can. J Bot. 52:645-649.
Dalai,  R.C.   1975.    Hydrolysis   products  of  solution  and  exchangeable
     aluminum in acidic soils.  Soil Sci. 119:127-131.
Davies, P. H. ,  and  W.  H. Everhart.  1973.  Effects of chenr'cal variations
     in aquatic  environments.  III.   Lead toxicity  to  rainbow trout  and
     testing  application  factor   concept.    EPA R3-73-Q11C.   U.   S.   EPA,
     Washington O.C. 80 pp.
Davis, J.M.   1980.   Acidic  precipitation.   External  Review  Draft.   U.S.
     EPA Environmental  Criteria  and  Assessment Office,  Research  Triangle
     Park, N.C.   91 pp.
Oaye,  P.G.   1980.   Attempts to acclimate  embryos  and alevins of  Atlantic
     salmon,  Sal mo  salar,  and rainbow  trout,  S.  gairdneri,  to  low  pH
     Can.  J. Fish. Aquat. Sci. 37:1035-1038.
Daye,  P.G., and E.T. Garside.  1975.  Lethal  levels of pH for brook trout,
     Salvelinus fontinalis (Mitchill).   Can. J. Zool.  53:639-641.
Daye,  P.G.,  and E.T.  Garside.   1976.   Histopathologic  changes  in  the
     surficial   tissues  of   the  brook   trout,   Salvelinus  fontinalis
     (Mitchill), exposed  to acuts  and chronic  levels  of pH.   Can.  J.  Zool.
Daye,  P.G.,  and  E.T.  Garside   1977.    Lower  lethal  levels  of pH for  em-
     bryos and  alevins of atlantic salmon,  Sal mo  salar  L.    Can.   J.  Zool.
Daye,  P.G.,  and  E.T.  Garside.  1979.   Development and survival of embryos
     and  alevins of  th<;  atlantic salmon,  Sal mo  salar  L. ,  continuously,
     exposed  to  acidic,  levels  of  pH  from fertilization.  Can.  J.  Zool;
Daye,  P.G.,  and E.T.  Garside.  1980a.   Structural  alterations in  embryos
     and alevins  of the Atlantic  salmon,  Salmo salar L., induced by con-
     tinuous or short-term exposure to  acidic  levels  of pH.   Can.  J.  Zool.
Daye,  P.G.,  and  E.T.  Garside.    1980b.   Development, survival and struc-
     tural  alterations of  embryos and  alevins  of Atlantic  salmon,  Salmo
     salar  L. ,   continuously  exposed   to  alkaline  levels  of  pH,   from
     fertilization.   Can. J.  Zool.  58:369-377.
DeVilliers,  J. ,  and   M.  Jackson.   1967.   Aluminous chloride  origin  of
     pH-dependent  cation  exchange capacity  variations.   Soil  Sci.   Soc.
     Amer. Proc. 31(5):614-619.
Dickson, W.   1978.   Some effects  of  the acidification  of Swedish lakes.
     Verh.  Internat. Verein.  Limnol.  20:851-856.
Dooge, J.C.I.   1973.   Linear Theory  of Hydrological  Systems.  USDA  Tech-
     nical Bulletin #  1468.   327  pp.
Doudoroff,  P.,  B.  Anderson,  G.   E.  Burdick,  P.  S.  Galtsoff, W.   B.  Hart,
     R. Patrick,  E.  T.  Strong,  E. W.  Surber, and V.  M.  VunHorn.   1951.
     Bio-assay methods  for the evaluation  of  acute t?\;city  of industrial
     wastes to fish.   Sewage  I rid.   Wastes 23:1380-1397.
Drablrfs,  0., and  A.   To!Ian (eds.).    1980.    Ecological  Impact  of   Acid
     Precipitation.     Proceedings  of   an    international    conference,
     Sandefjord,  Norway,  March 11-14,  1980.   SNSF Project,  Oslo,  Norway.
     383 pp.

Driscoll, C.  T.   1980.   Aqueous speciation of  aluminum  in the Adirondack
     region  of  New York  state,  USA.   pp.  214-215  In:   D.  Drablefs and A.
     Tollan  (eds.)  Ecological   Impact  of Acid Precipitation.   Proceedings
     of  an  international conference,  Sandefjord,  Norway,  March 11-14,
     1980.  SNSF Project, Oslo, Norway.
Driscoll, C.T., Jr., J.P. Baker, J.J.  Bisogni, Jr., C.L  Schofield.   1980.
     Effects  of  aluminum speciation  on fish  in  dilute  acidified waters.
     Nature 284:161-164.
