ornl
ORNL/TM-9258
 OAK RIDGE
 NATIONAL
 LABORATORY
                     Predicting Soil and Water Acidification

                          Proceedings of a Workshop
                                  Dale W. Johnson
                                  Ingvar S. Nilsson
                                  John O. Reuss
                                  Hans Martin Seip
                                  Robert S. Turner
OPERATED BY
MARTIN MARIETTA ENERGY SYSTEMS, INC.
FOR THE UNITED STATES
DEPARTMENT OF ENERGY

-------
     Printed in the United States of America. Available from
             National Technical  Information Service
                 U.S. Department of Commerce
       5285 Port Royal Road, Springfield, Virginia 22161
     NTIS price codes—Printed Copy: A04; Microfiche A01
This report was prepared as an account of work sponsored by an agency of the
United States Government. Neither the U nited StatesGovernment nor any agency
thereof, nor any of their employees, makes any warranty, express or implied, or
assumes any legal liability or responsibility for the accuracy, completeness, or
usefulness of  any information, apparatus, product, or process disclosed, or
represents that its use would not infringe privately owned rights. Reference herein
to any specific commercial product, process, or service by trade name, trademark,
manufacturer,  or otherwise, does not necessarily  constitute or  imply  its
endorsement, recommendation, or  favoring by the United States Government or
any agency thereof. The views and  opinions of authors expressed herein do not
necessarily state or reflect those of the United States Government or any agency
thereof.

-------
                                           EPA ERL-Corvallis Library
                                                  00003822
                                                ORNL/TM-9258

                                               430R85001
        ENVIRONMENTAL SCIENCES DIVISION
PREDICTING SOIL AND WATER ACIDIFICATION

           Proceedings of a Workshop
               Dale W. Johnson
                   Editor

            With contributions from
               Ingvar S. Nilsson
                John O. Reuss
               Hans Martin Seip
               Robert S. Turner
     A Workshop held in Knoxville, Tennessee
              March 26-29, 1984
                  Funded by
         The National Acid Precipitation
           Assessment Program by the
       U.S. Environmental Protection Agency
                  Hosted by
         Oak Ridge National Laboratory
             Oak Ridge, Tennessee
                                        LIBRARY
                             U* ENVIRONMENTAL PROTECTION AGENCY
                             OORVAUJ6 ENVIRONMENTAL RESEAHCMUffl
                                    200 SW 39IH OTWET^
                                 CCRVAl US OREGON 97339
      Date Published - January 1985
    OAK RIDGE NATIONAL LABORATORY
           Oak Ridge, Tennessee 37831
                 operated by
 MARTIN MARIETTA ENERGY SYSTEMS, Inc.
                    for the
       U.S. DEPARTMENT OF ENERGY
     under Contract No. DE-AC05-84OR21400

-------
                                  CONTENTS

                                                                          Page

WORKSHOP SUMMARY	   1

INTRODUCTION	   2

SOIL ACIDIFICATION	   3
  KEYNOTE ADDRESS:  Modeling the Effect of Acid Deposition on Soil
                         and Water Acidification (John O. Reuss)	   3
    Capacity-Intensity Concepts	   3
    The CO2-HCO3~ System	   4
    Soil-Solution Equilibria	   5
    Conclusions	  10
  DISCUSSION	  11
    Is Organic Sulfur Accumulation Important?	  11
    Is Sulfate Adsorption Reversible?	  12
    Can We Forecast Nitrogen Saturation?	  12
    What Regulates Nitrification?	  13
    Can Acidification Due to Vegetation Uptake and Humus
      Formation Be Quantified?	  13
    Are Natural Processes Self-Limiting?  	  14
    Is Weathering Stimulated by Acidification?	  14
    How Does Aluminum Mobilization Affect Vegetation Uptake?	  14
    Are Limiting Cations Conserved Even  in the Presence of
      Acid Precipitation?	  14

WATER ACIDIFICATION	  15
  KEYNOTE ADDRESS:  How Are Waters Acidified? (Hans Martin Seip)	  15
    Introduction	  15
    Aquatic Buffer Systems	  15
    Conclusions Reached at the Sandefjord Meeting	  15
    Soil-Water Interactions in the Catchment	  16
    Hydrology	  16
    Direct Effects	  17
    Mobile Anion Concept	  17
    Weathering and Cation Exchange	  18
    Sulfur Oxidation 	  19
    Prediction of Acidification	  19
    External and Internal Sources of Soil Acidification	  22
    Historic Trends in Lake Acidification and Relation to
      Activities in the Catchment	  23

                                       iii

-------
                              IV

   New Conclusions and Areas of Uncertainty 	 23
  DISCUSSION	 24

EXAMINATION OF BIRKENES AND GARDSJON DATA SETS	 26
  INTRODUCTION	 26

  ANALYSIS BY D. W. JOHNSON 	 26
  ANALYSIS BY I. S. NILSSON	 31

SOIL SENSITIVITY CRITERIA	 34

ACKNOWLEDGMENTS	 37

REFERENCES	 38

LIST OF ATTENDEES  	 42

-------
                              WORKSHOP SUMMARY
                JOHNSON, Dale W. (Editor). 1985. Predicting Soil and Water
                 Acidification. Proceedings of a Workshop. ORNL/TM-9258.
                Oak Ridge National Laboratory, Oak Ridge, Tennessee. 56 pp.
    A three-day workshop was held at the Hilton Hotel in Knoxville, Tennessee on March
27-29, 1984, preceded by a one-day tour of sites at or near ORNL. Funding for the work-
shop was provided by the National  Acid Precipitation  Assessment Program by the U.S.
Environmental Protection Agency.  One of the goals of this workshop was to develop a con-
sensus among the participants as to sensitivity criteria for acid deposition effects on both
soils and surface waters. As the meeting evolved, the workshop participants  spent most of
their time in a very productive discussion of important processes and  hypotheses regarding
soil and water acidification, primarily from the theoretical standpoint, using empirical data
to illustrate specific points. Only in the afternoon of the  last day were sensitivity criteria as
such discussed, but all of the preceding discussions clearly related to this issue as well. The
workshop discussions, including sensitivity criteria, are summarized in this document.
    A major  highlight of this workshop was a meeting of minds among aquatic and  terres-
trial scientists as to important mechanisms for surface water acidification. This paved the
way for  assessment activities, probably  in association with  modeling  efforts  such  as those
of John  Reuss, Nils Christophersen, and Jack Cosby. No such consensus or knowledge is
available for  forest effects, however,  because the important  mechanisms of forest  effects
are not known. A concensus was reached as to appropriate sensitivity criteria for soil acidi-
fication  and  aluminum mobilization, but there  was no consensus as  to  whether these
processes in themselves are responsible for reported  widespread forest dieback and decline.
Thus,  assigning forest effects sensitivity criteria at  this time would be premature at  best
and very possibly misleading.
    Two major areas of research were identified as most in need of further research: nitro-
gen (N) cycling (especially as affected by excess N  inputs) and soil weathering. Nitrogen
is emerging as a key nutrient in terms of not only the beneficial, effects of nitrogen deposi-
tion to N-deficient forests, but also the possible detrimental effects of excess nitrogen depo-
sition  on nutrient  imbalances and on nitrification and associated  soil  acidification. Soil-
weathering rate remains one of the least understood of the master variables controlling soil
acidification, even after many years of recognition of its great importance.
    A third in this series of workshops is being planned and will be hosted by Peter Dillon
in Ontario in  1985.

                                                                       Dale W.  Johnson

-------
                                   INTRODUCTION

                                   (Dale W. Johnson)

    One of the concerns relative to acid deposition is  its effect on soil and surface water
acidification. However,  controversy  surrounds  this issue because of  a gap  in the under-
standing of the processes whereby acidification occurs. In view of this problem, Prof. Folke
Andersson of the  Department  of  Ecology  and  Environmental Research,  The  Swedish
University of Agricultural  Sciences,  called an informal three-day workshop in Uppsala,
Sweden, in June 1982. The title of the workshop was  "Simulation Models of Water Aci-
dification." The  meeting was preceded  by a  two-day  tour of research sites in southern
Sweden. Participation was by invitation  only and was "restricted to scientists with experi-
ence in the field" in an effort to keep the group small and the discussions as meaningful
and as  factually  accurate as possible. Each participant  was asked to solicit travel expenses
from  his own sponsoring agency. At the conclusion of that workshop, it was agreed  that
another meeting  of the same type within a year or two would be productive. To that end,
an ad hoc organization was appointed by Prof. Andersson as follows:

Chairman                       Dale W. Johnson

Group Members and Areas of Responsibility
     Water acidification         Nils Christophersen
                                Hans Martin Seip
     Groundwater acidification   Gunnar Jacks
     Soil acidification           Dale W. Johnson
                                Ingvar S. Nilsson

Our objectives for the  1984 workshop were to (1) critically evaluate data sets from selected sites to
determine, insofar as possible, the degree to which  soils and surface waters  have been acidified by
acid  deposition; (2) review and evaluate models of soil and surface water acidification; and (3)
arrive at a consensus as to appropriate sensitivity criteria for soil and surface water acidification.

-------
                                    SOIL ACIDIFICATION

    The first day of the workshop was devoted primarily to discussions on soil acidification and
related  processes, although the  subjects of soil-water  interactions  and  water acidification  were
touched upon. The session began with a keynote address by John Reuss (summarized in the follow-
ing section), followed by a group discussion of several key questions relating to soil acidification by
both natural and anthropogenic mechanisms.

KEYNOTE ADDRESS:  Modeling  the Effect of Acid Deposition on Soil and  Water Acidification
(J.  O. Reuss, Colorado State University, Fort Collins, Colorado)

    The acidification  of soils  and  waters due to acid deposition involves a  number of  complex
processes. As might be expected in  a situation where the processes are not generally well understood
and the issue is of major economic importance, considerable controversy has resulted. Responsible
scientists differ substantially in  their interpretation  of the  available information,  and some have
questioned whether significant acidification of either soils or waters could take place as a  result of
the levels of deposition currently encountered (e.g., Rosenqvist et al. 1980, Krug and Frink 1983).
    In spite of the overall complexities, we found that a fairly simple model consisting of chemical
relationships  that are reasonably well known and accepted can be very useful in understanding the
probable responses of soil and water systems to acid deposition. Because of the interactions among
the processes described by these  relationships,  it is essential that they be considered simultaneously
and not applied in a piecemeal manner.
    Due to time limitations, the material that can be covered here is limited. After a  brief discus-
sion of the capacity-intensity concepts, I will discuss the CO2-HCO3~  equilibrium and soil-solution
equilibria, as well as the  effects of acid deposition on soil solution and surface water chemistry.

Capacity-Intensity Concepts

     Much of the literature concerning the effect of acid deposition  on soils and the soil-mediated
effects on surface waters has focused on the capacity of the soil to  adsorb the proton input. Con-
sider, for example, a system receiving annually 1 m of pH 4.2 rainfall and an equal amount of acid-
ity as dry deposition, for a total  of 0.125 eq  H+/m2. If this input falls on 30 cm of soil with a cat-
ion exchange capacity (CEC) of 0.15 eq/kg,  a bulk density of 1.2, and a 15% base saturation, a
comparison of the pool sizes scaled to the annual input of H+ would be about as follows:

                                   Annual H+ input — 1
                                   Exchangeable base —  65
                                   Exchange acidity — 365

     Obviously, it would require many decades (centuries in  deeper soils) for this input  to  bring
about a significant change in total soil acidity. A significant change in the exchangeable base cation

-------
pool due to replacement by acid inputs might occur in a few decades, but this reduction would be
mitigated by the release of these  cations by weathering processes. The reduction will  be further
mitigated in some acid soils where the replacement efficiency of the input acidity for exchangeable
bases may be substantially less than 1.0.

     From considerations such as these, many scientists concluded that soil effects due to acid depo-
sition or soil-mediated effects on surface water are likely to occur only on soils with very low CEC
and, therefore, on soils that are highly susceptible to changes in base saturation due to cation loss.
However, the capacity effects due to these changes in pool  size are not the only manifestation of
acid deposition inputs. Changes in soil solution composition  that may have a profound effect both
on terrestrial ecosystems and on surface  and groundwater  quality may occur without  significant
changes in these  pool sizes. It is on these so-called intensity factors that we will focus  during the
remainder of the presentation.

The CO2-HCO3~ System

     While the overall importance of the CO2-HCO3~  equilibrium in determining the properties of
both soil solutions and surface waters is well known, some of the implications,  particularly the role
of CO2 partial pressure (pCO2) in determining the alkalinity of the  drainage water,  are  often
neglected. In acid soils we can neglect the CO32~ ion so that the reaction  can be written as

   CO2  +  H2O ^ H+  + HCO3~  .                                                    (1)

From this reaction we obtain the equilibrium expression,

   (H+)  (HC03-)  =  Kc  • pC02  ,                                                      (2)

where the material in parentheses refers to  activities (or partial pressure in the case of pCO2). If
the concentrations are in microequivalents per liter (for our purposes, concentrations and activities
may be taken as equal) and the CO2 in percent, KC will be about 150. From Eq. (2) we find that at
0.03, 0.3, and 3% CO2 the product [(H+) (HCO3~)] is equal to 4.5, 45, arid  450, respectively. In
pure water a tenfold increase in CO2 results in both H+ and HCO3~ increasing by a factor of VTO
(i.e., 3.16) (Table 1). As a  result, increasing CO2 from 0.03% (near atmospheric) to 0.3% decreases
pH from 5.67  to 5.17, while at 3% the pH will be 4.67. In pure water the H+ and HCO3~ concen-
trations are equal and the alkalinity, defined  here as

   alkalinity = (HCOD  + (OFT)  - (H+)  ,                                         (3)

remains zero at all CO2 levels.
     Acid soils are buffered by internal processes, and pH changes with varying CO2 partial pres-
sures are  usually small. Because the product (H+)  • (HCO3~) must increase with the CO2 partial
pressure (Eq. 3.) and (H+) is fixed by soil processes, the response to changing CO2 levels in the soil
is mostly  in the HCO3~ concentration, and thus is reflected in  the alkalinity  of the soil solutions.
This is  illustrated in Table  1 (lines 4-12). For example,  at pH  5.67 and 0.03% CO2, the H+ and

-------
                      Table 1. The effect of pCO,on pH, H+, and HCO,~ In pure waters
                           and in soil solutions buffered by Internal soil processes

