&EPA
United States
Environmental Protection
Agency
Environmental Research
Laboratory
Corvallis OR 97330
EPA-600/3-80-024
January 1980
Research and Development
Geological and
Hydrochemical
Sensitivity of the
Eastern
United States to
Acid Precipitation
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/3-80-024
January 1980
GEOLOGICAL AND HYDROCHEMICAL SENSITIVITY OF THE
EASTERN UNITED STATES TO ACID PRECIPITATION
by
George R. Hendrey
Brookhaven National Laboratory
Upton, New York 11973
James N. Galloway Stephen A. Norton Carl L. Schofield
University of Virginia University of Maine Cornell University
Charlottesville, VA 22903 Orono, ME 04443 Ithaca, NY 14850
Paul W. Shaffer Douglas A. Burns
University of Virginia University of Virginia
Charlottesville, VA 22903 Charlottesville, VA 22903
Project Officer
Charles F. Powers
Terrestrial Division
Corvallis Environmental Research Laboratory
Corvallis, Oregon 97330
This study was conducted by Brookhaven National Laboratory
and Associated Universities, Inc. for the U.S. Environmental
Protection Agency in cooperation with the U.S. Department of
Energy under Interagency Agreement EPA 79-D-X-0672.
CORVALLIS ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CORVALLIS, OREGON 97330
-------
DISCLAIMER
One of the principal reasons for the preparation of this report for the
Environmental Protection Agency was to supply scientifically valid information
which could be incorporated into the EPA S02-Particulate Matter criteria
document, presently in the final stages of preparation. A strict requirement
pertaining to that document is that any scientific information used there must
be published (or at least in press) by January 1, 1980. Because of this
demanding time constraint, it was necessary that the contractor prepare this
report in a shorter time than would ordinarily be attempted, and that it be
published by EPA without undergoing peer review. We feel that early publi-
cation of these results in order to stimulate the broadest scientific dis-
cussion prior to completion of the criteria document justified waiving our
normally 'more rigorous prepublication review requirements. Publication,
however, does not signify that the contents necessarily reflect the views and
policies of EPA, nor does mention of trade names or commercial products con-
stitute endoresement or recommendation for use.
11
-------
FOREWORD
Effective regulatory and enforcement actions by the Environmental Protec-
tion Agency would be virtually impossible without sound scientific data on
pollutants and their impact on environmental stability and human health.
Responsibility for building this data base has been assigned to EPA's Office
of Research and Development and its 15 major field installations, one of which
is the Corvallis Environmental Research Laboratory.
The primary mission of the Corvallis Laboratory is research on the
effects of environmental pollutants on terrestrial, freshwater, and marine
ecosystems; the behavior, effects and control of pollutants in lakes and
streams; and the development of predictive models on the movement of pollu-
tants in the biosphere.
Tins report assembled detailed maps of the eastern U.S., showing bedrock
geology and surface alkalinity, and examined the relationship of the two as
indicators of environmental sensitivity to acid precipitation. Biological
impacts of acidification on aquatic systems were also summarized.
Thomas A. Murphy, Director
Corvallis Environmental Research Laboratory
-------
ABSTRACT
A new analysis of bedrock geology maps of the eastern U.S. constitutes a
simple model for predicting areas which might be impacted by acid precipi-
tation and it allows much greater resolution for detecting sensitivity than
has previously been available for the region. Map accuracy has been verified
by examining current alkalinities and pH's of waters in several test states,
including Maine, New Hampshire, New York, Virginia and North Carolina. In
regions predicted to be highly sensitive, alkalinities in upstream si'tes were
generally low, <200 (Jeq/£. Many areas of the eastern U.S. are pinpointed in
which some of the surface waters, especially upstream reaches, may be sen-
sitive to acidification. Pre-1970 data were compared to post-1975 data,
revealing marked declines in both alkalinity and pH of sensitive waters of two
states tested, North Carolina, where pH and alkalinity have decreased in 80%
of 38 streams (p < 0.001) and New Hampshire, where pH in 90% of 49 streams and
lakes has decreased (p < 0.001) since 1949. These sites are predicted to be
sensitive by the geological map on the basis of their earlier alkalinity
values. Thus this mapping of sensitive areas is validated by the observed
temporal trends. The map is to be improved by the addition of a soils com-
ponent.
This study was conducted by Brookhaven National Laboratory and Associated
Universities, Inc. for the U.S. Environmental Protection Agency in cooperation
with the U.S. Department of Energy under Interagency Agreement EPA 79-D-X-
0672.
:LV
-------
CONTENTS
Foreword. ... ... iii
Abstract . . iv
Figures . vi
Tables. . . ix
Acknowledgement . . x
I. Introduction . . ... 1
II. Preparation of Sensitive Areas Map . . . .... 5
III. Verification of the Sensitive Areas Map. ... 27
IV Temporal Trends in pH and Alkalinity: Validation of the Sensitive
Areas Map. . . 58
V Impacts of Acidification on Aquatic Biota. . . 74
VI. Regional Assessment of Acidification Impacts on Fish Population . . 83
Endnote ... . . . ... ... . . 91
References. ... 93
-------
FIGURES
Number Page
Il-la Unmodified tracing of all boundaries of geologic formations shown
on a portion of the North Carolina state geologic map. 8
11-lb Boundaries of rock formations in North Carolina having comparable
acid-neutralizing capacities .... 9
II-lc Smoothing of geologic contacts among areas of differing acid-
neutralizing capacities shown in Figure II-lb. . . ... .... 10
II-2a Unmodified tracing of all boundaries of geologic formations shown
on a portion of the New York state geologic map. . .... . 11
ll-2b Boundaries of rock formations in New York having comparable acid-
neutralizing capacities ... 12
II-2c Smoothing of geologic contacts among areas of differing acid-
neutralizing capacities in New York State. . . . . . 13
CI-3a Unmodified tracing of all boundaries of geologic formations shown
on a portion of the Tennessee state geologic map in a region of
dendritic drainage in flat-lying rocks and trellis drainage in the
valley and ridge provinces . . . 14
II-3b Boundaries of rock formations in Tennessee, having comparable acid-
neutralizing capacities, in the same region as Figure II-3a 15
[I-3c Smoothing of geologic contacts among areas of differing acid-
neutralizLng capacities in Tennessee, shown in Figure II-3b. . 16
II-4 Predicted susceptibility of surface waters in six regions of the
eastern United States to acidification as determined by bedrock
type 17-23
11-5 High altitude lakes (elevation > 600 m) in the Adirondack Mountains
of New York which have lost their fish populations ... 24
[II-l Location map of Virginia showing the Shenandoah National Park, the
original survey area and the smaller study area 26
TII-2a Key to bedrock geology in the study area in and adjacent to the
Shanandoah National Park ... ... 30
-------
Figures (cont.)