Dunshie,  D.   1979.   Evaluation of  the surface  water  resources  of  the
     Shavers  Fork  River.   Unpublished manuscript,  Eastern  Region,   U.S.
     Forest Service, Milwaukee, WI.  48 pp.
Edwards,  R.   W. ,  and  V.  M.  Brown.    1967.   Pollution  and  fisheries:   A
     progress report.   Wat.  Pollut.  Control,  Lond. , J.  Proc.  Inst. Sewage
     Purif. 66:63-78.
Edwards,  D.  J. ,  and S. Hjeldnes.  1977.  Growth and survival  of  salmonids
     in  water of different pH.  SNSF-project,  Norway, FR 10/77.   12 pps.
Elder,  J. F.  1978.   Copper cycles  and CuS04  algicidal  capacity in  two
     California lakes.  Environ. Manage. 2(1):17-31.
Ellerbe,  L.,  and G. Smith.  1963.  Apparent influence  of phosphate marl on
     site  index   of loblolly pine  in the  lower  coastal  plains  of  South
     Carolina. J. Forestry 61:284-286.
European  Inland  Fisheries  Advisory  Commission   (EIFAC).    1969.    Water
     quality  criteria   for european freshwater  fish-extreme pH values  and
     inland  fisheries.  Water Fas.  3:593-611.
Evans,  L.  S. , K.  f~,   Lewin, and F.  A. Vella.  1980.   Effect of  nutrient
     medium  pH  on  '.ymbiotic nitrogen  fixation  by Rhizobium  leguminosarum
     and  Pi sum sativum.   Plant and Soil 56:71-80.
Everhart, W.  H. , and R. A. Freeman.  1973.   Effects of chemical variations
     in  aquatic  environments:   Vol II. Toxic  effects  of aqueous  aluminum
     to  rainbow trout.  EPA-R3-73-011b.  U.S.  EPA, Washington, O.C.
Fahey,  T. J.   1979.   Changes  in  nutrient content of  snow  water during
     outflow  from  Rocky  Mountain  coniferous  forest.   Oikos 32:422-428.
Forbes,  R.D.  (ed.).   1955.   Forestry Handbook.  The Romld Press  Company,
     New  York.
Foy, C.  D., R. L. Chancy, and M. C. White.   1978.   The physiology  of  metal
     toxicity in plants.  Ann.  Rev. Plant Physiol.  29:511-556.
Fralish,  J. ,  and 0.  Loucks.   1975.    Site  quality evaluation models  for
     Aspen (Populus tremulodies Michx.) in Wisconsin.   Can. J. Forest Res.
Freedman,  B. ,  and T.  C.  Hutchinson.   1980,  Effects of  smelter pollutants
     on  forest  leaf litter decomposition near  a  nickel-copper smelter at
     Sudbury, Ontario.  Can. J. Bot.   58:1722-1736.
Fromm,  P. 0.   1980.   A review of  some physiological  and  toxicological
     responses  of  freshwater  fish  to acid stress.   Environ. Biol.  Fish.
Gambrell,  R.  P.,  R.  A.  Khalid, and  W. H. Patrick,  Jr.  1980.    Chemical
     availability  of mercury,   lead  and zinc  in  mobile  bay  sediment sus-
     pensions  as   affected  by  pH  and  oxidation-reduction conditions.
     Environ. Sci.  Technol. 14(4):431-436.
Geckler,  J.   R. ,  W.  B. Horning, T  M. Neiheisel,  Q.  H. Pickering,  E. L.
     Robinson, and  C.  E.  Stephan.   976.  Validity  cf  laboratory  tests for
     predicting copper  toxicity  in streams.  EPA 600/3-76-116.    U.G. EPA,
     Ouluth,  Minnesota.  192 po.

Gjessing,  E.  T. ,  A.  Henriksen,  M. Johannessen,  and  R.  F.  Wright.   1976.
     Effects of acid precipitation on freshwater  chemistry,  pp.  65-85 jjn:
     F.H.  Braekke  (ed.)  Impact ojf Acid Precipitation on Forest  and  Fresh-
     water Ecosystems i_n Norway.    SNSF Project, Oslo, Norway.
Glass, G.E.  1980.  Susceptibility of aquatic and terrestrial  resources .of
     Minnesota, Wisconsin and Michigan to  impacts from acid precipitation:
     Informational  requirements.   pp. 112-113   In:   0.   Drablris  and  A.
     Tollan  (eds.).  Ecological  Impact of  Acid Precipitation.  Proceedings
     of  an  international   conference,  Sandefjord,  Norway,  March  11-14,
     1980.  SNSF Project, Oslo, Norway.
Glass, G.E.,  and  O.L.  Loucks, eds. 1980.   Impacts  of airborne  pollutants
     on  wilderness areas  along  the  Minnesota-Ontario  border.    EPA 600/
     3-80-044.  U.S. EPA-Ouluth.    187 pp.