                                                  H+   HCOj-   Alkalinity

                      Line   CO,   K(CO2),   pH
Pure water
\
2
3
0.03
0.30
3.0
Soil solutions
4
5
6
7
8
9
10
11
12
0.03
0.30
3.0
0.03
0.30
3.0
0.03
0.30
3.0
4.5
45
450
(buffered)
4.5
45
450
4.5
45
450
4.5
45
450
5.67
5.17
4.67

5.67
5.67
5.67
5.17
5.17
5.17
4.67
4.67
4.67
2.1
6.7
21.2

2.1
2.1
2.1
6.7
6.7
6.7
21.2
21.2
21.2
2.1
6.7
21.2

2.1
21.2
212
0.67
6.7
67.2
0.21
2.1
21.2
0
0
0

0
19.1
210
-6.0
0
60.5
-21.0
-19.1
0
HCO3  concentrations are both 2.1 iteq/L, (the same as in water) and the alkalinity is zero. How-
ever,  increasing CO2 to  0.3  and 3%  increases the alkalinity of the soil solution to 19.1 and  210
/xeq/L, respectively. Increasing CO2 would have no effect on the alkalinity of water not buffered by
soil processes.
    A soil solution at pH S.67 will have positive  alkalinity if the CO: content, of the soil gases is
above 0.03%. At a solution pH of 5.17 the alkalinity will  be positive above 0.3% CO2, and at pH
4.67 it will be  positive above 3%. In this alkalinity-generating process, equal amounts of H+  and
HCO3~ are formed (Eq.  1),  but the HCO3~ concentration increases while the H+  is held constant
by soil buffering. We may think of the H+ as being consumed in the dissolution of soil minerals,
bringing A13+ into solution. The Al3"1" then displaces cations such as Ca2+, Mg2+, and  K+ from the
ion exchange complex. The net effect is the formation of bicarbonates of these cations  (i.e., alkalin-
ity).
    The changes  in alkalinity brought about in the soil  solution by variations  in soil pCO2  can
drastically affect the pH of the drainage  water. The  relationship between  alkalinity and pH  in
water that is not in contact with soil processes, as derived from Eqs. (1) and (2), is shown in Fig. 1.
The pH 5.17 soil solution (lines 7-9, Table 1) at CO2 levels of 0.03,  0.3, and 3% has  an alkalinity
of  —6, 0,  and 60.5 M^q/L,  respectively.  When this  water equilibrates with  atmospheric CO2
(0.03%), the pH will be 5.17 if the soil CO2 is 0.03%, 5.67 at 0.3% CO2, and  7.1  at 3% CO2. Thus,
at a  soil  solution pH of 5.17 the drainage water pH  would vary by nearly 1.9  units simply by
varying soil CO2 over a range that may commonly be found in the soil.

Soil-Solution Equilibria

    When an acid forest soil  is subjected to  acid deposition, the concentration of the strong acid
anion (SO42~, and in some cases NO3~) will increase. In most acid soils the natural concentrations
of these anions are very low;  a typical value for SC>42~ might be 20 peq/L. The HCO3~ is also low
due to the nature of the  CO2-HCO3~-H+ equilibria described above (e.g., at  pH 4.67  and 3% CO2
the concentration of HCO3~ will be  about 20 jteq/L). While organic anions can be an important
constituent of soil solutions and drainage waters, one of the characteristics of acid forest soils is low

-------
                 8
                               1
                 -100
-50
                                                       ORNL-OWG 84-1616
                                                     I
                                                            „.	0.3
                                                         •*
                                                                  0.03%
                                                                 — 3.0
              I
I
I
50
          100
         150
                                    ALKALINITY (^eqL")
             Figure 1. The relationship between the pH and alkalinity of water at 0.03, 0.3, and 3% CO2.
anion concentration. When such soils are subjected to acid deposition the increase in SO42  concen-
tration can be very significant, typically in the range of 100 to 300 neq/L. (The time required for
this increase in SO42~ can vary markedly; high-sulfate-adsorbing soils may not come into SO42~
equilibrium  for decades.) This increase  in anion concentration can have a very important influence
on the cation composition.  Not only will the total cation concentration in solution increase to main-
tain charge balance, but the relative amounts will also change. The most important of these changes
involves the H+,  Ca2+, and A13+ ions (we will use Ca  as a proxy for both Ca and Mg in this dis-
cussion). These responses can be described by the use of three  relationships familiar to many of
you, which in the interest of time I will present without discussing their derivation or their limita-
tions. The first is the relationship between Al3"1" and H+:
           - Ka(H+)3  ,
                                                              (4)
i.e., the activity of the A13+ ion is proportional to the third power of the H+ activity or, in negative
logarithm form,
    3  pH  - pAl
                                                              (5)
Values of KA in the soil may range from 7.0 or less to near 10.  Useful reference points are the
values of 8.04 and 9.66 given by Lindsay (1979) for gibbsite and amorphous A1(OH)3, respectively.
A plot of this relationship for a representative set of KA values is shown in Fig. 2. The implications

-------
                1000
                 800
                 600
                                                           ORNL-DWG 84-1612
              0)
             fO
             <  400
                 200
                                                                    I
                                 25
50
  75
L'1
                      100
125
   Figure 2. The reUtkwshlp betweca A|3+ ud H+,
coefficients are assumed to be 0.96 for H   and 0.70 for Al
3+
      3 pH - pAl values In the ruse of 8.0 to 9.5. The activity
of the shape of these curves are twofold.  First, if the value of KA is high the system is highly buf-
fered.  For example, if KA is 9.5 the H+ concentration  is unlikely to exceed about 30 j&q/L (pH
4.5),  as  any further  H+ will only result in increased A13+ in solution. Second, as shown  by the
upward  curvature  of the lines,  increasing anion  concentrations will  result in increases in  total
cations and in  the  proportion of A13+ relative to H+. This consequence can be stated as a general
principle, i.e., increasing solution concentration will increase the proportion of the cations with the
higher valence.
     The second  relationship is that between Ca2+  and A13+,  which I describe  below using the
equation of Gaines and Thomas (1953):
    (A13+)2
    (Ca2+)
        2+\3
                                               (6)
The parentheses denote activities in the solution phase, Kg is the ion exchange constant that reflects
the thermodynamic properties of the exchanger, and CaX and A1X are the fractions of the total
exchange sites occupied by the Ca+ and A13+ ions, respectively. Equation (6) states that the activ-
ity of A13+ in solution is proportional to the 3/2 power of the Ca2+ activity, i.e.,
           =  KB(Ca2+)3/2
                                               (7)

-------
                                                8

The proportionality constant KB  is a function of the degree of saturation of Ca and Al and of the
constant Kb (Eq.  6.)- These relationships tell us that if the solution concentration  increases due to
increased sulfate  or  nitrate from acid deposition,  the  Al3+:Ca2+ ratio in  solution will increase so
that the activity  of  A13+  remains proportional to the 3/2 power of the Ca2+ activity  (Fig. 2).
Again, an increase in solution concentration results in a shift toward the ion of higher valence. This
is an intensity response and occurs immediately as the concentration changes.
     The lines  in  the plot  of A13+ vs Ca2+ (Fig.  3) also curve upward, due to an increase in the
fraction of A13+ as the total concentration increases. For the Kg value used to construct Fig. 2, this
curvature increases markedly as the fraction of exchange  sites  occupied  by Ca is reduced below
about 0.2. The effect of increasing values of Kg is  similar to that of increasing Ca saturation. Thus,
increasing Kg has the effect of decreasing the Ca saturation at which the change in the Al3+:Ca2+
ratio in solution due to higher concentration becomes significant.
                                                            'ORNL-DWG 84-1615
                                                 KQ  "0.5
                     500
                                                                              600
    Figure 3. The relationship between A13+ and Ca2+ In soil solution, with the fraction of the exchange sites occupied by
 Ca2+ in the range of 0.05 to 0.20. Log of the Gaines-Thomas exchange coefficient is 0.05 and the activities of Ca   and
 AI3+ are 0.85 and 0.70, respectively.
     While the effect of an increase in solution concentration due to acid deposition will be a higher
proportion of A13+ in solution relative to Ca2+ (actually relative to all mono- or divalent cations),
the total  amount of all ions in solution will increase. Thus, the export of Ca2+ is accelerated, even
though the Al3+:Ca2+ ratio is reduced. If the exchangeable Ca pool is depleted due to increased Ca
loss over  time, the Ca saturation will be reduced and the proportion of A13+  in solution will be fur-
ther increased.

-------
    The third relationship is that between Ca2"1" and H+. Combining Eqs. (4) and (6) we obtain
the rather formidable-appearing Eq. (8),
   (Ca2+)
K2/3
                          CaX
                                             (8)
                   (A1X)2/3
This is more familiar to soil scientists in the form,
    pH  - 1/2  pCa
                                                                            (9)
where KL is the well-known "lime potential." Equation (8) simply states that in the soil solution the
Ca2+ activity is proportional to the square of the H+ activity, and that the proportionality is deter-
mined by the A13+-H+ proportionality constant (Ka), the ion exchange constant (Kg), and the frac-
tion of exchange  sites occupied by Ca and  Al. Figure 4 shows the H+-Ca2+  relationship in  soil
solutions of varying lime potential.  The response of the system to increased solution concentrations
from acid deposition  inputs is again to increase  the proportion of  Ca2"1"  (i.e., the  ion  of higher
valence) (Fig. 4). This figure serves to illustrate the concept of ion exchange buffering. For exam-
ple, at a lime potential of 3.00, a soil solution pH of 4.3 could only be attained if the Ca concentra-
tion were about 1000 jieq/L. As Ca saturation is reduced the lime potential decreases, allowing pH
                                                           ORNL-DWG 84-1614
                   4000
                0)
               •f
               CM
                o
                   2000
                                                          1
                                   325    3.00
                                                            275
                                    25
50        75
 H + ^eq LH
                                                     100
125
     Figure 4. Hie relationship between Ca2+ and H+ in soU solution for lime potential values in the range of 2.25 to 3.25.
 Calculations assume activity coefficients of 0.96 and 0.8S for H  and Ca  , respectively.

-------
                                               10

to decrease.  Eventually, the pH decrease will be buffered by bringing A13+ into solution  (Fig. 2)
and the pH at which this A13+ buffering occurs will be determined by KA.
    One of the problems  with using Eq. (6) is that  the ion exchange constant (Kg)  is difficult to
measure. However, if the  solubility of A13+ (Ka or KA), the degree of Ca and Al saturation, and
the lime potential are known, Kg can be calculated by  rearranging Eq. (8).   •

Conclusions

    It is a relatively simple matter to  combine the relationships in  Eqs. (2), (4), and (6) with the
charge balance requirement to obtain a set of simultaneous equations such that if the various con-
stants and the Ca and Al saturation of the exchange complex  are  known, the composition of the
major cations in solution can be calculated for any  combination  of anion strength and pCO2. Addi-
tional ions can be included using similar relationships. Several current models are based on similar
concepts (Reuss 1980, Christophersen et al. 1982, Chen et al. 1983). Calculations of this type pro-
vide a quantification of the "salt effect" mechanism proposed by  Seip (1980). The results  indicate
that in some  low base saturation soils the increase in the anion concentration in solution due to acid
deposition inputs may cause a significant increase in the A13+ concentration in the soil solution.
    The increase in Al can, under certain conditions,  have profound implications for both the plant
component of the ecosystem and the quality of the drainage water. The fact that A13+ is  toxic.to
many plants  is well known. Unfortunately, the levels at which various forest ecosystems will show
significant effects are not  well established. The A13+ concentration of the drainage waters is also
critical. Within the pH range  critical to many aquatic species, A13+ effectively acts as an  acid.
Thus, the alkalinity of the  water (Eq. 3) can be reduced by either H+ or A13+. Soil solutions with a
low but positive alkalinity may develop negative  alkalinity, resulting in a  substantial  depression in
pH of the drainage water.  This depression may occur as a direct result of  the acid deposition input
without the necessity for a reduction in base cation status. In other cases, A13+ will not be signifi-
cantly increased unless base saturation is reduced by cation export. Soils  with a large reservoir of
base cations, or those that release substantial cations as a result of mineral weathering processes,
may not develop increased  A13+ concentrations, even under prolonged exposure to acid deposition.

-------
                                        DISCUSSION

    Dale Johnson presented a list of key processes relating to soil acidification as well as a series of
questions that were addressed during the day (Table 2). For the purposes of this report, the discus-
sions are summarized according to their relationship to the questions listed in Table 2.

                 Table 2. Key processes and questions relating to soil acidification
                                 (presented for group discussion)

                                                            Questions
         Inputs
           Atmospheric sulfur

           Atmospheric nitrogen

           Naturally produced acids
         Responses
           Leaching
           Weathering
           Vegetation uptake and cycling
Is organic sulfur accumulation important?
Is sulfate adsorption reversible?
Can we forecast nitrogen saturation?
What regulates nitrification?
Can the acidification effects of vegetation
  uptake and humus formation be quantified?
Are natural processes self-limiting?
Is weathering stimulated by acid input?
How does aluminum mobilization affect
   vegetation uptake?
Are limiting cations conserved even with
   acid deposition?
Is Organic Sulfur Accumulation Important?

     Johnson initiated this discussion by stating that although sulfur is in the organic form in most
soils, SO42~ adsorption to Fe + Al oxides (or the lack of it) appears to be the dominant process of
SO42~  accumulation  in  areas receiving  elevated  atmospheric  sulfur  inputs.  He supported  this
hypothesis with two observations: First, forest ecosystems on  Spodosols in the northeastern United
States generally do not show a net SO42~ retention (iie., SO42~ outputs > inputs),  whereas forest
ecosystems on Ultisols in the southeastern United States generally do retain SO42~. This is consist-
ent  with  current  information on the relative abilities of these  two soil  orders  to  adsorb sulfate
(Johnson  and Todd 1983), although this does not imply that all Ultisols adsorb more SO42~ than
all Spodosols. He also noted that the picture is not at all clear for Inceptisols, where SO42~ adsorp-
tion properties appear to vary enormously.
     Second, lysimeter studies in Ultisols indicate little SO42~ retention in  organic-matter-rich sur-
face horizons but considerable retention in Fe  + Al-oxides-rich subsoils (Johnson et al. 1981; Kelly,
                                               11

-------
                                              12

in press; Richter et al.  1983). He noted, however, that SO42~ mineralization from surface soils on
Walker Branch Watershed, Tennessee, was very large (Johnson et al. 1982) and thus organic sulfur
transformations must be important and cannot be neglected.
    Johnson then asked Mark David, who has considerable experience in organic and inorganic sul-
fur transformations  in  Spodosols, for a  rejoinder.  David reviewed the sulfur studies he  and his
colleagues have conducted (David et al. 1983, 1984) which show large rates of sulfur incorporation
and turnover into organic matter. He basically concurred, however, that  net SO42~ retention in for-
est soils appeared to  be  governed by adsorption properties.