Number
lll-2b Depiction of bedrock geology in the study area in and adjacent to
the Shenandoah National Park 31
lll-3a Key to pH of streams in the Shenandoah survey area. Shaded areas
represent watershed areas upstream from sampling locations ->^
III-3b Depiction of streams in the Shenandoah survey area .35
lll-4a Key to alkalinity of streams in the Shenandoah survey area. Shaded
areas represent watershed areas upstream from sampling locations . . 36
III-4b Depiction of alkalinity of streams in the Shenandoah survey area . 37
iII-5 Changes in water chemistry in White Oak Run during the storm of
July 16, 1979. ... 39
1II-6 Geologically-based sensitive areas map of New Hampshire 41
1II-7 Geologically-based sensitive areas map of North Carolina 42
1II-8 Location and alkalinity of New Hampshire sites sampled during the
summer of!979 . 44
III-9 Location and alkalinity of North Carolina sites sampled during the
summer of!979 45
111-10 Location of New Hampshire sites where pH and alkalinity data was
available for the period 1976-1979 47
III-ll Location of North Carolina sites where pH and alkalinity data was
available for the period 1976-1979 48
111-12 Minimum alkalinity of counties in New Hampshire for the period 1976-
1979 49
111-13 Minimum alkalinity of counties in North Carolina for the period
1976-1979. . . . . . .50
111-14 Average alkalinity of counties in New Hampshire for the period 1976-
1979 .... . .... 51
111-15 Average alkalinity of counties in North Carolina for the period
1976-1979 . . . . . 52
1V-1 Distribution of alkalinity in samples from North Carolina fisheries
data, 1960-1963. . . . 63
1V-2 Distribution of alkalinity in samples from North Carolina field
trip, 1979 64
vii
-------
Figures (cont.)
Number Page
IV-3 Distribution of pH in samples from North Carolina fisheries data,
1960-1963 65
1V-4 Distribution of pH in samples from North Carolina field trip, 1979 . 66
IV-5 Comparative plot of alkalinity for North Carolina streams, 1960-
1964 versus 1979 67
IV-6 Comparative plot of pH for North Carolina streams, 1960-1964 versus
1979 68
IV-7 Distribution of pH in samples from New Hampshire fisheries data
1937-1939. . 69
IV-8 Distribution of pH in samples from New Hampshire field trip 1979 70
TV-9 Comparative plot of pH for New Hampshire streams, 1937-1939 versus
1979 71
IV-10 Calcium versus pH for headwater streams in North Carolina and New
Hampshire plotted over Henriksen's line. . . 73
VI-1 Acidification of Adirondack lakes in relation to pH measured in
1975 . . . 86
VI-2 Acidification of Adirondack lakes in relation to S04 measured in
1975 ... . . . . 87
VI-3 Nomograph for predicting lake pH and fish population status from
lake Ca levels and either regional acidification or precipitation
pH . . . . . 89
VI-4 Changes in the H+ ion concentrations in 36 Adirondack lakes from
the 1930's to 1975 in relation to Ca levels measured in 1975 . . . 90
V1L1
-------
TABLES
Number Page
11-1 States and scales of geologic maps used for preparing the maps of
sensitive areas. . . 7
III-l pH and alkalinity of streams in major bedrock formations of the Blue
Ridge in Central Virginia. ... ... . . 28
1II-2 Comparison of the geologically based map of sensitive areas (Figure
III 7) to county-by-county maps of alkalinity (Figures 111-13 and
III 15), for 49 of 100 North Carolina counties 55
JII-3 Comparison of the geologically-based sensitive areas map (Figure
III-6) to county-by-county maps of alkalinity (Figures 111-12 and
III 14) for all 10 counties in New Hampshire ......... . 55
1II-4 Comparison of the geologically-based binary number map of sensitive
areas (Figure II-4b) to county-by-county maps of alkalinity (Figures
111-13 and III 15), for 49 of 100 North Carolina counties 56
JlI-5 Comparison of the geologically-based binary code sensitive areas map
(Figure II-4a) to county-by-councy maps of alkalinity (Figures
111-13 and 111-15) for all 10 counties in New Hampshire 56
]V-1 A comparison of old and new data for headwater streams in North
Carolina ... . .... . . . .... 60
1V-2 A comparison of old and new data for New Hampshire streams and
lakes . 61-62
V-l Damages to aquatic biota likely to occur with increasing acidity . 82
V-2 Summary of damages to aquatic organisms with decreasing pH . . . 82
VI-1 Lake classification for Adirondack Mountain lakes on the nomograph
of Henriksen ... . .... . .... . . 91
VI-2 Fish population status in 24 Adirondack Mountain lakes, classified
according to the nomograph of Figure VI-3 . ..... 91
IX
-------
ACKNOWLEDGMENTS
The authors wish to thank Avis Newell for technical work in the field and
laboratory; Fred Liotta, Susan Hall, Lisa Thurlow, Roland Dupuis and Marilyn
Morrison for assistance in map preparation; and Ronnie Evans and Charlie Bores
for data base management, data reduction, and computer graphics. We apprec-
iate the generous cooperation and assistance by personnel of the Shenandoah
National Park, the North Carolina Department of Inland Fisheries, and the New
Hampshire Department of Fish and Game.
-------
I. INTRODUCTION
In recent decades, man's activities have led to significant changes in
atmospheric and precipitation chemistry. Among these activities, metal smelt-
ing and combustion of fossil fuels have resulted in atmospheric emissions of
sulfur and nitrogen oxides. The effect of these emissions has been a dramatic
increase in the acidity of rainfall, particularly in areas downwind of major
industrial areas. Although the phenomenon of acidic precipitation has been
developing on a hemispheric scale, only recently have workers tried to iden-
tify those areas in the United States receiving acidic precipitation1 and
associated heavy metals.67 While a reasonably good understanding of the
sources and distribution of acid precipitation has been developed,16'18'19'27'
/i>72>77 QUr un(}erstanding of the effects of acidic precipitation on terres-
trial and' aquatic systems is still fragmentary. Reductions of pH and alka-
linity of lakes affected by acidic precipitation have been well documented,
particularly in the northeastern United States, southeastern Canada, and
Scandanavia. In these same areas, declines in fish polulations and adverse
effects on other aquatic fauna have been widely observed.1'23'37'38'41'42'49'
57)68)86)91)92)103)104
Studies of the hydrobiological problems caused by . acidic precipitation
have been primarily qualitative. Biological aspects of the synoptic lake
surveys in Sweden,1 Norway,49 Canada,18 and the United States90 have concen-
trated on changes in presence or absence of fish and in the kinds and numbers
of species. Very little quantitative information is available concerning the
effects of acidification on biomass. This is especially true for the bac-
teria, and other microorganisms. There is strong evidence indicating that
processes such as phytoplankton production and microbial decompostion are
inhibited, but quantitative data are scarce. Changes in the availability of
inorganic nutrients and in nutrient recycling are hypothesized but have not
been demonstrated; effects on food chains or webs are not well known, and no
quantitative data are available concerning the effects on energy processing in
natural aquatic ecosystems.