Glass,  G.E. ,  O.L.  Loucks,  and  R.W.  Miller.   1980.   A  research plan to
     determine the  susceptibility of aquatic and terrestrial  resources to
     damage  from  atmospheric  pollutant  deposition.   U.S.  EPA-Ouluth,
Glass, G.E., T.G.  Brydges and O.L. Loucks  (eds.).   1981.  Impact  Assessment
     of  Airborne  Acidic  Deposition  on  the Aquatic  Environment  of  the
     United  States  and  Canada.   U.S.  Environmental  Protection  Agency,
     Environmental Research  Laboratory-Duluth.  152 pp.
Glass, G.E.,  L.J.  Heinis,  L. Anderson, C.  Sandberg,  J.  Use, J.  Rogalla,
     F. Boettcher and T. Rousch.    1981a.   pH and  alkalinity changes  during
     snowmelt  in   streams  and  lakes  of  the Boundary  Waters  Canoe Area
     Wilderness  (BWCAW)  and Voyageurs  National  Park  (VNP)  of  northern
     Minnesota.   Presented  before the Division of  Environmental  Chemistry
     of  the  American  Chemical   Society,  Atlanta,  Georgia,  March  1981.
Goldberg,  E.  0.   1975.   Entry, distribution  and  fate of heavy  metals  and
     organohalogens  in  the  physical  environment.   pp. 233-256 In: A.D.
     Mclntyre  and  C.F.  Mills (eds.)  Ecological Toxicity Research;  Effects
     of  Heavy  Metal  and Organohalogen Compounds.   Plenum Press,  New York,
Gordon, A. G.  1964.  The nutrition and growth, of ash, Fraxinus  excelsior,
     in  natural stands  in the English lake district as related  to  edaphic
     site  factors.  J. Ecology 52:169-187.
Gorham,  E. ,   and  W.   W.  McFee.   1980.    Effects  of  acid  deposition upon
     outputs   from  terrestrial  to  aquatic  ecosystems.   pp.  465-480  In:
     T.C.  Hutchinson  and M.  Havas (eds.)  Effects o_f Acid Precipitation p_n
     Terrestrial Ecosystems.  Plenum  Press, New York.
Goriny, C.,  and J. Jamaker (eds).  1972.   pp. 3-143  In:  Organic Chemicals
     _uj the  Soil Environment, Vol.1.  M. Dekker, Inc. , New York.
Gough,  L.  P.,  H.  T.  Schacklette,  and A.  A.  Case.   1979.   Element  concen-
     trations  toxic  to  plants, animals and man.  Geological  Survey Bulle--
     tin  1466.  U.  S.  Government  Printing Office,  Washington, O.C.  80  pp.
Grahn,  0.    1980.    Fish kills in two moderately acid lakes  due to  high
     aluminum  concentration.  pp. 310-311  In:    D.   Drableis  and  A.  Tollan
     (eds.)  Ecological   Impact of Acici Precipitation.   Proceedings  of  an
     international  conference,   Sandefjord,  Norway,  March  11-14,  1980.
     SNSF  Project, Norway.
Grande,  M. ,  and  S.  Andersen.  1979.  Toxicity  of  acid snowmel1" water  to
     brown trout  (Sal mo  trutta L.).   Vatten  35:27-32.
Grande,  M. ,   I. P.  Mun->z, and S.  Andersen.   1978.   Relative  tolerance  of
     some  salmonids  to  acid waters.    Verh.  Interna..  Verein.  Limnol.

Hagen, A., and  A.  Langeland.   1973.  Polluted snow in southern Norway and
     the  effect of  the meltwater  on  freshwater and  aquatic organisms.
     Environ. Pollut. 5:45-57.
Hall, R.  J. ,  and G.  E.  Likens.  1980.  Ecological effects of experimental
     acidification on  a stream  ecosystem.   pp. 375-376.   In:  D. Drablrfs
     and  A.  Tollan (eds.)  Ecological Impact  of  Acid  Precipitation.   Pro-
     ceedings  of an  international  conference,  Sandefjord,  Norway,  March
     11-14, 1980.  SNSF Project, Oslo, Norway.
Hannah,  P.  R.  1968.   Estimating site  index  for  whice  and black  oak  in
     Indiana from soil and topographical factors.  J.  Forestry 66:412-417.
Hart, W. B., P. Doudoroff,  and J. Greenbank.   1945.  The Evaluation of the
     Toxicity  of  Industrial  Wastes, Chemicals, and other  Substances  to
     Freshwater  Fishes.    Atlantic   Refining  Co., Philadelphia.   317 pp.