Is Sulfate Adsorption Reversible?

    The reversibility of SO42~  adsorption  and incorporation into organic matter  was discussed
without definitive conclusions, except that variability would no doubt be large.
    Egbert Matzner recounted studies at Soiling, West Germany, which indicate that the precipita-
tion of jurbanite (A1OHSO4) at  first caused net SO42~ retention in soils at that site.  Later, how-
ever,  as soil solution pH declined and jurbanite became unstable, soils began showing a net SO42~
and A13+  release. Clearly,  jurbanite formation  is a very  important consideration for soil SO42~
interactions that  can lead to patterns quite different from SO42~ adsorption, that is,  SO42~ desorp-
tion would not  be  expected to  occur  until solution  SO42~  concentrations drop or solution  pH
declines.

Can We Forecast Nitrogen Saturation?
    Helga Van  Miegroet and Jan Mulder were asked  to briefly review studies of nitrogen fluxes in
red alder (Alnus rubra) stands  in a relatively  unpolluted area of Washington  State, USA, and
coniferous  and deciduous stands  in The Netherlands. Red alder is a nitrogen-fixing  species, and it
appears to have  caused excess  nitrogen accumulation  and associated high rates of NO3~ leaching
after  40 years of site occupation. The net H+ generation from nitrification and nitrate leaching in
this stand is nearly 4 keq-ha~'-year~' (Van Miegroet and Cole, in press), a value quite compara-
ble to estimated  anthropogenic H+ input at  Soiling, West Germany (Ulrich 1980). Surface soils in
the red alder stand are  O.S pH units lower than those in  an adjacent Douglas-fir (Pseudotsuga men-
ziesii) stand, and according to calculations by John Reuss there is evidence of A13+ mobilization
during seasonal  peaks in NO3~ concentration (up to 2000 ^eq/L). Nonetheless, Dale Cole pointed
out that the stand is quite healthy for its age and there is every reason  to believe that  natural suc-
cession to conifers [Douglas-fir, western hemlock (Tsuga heterophylla)] will proceed normally.
    Jan Mulder then reviewed the work he and  his colleagues conducted in forests of The Nether-
lands receiving high  rates of nitrogen and sulfur inputs primarily as gaseous (NH3 and SO2) deposi-
tion to forest canopies.  These gaseous inputs result in large inputs of (NH4)2SO4 to the forest floor
in throughfall (Van Breemen et al. 1983). The high  rates of NH4+ stimulate  nitrification in the
soil, and the net result  is large exports  of soil cations, with both SO42~  and NO3~ leaching. Coin-
cidentally, the magnitude of the leaching rates in these forests  was similar to that in the natural red
alder stand.
    Egbert  Matzner   reviewed  the nitrogen  situation  at   Soiling,  where  nitrification  pushes
(presumed to be caused by favorable conditions  for nitrification) during warm, dry summers cause
A13+  mobilization and  damage to tree  roots. By comparing different sites with varying degrees of
soil acidification, it  was concluded that the  high base saturation of the exchange sites reduces the

-------
                                              13

risk of root damage following acidification pushes. Reduction of base saturation by SO42~ leaching
therefore leads to increasing root damage by acidification pushes.
    None  of these investigators could provide a  framework  for  forecasting nitrogen  saturation,
although each knew at least the approximate magnitude of nitrogen inputs at their N-saturated
sites (~80 kg-ha~'-year~' at the red alder site, 65 kg-ha"1-year"1 at The Netherland sites, and
40 kg-ha~'-year~' at the Soiling site). GOran Agren proposed  a model for determining nitrogen
saturation based on his nitrogen productivity model (Agren 1983). The model relies on foliar nitro-
gen content (which varies with foliage biomass as well as N concentration) as a index of degree of
nitrogen saturation. Assuming 20% of incoming nitrogen is taken  up by trees  (based on fertilizer
studies), Agren can calculate the actual time needed  to reach nitrogen saturation  for a given  site.
Immobilization of nitrogen  in soil organic matter  is not considered, however.  Johnson  questioned
whether 20% tree  uptake would  apply to atmospheric nitrogen inputs,  which are spread fairly
evenly through time, as opposed to one-shot fertilizer nitrogen inputs.

What Regulates Nitrification?

    This much-debated question was addressed by Vitousek et al.  (1979) among others and a full
review of it  was  not  possible here. Paul  Bulgden had indications  of nitrification  inhibitors at his
sites in Belgium,  whereas others (Johnson, Van Miegroet and Cole, Mulder) suspected NH4+  sup-
ply as the major factor. There is no doubt that inhibitors exist  at some sites, whereas NH4+ supply
is important at other sites.
    Once again the question of slow, steady atmospheric nitrogen inputs vs one-shot fertilizer appli-
cations was raised.  Johnson discussed results of a fertilizer study where quarterly applications of 25
kg/ha of nitrogen  as  urea  stimulated more nitrate leaching than annual  100-kg/ha applications.
Slow, steady inputs of NH4+ are probably optimal for building populations  of nitrifying organisms
in soil, and thus  nitrogen saturation  (defined as a state where NO3~ export ^ N input) may be
more rapidly approached with atmospheric  nitrogen  inputs than equivalent inputs of nitrogen  as
one-shot fertilizer applications.

Can Acidification Due to Vegetation Uptake and Humus Formation Be Quantified?
    Several  schemes  have  been proposed  for  quantifying  the acidification effects of vegetation
uptake and humus formation (as well as natural leaching), all of which involve  the same  basic
assumptions  (Ulrich 1980, Nilsson et al.  1982). There appeared  to be general agreement as to the
methods to be used to estimate H+ production from natural leaching (by  net anion production) and
humus formation  (estimate  organic  matter accumulation and  multiply  by  an  average cation
exchange capacity for humus, usually near 2-3 keq/kg), but the methods used to estimate acidifica-
tion from plant uptake are  less clear. This is due in part to uncertainties about the form in which
nitrogen is taken up,  as NH4+ (acidifying) or as NO3~ (alkalizing). For the other major elements,
the form of uptake is known (Ca2+, K+, Mg2+, SO42~~). It was  agreed that if  nitrogen is mineral-
ized from  organic  matter  and subsequently taken up by  plants,  the  net  acidifying effect is zero.
Thus,  only external inputs of NH4+ and NO3~ need be considered in regard to acidification. Simi-
larly, organic cation  mineralization  followed by uptake would have no net effect on acidification
(Van Breeman et al. 1983). However, the proportions of cation uptake from organic vs inorganic
soil sources are seldom known.

-------
                                              14

Are Natural Processes Self-Limiting?

    This question has been raised several times prior to the workshop by John Reuss.  He pointed
out that natural  leaching agents are fundamentally different from atmospheric inputs  in that the
former are weak acids (carbonic and organic) and therefore will tend to "turn themselves off as
soils acidify.  This point brought little debate, but it was not clear how acid these natural leaching
agents (especially organic acids) could make soils and soil solutions before turning themselves off.

Is Weathering Stimulated by Acidification?

    This question was centered on the possible compensation for accelerated acid input by an  accel-
erated rate  of exchangeable  cation  replacement  by  weathering.  Egbert  Matzner  noted that
estimated rates of weathering for the Soiling site were close  to estimates of weathering of soils over
geologic  time periods (since  deglaciation) in the  same region, suggesting that  weathering had not
accelerated to compensate for acid deposition. Arne  Stuanes was  asked to review his results from
the Norwegian acid irrigation studies in  this context. He reported that weathering (as  determined
by the mass balance approach, where inputs  and outputs are compared to the actual changes  in soil
exchangeable cations) had been accelerated by accelerated  cation leaching, but the  compensation
was not total because decreases in exchangeable base cations did occur with acid treatments.  Helga
Van Miegroet reported that the weathering rates in the  red alder stand must have been  accelerated
considerably  as compared to those in the Douglas-fir stand because, for example, the exchangeable
Mg2+ supplies in the red alder soil equalled only a 3-year supply at current leaching rates.
    It seemed to be the  group's conclusion that weathering could indeed be stimulated by increased
acid input, but that the rates of weathering will be quite variable and are usually not known.  It  was
clear  that determining weathering rates and the development of additional techniques  for the  deter-
mination of weathering rates constitute pressing research needs.

How Does Aluminum Mobilization Affect  Vegetation Uptake?

    The question of A13+ damage to tree roots and other hypotheses regarding causes  of tree  die-
back  and forest decline  were not  discussed during this  workshop. When  asked about this, Egbert
Matzner replied that the subject was so involved and complex that it would have to be (and will be)
addressed in  another workshop devoted entirely to it. The group agreed and thus this issue was not
considered per se, but only indirectly in the context of soil acidification.

Are Limiting Cations Conserved Even in the Presence of Acid  Precipitation?
    Johnson noted that trees take up cations individually for individual reasons and that we cannot
use the sum  of base cations as a particularly useful index of site nutrient cation status.  Even when
acid deposition is causing net export of base cations, not all base cations are necessarily being lost
(i.e., Ca shows a balanced budget, while K, Na, and K show a net export from the Fullerton site on
Walker Branch).  It would seem  that, from both a chemical  and biological perspective,  limiting
cations are likely  to show a  smaller net loss from a  system  than nonlimiting cations. However, an
argument can be  made  that even  small  additional losses of limiting  cations are more  detrimental
than greater losses of nonlimiting  cations. This, as well  as the entire area of acid deposition effects
on the cycling of individual cations, needs further investigation.

-------
                                 WATER ACIDIFICATION

KEYNOTE ADDRESS: How Are Waters  Acidified?   (Hans Martin Seip.  Central  Institute for
Industrial Research, Oslo, Norway)

Introduction
    Though I was asked to discuss aquatic acidification, most of my talk  will be on soil-water
interactions. The most interesting processes  related to water acidification go on in the catchment. I
will therefore come back to many of the same processes that we discussed yesterday but from a
slightly different point of view.
    To what extent  has freshwater acidification occurred? We all know about the difficulties in
comparing recent and old measurements of pH and  alkalinity. Perhaps the biological  changes pro-
vide the most convincing evidence for acidification (e.g., see Overrein et al. 1980, Swedish  Ministry
of Agriculture 1982).
    Most likely many of the lakes in southernmost Norway have always had a very low alkalinity.
In such cases, pH shifts of 0.5 to 1 unit will have dramatic biological consequences. Studies of dia-
toms in sediments have also been used to estimate pH changes in lakes. I will return to this  later.

Aquatic Buffer Systems
    Acidification is by  definition a change in  the H+ concentration, but it is often convenient to
talk about change in alkalinity  or  acid neutralizing capacity  (ANC). One  possible  definition of
ANC in a system with aluminium and organic acid is

    [ANC] = [HCO3~]  + 2[CO32~]  +

                [A1(OH)2+]  +  2[A1(OH)2+]  + 4[A1(OH)4-]
                + [RCOO-] +  [OH~]  -  [H+]
   [ANC] = Jf_n  ft d(PH) ,

where /9 is the buffer intensity.
    Organic  acids will usually buffer over a broad pH range, with maximum between 4 and 5. In
the example,  the bicarbonate system starts to dominate at pH slightly less than 6 (equilibrium with
atmospheric CO2 is assumed). Aluminum dominates at low pH values. The buffering caused by the
sediments may also be of considerable importance.

Conclusions Reached at the Sandefjord Meeting

    I will discuss the causes of acidification by going  back to the Sandefjord meeting held 4 years
ago. My discussion there concerned three conceptual models of water acidification (Seip 1980): (1)
                                             15

-------
                                              16

the model that is based on a direct effect (i.e., assuming that a substantial fraction of the precipita-
tion reaches rivers  and lakes essentially unchanged); (2) the model  that emphasizes the increased
deposition of mobile anions  (in particular SO42~);  and (3) the model that is based on effects on
freshwater  through a change in soil  acidity brought about by  acid  precipitation and other causes
(e.g., changed vegetation).
     The conclusions  I reached at the Sandefjord  meeting, mainly based on  Norwegian studies,
were:
(1) The fraction of the precipitation reaching rivers and lakes essentially unchanged is normally
    small. Thus the direct effects are small, but not negligible.

(2) Changes in the soil may occur as a result of acid precipitation. This may contribute to freshwa-
    ter acidification. However,  in my opinion, these effects are also likely to be small, but not negli-
    gible.
(3) Though there  seems to be no consistent  pattern of changes  in  land  use that may explain
    regional acidifications, these factors are probably contributing in some areas.

(4) The easiest way to understand the acidification of freshwater is by considering the increased
    sulfate deposition. The increased concentrations of mobile anions explain at least a substantial
    part of the observed acidification.

     I'll leave this subject  with the question, "To what extent has recent research strengthened or
weakened these conclusions?"

Soil-Water Interactions in the Catchment
     A catchment is a very complex system, and a large number of  factors/processes may possibly
play a role in connection with acidification:

— Hydrology
— Deposition of mobile anions
— Canopy interactions
— Carbonate system (pCO2)
— Organic and other weak acids
— Cation exchange
— Adsorption  of anions in soil
— Acid hydrolysis  of minerals
— Oxidation and reduction reactions (S and N compounds)
— Uptake and release of ions by vegetation.

Hydrology
     It is evident that soil-water  interactions  depend critically  on  the hydrology.  A  satisfactory
knowledge of the hydrology is therefore extremely important in understanding acidification. Four
years  ago  we did some work on the Birkenes model. The  hydrologic model consists simply of two
reservoirs, one representing the upper soil horizons  and the other the mineral soils (Christophersen
et  al.  1982). Even at  that time we were aware that this model was too simple. Studies of natural
isotope ratios (in particular, 18O/16O) may be used  to estimate the amount of "old water"  from the
catchment in the runoff. Studies by  Rodhe (1981) show that even during snowmelt the fraction of

-------
                                              17

melt water generally is fairly small. Our first I8O studies at Birkenes indicated that at high flow the
model gives too much water from the upper reservoir to the stream. In cooperation with Peter Dil-
lon and his group we  recently tried to modify the Birkenes model for an  inflow stream  to Harp
Lake (Harp 4). We then introduced "piston flow" [e.g., water coming from the upper reservoir may
push out an equivalent amount from the B-reservoir (Seip et  al. 1984)]. Other possible improve-
ments are to use a model based on the variable source area concept to introduce channel flow.
     In  many  catchments (e.g., Birkenes), the most  critical pH values are observed at high dis-
charge such as during  snowmelt.  With our  simple  Birkenes model (also after introducing piston
flow), the reason is partly that the runoff contains a large fraction of water from the upper reser-
voir.
     At  the meeting in Bolton Landing (Acidification: Natural or Anthropogenic?) this past fall,  Al
Lefohn  presented snowmelt  results  from  several areas. He argued that acid episodes are a natural
phenomenon. This is most likely true, but addition  of sulfur  anions, and  sometimes NO3~, will
cause the episodes to be more critical and  perhaps more frequent.