The overall objective of the sensitive areas project has been to evaluate
the eastern United States and determine and map which areas are vulnerable to
adverse impacts from acid precipitation. In addition, a review of the bio-
logical consequences of freshwater acidification is presented and a regional
assessment of effects on fish, which provides a basis for predicting such
impacts, is described.
A. Geologic Controls
Impact to aquatic ecosystems is largely based on the chemical charac-
teristics of the bedrock. Limestone terrains yield "infinite" acid-neutral-
izing (buffering) capacity to acidic precipitation whereas granites (and
1
-------
related igneous rocks), their metamorphic equivalent, and noncalcareous sand-
stones yield minimal buffering. Several hundred common minerals comprise the
huge variety of rocks which comprise the crust of the earth, the surficial
overburden, and soils. Of interest in terms of susceptibility of a landscape
to acidification from precipitation is the total capacity of the minerals, no
matter what their configuration, to assimilate protons (H+ ). Specific
considerations are solubility of the phases and the kinetics of solution.
Certain minerals are soluble and rapidly dissolved but yield no buffering
capacity, e.g. :
NaCl .. - Na+ .+ Cl- K = 10+1-58 = [Na+] [C1-] (1)
xl aq aq sp
Others are relatively insoluble and yield no buffering capacity, e.g.
Si02 n + 2H20 = H4Si04 K = 10-4-° = [H4Si04] (2)
^xl ^ aq 4 4aq sp * *
quartz n
Common rock-forming minerals, when placed in water, yield pH's in excess
of 7 (commonly in excess of 8) but the kinetics are sufficiently slow so that
the total potential H+ consuming capacity of these minerals is not realized,
e.g.:
2NaAlSi308 , + 2H+ + 9H20 = (3)
Xlalbite aq aq
= 2Na + Al2Si205(OH)4xl + 4H4Si04
H kaolinite
,
K ~ 10
sp
[Na+]2/[H+]2
H4Si04
Assuming an initial pH of 4.0 and buffering from the C02 - H20 system (HC03 £
10-5*7), the expected pH after complete reaction would be approximately 12.
This is never achieved.
Additional H sinks available in soils include Al and Fe hydroxides and
silicates, e.g.:
Al2Si205(OH)4 . +6H+ = 2A1 +3 + 2H4Si04 + H20 (4)
t /: t>v 4xl. n. .. aq aq * *aq * aq
kaolinite M H 1-1
AT ^u% + 3H+ = A1+3 + 3H20 (5)
Al(OH)3xl aq aq * aq
gibbsite
FeO(OH) , + 3H+ = Fe+3 + 2H20 (6)
Xl ..u-4- a°t a1 a(i
geothite ^
-------
These minerals are either developed in sub-soils (B-horizon) in young soils
(and thus not necessarily available to overland flow) or dominate in mature/
old soils where they may provide extensive acid neutralizing capacity.
The most important mineral for neutralizing acidic waters is calcite (or
nearly any other carbonate mineral). The solution at low and at intermediate
pH of this mineral is given by:
CaC03 + 2H+ = Ca++ + H?COo at low pH and (7a)
0 aq aq z Jaq
2CaCOo + 2H =2 Ca+ + 2HCOq at intermediate pH. (7b)
aq aq aq
Further addition of acid to the resulting (7b) solution (2Ca++ + 2HC03)
will result in the protenation of the HC03 :
HC03 + H+ -> H2C03 . (8)
J aq aq aq
These reactions (7 and 8) are rapid and for each mole of CaC03 consumed, 2
moles of H are consumed before the pH is reduced significantly below 5.
Reactions 4, 5, and 6 or similar reactions consume large amounts of H ions at
pH's below 5 and may be the primary acid-consuming mechanism for very acidic
solutions. They do not contribute HC03- for buffering except by virtue of the
dissociation of H2C03 as the pH is raised by the reaction:
H2C03 -* H+ + HC03 .
aq aq aq
The amount of HC03- produced will be a function of the total C02 in the
system. It should be noted that the gain or loss of molecular C02 to the
water, by itself, with no addition of cations to balance HC03- production,
will not change alkalinity although pH does change.
Thus we have a spectrum of re.sponse of minerals (and rocks) to the
changing acidity of atmospheric precipitation. Accordingly, rocks may be
classified by the buffering capacity or acid-neutralizing capacity which they
render to surface waters.
Additional controls on acidification of aquatic ecosystems include:
hydrologic characteristics of the terrain (i.e., overland flow versus ground-
water flow, soil porosity/permeability, residence time of water in the soil,
distribution of precipitation through time, type of precipitation, thickness
of soil), types of soils [residual, glacial (till, ice-contact stratified,
etc.), aeolian, lacustrine, alluvial, etc.], mineralogy and age of soil.
Any prediction about the vulnerability of a terrain to acidification from
precipitation based solely on bedrock geology must be tempered with considera-
tion of these other factors. For example, Florida is underlain by highly
calcareous and phosphatic rocks suggesting that acidification of lakes and
streams is highly unlikely. However, many of the soils (particularly in
northern Florida) are very mature, highly leached of CaC03, and as a result,
3
-------
acidification of some lakes with minimal deep groundwater influx has occurred
(P. Brezonik, pers. comm. 1979). Conversely, there are some areas in Maine
underlain by granite and receiving precipitation with an average pH of about
4.378 where lakes have not been acidified because of lime-bearing till and
marine clay in the drainage basins of the lakes.
B. Mapping of Areas Sensitive to Acid Precipitation
Other investigations33'64'71 have defined areas of the United States and
Canada as susceptible to impacts from acidic precipitation, from interpreta-
tions of large-scale geologic maps. These areas were largely concentrated in
Precambrian terrains (the Canadian shield, the Adirondack Massif), the
Precambrian core of the Appalachian Mountains, and in younger igneous rocks of
the Northwest and Rocky Mountains.