Harvey,  H.H.   1975.   Fish  populations  in  a  large group of acid-stressed
     lakes.  Verh.  Internat. Verein. Limnol.   19:2406-2417.
Harvey,  H.H.    1979.   The  acid  deposition  problem and  emerging research
     needs in the toxicology of  fishes.   Proc. Fifth Ann. Aquatic Toxicity
     Workshop,  Hamilton, Ontario,  Nov.  7-9,  1978.  Can.  Fish.  Mar.  Serv.
     Tech. Rep. 862:  115-128.
Harvey, H.H.   1980a.   Widespread and diverse changes in the biota of North
     American  lakes  ana rivers  coincident  with acidification,  pp. 93-98
     In:  0.  Drabltfs  and A.  Tollan  (eds.) Ecological  Impact of Acid Preci-
     pitation.   Proceedings of  an  international  conference,  Sandefjord,
     Norway, March 11-14, 1980.  SNSF Project, Oslo, Norway.
Harvey, H.H.   1980b.   Continuing changes in the  fish fauna of the LaCloche
     mountain lakes.   Int.  Conf. Ecol. Impact  of Acid Precip., Sandefjord,
     Norway, March 11-14, 1980.  SNSF Project, Oslo,  Norway.
Harvey,  H.H.,  R.C.  Pierce,  P.J. Dillon, J.R.  Kramer,  and D.M. Whelpdale.
     1981.  Acidification in the Canadian Aquatic  Environment:   Scientific
     Criteria  for  Assessing the  Effects  of  Acidic  Deposition on Aquatic
     Ecosystems.   Publication  NRCC  No.  18475.   National  Research Council
     of Canada, Ottawa, Ontario, Canada.  369  pp.
Heilman,  P.  E.   1968.   Relationship  of  availability  of  phosphorus  and
     cations  to forest succession  and  bog  formation  in interior Alaska.
     Ecology 49:331-336.
Henriksen, A.   1980.   Acidification of  freshwaters:   a large scale titra-
     tion.   pp. 68-74 In:    D.  Drablrfs  and  A.  Tollan  (edsi) Ecological
     Impact  of  Acid Precipitation.   Proceedings of  an  international  con-
     ference, Sandefjord,  Norway,  March 11-14,  1980.   SNSF Project, Oslo,
Herbert, D. W. M.  1965.  pp.   173-195 In:    Pollution and Fisheries.  Ecol.
     Ind. Soc., 5th Symp. British Ecol.  Soc. Blackwell Scientific, Oxford.
Herrmann,  R. ,  and  J.  Baron.   1960.   Aluminum  mobilization  in  acid stream
     environments, Great Smoky Mountains  National Park, USA.  pp. 218-219
     In:   D.  Drabltfs  and A. Tollan  (eds.)  Ecological  Impact of Acid Pre-
     cipitation.   Proceedings  of an international conference, Sandefjord,
     Norway, March 11-14, 1980.  SNSF Project, Oslo, Norway.
Hewitt,  E.  J.   1952.   A   biological  approach  to the  problems  of  soil
     acidity. Int.  Soc. Soil Sci. Trans. Dublin  1:107-118.
Hutchinson,  G.  E.   1957.   A Treatise on Limnology.  Volume 1.   John Wiley
     and  Sons,  New York.  1015 pp.
Inhaber,  H.   1974.   Environmental  quality:   Outline  for a national index
     for  Canada.  Science 186:798-805.
Jeffries,  O.S., C.M. Cox,  and  P.J.  Dillon.   1979.   Depression  of pH  in
     lakes and  streams  in  central  Ontario  during  snowmelt.  J. Fish.  Res.
     Board Can. 36:640-646.

Johannessen, M.   '.980.   Aluminum,  a buffer in acidic waters?  pp.  222-223
     In:  D. Drablrfs  and A.  Tollan (eds.)  Ecological  Impact of  Acid  Pre-
     cipitation.  Proceedings  of  an international conference, Sandefjord,
     Norway, March 11-14, 1900.  SNSF Project, Oslo, Norway.
Johansson, N. , P. Runn and G. Milbrink.   1977.  Early development of three
     salmonid species in acidified waters.   Zoon  5(2):127-132.
Johnsen,  I.,  and B.  Freedman.   1980.   Identification  of sensitive sites
     and  soils.   pp.  611-615   In:   T.C.  Hutchinson and M.  Havas (eds)
     Effects  of  Acid  Precipitation  on  Terrestrial  Ecosystems.  Plenum
     Press, New York.