Direct Effects

     Having discussed the hydrology it seems convenient to comment on what I call direct effects. I
still  think my  conclusion from Sandefjord is valid for most areas, but I realize that  in some places
the lake water composition may come close to that of the precipitation. In particular, this is true for
some seepage lakes (i.e., lakes without surface inlets or outlets) such as  those  in  Wisconsin
described  by Jerry Schnoor  and J.  M. Eilers. Also, other lakes that occupy a large  fraction of the
catchment, say 30 to 40% or more, will  of course be considerably influenced  by direct deposition
[e.g., Lake  GardsjOn in southern  Sweden (Nilsson  1984a)]. In  most cases, however, the contact
between water and soil will be sufficient  to cause a great modification of water chemistry. In the
SNSF project, we did some experiments with radioactive tracers to study this interaction (Overrein
et al. 1980). The I8O studies  I mentioned seem  to  strengthen the conclusion about considerable
soil-water interaction even during snowmelt.
     Data on  streamwater  chemistry  during snowmelt from  the  Muskoka-Halibourton  area  in
Ontario point  in the same direction. The  variation in sulfate concentration during snowmelt is sur-
prisingly small. Because it is well known that most of the ionic impurities leave the  snowpack with
the first meltwater,  much of the sulfate  must come  from the  soil.  (Of course, the contact in the
beginning of the snowmelt could be small, but model calculations indicate that this is  not the case.)

Mobile  Anion  Concept

     Mobile anions are important as vehicles for cation transport  through a catchment. In the origi-
nal Birkenes model we ignored  all  anions except SO42~. The stream is so acid that  bicarbonate is
usually  negligible. (We have now included bicarbonate in a version of the model that we apply to
Harp 4, see section  on Prediction of Acidification). The argument for ignoring Cl~ was that it is
roughly compensated by Na+, both in precipitation and runoff. This approximation has been criti-
cized, and it  is true that precipitation episodes  with particularly  high  concentrations of sea-salts
may give  acid surges in the runoff. This has been observed on the west coast of Norway (SNSF-
project), and  I  believe it has  also been  found  in southwestern Sweden (neighborhood  of Lake
Gardsjdn). A recent  study in Scotland (Galloway area) seems to show the same effect.

-------
                                              18

    Nitrate is normally low in Norwegian freshwaters. The snowmelt period may be an exception,
as in  the Storgama catchment in southernmost Norway where sulfate is the dominant anion,  but
nitrate cannot always be neglected.  We usually  find high concentrations of sulfate and nitrate
simultaneously. This is different from the situation found for Woods and Panther lakes (Adirondack
Mountains) as shown  by Galloway et al. (1980).
    For Dart Lake (Adirondack Mountains), Driscoll and Schafran found no statistically signifi-
cant correlation between H+  + Al  and sulfate, even though sulfate is the dominant anion. The  H+
+ Al was, however,  strongly correlated with nitrate. I  do not know if  the correlation  reflects  a
causal relationship; Ingvar Nilsson recently observed similar trends in the  soil solution of lysimeters
(Nilsson 1984b). He suggested a mechanism involving basic aluminum sulfate [A14(OH)10SO4] that
may explain  the  results. Because the NO, emissions seem to increase quite rapidly, there  is no
doubt that  nitrate represents a potential problem.
    Organic  anions may  also deserve  a more detailed  discussion  than  I gave at the Sandefjord
meeting. Krug and Frink (1983) recently criticized the "mobile anion" concept. They stated that an
increase in input of H2SO4 would  lead to a decrease in organic anions and therefore not  to an
increase in the sum of anions. Actually, the logic  in their paper is not very clear. They argue that
acid input  has little effect on pH of the soil and water, but the concentration of organic anions  will
only decrease if the percolate becomes more acid. Furthermore, there is  no experimental evidence
that a possible decrease in organic anions should be equivalent to the sulfate increase. Work by Dil-
lon, LaZerte, and others in the Muskoka-Halibourton area shows that the concentration of organic
anions is usually low during snowmelt, which is the period with the lowest pH values.

Weathering and Cation Exchange
    We probably all agree that weathering and cation  exchange are key processes in estimating
effects of acid deposition. We may also consider acid hydrolysis and oxidation processes.  Hydro-
lysis of minerals and cation  exchange  are often treated together because these processes usually
result in consumption of H+  and production of Ca + Mg. It is useful, however, to make a distinc-
tion because  the  hydrolysis  of minerals always consumes H+ (with the  exception of quartz)  and
cation exchange may actually acidify  the percolate. Furthermore  cation exchange is probably  a
much faster process than hydrolysis.
    Some  recent estimates  of H+ consumption (in meq/m2) during the weathering process  are
given below
Van Breemen     World average                         310
 etal. (1984)     Range for  20 catchment                0-1590
                  Catchments with low acidification rate  40-350

 Schnoor et al.    Seepage lakes                          23
                  Drainage lakes                          80
    The rate constant for hydrolysis of minerals is expected to vary  with acidity, e.g.,

         r  = k[H+]a ,

where k and a are constants. Estimates of a are, however, very uncertain.

-------
                                            19
Sulfur Oxidation
    Even if we agree that sulfate is the most important anion in connection with acidification, it
does not necessarily mean that its source is anthropogenic. Mineralization of organic sulfur or oxi-
dation of other sulfur compounds (e.g., FeS2) is often important. This process had to be introduced
in the Birkenes model at an early stage. Also this sulfur may, of course, have atmospheric deposi-
tion as its origin. In the SNSF  project, we  did some experiments with  radioactive  sulfur  showing
that  sulfate supplied  to a mini-catchment will take part in  a number of complicated processes
(Overrein et al.  1980).  Work  by Mitchell, David, and others on soil from  the Adirondack Moun-
tains confirms the importance of sulfate formation through organic sulfur mineralization.

Prediction of Acidification

    Two recent attempts to predict changes  in H+ (or in H+  4- Al ions) are based on the mobile
anion concept.
    Henriksen and Wright developed an empirical model (Henriksen 1982, Wright  and Henriksen
1983). They consider a  simplified system;  for example, organic acid is neglected. Strong acid (SA)
may be expressed as

   SA =  H+ + 2Aln+ ~ HCO3~  .
For a change in nonmarine sulfate concentration in the water, a charge balance gives

   ASA = ASO4*  -  A(Ca* + Mg*)  .
A factor, F, is then defined

   F  = A(Ca* + Mg*)/ASO4*


   ASA - (1 -  F)  AS04* .

For southernmost Norway, they find  that F  < 0.4 and suggest that the most likely value  is F =
0.2.
    Here F is the ratio of increase in  nonmarine Ca + Mg to the increase in  ASO4. An upper
limit  for F was  obtained  by assuming that the Ca + Mg concentration in lakes in southernmost
Norway was negligible in industrial times. Wright and Henriksen (1983) give an estimate for water
quality in the "old days" that seems quite reasonable. They also estimated that a  30% reduction in
sulfate deposition would restore chemical conditions  such that 22% of the  lakes  now experiencing
fishery problems should be able to support fish. If, however,  a  soil acidification due to factors other
than  acid deposition occurred  in this region, the formulas cannot be used without modifications for
predicting changes in water quality due to changes in  deposition.
    Christophersen et al. (1984) tried to make a prediction model based  on the Birkenes model.
While modifying the model for the Storgama catchment, we noticed that fairly satisfactory agree-
ment could be obtained  for most of the year by using  only data for the upper reservoir.

-------
                                             20

    To use the model for low SO4 loadings we had to include bicarbonate. We then had four equa-
tions as given below. Actually this is the model presented by  Reuss yesterday, just formulated dif-
ferently.
            2[M2+]  +  3[A13+]  -  2[S042-]  +  [NCV] + [HCOj"]


                   KG


   [M3+] [H+r3  =  "KH,


   [H+] [HC03-] = K PC02 ,

where M2+ is  the sum of Ca2+ and Mg2+. All  of the theoretical curves in Fig. 5 were calculated
with "Its,, =  10* •'. Curve (I) was obtained with pCO2 = 0 and KQ = 10~221; curve (II) with a
pCO2 in soil equal to 100 times that in the atmosphere, a pCO2 in water equal to two times that in
the atmosphere, and KG =  10~2-15; and curve (HI) corresponds to the same CO2 pressures, but K0
= io-2-37.
    In Fig. 5 we compare simulated values of M2+ [= Ca2+  + Mg2+], A13+, and H+ with obser-
vations. Horizontal bars represent the median of observations; 75th  and 25th percentiles are also
given. The model seems to show that, at least for this catchment, the mobile anion concept is use-
ful. If we want  to use the model for prediction at very  low values of [804] + [NOs], organic
anions may become important.

The model predicts that the main changes are in A13+ (a constant level of organically complexed Al
is added) and M2+, with fairly small changes in  [H+] (F-factor = 0.45-0.65) except when [SO42~]
+ [NO3~] becomes very small. This is not in good agreement with the Henriksen- Wright model. It
is, however, important to remember that the Birkenes model  was developed to describe day-to-day
or seasonal variations. If soil acidification does occur, our model may also give an F-factor equal to
0.2. Comparison thus seems to indicate some degree of soil acidification. Peter Chester reached a
similar conclusion on the basis of the lake data from southernmost Norway.
    Often soil chemists point out that the amount of acid in  the precipitation is low compared to
that amount needed to change the pH or base saturation in the soil.  We must, however, remember
that the water does not flow homogeneously through the soil,  but mainly follows channels or larger
pores. This does not mean that there is no soil-water interaction, but the amount of soil in contact
with the main flow of water may be just a small fraction of the total amount. A counter-argument
may be, however, that studies  of 18O/16O ratios generally indicate long residence  times and, thus,
perhaps better distribution of the water in the soil.
     Experiments and field  studies going on at  present may  contribute greatly to solutions  of the
questions mentioned. I refer partly to data from the Canadian-Norwegian-Swedish RAIN project
and partly to the very interesting  changes that  seem to take place in lakes around Sudbury,
Ontario. The emissions from INCO at Sudbury have now been reduced to less than 50%  of the
values in the mid-1970s.

-------
                                               21
                                                           i   i  _T_._	in
                              50
100           150           200
           [so2-]«[NOS] ,i
                A At as Al3',
             50
                              50
                A  M2*,
            150
            100
             50
100           150           200

              [SO^fNO^peql-1
                              50
100
150
                              200
    Fig. 5. Medians of obscncd [H+k IAI3+l ud [M2+] with the 75th and 25th percentiles as tmetiaas of [SO42  ) +
[NO3  ). All available data for the Storgama catchment (n = 424) from 1974 to 1982 are included with three theoretical
curves (see text). Only total Al concentrations have been measured. We have therefore added 10 iicq/L to the computed
values to take organic complexes into account.

-------
                                               22
     Before leaving the  subject of prediction, I should  like to mention  some very  recent model
results by Rustad, Christophersen, and Seip. We carried out calculations for Harp 4 including those
for the anions SO42~ and HCO3~ and the cations M2+ [= (Mg2+ +  Ca2+)], H+, and A13+. The
hydrologic model was mentioned previously. The pCO2 in soil was assumed to be a function of soil
temperature  only. In Fig.  6,  I  plotted results for  H+ during a snowmelt period. (The peaks are
somewhat higher than the observed values.) Then we tried a calculation with reduced  sulfur deposi-
tion and reduced S-reservoirs, such that  the SO4 concentration in runoff was reduced by roughly
50%. The pH change is nearly one unit. This is in agreement with some preliminary calculations by
Seip and  Rustad (1984) showing that, in the pH  range we are discussing here, quite large shifts
may occur for moderate changes in deposition.
                                Simulation results     Harp
                                  Snowmelt  1979
                      r1)
               15
              10
                             March
April
   Fig. 6. Simulated daily H+ concentrations for Harp 4 daring snowmeh. Open circles correspond to present 804   con-
centrations in runoff. Filled circles were obtained for approximately a 50% reduction in the sulfate concentrations.
External and Internal Sources of Soil Acidification
     I mentioned earlier experiments showing that acid precipitation may result in soil acidification.
There are also  observations (e.g.,  from Sweden and  Germany) showing that the soil has become
more acid during recent decades. Nilsson (1983) discussed increases in soil acidity in Swedish forest
soils  and  concluded that "To date  no unequivocal evidence exists that points to a soil acidification
mainly caused by  atmospheric deposition.  The  tree species replacement and ion accumulation in
plant biomass and  humus seem to  be the most important causes." In  my view, atmospheric deposi-
tion has caused  soil acidification in some areas in Germany.
     In humid, temperate climate,  soil acidification  is a natural process, but usually at a slow rate.
Various anthropogenic activities may accelerate the process. If pine  trees are planted on an area
previously covered  by  grass, the  soil  will  certainly become more  acid.  Nilsson et al.  (1982)
estimated  net rates of acidification as the result of  excess cation accumulation in trees and  humus.
Table 3 is taken from  a recent study carried out for  EEC. The work by Nilsson et al. is probably
the basis  for some of the numbers. These numbers  seem fairly reasonable to me. I think, however,
the table may be a topic for further discussion.

-------
                                              23
                                 Table 3. Sources of acid Input to soil
                                (EuTironmental Resources Limited 1983).
                                        Values in percent
Forest


Fertiliser/nitrification
Biomass growth
Tree harvesting
Humus decomposition/
organic acid production
Soil respiration
Acid precipitation
Agricultural
land
65-80




10-20
5-20
Peat
bog



70-90


10-30
Spruce
med. soil

20-40
10-15
20-40


25-50
Pine
poor soil

5-10
0-10
10-15

0-5
60-80
Historic Trends in Lake Acidification and Relation to Activities in the Catchment

    Historic trends in acidification of lakes may be obtained by studying the diatoms in the sedi-
ments. There are  some difficulties,  however  (e.g.,  what is  the  importance of  bioturbation).
Nevertheless, it may be of interest to look at some results from various parts of the world.
     In Lake Gardsjb'n, a slow acidification was found during the centuries up to a few decades ago
 (Renberg and Hellberg 1982). There must certainly have been  vegetation changes, etc., during this
 period, but the pH decrease seems to be quite smooth almost all of the time.  (There  is, of course, a
 long period between the layers used to determine pH.) Charles (1984) studied a sediment core from
 Big Moose Lake in the Adirondack Mountains. From about 1800 until about 1950 the inferred pH
 of the lake was  —5.7. After  1950 the value  dropped steadily to about 4.7. It is known that there
 were major logging operations around  Big Moose  Lake in the last century,  but lake pH does not
 seem to have been affected. Also, the inferred  pH for Round Lock  of Glenhead, Galloway, shows
 small variations  for centuries; acidification seems to have started  around  1850. Flower and  Battar-
 bee (1983) compared  the acidification in Round  Loch of Glenhead (no afforestation)  to  that in
 Lock Grannoch (=70% afforestation). The acidification is similar in these lakes, though the details
 in the diatom patterns are somewhat different.