In the United States surface waters acidified from atmospheric deposition
have now been identified from New England and New York to Florida, and in the
Boundary Waters Canoe Area, Minnesota. Acidic precipitation exists (with pH
< 5.6) in all states contiguous to the Mississippi River and east to the
Atlantic Ocean78 as well as in California72 and in Washington.6 Because of
this widespread aspect of acidic precipitation, it is important to understand
the natural characteristics of the landscape which render an area's aquatic
ecosystems susceptible to impact. Three major factors can be identified as
important. They are: meteorology, pedology, and geology. This report
stresses the geological control of sensitivity.
C. Water Quality Data Collection
The utility of the geologically-based maps of sensitive areas is limited
by the accuracy with which they identify areas in which surface waters are
actually sensitive. We define sensitive waters as those with alkalinity
values less than 500 (Jeq (micro equivalents) of alkalinity per liter. The
accuracy of the maps was checked agains actual measurements of lake and stream
water chemistry.
Data from existing records in many states of the eastern U.S., from the
STORET system, and from field measurements conducted in New Hampshire and
North Carolina in 1979 are combined in the Acidification Chemistry Information
Database (ACID) at Brookhaven National Laboratory. Data is registered by
state, county, station code, station name, data source code, latitude, longi-
tude, and sample date. Data sets include pH, alkalinity ((Jeq/1) , Ca (fjeq/1) ,
conductivity (microSiemens), color (mgPt) and fish status. Information on
methods used for chemical determinations is also stored in ACID.
-------
II. PREPARATION OF SENSITIVE AREAS MAP
An analysis of the vulnerability of aquatic ecosystems to acidification
from precipitation must start with the bedrock geology. The scale of vari-
ability of rock types is such that one must literally look at the geology on a
drainage basin basis to make predictions about individual lakes and streams.
Igneous rocks normally have maximum dimensions less than 10 km. Folded/
faulted metasedimentary/sedimentary rocks may have essentially 1-dimensional
distribution. Flat lying sedimentary rocks may be widely distributed but
topographic relief commonly intersects many rock types over short distances.
Small amounts of limestone in a drainage basin exert an overwhelming
influence on water quality in terrains which otherwise would be very vulner-
able to acidification. Consequently, in the regional analysis of vulner-
ability, areas with intimate mixtures of rocks of varying acid neutralizing
capacity have been classified according to the more influential rock types.
Analysis of geologic terrain has been undertaken on the scale of the most
recent state geologic maps (Table II-l) Rock formations have been classified
according to their potential buffering capacity. This judgment was based on
map explanations, various stratigraphic lexicons61'62 and S. Norton's personal
knowledge of the geology of areas.
Rock classification was as follows:
Type I - Low to no buffering capacity (labeled 1 on maps)
(Widespread impact from acidic precipitation expected.)
Granite/Syenite or metamorphic equivalent
Granite gneisses
Quartz sandstones or metamorphic equivalent
Type II - Medium/Low buffering capacity (labeled 2 on maps)
(Impact from acidic precipitation restricted to first and second
order streams and small lakes. Complete loss of alkalinity
unlikely in large lakes.)
Sandstones, shales, conglomerates or their metamorphic equivalent
(no free carbonate phases present).
High grade metamorphic felsic to intermediate volcanic rocks.
Intermediate igneous rocks.
Calc-silicate gneisses with no free carbonate phases.
Type III - Medium/High buffering capacity (labeled 3 on maps)
(Impact from acidic precipitation improbable except for overland
run-off effects in areas of frozen ground.)
Slightly calcareous rocks.
Low grade intermediate to mafic volcanic rocks.
Ultramafic rocks.
Glassy volcanic rocks.
-------
Type IV - "Infinite" buffering capacity (labeled 4 on maps)
(No impact to aquatic ecosystems.)
Highly fossiliferous sediments or metamorphic equivalents.
Limestones or dolostones.
The development of vulnerability maps for 3 types of terrains is illus-
trated in several steps. Figure Il-la represents, unmodified, a tracing of
all boundaries of geologic formations shown on part of the North Carolina
state geologic map. This area is underlain by Precambrian and possibly lower
Paleozoic rocks which are highly folded, faulted and metamorphosed in the core
of the Appalachian Mountains. Figure Il-lb differs from Il-la in that bound-
aries between contiguous formations with comparable acid-neutralizing capaci-
ties are not shown. Figure II-lc is modified from Il-lb by the smoothing of
contacts, favoring rocks with higher buffering capacity, and the deletion of
small areas or linear outcroppings of low buffer capacity rocks.
Figure II-2a represents, unmodified, a tracing of all boundaries of
geologic formations shown on part of the New York state geologic map. This
Precambrian terrain is characterized by high grade metamorphic rocks intruded
by many igneous rocks, largely granitic. Figures II-2b and II-2c are devel-
oped by the same techniques as Figures Il-lb and II-lc.
Figures II-3a, b and c represent similar development for a portion of the
state of Tennessee in an area of dendritic drainage in flat-lying sedimentary
rocks (western half) and trellis drainage in valley and ridge topography in
the highly folded Paleozoic section (eastern half).
The state geologic maps already represent a smoothing and generalization
of geologic boundaries. Most state maps utilized in this synthesis have been
drawn at a scale of 1:250,000 or 1:500,000 from maps at a scale of 1:24,000 or
1:62,500 with the loss of much small detail. At a scale of 1:250,000 the
largest geologic unit that is commonly deleted is approximately 0.2 miles in
diameter or width.
"Zero"-dimensional rocks of this size (generally intrusions or erosional
remnants of flat lying strata) will not normally affect the chemistry of a
large number of streams or lakes. However, one-dimensional rock map units
(generally the intersection with the land surface of moderately to steeply
dipping strata or dikes or metamorphic equivalents) commonly affect large
numbers of primary and secondary streams and lakes which lie in topographic
lows occupied by certain strata. Thus, thin limestones in the Valley and
Ridge topographic province (Figure II-3a) may not be shown on a scale of
1:500,000 (and thus not on these maps) but may dominate the area's water
chemistry and vulnerability to acidic precipitation.
The geologically based sensitivity maps may have also been organized on a
county basis to make these data compatible with existing data bases for soils,
agriculture, forestry, and other land uses. (A new soils map, being prepared
by a separate project74 will also be organized on a county basis so analyses
of county susceptibility to acidification of surface waters may be undertaken.)
Each county in the study area states (Table II-l) was analyzed by planimetry
for percentage of area underlain by rock types I and II. These data are shown
for the eastern United States on Figures II-4. Although individual drainage
6
-------
Table II-l.
States and scales of geologic maps used for preparing the maps of
sensitive areas.