Johnson,  D.  W.    1980.   Site susceptibility to leaching  by H2S04  in  acid
     rainfall.   pp. 525-535  In:    T.C.  Hutchinson  and  M.   Havas (eds)
     Effects  of  Acid  Precipitation  ojn  Terrestrial  Ecosystems.  Plenum
     Press, New York.
Johnson, D.W., D.W. Cole, S.P.  Gessel, M.J. Singer and R.V. Minden.  1977.
     Carbonic  acid  leaching  in  a  tropical,  temperate,  subalpine,   and
     northern forest soil.  Arctic and Alpine Research 9:329-343.
Johnson,  N.M.   1979.   Acid rain:   Neutralization within  the Hubbard Brook
     ecosystem and regional implications.  Science 204:497-499.
Katz,  M.    1971.   Toxicity  bioassay  techniques  using  aquatic  organisms.
     pp. 763-800   In:   L. L.  Ciaccio   (ed.)  Water  ar.d  Water   Pollution
     Handbook, Vol. 2. Marcel Dekker, Inc., New York.
Kramer, J.  R.  1976.  Geochemical and  lithological factors  in acid precipi-
     tation,  pp.  611-618  In:   USDA  Forest Service  General  Tech. Report
Last,  F.T.,  G.E.  Likens, B.  Ulrich, and L. Wallete.  Acid precipitation --
     progress and problems.  Conference summary,  pp. 10-12  In:  D.  Drablds
     and  A.  Tollan (eds.) Ecological Impact  of Acid Precipitation.   Pro-
     ceedings  of an  international  conference, Sandefjord,  Norway, March
     11-14, 1980.  SNSF  Project, Oslo, Norway.
Lett.  P.  F. ,  G.  J.  Farmer, and F. W.  H. Beamish.  1976.   Effect of copper
     on  some  aspects   of  the  bioenergetics  of  rainbow  trout  ( Salmo
     gairdnerii). J. Fish. Res. Board Can.   33(6):1335-1342.
Levins, R.   1968.   Evolution  j_n  Changing  Environments.   Princeton Univ.
     Press, Princeton, NJ.  120 pp.
Lewis,  W.  M. ,  and M. C. Grant.  1979.  Changes in the output  of ions  from
     a  watershed  as  a  result  of  the acidification  of precipitation.
     Ecology 60:1093-1097.
Light,  P.   1341.   Analysis  of  high   rates  of snowmelting.    American
     Geophys. Union Trans., Pt. 1:195-205.
Likens, G.  E. ,  F.  h,. Bormann,   R.  S.  Pierce,  J.  S. Eaton, and N.  M. John-
     son.   1977.   Biogeochemistry  of  a  Forested   Ecosystem.    Springer-
     Verlag, New York.   146 pp.
Likens, G. E. , R. F. Wright, J. N. Galloway and T. J. Butler.  1979.   Acid
     rain. Sci. Amer. 241(1):43-51.
Linnartz, N. E. 1963.  Relation of soil  and topographic characteristics to
     site quality for southern pines  in the Florida parishes of  Louisiana.
     J. Forestry 61:434-438.
Lloyd,  R.   1960.   The  toxicity of zinc sulphate to rainbow  trout.   Ann.
     Appl. Biol.  48(l):84-94.
Lloyd,  R.    1961.  The toxicity of mixtures of zinc and copper  sulphates to
     rainbow  trout (Salmo  qairdneri i Richardson).   Ann. Appl.   Biol.  49:

Loucks, O.L.,  T.V.  Armentano,  R.W.  Usher, W.T. Williams, R.W. Miller,  and
     L.T.K.  Wong.    1980.    Crop  and  Forest  Losses  Due  to  Current  and
     Projected  Emissions  from Coal-Fired  Power Plants  in  the Ohio  River
     Basin.   Report prepared for  Ohio River  Basin Energy Study  (ORBES).
     U.S.  Environmental Protection Agency, Washington, D.C.
Mader,  D.  L.  1963.   Soil  variability:   a  serious  problem  in  soil-site
     studies  in the  northeast.   Soil  Sci.  Soc.  Arr.er.  Proc. 27:707-709.
Magistad,  O.C.   1925.   The  aluminum content of  the soil solution  and  its
     relation  to soil  reaction  and plant growth;   Soil Sci. 20:181-226.
Mancy,  K.  H. ,  and  H.  E.  Allen.   1977.  A controlled bioassay system  for
     measuring   toxicity  .of   heavy  metals.    EPA-600/3-77-037.   U.S.
     EPA-Duluth, Minnesota.   Ill pp.
McCormick,  L. ,  and K.  Steiner.    1978.   Variation  in aluminum  tolerance
     among six  genera of trees.   Forest Sci.  24(4):565-568.