 New Conclusions and Areas of Uncertainty
     Going back now to the conclusions I made 4 years ago (section  3, Conclusions Reached at the
 Sandefjord Meeting), I do not think the modifications need to be too large. I did perhaps  at that
 time underestimate the importance of changes in soil properties, such as base saturation due to acid
 deposition. As mentioned, the soil changes along macropores may be greater  than those for  the soil
 on the average.  Also one should perhaps pay more attention to nitrate than I did, though I still
 think  the  first goal should be to reduce sulfur emissions.  In Sandefjord I mainly discussed  quite
 acid systems, but also in these cases bicarbonate should be included, especially to be able to make
 predictions for low levels of sulfate deposition.

     Very  briefly I would say that the main  effort now should be directed toward obtaining better
 prediction models for responses to changed emissions of SO2 and NO,.  This will, however, include a
 number of tasks, for example, obtaining better:  (1) estimates of various sources of soil acidification,
 (2) understanding of the hydrology, (3) estimates of weathering rates and their pH dependency, (4)
 understanding of the importance of nitrate vs sulfate, (5) measurements of pCO2 in  soils, including
 seasonal variations, and (6) understanding of  the aluminium chemistry in soil  and water.

-------
                        DISCUSSION (Summarized by D. W. Johnson)

    Following the keynote address by Seip, there were brief presentations by Bob Goldstein, David
Lam,  Jack Cosby, Dick Wright, Michael Hauhs, Peter Dillon, and Richard Skeffington, which are
summarized below along with the discussions.
    Bob Goldstein reviewed the Intergrated Lake-Watershed  Acidification Study (ILWAS) model
(Chen et  al. 1983, Chen  et  al. 1984)  and the results of recent applications of the model  to the
ILWAS watersheds  to  examine the role of hydrology, cation exchange, weathering, and climatic
variables in the acidification of lakes (Goldstein et al.  1984). He emphasized that the relative rout-
ing of water through different flow paths within a watershed is a major determinant of lake alkalin-
ity and lake vulnerability from acidification by atmospheric deposition. This is supported by results
from  the Regional Integrated  Lake-Watershed Acidification Study (RILWAS) as well as ILWAS.
The reader is referred to the above-cited publications for details.
    David Lam presented  surface water and groundwater model results  which appeared to explain
the observed spatial gradient of the stream alkalinity in the Turkey Lakes watershed near Sault Ste
Marie, Ontario.
    Jack  Cosby reviewed data and modeling results for the White Oak Run watershed in Virginia,
where distinct seasonal peaks in alkalinity  and base cation concentrations occur. The  peaks were
thought to be the result of seasonal variations in soil pCO2, and some back-calculations were made
to estimate what soil CO2 pressures must have been. Johnson suggested that soil pCO2 itself can be
modeled,  given CO2 evolution rates, soil porosity, and  soil water content (e.g., see DeJong and
Schappert 1972). He also suggested that rapid, event collection of lysimeter water could be used to
back-calculate soil pCO2,  because the  kinetics of CO2 loss from waters in collection  vessels was
apparently favorable. Calculations of kinetics need checking, however.
    Michael Hauhs presented hydrology,  soil solution, and stream chemistry results from  a new
study at Lange Bramke in West Germany which may relate to both  water acidification and forest
decline. He noted increasing  NO3~ and A13+ concentrations on  the north slope and  decreasing
Ca2+  and Mg2+ concentrations in soil solutions on the south  slope of the watershed. Symptoms of
forest dieback were also noted on the southern slope of the watershed, and the dieback was associ-
ated  with Mg2+  deficiency.  He hypothesized that base  cation depletion  occurred due to SO42~
leaching on the south  slope and that A13+ mobilization by nitrate is causing root damage on the
north slope.
    Dick  Wright  reviewed the Henriksen  model  and the  modifications and applications of it to
North American surface waters (Wright et al.  1983). The reader is  referred to that document for
details which will not  be repeated here. Johnson  questioned  the  very low values for the F-factor
used,  the  F-factor being the change in base cation concentration per unit change in sulfate concen-
tration. Wright reported an average  value of 0.4, which would imply either extremely acid soils or
surface runoff  with  little  contact with soil exchange sites. Reuss also  questioned  the  low value,
pointing out that the value must be representative of only the most sensitive  systems  which have
already become acid and that the same factor is probably not applicable to those less sensitive sys-
tems which have not yet become acid.

                                              24

-------
                                               25

    Peter  Dillon reviewed  results of surface  water chemistry in the Sudbury area, where sulfur
emissions were reduced by 80% from the early 1970s to the early 1980s. Recovery of surface water
pH and alkalinity was very rapid  and was associated with decreases in sulfate concentration. A sur-
face water pH rise  of 1  unit  within  5  years  was not uncommon. The  results are  consistent  with
those of soil SO42~  adsorption studies which show  that Spodosols  (Podzols)  have relatively  low
adsorption  capacities.  However, sulfate  adsorption studies have not been conducted on soils of the
area and it is not known whether  this or some other factor (e.g.,  hydrology) is the major reason for
the rapid recovery. Dillon also noted  that the F-factor, or the change in base cation concentration
per unit change in sulfate concentration, was  approximately 0.75 for this area, considerably higher
than the 0.4 value given as an average for sensitive regions  by  Wright et al. (1982). Dillon also
noted that  weathering must have been significant even in these  granitic soils, because calculations
indicated that it  would have taken only 8  to 10 years to deplete  the entire soil exchangeable cation
reserve.

     Richard Skeffington reviewed studies in the United Kingdom where  afforestation of pasture
land has caused  increases in concentrations of major ions, including A13+, and resulted in the death
of trout. At  Loch Grannoch (southwest Scotland), diatom records indicated that there was a sharp
decline  in lake  pH  after  sheep  were  removed from the  pasture  (Flower  and  Battarbee 1983).
Hypotheses for the  changes include increased deposition (especially cloud moisture) onto the forest
canopies, decreased water  flux (increased evapotranspiration) causing  increased  concentrations in
waters,  sulfur mineralization  due  to plowing prior to afforestation, and increased organic acid
export due to plowing (Binns 1984).

-------
               EXAMINATION OF BIRKENES AND GARDSJON DATA SETS
INTRODUCTION

    The first item on the agenda was an examination of data sets from Birkenes watershed in Nor-
way and Gardsjo*n watershed in Sweden by Dale Johnson and Ingvar Nilsson. The idea for this ses-
sion was to try to resolve differences of opinion as to causes of surface water acidification by focus-
ing on a common  data set  and discussing  important processes in the  context of this data set.
Synopses of the individual presentations are given below.

ANALYSIS BY DALE W. JOHNSON

    My approach to this analysis  was to use selectivity coefficients of the Gaines and Thomas
(1953) type to calculate changes in surface water chemistry with arbitrary changes in SO42~ con-
tent. The basic equations for A13+ — M2+  or M+ exchange are:

    (Al3+)2[m2+l3   =         ,+   _   (Al3+)[m2+]3/2                                  -(1)
    (M2+)3[A13+]2     W"  l     J      (m2+)3/2Q,'/2  '

   K2  =

                                                                                       (4)
where () denotes exchange phase, in equivalent fractions,
      [] denotes solution phase (in /teq/L),
      Q = selectivity coefficient,
      K = exchange coefficient assuming no change in base saturation.

Given these equations along with the charge balance equation, one can take stream chemistry data,
assume the solutions to have been in equilibrium with a soil of unknown properties, and then calcu-
late K values. This is done below using the  current concentrations in Birkenes streamwater (Table
4; Christophersen et al.  1982).
                                            26

-------
                                          27

                 Table 4. Blrkenes streamwater concentrations: Actual and calculated

                                    Predicted concentration if SO4   = 30 fieq/L

                                  No % BSa change            2X % BSa change
Ion
H+
pH
A13+
Na+
K+
CaI+
MgJ+
NH4+
2 cations'
SO42~
HCO3~
Cl
NO3
2 Anions
isiirrcm
cone.
33
4.48
71
123
7
67 (107)
40
<5
341
181
0
123
7
282
Na+ changes
25
4.60
32
94
5
63


219
30
0
123
7
160
Na+ constant
22
4.66 .
21
123
5
48


219
30
0
123
7
160
Na+ changes
20
4.70
20
75
4
99


218
30
0
123
7
160
Na+ constant
27
4.78
11
123
4
66


221
30
0
123
7
160
             Base saturation.
 The K values for the Birkenes watershed (from Table 4) are as follows:


                               =  2.62 X 10"                                          <5>
 lCn2+  +  Mo2+l3               fK + l3                                                f*\
Ik§	±_Mg—L  =  547       _L*_J_ = 40  .                                      (6)
                          243        l"-  J    =48
         [A13+]2                     [A13+]

We can then express total cations as a function of A13+:

     Total cations = 341  = H+  + A13+ + Na+ + Ca2+ + Mg2+ + K+
                           = (506 [A13+])'/3 + A13+ + (2.62 X 104[A13+])'/3
                           + (243 [A13+]2)'/3 + (4.8 [A13+])'/3
                           = 39.4 [A13+]'/3 + 6.24 [A13+]2/3 + A13+ .
    At this point, A13+ can be solved for iteratively to match any total cation value. For our pur-
poses,  I assumed a  SO42~  reduction  from  181  to 30  jteq/L  (decrease of 151  neq/L)  and a
corresponding decline in total cations (341 to 219 /ieq/L, ignoring the charge balance problems if
Al is considered to be A13+ in this water. If Al is A13+ or any charge greater than zero,  an anion
deficit exists, implying the presence of organic acids which may control pH.  I will not pursue this
controversial matter further in this exercise, however). The results of this calculation are given in
the third column of Table 4. Once an appropriate A13+ value  is obtained to match total cations, all
other cations are solved for. The results show a very small change in pH, a 55% reduction in A13+,
a 41% reduction in Ca2+ + Mg2+  (which are bulked as m2+  in this calculation), and a 24% reduc-
tion in Na+. An argument can be made that Na+ (as well as Cl~) will not decline but remain in a
constant steady-state condition due  to the minimal influence of soil exchange sites on Na+. We can

-------
                                             28

hold Na+ constant and resolve for A13+ and the other cations. This result, shown in column 4 of
Table 4, indicates a further  reduction in all cations (except Na+) but little additional  change in
pH.
    Now I  consider a change in  base  saturation (BS). To estimate potential  change,  I took soil
chemical data, and what physical data were available [from Frank (1970) (an assumed bulk density
of 1.2 g/crn3)], calculated soil base cation (BC) content, and divided this by net base cation export
from Christophersen et al. (1982). The results (Table 5) show a value  of  ~20 years,  so let us
assume  that %  BS has declined by 50% in 20 years (i.e., was 2 X its current value 20 years ago).
This is a worst case scenario since weathering is assumed to be zero.
Table 5. Soil contents vs input-output at Blrkenes
(Assumes bulk density — 1.2 g/cm3)

Soil
Iron Podzol
Iron-humus Podzol
Humus Podzol
2 BC 2 A1
keq/ha
17 85
17 180
18 79
2 BC + Al

102
197
97
(2 BC") +
BC export

19
19
20
G
(2 BC + Alat) +
BC export
Years
113
219
108
              °BC = base cation.                     _
              'Cation output minus cation input = 0.9 keq ha   year  .
    We  must then calculate Q values (neglecting exchangeable K+ and  Na+ as minor or non-
changing components) as shown below:

    (A?;?m;!i;  -  Q - (o-88)23(i°7>;  =  2.58  x  & .                               w
    (m2+)3[Al3+]2     V     (0.09)3(71)2
We then double % BS and calculate K:
Then the new K values are used to express total cations as a function of A13+  as in Eq. (7). The
results for a 50% change in base saturation with Na+ changing and Na+  constant are given in
columns 5 and 6 of Table 4.
     Clearly, it is not possible in theory to raise pH to acceptable levels (even using this worst-case
assumption) unless other mechanisms are involved (although as Reuss noted,  A13+ can change con-
siderably). This is the situation we have been debating for several years, and the calculations here
show changes of similar magnitude to those I  made by "adding"  SO42~  to unpolluted waters in
Alaska (Johnson  1981). The key to the problem is the change in alkalinity from positive to negative
values (or vice versa, Reuss and Johnson, in press). So far I have treated Birkenes streamwater as if
it were soil solution, but soil solution is  subject to much higher pCO2. Can alkalinity develop in a
soil solution of pH 4.78 (column  5, Table 4)? Let us assume that soil pCO2  is 0.3 X 103 Pa (0.03
atm), a not-unreasonable value. Then

-------
   [HC03-]  -
                                              29
                 KrKh  pC02  _  (10-7-74)(O.Q3)
                                                                      (10)
                     [H+]
                =  33 X  10~6
                    jQ-4.78

                   =  33 Meq/L  ,
   Alkalinity  =  HCO3~  -  H+ - A13+

                =  33  -  17  - 11  =  5
                                                                      (11)
Thus, the soil solution of pH 4.78 could have positive alkalinity and experience pH rise (and conse-
quently A13+ precipitation) upon degassing when it enters streamwater, as described by John Reuss
and Hans Martin Seip in their keynote addresses and by Reuss and Johnson (in  press). In short, if
the added SO42~ to Birkenes soil solutions caused a salt-effect pH shift, such that slightly positive
alkalinity became negative, a fairly large change in stream pH could result. Whether this actually
occurred at Birkenes remains to be demonstrated (especially if organic acids are present).
    An analogous analysis was carried out for the Gardsjb'n inlet (Table 6). In  this case, develop-
ment of positive alkalinity by removing SO42~ is more likely, because current A13+ and H+ concen-
trations are lower than those  at Birkenes:
    ru™ -i  _  K'-Kh PC°2  _  (10~7-74)(0.03)
    [HC03  ]	—-              W_M
67  X  1(T6
                                         67
    Alkalinity = HCO3 - H+  - A13+

                = 67  -  8 - 3  =  56 jieq/L  .
                         Table 6. Gardsjon Inlet concentrations: Aetna! and calculated

                                        Predicted concentration if SO42~ = 30 peq/L
                                       No% BS° change
                                            2X%BS° change
Ion
H+
PH
A13+
Na+
K+
Ca2+
Mg"
NH4
2) Cations
S042~
HCO3
Cl
NO3~
2 Anions
^UIICIll
cone.
19
4.7
32
209
14
94
91 (185)
3
462
181
0
232
5
418
Na+ charges
14.6
4.84
14.5
262
11
109


310
30
0
232
5
267
Na change
12
4.92
8
209
9
73


311
30
0
232
5
267
Na+ change
12
4.92
8
126
8
162


316
30
0
232
5
267
Na+ constant
8
5.09
3
209
6
84


310
30
0
232
5
267
                 °BS = base saturation.
                                                                      UBRAHV
                                                          1X9. ENVIRONMENTAL PROTECTION AGENCY
                                                         CORVAUJS ENywOMMBTOUL RESEARCH LAB
                                                                 2uO SW SSlM STREET
                                                               CORVMU
-------
                                               30

It should be noted, that the lake itself has lower H+ and A13+ concentrations than the streamlets
entering it.
     In  summary, the key to the issue of surface water pH change according to this model is the
change  in soil solution alkalinity from positive to negative (or vice versa) values, as pointed out by
Reuss. Soil solution alkalinity is determined by pCO2 and H+ and A13+ vs base cation selectivity
coefficients, none of  which are conveniently measured  in a survey mode.  Surface water alkalinity
itself may provide a crude index, however.