Alabama
Arkansas
California
Connecticut
Delaware
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
1:500,000
1:500,000
1:750,000
1:500,000
1:300,000
1:3,125,000
1:500,000
1:500,000
1:1,000,000
1:500,000
1:500,000
1:1,250,000
1:500,000
1:250,000
1:250,000
1:2,500,000
1:1,000,000
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Oregon
Pennsylvania
Rhode Island
South Carolina
Tennessee
Vermont
Virginia
Washington
West Virginia
Wisconsin
1:500,000
1:500,000
1:250,000
1:250,000
1:250,000
1:500,000
1:500,000
1:500,000
1:250,000
1:250,000
1:250,000
1:250,000
1; 250, 000
l': 500, 000
1:500,000
1:250,000
1:1,000,000
-------
0
35
20 kilometers
81
SOUTH CAROLINA
Figure Il-la. Unmodified tracing
of all boundaries of geologic
formations shown on a portion of
the North Carolina state geologic
map. Numbers refer to buffering
classification. Dashed lines are
county boundaries.
-------
--36
35
20 kilometers
SOUTH CAROLINA
' Figure Il-lb. Boundaries of rock
I formations in North Carolina
1 having comparable acid-neutraliz-
i ing capacities. Numbers refer to
buffering classifications. Dashed
lines are county boundaries.
-------
36 +
35
81
SOUTH CA^.NA
2O kilometers
_>_>
I Figure II-lc. Smoothing of geo-
I logic contacts among areas of
/ differing acid-neutralizing capa-
| cities shown in Figure II-lb
Numbers refer to buffering class-
"""
10
-------
1O kilometers
Figure II-2a.
Unmodified tracing of all boundaries of geologic formations shown on a portion of the New
York state geologic map. Dashed lines are county boundaries.
-------
75°15
43»30'
NEW YORK
_10 kilometers
Figure II-2b.
Boundaries of rock formations in New York having comparable acid-neutralizing capacities
Dashed lines are county boundaries.
-------
Figure II-2c. Smoothing of geologic contacts among areas
of differing acid-neutralizing capacities in New York
State. Areas of the Adirondack Mountains, shown in Figure
II-2b, are located in the center of the northeastern third
of the state. Numbers refer to buffering classification.
Dashed lines are county boundaries.
-------
35°30
TENNESSEE
35°00'
35°30'
m
z
z
m
co
t/i
m
m
3 5° 00-
86°00'
ALABAMA
GEORGIA
85° 15'
Figure II-3a. Unmodified tracing of all boundaries of geologic formations shown on a portion of the
Tennessee state geologic map in a region of dendritic drainage in flat-lying rocks and
trellis drainage in the valley and ridge provinces. Dashed lines are county boundaries.
-------
35°30
TENNESSEE
UJ
UJ
\f)
cO
UJ
z
z
UJ
35°00'
35*30'
35° OO'
86°OO'
ALABAMA
0
.Smites GEORGIA
85*15'
Figure II-3b. Boundaries of rock formations in Tennessee, having comparable acid-neutralizing capac-
ities, in the same region as Figure II-3a. Numbers refer to buffering classification.
Dashed lines are county boundaries.
-------
35°30
TENNESSEE
35°00'
35°30'
35°00'
86° 00'
ALABAMA
0
Smiles GEORGIA
85°15'
Figure II-3c. Smoothing of geologic contacts among areas of differing acid-neutralizing capacities in
Tennessee, shown in Figure II-3b. Numbers refer to buffering classification. Dashed
lines are county boundaries.
-------
Figure II-4. Predicted susceptibility of surface waters in six regions of
the eastern United States to acidification as determined by
bedrock type. Data are organized on a county-by-county basis.
The first character of the 2-digit code represents the decade-
percent of Type-I rock (0 = 0-9%, 1 = 10-19%, 2 = 20-29%, etc.)
and the second character represents the decade-percent of rock
type II. The states included are (a) Maine, New Hampshire,
Vermont, Massachusetts, Rhode Island, Connecticut, and New
York; (b) Pennsylvania, New Jersey, Delaware, Maryland, West
Virginia, Virginia, and North Carolina; (c) Minnesota (exclud-
ing western-most counties), Wisconsin, Michigan and Iowa; (d)
Missouri, Illinois, Indiana, Ohio, and Kentucky; (e) Arkansas,
Mississippi, Alabama, and Tennessee; and (f) South Carolina,
Georgia and Florida.
17
-------
CO
-^03
^
i04 "I '24
l^^M-'o,:,'"-' I .'""Y 5 '! ,«-"«i'/isl₯^/-.5!z;J< ^r'&^'^»
Figure II-4a
-------
r;"
r
-0-V---! ".-^qj.,, i ., '; T°,4o
CR.^OBO | WABRIN p MC KE»N 1 " ! _ _-_--! - /
(,_ .»..-. ^.^i...^ .^ "o i 3 - -j>"-A i ''/X
K^i^^Fvj0^! ^r-;-;rc.Lc -0;4r^?^\i ^;,^
u* --fv-o. r0r-v.-----^»:-v---;;.;VL-^I^;.^^^
, ,,-.. °i?j
, ,j- i ' ^*fNt i
- -T^i^\ \"'"; \^_
[ ;-*.' iV-./1- ry^-.""",-^; .--r-.r /-s^«^ Moe />~..: i?v<
k~-"-' y -l^"10"; ^'f ,03 ,r"Co6"--s i:::,,'-}-- \ >'-;,s&*<> ^.-.rr.V \ /> Tvr
^ ^'.UTLt.iV ,Jo'" i ' 3^ SitNTBE^ ^:"/.«Hi^ ,->y^^ r-;«° a ".A«,-',« l?'*;S
^ jr. ff / ^--JN"TJ-Y" ^ rS'^^^^%5^
lS\-*r vr'/./."l°"~V;v>'~/ ( ^~'"A--U^-^ ' n4*':"< ' _/' -os'- 4>>"»»"/|0^y ^,
^%,^j^.o'--'-;.".'.'ew-^-'i v,--:> .,..ST ej^^/^6
)i| *., ~, t ' " /., '--,-- /O 2 O H , ^ 'r-lAjMARRISBUPrti ^- X '°"«, THE WTO»,*J ^ ^ . ,
/l,.>\hr^--/ --..j/ >.'~--S:v ._ ^.^s^-^^t
/.,."-- x- 27 _ -, ,5LJ"LSLP_9_ _*>-:' oV"
.; ^s^i\ ""'.'"- -rl.,.^Mi ,.. in?..
Figure ll-4b.
19
-------
r-o
o
-_J-\
p^-
1 \
ry \
*tes
J F & ' f-A'f ,_
-_.'' ^ , Sftr .