McFee, W.W.   1980.   Sensitivity of soil regions  to  long-term  acid precipi-
     tation.     U.S.  EPA  Report  No.   600/3-80-013.   U.S.  EPA-Corvallis.
McFee,  W.  W. ,  J.   M.   Kelly  and R.  H.  Beck.    1976.   Acid  precipitation
     effects  on soil  p1-! and base saturation  of exchange sites,   pp. 725-
     735 In:   USDA Forest Service Gen. Tech. Report  NE-23.
McLeay, D.   J. ,  C.  C.  Walden  and  R.  J.  Munro.   1979.   Effect  of pH  on
     toxicity  of kraft pulp  and paper mill effluent to salmonid  fish  in
     fresh and  seawater. Water Res. 13(3):249-254.
Menendez,   R.   1976.   Chronic effects of reduced pH on brook  trout (Salve-
     linus fontinalis).  J.  Fish Res. Board Can.   33(1):118-123.
Miller, P.R.    1980.  Proceedings of Symposium  on Effects of Air Pollutants
     on Mediterranean and Temperate Forest Ecosystems, Riverside,  Califor-.
     nia,  June  22-27, 1980.   Pacific Southwest Forest and Range Experiment
     Station,  Berkeley, California.  256 pp.
Mogren, E.W.,  and K.P.  Dolph. 1972. Prediction of  site index  of lodge-pole
     pine from  selected environmental factors.   Forest Sci. 18(4):314-315.
Mount, D.  I.    1966.  The effect of total hardnpss  and pH on acute toxicity
     of zinc  to  fish.   Air Water Pollut. Int.  J. 10:49-56.
Mount,  D.  I.,  and  C.   E. Stephan.   1967.   A  method for establishing  ac-
     ceptable  toxicant  limits  for  fish  malathion  and  the  butoxyethanol
     ester of  2,4-D. Trans.  Amer. Fish. Soc.    92:185-193.
Muniz,  I.   P., and  il.  Leivestad.   1980.   Toxic effects of aluminum on  the
     brown trout, Sal mo trutta.   pp. 320-321 In:   D.  Drablrfs  and  A.  Tollan
     (eds.)  Ecological  Impact of  Acid Precipitation.   Proceedings  of  an
     international  conference,  Sandefjord,   Norway,  March  11-14,  1980.
     SNSF Project,  Oslo, Norway.
Munns,  D.N.   1965.   Soil  acidity and growth  of a  legume.  II.   Reactions
     of aluminum and  phosphate  in solution and  effects  of aluminum,  phos-
     phate,  calcium,  and pH  on  Medicago  satlva L. and Trifolium  subter-
     raneum  L.  in  solution  culture.   Aust.  J.  Agric.  Res. 16:743-755.
Norton, S.  A.   1976.   Changes  in  chemical  processes  in  soils  caused  by
     acid  precipitation.    pp. 711-724  In:   USDA  Forest  Service  General
     Tech. Report NE-23.
Nriagu, J.  L. , and R.  D. Coker.   1980.   Trace metals in humic and fulvic
     acids from Lake  Ontario sediments. Environ.  Sci. Technol. 14(4):443-
     446.                                                  '
Overrein,   L.N., H.M.   Seip,  and  A.  Tollon^   1980.   Acid precipitation  --
     effects   on   forest  and  fish:   Final    Report  of  the   SNSF-Project
     1972-1980.  Res. Report 1980, Blindern, Oslo,  Norway.  175 pp.

Packer, R. K. ,  and W.  A. Dunson.   1970.   Effects of low environmental pH
     on  blood  pH   and   sodium  balance  of  brook  trout.   J.  Exp.  Zqol.
Peltier, W.   1978.   Mechods  for measuring the acute toxicity of effluents
     to  aquatic  organisms.    EPA  600/4-78-012.    U.S.  EPA,  Cincinnati.
     52 pps.
Petersen, L.   1980.  Sensitivity of different soils to acid precipitation.
     pp. 573-577 In: T.C.  Hutchinson and M. Havas  (eds.)  Effects  of Acid
     Precipitation  on  Terrestrial  Ecosystems.    Plenum  Press,  New York.
Rahel,   F.  J., and  J. J.  Magnuson.   1980.   Fish in naturally acidic lakes
     of  northern Wisconsin,  U.S.A.   pp. 334-335  In:   D.  Drablrfs  and A.
     Tollan  (eds.)  Ecological  Impact .of Acid Precipitation.  Proceedings
     of  an  international  conference,   Sandefjord,  Norway,  March  11-14,
     1980.  SNSF Project, Oslo, Norway.
Rapport, D.   1981.   Environmental  transformation  in the lower Great Lakes
     aquatic ecosystem:  A case study of the stress-response  environmental
     statistical system.   Discussion paper for the senior advisors to  the
     Economic Commission for Europe.  Document No.  CES/455.