-------
ANALYSIS BY INGVAR S. NILSSON
    The following is a summary of a forthcoming  paper with the title, "Why is Lake GardsjOn
acid?—An evaluation of processes contributing to soil and water acidification" (Nilsson 1984a). It
will be  published in the Ecological Bulletin, Volume 37 (issued by the Swedish  Natural Science
Research Council).
    The average pH in the lakewater was 4.6 before liming, which  took place in April 1982. Three
studies  have shown  that the lake was  considerably  less acid some  decades ago. Renberg and
Hellberg (1982) showed that there has been a sudden increase of acid-tolerant diatoms in the lake
sediments,  at a level which has been dated to a few decades B.P. Further evidence comes from an
inventory of chironomid remnants in the sediments (Henrikson and Oscarsson 1984), also showing  a
successive increase of acid-tolerant species. Viable fish populations were reported  as  late  as  1949,
and summer pH in the epilimnetic water was about 6.2 (Anonymous 1949, Hultberg 1984b).
    The drainage area of Lake GardsjOn is 74.3 ha, the lake area being  31.2 ha. The morainic soil
has a clay  content of 4 to 11%, with a mean of 5% (Melkerud 1983, Olsson et al. 1984). It is char-
acterized as an iron  humus Podzol or an iron Podzol, depending on the topographical  location and
the concomitant water regime. The average soil depth is only SO cm. The vegetation is dominated
by mixed coniferous forest. Norway spruce (Picea abies Karst.) is the most important tree species.
According  to Olsson (1984), the site had  been covered by forest  for several hundred years, even
during the late 19th century and early part of the  present century, when extensive areas in this part
of Sweden were still occupied by Calluna heathland. Forest cutting according to a selected system
has been practiced, however, with cattle being allowed to graze in the forest up to 1950. A large
number of dead junipers (Juniperus communis L.) are found in the forest today, which is a strong
indication  of a previously more open forest. Some afforestation and clear-cutting has been under-
taken within the catchment.
     A proton budget was constructed to evaluate the relative importance of the present acid deposi-
tion compared to internal proton sources (Table 7). There is a large amount of uncertainty in some
of the estimates, for instance in the estimate of the weathering rate. (The total deposition of metal-
lic cations  is difficult to quantify).

                        Table 7. Proton budget (keq H+ ha~' year"') for the watershed
Wet deposition
SC>2 deposition
NH4+
NO3~

Accumulation of base
cations in the biomass
Dissociation of organic
acids including complex
formation
Sum
0.5
0.5
0.6-0.8
0.0

0.6

0-0.6


2.6-3.0
H+ output
NH4+ output
NOf deposition
Accumulation of anions
in the biomass

Weathering



Sum
0.5
0.0*>
0.6-0.8
0.0


1.6-2.3



2.7-3.6
                    ''The value is <0.05-
                                              31

-------
                                               32

    One proton  source  is not included in the table, namely the formation of organic aluminium
complexes  which precipitate before they reach the stream. According to some  rough calculations
this process might  account for another 0.2 to 0.6 keq-ha~'-year~'. If this source is included, the
external input of free and potential acidity would account for 30-50% of the total proton load.
    The proton  budget  indicates the extent of the current soil acidification,  as do lime potential
data presented by Nilsson and Bergkvist (1983), which showed that the  proton  flux in the deposi-
tion was of importance for the acidification of the uppermost organic soil layer, as the average lime
potential in the stand precipitation was lower than that of the soil solution in the humus layer (1.95
against 2.11).  Lime potential is here defined as pH-0.5  p(Ca+Mg).  A soil chemical inventory
showed that exchangeable calcium and magnesium make up 35 and 14%, respectively,  of the effec-
tive cation exchange capacity in  the humus layer,  while the corresponding figures for the B horizon
are 4 and  2%  (Olsson et al. 1984). Although most of the net output of  base  cations in the runoff
water should result from weathering processes (cf. Bache 1983), a certain net loss of exchangeable
base cations is likely in  the humus layer because of the lime potential gradient and because of the
relatively high calcium  and  magnesium  saturation.  Such a  loss should  be  of considerably  less
importance in the mineral soil, at least in absolute terms, as the selectivity for an individual cation
usually increases with a decreasing share of the cation exchange capacity (CEC) (Wiklander and
Andersson 1972, Wiklander 1980).
    As pointed out by Seip (1980) and others, soil acidification per se will not cause surface water
acidification unless there is also an increased flux of mobile anions.
    An important  question is whether an increasing  flux of sulfate, for example, could be a suffi-
cient explanation of the  fairly recent lake  acidification. One way to deduce the pH value in the lake
water  at a  different sulfate concentration is to insert the  present proportionality constant between
H+ and SO42~.  A 50% reduction of the sulfate  concentration would, according to this simplistic
analysis, increase the pH  to 4.9  to 5.0. This pH  interval is still a bit too low  for the carbonic acid
system to become important. It is evident  both from this approach, and from the more sophisticated
one used by Dale Johnson, that  soil  acidification has played a crucial role in  the lake acidification
process.
    The actual  significance  of  the  present atmospheric deposition can  be further illustrated by
making a  proton budget for  the lake, which is shown in Table  8. It is  essentially based on data
reported by Grennfelt et al. (1984) and Hultberg (1984a). The atmospheric proton flux, as in the
                                    Table 8. Proton budget for Lake
                                    Gardsjtin, based on the lake area.
                                       The contribution from the
                                       drainage area is estimated
                                      from extrapolation of fluxes
                                      measured in the three small
                                       ganged microcatchments
H+(keqha 'year ')
Wet deposition
SO2 deposition
Main inlet
Drainage area
Main outlet
Calculated
Proton retention
0.50
0.43
0.44
0.74-1.15
0.94
1.17-1.58


-------
                                             33

"terrestrial" proton budget, consists of H+, as measured in the wet deposition, plus the H+ result-
ing from the conversion of dry-deposited SO2 to H2SO4. Depending on how the flux from the drain-
age area is estimated (the lower value of the range being obtained from an extrapolation of the H+
flux measured in  a partly clear-cut subwatershed), the  direct deposition on  the lake surface
accounts for 37 to 44% of the proton input to Lake GardsjOn.
    The recent acidification of Lake Gardsjtin is likely to have been caused by  a combination of
direct atmospheric deposition on the  lake  surface and an  increased flux  of  protons and  aluminum
from the drainage area. This latter increase seems to be caused by an increase of the sulfate flux
and by  soil acidification, both  processes being significantly influenced by the atmospheric deposi-
tion.

-------
                                 SOIL SENSITIVITY CRITERIA

                                 R. S. Turner and D. W. Johnson
                                 Oak Ridge National Laboratory
                                      Oak Ridge, Tennessee

     During the  session on  "Soil Acidification," John Reuss listed several  effects of soil acidifica-
tion, including cation depletion, aluminum mobilization, change in soil pH (ApH/AH+), and base
cation loss  from calcareous soils.  Egbert Matzner suggested using a  broader measure such as  a
decrease in  acid  neutralizing capacity. It was suggested that that would involve a capacity factor so
large as to make  no soils  sensitive. What  are needed are more-specific capacity factors such as
exchangeable cations or intensity factors such as solution  pH depression and aluminum mobiliza-
tion. Jeff Lee pointed out that looking at specific soil acidification factors may be too narrow; what
are needed,  at  least  for assessment purposes, are criteria  that  would tell us "what are  sensitive
forests." At the end of the discussion Dale Johnson agreed  to outline his thoughts on sensitivity cri-
teria (Table 9) that could be used as a basis for further discussion.
     Results of the discussion at the "Soils Sensitivity Criteria" session are summarized in Table 10.
The consensus was that the sensitivity criteria could be broken down into capacity and intensity fac-
tors. However, it was pointed out that, to be useful, capacity factors had to  be expressed as capacity
over rate. For a capacity factor to be a  good indicator of sensitivity, we need to know how  long  it
                                       Table 9. SensitMty criteria
                            Receptor
    Solid phase
 Solution phase
                     Soils (solid phase)
                       Capacity factors
                         Reduction in % BS
                         S(>42  -retention
                         NC>3 -retention
                      Waters (soil solution and
                      surface water)
                       Intensity factors
                         A13+ mobilization
                         pH depression

                      Organisms
                       Forest damage
                       Fish kill
Low CEC, moderate
% BS (Wiklander
1980), weathering
Fe + Al oxides,
organic matter
Nitrification potential,
C/N ratio, forest
production
Low%BS
% BS, pCO2)
selectivity coefficients
Ionic strength
Alkalinity
      — Mechanisms unknown —
                    Water chemistry?
                                                34

-------
                                                   35
                                        Table 10. Sensitivity criteria
                                    as revised during the workshop session
                       Soil (solid phase) capacity factors—all need to be expressed
                       as capacity/rate to give time until "damage"

                       Reduction in              Low CEC (independent of CEC over the
                       base saturation (BS)        long term)
                                               Moderate initial BS (independent of
                                                initial BS over the  long term)
                                               Low weathering (base cation replace-
                                                ment) rate

                       Reduction in cation        pH
                       exchange capacity (CEC)   Clay mineralogy
                       SC>4    retention          Fe and Al oxide crystallinity
                       (inorganic)               Organic matter

                                               P" 2-
                                               SC>4   concentration in solution
                       Organic S retention        Microbial processes
                                               High organic matter buildup
                       SO42~ reduction         Anaerobic conditions
                       NOjT retention           Nitrification potential
                                               Nitrification inhibitors
                                               C/N ratio (weak predictor)
                                               Forest productivity
                                               Forest type
                                               Litterfall rate
                                               Decomposition rate
                                               Climatic factors
                       NH4+ retention          CEC
                       Soil solution and surface water intensity factors—these
                       factors potentially cause "damage* when one or more of the
                       capacity factors is exhausted

                       pH depression            pCO2
                                               BS
                                               Cation selectivity coefficient
                                               Low alkalinity alone not suitable
                                                surrogate
                       A13+ mobilization         Low BS, low pH
                                               High ionic strength  (mobile anions)
                                               Presence and solubility of Al mineral
                                               Organic matter complexing
                                               Paniculate transport to waterbody
will take to exhaust the capacity, or how long it will be until enough capacity is exhausted that the
intensity factors will be affected. This led to discussions of little known variables such as weathering
rates, biological uptake,  and  cycling rates of  some  of the elements. Thus, it was recognized that,
while we can identify  numerous capacity factors that may affect sensitivity of a soil to acidic depo-
sition, we do not know enough about the biogeochemical process rates associated with those factors
to be able to tell how  near we are to exhausting the  capacity of many of the  factors  listed. The spe-
cific research needs mentioned included effects on weathering by organic and mineral acids, biologi-
cal interactions  of sulfur, and a  better clarification  of nitrification mechanisms. Also needed is a
better understanding of hydrologic flow routes through the  soil,  and potential  differences between
macropore and bulk soil chemistry and rates of change.

-------
                                              36

    Very limited discussion was directed toward specifying soil sensitivity criteria for forest damage
or for fish kills. Because so little is known about the mechanistic relationships between forest dam-
age and  acidic deposition, participants were reluctant to suggest which soil factors would be useful
sensitivity criteria.  The consensus  was that  the subject should be discussed more  fully at another
workshop. The most direct indicator of hazard  to fishes was thought to be the ratio of Al to Ca in
surface  water.  Concentrations of these elements could be dependent, to one degree or another, on
most of the soil sensitivity factors listed.
    The sensitivity criteria listed in Tables 9 and 10 are based on known soil chemical and  physical
properties (as they relate to the movement of SO42~ and NO3~ through soils) and  on the rate at
which soil and soil solution acidity will change  in response to  SO42~ and NO3~ leaching. A reduc-
tion in base  saturation occurs most easily in soils with a moderate %  base saturation (%  BS) (so
that  the yield  of base cations per unit H+  input, or M/H+, is  near 1.0) and low exchangeable
reserves, which translates  into moderate % BS  and low CEC. This is essentially the view  espoused
by Wiklander (1980). According to the best available evidence, SO42~  retention of is related to Fe
+ Al oxide and organic matter content (Johnson and Todd 1983). Retention of NO3~ is governed
by ecosystem nitrogen status, which in turn is  related to tree  nitrogen uptake, soil microbial nitro-
gen demand, and nitrification potential (Vitousek et al.  1979). Very broad sensitivity criteria would
include  tree  nitrogen uptake by forest type  and soil C/N ratio overlaid on atmospheric deposition
rate.
    For waters, we are interested in  solution H+ and A13+ concentration, both of  which are inten-
sity factors determined both  by capacity factors (% BS, selectivity coefficients) and intensity factors
(pCO2, ionic strength, alkalinity). See the keynote address by Reuss for details.
    The sensitivity criteria discussed up to this point relate to soil and solution chemistry  only and
not to biological effects. In terms of fish, the consensus was that the relationship between  chemical
and biological  effects are relatively straightforward (i.e., fish damage = f (water chemistry). In
terms of forest damage, we  do not know appropriate toxicity  levels  for A13+  yet nor, in a broader
sense, do we know that A13+ mobilization is the cause of observed  forest dieback and decline. Thus,
the link between acid deposition and forest damage (if any).has not been established and, therefore,
sensitivity criteria cannot be assigned  except insofar as they represent hypothesized mechanisms. In
terms of the latter, the A13+ and  N-saturation  hypotheses would be defined as in Tables 9 and 10.
It must be emphasized, however, that the assignment of forest damage sensitivity criteria is prema-
ture at best and misleading at worst  until mechanisms of forest damage are established. The latter
will require more research and more  time to sort out; it has  taken well over  a decade to  sort out
appropriate mechanisms for water acidification.