7.*"-;> - t'jf-.-1'h I 0.9
%./» ^C"'- - . .I-; oouc^
;1: *V .-'/(. - . ,Ufl I /
1 ol 36 jjj,
j.^.L: J-q-
-i. . 4tK Y""."
' L-V*
| ^H1PP {»!,
':;.*&"" r" s-.,^,-^f";,,;.-; ^9'"i "^rM. 09,1 ; ^v^B'ic-,-0,-i> 6» ' t u^----- i <^f^^>f^"J'K^^---vt*\>*-r^.
K^-.JJN FiJ k'-&:&-& A" L^L{:.:r>.y«Jr,;^:^^ :" l:^^^ - T^A' 1* /^V,^^^
«^-i-2Kii"w=- fi ^r ^^!«^^^ *s^ S-.
S%i^-:V^^^B^^ in '<^r4^u. ^
«5-pT<^^^^\:]&r^v^\Vo>'^ »"-^ rX-^-4-'1'" * \ ; V"""""?/>-^- ^ -,*;R-f J-r'^
:^]:i:SfSSpf ?_^ ^ ^tet^r^-
v^ ''i .jj\^r.'.»:,".,vr N^-^-f.r ^ ir «a.',i .-« ,-L,-V-1--,VT^5-,-r°, "i :»..^!''^ f./'4^C7&f &"-;' i-^ ...*»« t1 -X
-VeSS ! %,'f 'iV,;!.:?*j'-^'li/Jif;«IB, , ;; !-;^:- i -,-^r! . J* ->44^i -; '^J> /=7 ^.^;l^MF=^i-'-;r--^Sfo "\
^
.^««-. -,
IX r- J '
^JA^ O*
Tr.
------ " ""
!^;;t;kS:^;.^r i^[ri'7^cH^Q:'J-r[ 5;~"~1rH°''u"i-J olTi. i^:/^-"
^-----t-^.4-----u.^h~--v--r-v-^4-r::j««.ri^y^>^^ i~~v.t-t3
*" o'^,',": i "' f'^,7.! ." '! """1h"Jd* i;-»":Me"""-L._ ._,.i f -^- ' °4..- i -r:^k.. i'f""°"!-^«HI*;»^w.uK
V--^-i^i;$;t.:.ri:±i^p/i----|---^^ [^tr^i ^h^^ait-^J
? "~.. °1 ! A.^J X I.OM^L^KSOH-S.
-'".», i «,,,,. i >,.. ;.-,,., K_^i:»±-__l
"T
-T-' 1 tr--
""i w.,, -i 01 i e's,
"- _ 1 ' i*? " '" ^ "' l Srs Ml , ! """""""I
^j " "~'"V~r'51~~f +---i"-^K>fE-^ i,-v; r--
r"/"ss; .r.''°i».o,sc'»;-)"«~|;:.rrf"H*s'"1'"
* f - r "'-^- -J -r-- r-J -r-J -r -i'"->-'~::^
l.,>t|,^ ;;_ ;:!_ | r___
j 77-4--'----1 - - A- -> -- + :
.*f"fON-r1 ,,,,_, I i ', , ,^_ J '
Figure ll-4c
-------
r;;- -
"^'^"X \ /,.^?V~ V .-' o-C^
'"> *--fsto^\' x^\ Ai'C"5/ ^ |,,"""^'-(1
^>* i'T^^^-v-S^-^^x
S5^^-fc^'^s^-:^v^"?i
--' -1r4'""^"!\Q ^--c-i -,....^-i-^»f' a»,^vj^.so.Hvo1
Figure II-4d
-------
.f'.lrTTJ
_, n ,^-J,^ry"' J ^...'^i !
"T^ A cu.la*M*"'^ jS7 .»«»!- - "-x^hi.r /"">.
, -^mu^ -^ ..^#f/^.^_ « ' /V»V^^
1 ^ SV' rurti*1
»".'i/.-£k./
- --,- , ^?^$r";:o°91 o6seLL^
^ r^:o..i.oz i ;u;;;;;r-x./
^H^-T-1-; r-i _p ^r-1 v)
^-.rI'/T-.r-f^
J El MOBILE ^RV: \
Jl:^f . \
^_iii2iil-T--N^
Figure II-4e
-------
,~o5"7"~
", ^r
\;i/ SA*.,A " '...,,.
V!?| ROSA go1 -' '
*- ; i " ' j-r^
^TTZ-
,i -1 -- itxi .
\- - ' _ \'HJ Iftt'TtitH
PAi.!"], BTACH
C O L L '
1 D ** " fi C
Figure II-4f.
23
-------
NEW YORK __ _.
NO
-P-
NEW YORK
10 kilometers
Figure II-5- High altitude lakes (elevation > 600 m) in the Adirondack Mountains of New York which
have lost their fish populations (A) are shown superimposed on a map of rock-types I and
II. Lakes which still have fish are indicated (P).
-------
systems or lakes cannot be analyzed in this fashion, regions of high vulner
ability can be identified. The first digit of the binary number for each
county represents the percentage (0-9% = 0, 10-19% = 1, etc.) of rock type I
in the county. The second digit represents a similar decade value for rock
type II. For example the number 23 indicates that 20-29% and 30-39% of the
county is underlain by rock types I and II, respectively Rock types III and
IV are not represented in these maps.
The predictive accuracy of the maps can be established by comparing the
geology-based maps against surface water chemical data in areas receiving
acidic precipitation as discussed in Section III of this report or by looking
at the biological response of aquatic ecosystems to acidic precipitation. A
number of lakes in the Adirondack Mountains have been examined in terms of
lake pH and fish populations,92 as discussed in Section VI of this report, and
a close correlation between the two was found. Figure II-5 depicts those
lakes which have lost their fish populations and the rock type underlying the
area. A geographic correlation exists between fish absence and the existence
of rock types I and II in this region. Topography and orographic precipi-
tation are factors contributing to the distribution of lakes in which fish are
present or absent.
Correlations also exist between acidification of surface waters and
bedrock geology (Section III) in a number of states and areas which have thin
soils, including New England.22 This suggests that bedrock geology exerts a
very strong influence on the regional extent of acidification of aquatic
ecosystems in response to atmospheric loading of acid. Regionally, soils and
vegetative types are of secondary importance. However, locally, soils may
overwhelm bedrock influences.
25
-------
Shenandoah
National Park
Figure III-l.
Location map of Virginia showing the Shenandoah National Park, the original survey area
(large, lightly stippled area), and the smaller study area (darkly shaded). The small
study area is the same area as shown in Figure III-2 to 111-4.