Rapport, D. ,  and A.  Friend.   1979.  Towards a Comprehensive  Framework  for
     Environmental  Statistics:   A  Stress  Response Approach.  Publication
     No. 11-510.  Statistics Canada, Ottawa, Ontario, Canada.
Rapport,  D.  J. , and H.  A.  Regier. ' 1980.  An  ecological  approach to  en-
     vironmental information.  Ambio 9(1)22-27.
Rapport, D.  J.,  H.  A.  Regier, and C. Thorpe.  1980.  Diagnosis, prognosis
     and treatment  of  ecosystems under  stress.   In:  G.  Barret and Rosen-
     berg (eds.) Stress  Ecology, Wiley (in press).
Ratsch, H. C.  1974.  Heavy-metal accumulation in  soil and vegetation from
     smelter  emissions.   EPA-660/3-74-012.  U.S.  EPA,  Corvallis.   23  pp.
Reuss,   J.O.   1975.   Chemical-biological  relationships  relevant  to eco-
     logical  effects of  acid rainfall.   Report No.  EPA-650/3-75-032.  U.S.
     Environmental Protection Agency, Corvallis, Oregon.
Reuss,   J.   1976.   Chemical  and biological  relationships  relevant to  the
     effect  of  acid rainfall  on the  soil-plant  system,   pp. 791-813  In:
     Proc. of the  First  Int. Sym.  on  Acid Precip. and the  For. Eco. USOA
     For. Ser. Gen.  Tech. Rep. NE-23.
Reuss,   J.  1978.   Simulation of nutrient  loss  from soils  due to rainfall
     acidity. EPA 600/3-7o-053.
Richards,  N.  H. ,  and  E.  L.  Stone.  1964.  Application  of soil  survey  to
     planting site  selection:   an  example from  the  Allegheny uplands  of
     New York.   J. Forestry 62:475-480.
Runn,  P. ,  N.  Johansson and G. Milbrink.  1977.   Some effects of low pH on
     the  hatchability   of  eggs  of  perch,  Perca  fluviatilis  L.  Zoon.
Sa'uter,  S. ,  K.  S.  Buxton,  K.  J.  Macek,  S.  R.  Petrocelli.   1976.   Effects
     of exposure to  heavy metals on  selected freshwater fish:  Toxicity of
     copper,   cadmium,  chromium and  lead  to eggs  and  fry  of  seven fish
     species.  EPA 600/3-76-105. U.S. EPA, Ouluth,  Minn.  .75  pps.
Schindler, O.W., R.W. Newbury,  K.G.  Beaty, and P.  Campbell;   1976.  Natural
     water  and   chemical  budgets  for a  small  Precambriah  lake  basin  in
     central  Canada.  J.  Fish.  Res.  Board Can. 33:2526-2543.
Schnitzer, M.    1980.    Effect  of  low  pH  on  the  chemical  structure  and
     reactions  of  humic  substances,  pp. 203-222  In:  T.C.  Hutchinson  and
     M.  Havas, (eds.)  Effects  o_f  Acid  Precipitation cm  Terrestrial  Eco-
     systems. -Plenum Press, New York.

Schofield, C.  L.   1980.   Processes limiting fish populations  in acidified
     lakes,  pp. 345-356  In:   O.S.  Shriner, C.R. Richmond, and S. E.  Lind-
     berg  (eds.) Atmospheric  Sulfur Deposition:  Environmental Impact  and
     Health Effects.  Ann Arbor Science; Ann Arbor, Michigan.
Schofield,  C.L.    Personal  communication.   Dept.  of  Natural Resources,
     Cornell University, Ithaca, New York.
Schofiald,  R. ,  and A.  Taylor.   1955.   The measurement of soil pH.  Soil
     Sci. Soc. Amer. Proc.  19:164-167.
Seip, H.  M. , G. Abrahamsen,  N.  Christophersen, E. T.  Gjessing and A.  0.
     Stuanes.  1980.   Snow and  meltwater  chemistry  in mini-catchments.
     SNSF Report IR 53/80.  Norway.  51 pps.
Spehar,   R.  L.   1976.   Cadmium and zinc  toxicity  to flagfish, Jordanella
     floridae.  J.  Fish. Res.  Board Can.  33:1939-1945.
Sprague,  J.  B.   1969.    Measurement  of  pollutant toxicity  of fish:  I.
     Bioassay methods for acute toxicity.  Water Res.   3:793-821.
Sprague,  J.  B.   1970.   Measurement of  pollutant  toxicity  to fish:   II.