-------
                                 ACKNOWLEDGMENTS

    This research was funded as part of the National Acid Precipitation Assessment Program by
the U.S. Environmental Protection Agency under Interagency Agreement No. 40-1353-83 with the
U.S. Department of Energy under  Contract  No  DE-AC05-84OR21400 with Martin  Marietta
Energy Systems, Inc.
    J. O. Reuss' contribution was sponsored in part by the U.S. Environmental Protection Agency.
The research described in this article has been funded in part by the EPA/NCSU Acid Precipita-
tion Program (a cooperative agreement)  between the U.S. Environmental Protection Agency and
North Carolina State University.  It has not been subjected to EPA's required peer and policy
review and  therefore does not necessarily reflect the views of the Agency and no official endorse-
ment should be inferred. Publication No. 2369, Environmental Sciences Division, ORNL.
                                            37

-------
                                      REFERENCES
Agren, G. I. 1983. Nitrogen productivity of some conifers. Can. J. For. Res. 13: 494-500.
Anonymous. 1949. Lake register. Drottningholm (In Swedish).
Bache, B. 1983. The implications of rock weathering for acid neutralization, pp.  175-187. IN Eco-
    logical Effects of Acid Deposition. Report PM 1636. National Swedish Environment Protection
    Board, Solna.
Binns,  W. O. 1984. Acid Rain and Forestry. UK Forestry Commission Research and Development
    Paper 134. Forestry Commission, Edinburgh.
Charles,  D.  1984.  Recent pH  history of  Big Moose Lake (Adirondack Mountains, New  York,
    USA) inferred  from sediment diatom assemblages, pp. 21-28.  IN 22nd Congress of Interna-
    tional Association of Theoretical and Applied Limnology, Lyon,  France.
Chen, C. W., S. Gherini, R.  J. M. Hudson, and J. O. Dean. 1983. The Integrated Lake-Watershed
    Acidification  Study. Vol. 1: Model Principles and Procedures. Rep. No. EA-3221.  Electric
    Power Research Institute. Palo Alto, California.
Chen, C. W., S. A. Gherini, J. D. Dean, R. J. M. Hudson, and R. A. Goldstein. 1984. Development
    and  calibration  of the integrated lake-watershed  acidification  study  model,  pp.  175-203. IN
    J. L. Schnoor (ed.),  Modeling of Total Acid Precipitation Impacts. Butterworth,  Stoneham,
    Massachusetts.
Christophersen, N., H. M. Seip, and R. F. Wright. 1982. A model for streamwater chemistry at
    Birkenes, Norway. Water Resour. Res. 18: 977-996.
Christophersen, N.,  S.  Rustad, and H. M.  Seip. 1984. Modelling streamwater chemistry with
    snowmelt. Phil. Trans. R. Soc. London, B 305: 427-439.
David, M. B., M. J. Mitchell, and J. P. Nakas. 1982. Organic and inorganic sulfur constituents of a
    forest soil and their relationship to microbial activity. Soil Sci. Soc. Am. J. 46: 847-852.
David, M. B., M. J. Mitchell, and S. C.  Schindler. Dynamics of organic and inorganic sulfur consti-
    tuents in  hardwood forest soils.  IN Proc.,  Sixth  North American  Forest Soils Conference,
    Knoxville, Tennessee,  June  1983 (in press).
DeJong,  E., and H.  J. V. Schappert.  1972. Calculation of  soil  respiration and activity from CO2
    profiles in the soil. Soil Sci. 112: 328-333.
Driscoll,  C.T., and  J. J.  Bisogni. 1983. Weak acid/base  systems in  dilute acidified lakes and
    streams of the Adirondack region  of New  York State, pp. 53-72.  IN J. L.  Schnoor (ed.),
    Modeling of Total Acid Precipitation Impacts. Butterworth, Stoneham, Massachusetts.
Environmental Resources  Limited.  1983. Acid  Rain. A Review of the Phenomenon  in the EEC &
    Europe. Graham & Trotman for the Commission of the  European Communities.
Flower, R. J.,  and  R. W.  Battarbee, 1983. Diatom evidence for recent acidification of two Scottish
    lochs. Nature 305: 130-133.
Frank, J. 1980. Soil  survey at Birkenes, a small catchment in Aust-Agder County, Sourthern Nor-
    way. TN 60/80. SNSF-Project, Oslo, (In Norwegian).
                                             38

-------
                                             39

Gaines, G. L., and H. C. Thomas. 1953. Adsorption studies on clay minerals: II. A formulation of
    the thermodynamics of exchange adsorption. J. Chem. Phys. 21:714-718.
Galloway, J.  N., C. L. Schofield, G. R. Hendry, N. E. Peters, and A.  H. Johannes.  1980. Sources
    of acidity in three  lakes during snowmelt. pp.  264-265. IN D. Drabl0s and A.  Tollan (eds.),
    Ecological Impact of Acid Precipitation. SNSF-project, NIVA, Oslo.
Goldstein, R. A., S. A. Gherini, C. W. Chen, L. Mok, and R. J. M. Hudson. 1984.  Integrated acid-
    ification  study (ILWAS): A mechanistic ecosystem analysis. Phil. Trans. R. Soc. Lond. Ser. B
    305: 409-425.
Grennfelt, P.I., S. Larsson, P.  Leyton, and B. Olsson. 1984. Atmospheric deposition to the  Lake
    Gardsjb'n area. IN  F. Andersson and B. Olsson (eds.), Lake Gardsjb'n—An Acid Forest  Lake
    and its Catchment. Ecological Bulletin, Volume 37,  Swedish Natural Science  Research Coun-
    cil, Stockholm (in press).
Henriksen, A. 1982. Changes in base cation  concentrations due to freshwater acidification.  Acid
    Rain Research Rep. 1/1982.  Norway Institute  Water Research, Oslo.
Henrikson, L., and H. G. Oscarsson. 1984.  History of the acidified Lake Gardsjb'n;  the development
    of chironomids.  IN F. Andersson  and  B. Olsson (eds.), Lake Gardsjb'n—An Acid Forest  Lake
    and its Catchment. Ecological Bulletin, Volume 37.  Swedish Natural Science  Research Coun-
    cil, Stockholm (in press).
Hultberg, H.  1984a. Effects on land-water chemistry and mass transports by S- and N-deposition
    to the acid Lake Gardsjb'n catchment, SW Sweden. IN  F.  Andersson and B.  Olsson (eds.),
    Lake Gardsjdn—An Acid Forest Lake and its Catchment. Ecological  Bullentin, Volume 37.
    Swedish  Natural Science Research Council, Stockholm (in press).
Hultberg, H.  1984b. Historical changes in fish populations and water  chemistry in Lake Gardsjb'n
    and lakes in a near area during the last century. IN F. Andersson and  B.  Olsson (eds.),  Lake
    Gardsjb'n—An Acid Forest Lake and its Catchment. Ecological Bulletin, Volume 37. Swedish
    Natural  Science Research Council, Stockholm (in press).
Johnson, D.  W. 1981. The natural acidity of some unpolluted waters in southeastern  Alaska and
    potential impacts of acid rain. Water Air Soil Pollut. 16:243-252.
Johnson, D. W., and D. E. Todd. 1983. Some relationships among Fe,  Al, C, and SO42~ in a vari-
    ety of forest soils. Soil Sci. Soc. Am. J. 47: 792-800.
Johnson, D. W., G. S. Henderson, D. D. Huff, S. E. Lindberg, D. D. Richter, D. S.  Shriner, D. E.
    Todd, and J. Turner. 1982a.  Cycling of organic and  inorganic sulphur in a  chestnut oak forest.
    Oecologia 54: 141-148.
Kelly, J. M.  1984. Power plant influences on bulk precipitation, throughfall, and stemflow nutrient
    inputs. J. Environ. Qual. 13: 405:409.
Krug, E.C., and C. R. Frink. 1983. Acid rain on acid soil: A new perspective. Science  221: 520-525.
Lindsay, W.  L. 1979. Chemical Equilibria in Soils. John Wiley and Sons,  New York. 449 pp.
Melkerud, P.A.  1983. Quaternary deposits  and bedrock outcrops in an  area around Lake Gardsjon,
    southwestern Sweden, with physical mineralogical and geochemical investigations.  Reports  in
    Forest Ecology and Forest Soils 44. Department of Forest Soils, Swedish University of Agricul-
    tural Sciences, Uppsala.
Nilsson, S.I.  1983. Effects on soil chemistry  as a consequence of proton  input,  pp. 105-111. IN B.
    Ulrich and J.  Pankrath (eds.), Effects  of Accumulation of Air Pollutants in Forest Ecosystems.
    D. Reidel Publ.  Co., Hingham, Massachusetts.

-------
                                              40

Nilsson, S. I. 1984a. Why is Lake GardsjBn acid?—An evaluation of processes contributing to soil
    and  water acidification. IN F. Andersson and  B.  Olsson (eds.), Lake  GardsjOn—An Acid
    Forest Lake and  its Catchment.  Ecological  Bulletin, Volume  37.  Swedish Natural  Science
    Research Council, Stockholm (in press).
Nilsson, S. I. 1984b. The interaction between mobile anions and aluminium in a shallow podzol. IN
    Proceedings from  Acid Rain and Forest Resources Conference, Quebec  City, June 1983  (in
    press).
Nilsson, S. I., and B. Bergkvist.  1983. Aluminum chemistry and acidification processes in a  shallow
    Podzol on the Swedish west  coast. Water Air Soil Pollut. 20:311-329.
Nilsson, S. I., H. G. Miller, and J. D. Miller.  1982.  Forest growth as a possible cause of soil and
    water acidification: An examination of the concepts. Oikos 39: 40-49.
Olsson, B. 1984. Land-use history of the Lake GardsjSn area, southwestern Sweden. IN F. Anders-
    son and B. Olsson (eds.), Lake GardsjOn—An Acid Forest  Lake and its Catchment. Ecological
    Bulletin, Volume 37. Swedish Natural Science Research Council, Stockholm (in press).
Olsson, B., L. HallbMcken,  S. Johansson, P.  A. Melkerud, S. I. Nilsson, and T. Nilsson. 1983. The
    Lake Gardsjon area—Physiographical and biological features. IN  F. Andersson and B. Olsson
    (eds.), Lake Gardsjftn—An  Acid Forest Lake and its Catchment.  Ecological Bulletin, Volume
    37, Swedish Natural Science Research Council, Stockholm (in press).
Overrein, L. N., H. M. Seip, and A. Tollan. 1980. Acid Precipitation—Effects on Forest and Fish.
    Final Report of the SNSF Project 1972-1980. NIVA, Oslo.
Renberg, I., and T. Hellberg. 1982. The  pH history of lakes in  south-western Sweden as calculated
    from subfossil diatom flora of the sediments. Ambio 11: 30-33.
Reuss, J. O. 1980. Simulation  of soil nutrient losses resulting from rainfall acidity. Ecol.  Model.
    11:15-38.
Reuss, J. O., and D. W. Johnson. 1984. Effect of soil processes on the acidification of water  by acid
    deposition. J. Environ.  Qual. (in press).
Rodhe, A. 1981. Spring flood. Meltwater or groundwater? Nordic Hydrol. 12: 21-30.
Rosenqvist, I. Th., P. Jorgensen, and  H. Rueslatten. 1980. The importance of natural H+ ion pro-
    duction for acidity in soil and water, pp. 240-241. IN D. Drablos and A. Tollan (eds.),  Ecolog-
    ical Impact of Acid Precipitation. Jobs. Grefslie Trykkeri A/S, Mysen, Norway.
Seip,  H. M.  1980.  Acidification  of freshwater—sources and mechanisms,  pp. 358-365.  IN D.
    Drablos and A. Tollan (eds.), Ecological Impact of Acid Precipitation. Jobs. Grefslie Trykkeri
    A/S, Mysen, Norway.
Seip,  H. M., and S. Rustad. 1984.  Variation in surface water pH with changes in  sulphur deposi-
    tion. Water Air Soil Pollut.  21: 217-223.
Seip,  H. M., R. Seip, P. J. Dillon, and E. de Grosbois. 1984. A model of sulphate concentration in
    a small stream in the Harp lake catchment, Ontario.  Can. J. Fish. Aquat. Sci. (in press).
Swedish Ministry of Agriculture. 1982. Acidification Today—and Tomorrow.  Swedish Ministry of
    Agriculture  Environment   '82  Committee,  Swedish  Ministry  of  Agriculture,  S-103  33
    Stockholm.
Ulrich, B.  1980. Production and consumption of hydrogen ions in the ecosphere. pp. 255-282.  IN
    T. C.  Hutchinson and M.  Havas (eds),  Effects of Acid Precipitation on Terrestrial  Ecosys-
    tems. Plenum Press, New York.