-------
III. VERIFICATION OF THE SENSITIVE AREAS MAP
The usefulness of the geological delineation of sensitive areas is wholly
dependent upon the accuracy with which the geological mapping identifies those
areas correctly. To verify the accuracy of the maps, three types of studies
were conducted covering geographic regions on three scales. First, an inten-
sive investigation of small headwater streams along a seventy-five mile
stretch of the ridgeline of the Blue Ridge Mountains in Shenandoah National
Park, Virginia was conducted. Streams were sampled over a variety of geo-
logical bedrock types. This work was to verify the assumption that stream
water chemistry is strongly influenced by bedrock in regions with thin soils.
Second, extensive surveys of low alkalinity lakes and streams in New Hampshire
and North Carolina mountainous regions were conducted in the period June-
August 1979 so that two sets of highly reliable water quality data would be
available. Stations were sampled which were located in regions expected to be
sensitive and for which we could obtain data taken prior to 1970. These data
sets were then to be used to test the map. The third method of verifying the
maps was a geographically large-scale comparison between stream water quality
derived from Federal, State and local records and the sensitivity predicted by
the maps.
A. Intensive Study
Within the overall context of the sensitive areas project, this section
focuses on the Blue Ridge of central Virginia.93 Within this area, pH and
alkalinity were measured in headwater streams, and bedrock geology and soils
data (where available) were determined from the literature. This work repre-
sents both an integrated example and a test of the sensitivity model described
in other portions of this report. By working on a small scale, local varia-
tion in bedrock geology or soil properties can be considered. In addition, by
sampling extensively within a small area, one can determine variability of
stream chemistry within a single geological formation, while repeated sampling
allows one to assess the stability of individual streams and their response to
precipitation events.
1. Site Description.
The study area was located in the Blue Ridge of central Virginia (Figure
III-l). Sampling sites extended over approximately 75 miles along both sides
of the ridgeline, with major effort centered along a 20 mile section at the
southern end of the Shenandoah National Park. With the exception of data
presented in Table III-l, discussion will be limited to work carried out
within the smaller area (hereafter referred to as the study area). Chemistry/
bedrock trends observed within the large area were consistent with those
occurring in the study area. Within the study area, the Blue Ridge occurs as
27
-------
Table III-l.
N3
00
pH and alkalinity of streams in major bedrock formations of the Blue Ridge in Central
Virginia.
Formation
Pedlar
Catoctin
Lovingston
Marshall
Old Rag
Antietam
Hampton +
Antietam
various
Major Rock Type
grandodiorite
meta-balalt
biotite quartz
monzonite gneiss
biotite quartz
monzonite gneiss
granite
quartzite
quartzite,
shale
limestone,
dolomite
Age
pc
PC
PC
pc
pC
C
C
0
Number of
Samples
22
18
6
3
9
10
19
5
pH
mean
6
7
6
7
6
5
5
8
.9
.0
.8
.0
.5
.7
.7
.1
5.
6.
6.
6.
5.
5.
5.
7.
Alkalinity (|Jeq/l)
range
8- 7
5- 7
25-7
8 -7
7 -6
25-6
1 -6
65-8
.7
.65
.15
.0
.8
.1
.3
.3
mean
188
265
227
254
96
7
11
3600
range
40-
115-
198-
205-
39-
380
655
270
305
195
-10-17.5
-5-
35
3360-3950
, For a better description, refer to text.
pC = Precambrian, C = Cambrian, 0 = Ordivician.
Number refers to number of sampling locations. Where samples were taken at one location on more than
one date, an average pH and alkalinity for that location are used. Streams passing through more than
one bedrock formation are not included in these calculations, except for streams passing through both
the Hampton and Antietam formations.
1 [Jeq/1 = 0.05 ppm CaC03.
-------
a single major ridge line 2500-4000 feet above sea level, with smaller discon-
tinuous ridges to the east and west. The western slopes are drained by
numerous small streams which feed the south fork of the Shenandoah River,
while the eastern side is drained by three river systems, the north and south
forks of the Moorman's River and Doyles River. Oak-hickory forest covers most
of the study area, with pine and hemlock occurring in many of the deeper
stream valleys. Nearly all of the sampling stations are in or adjacent to the
Shenandoah National Park, and much of the area is designated as wilderness;
thus human impact on stream chemistry is negligible.
Soils within the study area fall into two groups; western slopes have
coarse thin soils formed from weathering of sedimentary bedrock, while kaolin-
itic clays are present in large quantities in soils of the eastern slopes. In
both cases, soil pH is low (4.5-5.5), and cation exchange capacity, percent
base saturation, and organic content are low in all cases.5'75'87
Bedrock geology in the study area (Figure III-2a, 2b) and other areas of
the Shenandoah National Park is based on the work of Gathright,36 while bed-
rock in other parts of the survey area are based on Werner97 and Bartholomew.
Within the study area, five major bedrock formations occur; these are listed
below starting with the oldest and eastermost formation:
1. Pedlar Fm -- Massive or layered granodiorite or quartz monzonite.
Usually fine to medium-grained, composed largely of plagioclase and
plagioclase-feldspars and quartz.
2. Catoctin Fm -- Thick (2000 ft) beds or flow of metamorphosed basalt,
composed primarily of feldspar in a matrix of chlorite, actinolite,
epidote and pyroxene. Commonly includes cavities containing quartz and
calcite.
3. Weverton Fm -- Thin formation (100-500 ft). Pebbly quartzite beds
cemented locally by iron oxides. Interlayered with phyllites.
4. Hampton Fm (Harpers) -- Thick beds (1800-2200 ft) of metamorphosed sand-
stones and shale regularly interbedded, locally iron-cemented.
5. Antietam Fm (Erwin) -- Thick (700-1000 ft) bedded, very resistant
quartzite.
In general, the Pedlar and Catoctin formations occur in the eastern slopes of
the Blue Ridge, while the three clastic rock formations are found on the
western slopes. Exceptions occur at Madison Run and at the southern end of
the study area; the significance of these will be discussed later.
In addition to these five formations, several others are found in the
initial survey area:
1. Lovingston Formation -- augen gneiss, in gradational contact with the
Marshall Formation. Composed primarily of potassic feldspar, quartz,
oligoclase-andesine, and biotite.
29
-------
Antietam (Erwin) Formation. Cambrian.
Light-gray to white very resistant quartzite
Hampton (Harper's) Formation. Cambrian.
Metamorphosed shale and siltstone, inter-
bedded with dark, resistant quartzite
Weverton Formation. Cambrian.
Conglomeritic quartzite and shale
Catoctin Formation. Precambrain.