     Utilizing and applying bioassay results.   Water Res.  4:3-32.
Sprague,  J.  B.   1971.   Measurement of  pollutant  toxicity to fish:  III.
     Sublethal  effects  and "safe"  concentrations.   Water Res. 5:245-266.
Spurr,  S.  H.   1952.    pp.  299-320  In:   Forest Inventory.   Roland  Press
     Company, New York.
Spurr,  S.  H.  1956.   Soils in relation  to  site index curves,  pp.  80-85
     In:  Proc. Soc. Amer. For. Meeting, 1955.
Spurr,  S.  H.   1964.   pp.  126-146  In:    Forest  Ecology.   Ronald  Press
     Company, New York.
Turner,   R. ,  and J.  Clark.  1965.   Lime  potential  and  degree  base satura-
     tion of soils.  Soil Sci. 99(3):  194-199.
Tyler,  G.   1972.    Heavy  metals pollute  nature,  may reduce  productivity.
     Ambio 1(2):52-59.
U.S.  Environmental  Protection  Agency.   1972.    Proposed recommended bio-
     assay procedure for egg and fry stages of  freshwater  fish.  U.S. EPA,
     National Water Quality Laboratory, Ouluth, Minn.
U.S. Environmental  Protection  Agency.   1976.   Quality  criteria for water.
     U.S. EPA, Washington, D.C.  256 pp.
U.S.  Environmental  Protection  Agency.   1978. .National  Air Pollutant
     Emissions Estimates, 1940-1976.  EPA-450/1-78-003.
Van  Hook,  R.  I. ,  W. F. Harris and G.  S. Henderson.  1977.  Cadmium,  lead,
     and  zinc  distributions   and  cycling  in  a  mixed  deciduous   forest.
     Ambio 6(5):281-286.
Warner,   R.  E.   1967.   Bioassays  for  microchemical environmental  contami-
     nants  with  special reference to water  supplies.   Bull.   World Health
     Organ. 36:181-207.
Wiklander,  L.   1980.   The sensitivity of  soils   to  acid precipitation.
     pp. 523-567 In:   T.C.  Hutchinson and M. Havas (eds.) Effects of Acid
     Precipitation  on  Terrestrial  Ecosystems.   Plenum Press,  New  York.
Wollast,  R. ,  G.  Billen, and F.  T.  Mackenzie.   1975.    Behavior of mercury
     in  natural  systems  and  its  global  cycle.   pp.  145-166  In:  A.D.
     Mclntyre  and  C.F.  Mills  (eds.)  Ecological  Toxicology  Research:
     Effects  o Heavy  Metal  and Qrganohalogen Compounds.   Plenum Press,
     New York.

Wright, R. F., N. Conroy, W. T. Dickson, R. Harriman, A. Henriksen,  and  C.
     L.  Schofield.    1980.   Acidified  lake  districts   of  the  world:    A
     comparison  of  water chemistry of  lakes  in southern Norway',  southern
     Sweden, southwestern Scotland,  the Adirondack Mountains of New York,
     and  southeastern  Ontario,   pp.  377-379  In:  0. Drablefs and A.  Tollan
     (eds.)  Ecological  Impact of  Acid  Precipitation.   Proceedings of  an
     international  conference,  Sandefjord,  Norway,  March 11-14,  1980.
     SNSF Project, Oslo, Norway.
Wright, R.  F. ,  and  H.  Oovland.  1977.  Regional  surveys of the chemistry
     of  the snowpack  in Norway  late winter  1973,  1974,  1975  and 1976.
     SNSF Project, FR 12/77.  29 pp.
Wright, R.  F.,  H.  Dovland,-C.  Lysholm, and B.  Wingard.   1978.   Inputs and
     outputs of  water  and  major ions at  9 catchments in southern  Norway,
     July 1974-1975. SNSF Report TN 39/78.  60  pp.
Wright,  R.   F. ,  and A.  Henriksen.   1978.   Chemistry  of  small  Norwegian
     lakes,  with   special   reference  to  acid  precipitation.    Limnol.
     Oceanogr. 23:487-498.
Wright,  R.   F. ,  and Snekvik.  1978.   Acid precipitation:   chemistry and
     fish   populations   in   700   lakes  in   southernmost   Norway.   Verh. ..
     Internat. Verein.  Limnol.  20:765775.
Zimmerman,  A.P.,  and  H.H.  Harvey.   1979.   Sensitivity  to acidification  of
     waters  of Ontario and neighboring states.   Final  Report for  Ontario
     Hydro.   University of  Toronto, Toronto,  Ontario,  Canada.   118 pp.

Page Intentionally Blank