-------
                                             41

Van Breeman, N., P. A. Burrough, E. J. Yelthorst, H. F. Van Dobben, Toke de Wit, T. B. Ridder,
    and H. F. R. Reijnders.  1983. Soil acidification from atmospheric ammonium sulphate in for-
    est canopy throughfall. Nature 299: 548-550.
Van Breemen, N., C. T. Driscoll, and J. Mulder. 1984. Acid deposition and internal proton sources
    in acidification of soil and waters. Nature 307: 597-604.
Van Miegroet, H., and D. W. Cole.  1984. The  impact of nitrification on soil acidification and cat-
    ion leaching in a red alder (Alnus rubra Bong.) ecosystem. J. Environ. Qual. (in press).
Vitousek, P. M.,  J. R. Gosz, C.  C.  Grier, J. M. Melillo, W. A. Reiners, and R. L. Todd. 1979.
    Nitrate losses from disturbed ecosystems. Science 204: 469-474.
Wiklander, L. 1980. The sensitivity of soils to acid precipitation, pp. 553-567. IN T. C. Hutchinson
    and M. Havas (eds.), Effects of Acid Precipitation on Terrestrial Ecosystems. Plenum Publish-
    ing Corporation, New York.
Wiklander, L., and A. Andersson. 1972. The replacing efficiency of hydrogen ion in relation to base
    saturation and pH. Geoderma 7:159-165.
Wright, R.  F., and A. Henriksen. 1983. Restoration of Norwegian  lakes by reduction in sulphur
    deposition. Nature 305: 422-424.
Wright, R. F., N. Conroy, W. T.  Dickson, R. Harriman, A. Henrickson, and C. L. Schofield. 1980.
    Acidified lake districts of the world: A comparison of lake chemistry of lakes in southern Nor-
    way, southern Sweden, the Adirondacks of New York, and southeastern Ontario, pp. 380-382.
    IN D.  Drablos and A. Tollan (eds.), Ecological Effects of Acid Precipitation. Johs  Grefslie
    Trykkeri A/S, Mysen, Norway.
Wright, R. F. 1983.  Predicting acidification  of  North  American  lakes. Acid  Rain Res.  Rep.
    4/1983. Norwegian Institute  for Water Research, Oslo.

-------
                              LIST OF ATTENDEES
Goran I. Agren
Swedish University of Agricultural Sciences
Uppsala, Sweden

Ernesto Bosatta
Swedish University of Agricultural Sciences
Uppsala, Sweden
Nils Christophersen
Central Institute for Industrial Research
Forskningsv. 1, Blindern
Oslo 3, Norway
Roger Clapp
Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee 37831

Dale W. Cole
College of Forest Resources
University of Washington
Seattle, Washington 98195

Christopher S. Cronan
Land and Water Resources Center
University of Maine
Orono, Maine 04469

P. J. Dillon
Ontario Ministry of Environment
 Box 39
Dorset, Ontario POA 1EO
Canada
Jerry W. Elwood
 Environmental Sciences Division
 Oak Ridge National Laboratory
Oak Ridge, Tennessee 37831
Folke O. Andersson
Swedish University of Agricultural Sciences
Uppsala, Sweden
Paul Buldgen
University of Liege
Department of Botany
Sart Tilman, B-4000 Liege
Belgium

Robbins M. Church
U.S. Environmental Protection Agency
200 SW 35th Street
Corvallis,  Oregon 97377
 B. J. Cosby
 Department of Environmental Sciences
 Clark Hall
 University of Virginia
 Charlottesville, Virginia 22901

 Mark B. David
 U.S. Environmental Protection Agency
 200 SW 35th Street
 Corvallis, Oregon 97333

 Charles Driscoll
 238 Hinds Hall
 Department of Civil  Engineering
 Syracuse University
 Syracuse, New York 13210

 Neil Foster
 90 Florwin Drive
 Canadian Forestry Service
 Sault Ste. Marie
 Ontario, Canada
                                             42

-------
                                           43
James Galloway
Clark Hall
University of Virginia
Charlottesville, Virginia 22903

Michael Hauhs
Institute of Soil Science and Forest
 Nutrition
Busgenweg 2
34 Gottingen, West Germany

 John W. Huckabee
 Electric Power Research Institute
 3412 Hillview
 Palo Alto,  California 94091
 Dale W. Johnson
 Environmental Sciences Division
 Oak Ridge National Laboratory
 Oak Ridge, Tennessee 37831
 Jeffrey Lee
 U.S. Environmental Protection Agency
 200 SW 35th  Street
 Corvallis, Oregon 97333

 John Malanchuk
 Acid Deposition Assessment Staff
 RD-676
 U.S. Environmental Protection Agency
 Washington, D.C. 20460
 Jan Mulder
 Department of Soil  Science and Geology
 Agricultural University of Wageningen
 Duivendaal 10
 Wageningen, The Netherlands
 Stephen  C. Nodvin
 University of California-Riverside
 Route 1, Box  198
 Mammoth Lakes, California 93546
 Dan D. Richter
 Environmental Sciences Division
 Oak Ridge National Laboratory
 Oak Ridge, Tennessee 37831
Robert Goldstein
Electric Power Research Institute
P.O. Box 10412
Palo Alto, California 94303
George M. Hornberger
Department of Environmental Science
Clark Hall
University of Virginia
Charlottesville, Virginia 22903

 Hans Hultberg
 Swedish Environmental Research Institute
 P.O. Box 5207
 Gothenburg, Sweden
 J. M. Kelly
 TVA/ORNL Watershed Program
 Oak Ridge National Laboratory
 Oak Ridge,  Tennessee 37831 .
 David Lam
 Environment of Canada, CCIW
 867 Lakeshore Road
 Burlington, Ontario LTR 4A6
 Canada

 Egbert  Matzner
 University of Gottingen
 Busgenweg 2
 Gottingen, West Germany 0551

 Ingvar Nilsson
 Department of Ecology and Environmental
  Research
 Swedish University of Agricultural Science
 Uppsala, Sweden S-75007
 John O. Reuss
 Department of Agronomy
 Colorado State University
 Fort Collins, Colorado 80523
 Hans M. Seip
 Central Institute for Industrial Research
 Forskningsv. 1, Box 350
 Blindern, Oslo 3, Norway

-------
                                            44
Helgan Van Miegroet
College of Forest Resources, AR-10
University of Washington
Seattle, Washington

Ulf Wahlgren
Swedish Environmental Research Institute
Vikingavagen 5
Sollentuna, Sweden

Richard Skeffington
Central Electricity Generating Board
Kelvin Avenue
Leatherhead, Surrey KT22
England

Robb S. Turner
Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee 37831
Webb Van Winkle
Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee 37831
Dick Wright
Clark Hall
University of Virginia
Charlottesville, Virginia 22903

Arne O. Stuanes
Norwegian Forest Research Institute
P.O. Box 61
N-1432 AAS-NLH
Norway
Hans Van Grinsven
Department of Soil Science & Geology
Wageningen
The Netherlands

-------
                                                                         ORNL/TM-9258
                               INTERNAL DISTRIBUTION
   1. S. I. Auerbach                                     17.  R. J. Olson
   2. E. A. Bondietti                                     18,  D. E. Reichle
   3. R. B. Clapp                                        19.  D. S. Shriner
   4. N. H. Cutshall                                     20.  R. S. Turner
   5. J. W. Elwood                         ,             21.  W. Van Winkle
   6. S. G. Hildebrand                                   22.  Central Research Library
7-11. D. W. Johnson                                  23-37.  ESD Library
  12. J. M. Kelly                                     38-39.  Laboratory Records Dept.
  13. G. M. Lovett                                      40.  Laboratory Records, ORNL-RC
  14. R. J. Luxmoore                                    41.  ORNL Patent Office
  15. R. J. Norby                                        42.  ORNL Y-12 Technical Library
  16. J. S. Olson
                              EXTERNAL DISTRIBUTION

     43. Goran I. Agren, Swedish University of Agricultural Science, Uppsala, Sweden S-75007
     44. Folke Q. Andersson, Swedish University of Agricultural Science, Uppsala, Sweden S-
         75007
     45. Ann Bartusky, EPA/NCSU  Acid Precipitation Program, North Carolina State Univer-
         sity, Raleigh, NC 27606
     46. Ernesto Bosatta, Swedish University of Agricultural Science, Uppsala, Sweden S-75007
     47. Paula Buldgen, University of Liege,  Department of Botany, Sart Tilman,  B-4000 Liege,
         Belgium
     48. J. Thomas Callahan, Associate Director, Ecosystem Studies Program, Room 336, 1800 G
         Street, NW, National Science Foundation, Washington, DC 20550
     49. Nils Christophersen, Central Institute for Industrial Research,  Forskningsv.  1, Blindern,
         Oslo 3, Norway
     50. Robbins M. Church, U.S.  Environmental Protection Agency, 200 SW,  35th Street, Cor-
         vallis, OR 97377
     51, Dale W. Cole, University of Washington,  College of Forest  Resources, Seattle, WA
         98195
     52. B. J. Cosby, University of Virginia, Clark  Hall, Department of Environmental Science,
         Charlottesville, VA 22901
     53. Ellis Cowling, School of Forest Resources, North Carolina State University, Raleigh, NC
         27650
     54. Christopher S. Cronan, University of Maine, Land and Water Resources Center, Orono,
         MA 04469
     55. R.  C.  Dahlman, Carbon  Cycle  Program Manager,  Carbon Dioxide Research Division,
         Office  of  Energy Research,  Room J-311,  ER-12, Department of Energy, Washington,
         DC 20545

                                            45

-------
                                      46

56.  Mark B. David, U.S. Environmental Protection Agency, 200 SW, 35th Street, Corvallis,
    OR 97333
57.  P.  J. Dillon, Ontario Ministry of Environment,  Box 39, Dorset, Ontario, Canada POA
    1EO
58.  Charles Driscoll, Syracuse University, 238 Hinds Hall, Department of Civil Engineering,
    Syracuse, NY 13210
59.  Ivan  Fernandez, University of Maine, Orono, ME 04469
60.  G. J. Foley, Office of Environmental Process and Effects Research, U.S. Environmental
    Protection Agency, 401 M Street, SW, RD-682, Washington, DC 20460
61.  Neil  Foster, Canadian Forestry Service, 90  Florwin  Drive, Sault Ste. Marie, Ontario,
    Canada
62.  James Galloway, University of Virginia, Clark Hall, Charlottesville, VA 22903
63.  J. H. Gilford, U.S. Environmental Protection Agency, Office of Toxic Substances, 401 M
    Street, SW, TS-796, Washington, DC 20460
64.  Robert Goldstein, EPRI, P.O. Box 10412, Palo Alto, CA 94303
65.  Michael Hauhs, Institute of  Soil  Science and Forest Nutrition, Busgenweg  2, 34 Got-
    tingen, West Germany
66.  George M.  Hornberger,  University  of  Virginia, Department of Environmental Science,
    Clark Hall,  Charlottesville, VA 22903
67.  J.  W.  Huckabee, Project  Manager,  Environmental  Assessment Department, Electric
    Power Research Institute, 3412 Hillview Avenue, P.O.  Box 10412, Palo Alto, CA 94303
68.  Hans Hultberg, Swedish  Environmental Research Institute, P.O. Box 5207, Gothenburg,
    Sweden
69.  F.  A.  Koomanoff,  Director, Carbon  Dioxide  Research  Division,  Office  of  Energy
    Research, Room J-311, ER-12, U.S.  Department of Energy,  Washington, DC 20545
70.  David Lam, Environment of Canada, 867 Lakeshore Road, Burlington, Ontario, Canada
    L7R4A6
71.  Jeffrey Lee, USEPA, 200 SW, 35th Street, Corvallis, OR 97333
72.  Rick Linthurst, Kilkelly Environmental Associates, P.O. Box 31265, Raleigh, NC 27622
73.  John Malanchuk, USEPA, Acid Deposition Assessment Staff, RD-676, Washington, DC
    20460
74.  Egbert Matzner, University of Gottingen, Busgenweg 2, Gottingen, West Germany 0551
75.  Helen  McCammon,  Director,  Ecological Research  Division, Office of Health and
    Environmental Research, Office of  Energy Research, MS-E201, ER-75, Room  E-233,
    Department of Energy, Washington,  DC 20545
76.  Harold A. Mooney, Department  of Biological Sciences, Stanford University, Stanford,
    CA 94305
77.  Jan Mulder, Department of  Soil Science  & Geology, Agricultural University, Wagen-
    ingen, Duivendaal 10, Wageningen, The Netherlands
78.  Ingvar  Nilsson, Dept. of Ecology and Environmental Research,  Swedish University of
    Agricultural Science, Uppsala, Sweden S-75007
79.  Dr. Stephen C. Nodvin, University of California-Riverside, Route  1, Box 198, Mammoth
    Lakes, CA 93546
80.  Williams S. Osburn, Jr.,  Ecological Research Division, Office of Health and Environmen-
    tal Research, Office of Energy Research, MS-E201, EV-33,  Room F-216, Department of
    Energy, Washington, DC 20545

-------
                                           47

     81.  E. M. Preston, Corvallis Environmental Research Laboratory, U.S. Environmental Pro-
         tection Agency, 200 SW 35th Street, Corvallis, OR 97333
     82.  Irwin Remson, Department of Applied Earth  Sciences, Stanford  University, Stanford,
         CA 94305
  83-87.  John  Reuss, Department  of Agronomy,  Colorado State  University,  Fort Collins, CO
         80523
     88.  Gerald Schnoor, Engineering Bldg., University of Iowa, Iowa City, IA 52242
     89.  Hans M. Seip, Central Institute for Industrial Research, Forskningsv.  1, Box 350, Blin-
         dern, Norway
     90.  Richard Skeffington, Central Electricity Generating Board, Kelvin Avenue, Leatherhead,
         Surrey KT22, England
     91.  R. J. Stern, Director, Office of Environmental Compliance, MS PE-25, FORRESTAL,
         U.S. Department of Energy, 1000 Independence Avenue, SW, Washington, DC 20585
     92.  Arne O. Stuanes, Norwegian  Forest  Research Institute, P.O. Box 61, N-1432 AAS-
         NLH, Norway
     93.  Hans Vans  Grinsven,  Department of Soil  Sciences and Geology, Wageningen, The Neth-
         erlands
     94.  Helgan Van Miegroet, University of Washington, College of Forest Resources, AR-10,
         Seattle, WA
     95.  Ulf Wahlgren, Swedish Environmental Research Institute, Vikingavagen 5,  Sollentuna,
         Sweden
     96.  Raymond G. Wilhour, Chief, Air Pollution Effects Branch,  Corvallis  Environmental
         Research Laboratory, U.S. Environmental Protection Agency, 200 SW 35th Street, Cor-
         vallis, OR 97330
     97.  Frank J. Wobber, Division of  Ecological  Research, Office of Health and Environmental
         Research, Office of Energy Research, MS-E201, Department of  Energy, Washington,
         DC 20545
     98.  M. Gordon  Wolman, The  Johns  Hopkins University,  Department of Geography and
         Environmental Engineering, Baltimore, MD 21218
     99.  Robert W.  Wood, Director, Division of Pollutant Characterization  and Safety Research,
         U.S. Department of Energy, Washington, DC 20545
    100.  Dick Wright, University of Virginia, Clark Hall, Charlottesville, VA 22903
    101.  Office of Assistant Manager for Energy Research and Development, Oak Ridge Opera-
         tions, P.O. Box E, U.S. Department of Energy, Oak Ridge, TN 37831
102-128.  Technical Information Center,  Oak Ridge, TN 37831

-------