Dark green metamorphosed basalt inter-
bedded with purple phyllites, trace calcite
Pedlar Formation. Precambrian.
Coarse-grained massive to banded
granodiorite
Heavy black line denotes east-west
watershed divide
Scale in miles
'
-------
38°00'
Figure III-2b.
78°45'
Depiction of bedrock geology in the study area in and adjacent
to the Shenandoah National Park.
31
-------
2. Marshall Formation -- gray to green, medium grained gneissic granite, in
gradational contact with the Pedlar Formation. Major constituents are
potassic feldspar, quartz, oligoclase-andesine, and biotite.
3. Old Rag Granite -- massive plutonic granite, in gradational or fault
contact with the Pedlar Formation. Light gray in color, composed mainly
of blue quartz and perthitic microcline.
4. Various Limestone and Dolomite Formations -- light gray, mostly fine
grained limestone and dolomite. Beds are a few inches to ten feet thick.
Occurrence is at the base of the Blue Ridge in the Shenandoah Valley.
2. Methods and Materials.
Field samples were collected in 250 ml polyethylene bottles and stored on
ice until returned to the laboratory, then refrigerated until analysis. pH
was measured potentiometrically in the field; temperature and conductivity
were also measured in the field for each sample. For sequential sampling,
samples were collected using an automatic sampler (model S-4040, Manning
Environmental Corp.). These samples were stored on ice for return to the
laboratory, and pH and conductivity were measured in the laboratory.
Alkalinity and acidity were determined using a double endpoint titration
for alkalinity (to pH = 4.5 and 4.2) and single endpoint (pH = 8.3) for
acidity.3 Titrants were 0.005N HC1 and 0.005 N NaOH for the respective proce-
dures. Major cations were analyzed by atomic absorption spectrophotometry
using an Instrumentation Laboratories 751 spectrophotometer. Major anions and
silica were analyzed using standard procedures adopted for Technicon II Auto-
analyzer. 32
3. Results and Discussion.
Alkalinity and pH data for the large survey area are summarized in Table
III-l, while data for the smaller study area around Shenandoah National Park
are presented in Figures III-3 and III-4. Soil chemistry and vegetation are
generally similar throughout the survey area, and precipitation chemistry does
not vary significantly in the survey area; mean annual pH is approximately
4.2.34 Consequently, bedrock geology is interpreted to be the primary factor
determining stream chemistry in Virginia Blue Ridge; this conclusion is
carried through the remainder of this discussion.
Data presented in Table III-l indicate a wide range of pH and alkalinity
regimes for the Virginia Blue Ridge. As should be expected, pH and buffer
capacity are highest in those streams originating in limestone and dolomite
bedrock. These carbonate rocks react readily with both weak and strong acids,
resulting in high bicarbonate concentrations in solution and thus the high pH
and alkalinity observed. Only a few such streams were included in this sur-
vey, their purpose being to put the relatively low buffer capacity of other
systems into perspective.
32
-------
Streams from the four Precambrian, metamorphic formations (Pedlar,
Catoctin, Lovingston, and Marshall) have similar pH and alkalinity ranges,
with near neutral pH and alkaliriities of about 200 |Jeq/l (10 ppm as CaC03) .
While these levels are far below those of the carbonate streams, buffer capac-
ities in this range are sufficient to maintain a stream in good condition
except during extreme episodes of acid input. Even in such cases, the changes
should be short-lived and relatively minor. Old Rag granite, while quite
similar to the Pedlar and Lovingston formations, imparts a significantly lower
buffer capacity to streams flowing from it. Long term acidification appears
unlikely, but buffer capacity is sufficiently low that these streams probably
undergo moderate to substantial pH shifts during acid precipitation events
such as those observed in Madison Run (discussed below).
Streams in the clastic bedrock formations (Antietam and Hampton and to a
lesser extent, Weverton) are clearly the most vulnerable to both short and
long term acidification. The pH of these streams is well above that of inci-
dent precipitation, indicating that some neutralization occurs in the water-
shed, but the buffer capacity of streamwater is minuscule, at best.
Figures III-3 and III-4 show trends similar to those discussed above, and
clearly show the effects of different bedrock formations on the eastern and
western slopes of the Blue Ridge in the study area. These figures also show
the consequence of a stream passing through several bedrock formations; the pH
and alkalinity of such streams reflect the most reactive bedrock in their
watersheds. Madison Run (upper left corner) and several streams at the bottom
left side of the figures have much higher pH and alkalinity than other streams
on the western slopes; these differences are presumably attributable to the
presence of Catoctin bedrock in their watersheds.
Many of the streams in the survey area were sampled several times during
the summer to determine whether there were substantial changes in pH and
alkalinity as a result of seasonal variability or changes in stream discharge.
With the exception of Madison Run (see below) there seemed to be relatively
little change in chemistry of the streams involved.
It should be noted that in all streams studied the carbonic acid-
bicarbonate system was the dominant buffer system. Alkalinity titrations
showed pH behavior indicative of bicarbonate as the principal buffer, and
there were no data to suggest the presence of any other buffer.
Following the initial survey work in the Blue Ridge, Madison Run was
chosen for more intensive study during the remainder of the study period.
This stream has major tributaries flowing out of the Catoctin formation and
from clastic bedrock (Weverton and Hampton formations). During initial sam-
pling these tributaries had pH and alkalinity values typical of the respective
bedrock formations. Two kinds of sampling were done; regular (at least
bi-weekly) sampling at several stations on the main stream and tributaries,
and also sequential samples (usually 2 hour intervals) taken at various
stations during both dry periods and during and following individual rain-
storms. Sequential sampling during storms has allowed us to observe the
response of these streams to acid precipitation episodes.
Madison Run is a typical watershed on the western slope of the Blue
33
-------
pH = 5.0-5.5
pH = 5.5-6.0
pH = 6.0-6.5
pH greater than 6.5
(sample station = 0)
Scale in miles
Figure III-3a.
pH of streams in the survey area. Shaded areas represent
watershed areas upstream from sampling locations.
34
-------
38°00
78°45'
Figure III-3b. Depiction of streams in the Shenandoah survey area.
35
-------
Alkalinity greater than 250
(sample station = 0)
Scale in miles
1 Vz 0 1 2 3
_
(HH H H H I - 1
_
I I
Figure III-4a.
Alkalinity of streams in the survey area. Shaded areas repre-
sent watershed areas upstream from sampling locations.
36
-------
38°00
Figure III-4b.
78 45
Depiction of alkalinity of streams in the Shenandoah survey
area.
37
-------
Ridge, covering about six square miles. Vegetation cover is virtually com-
plete, consisting mostly of oak forest and extensive blueberry coverings.
Soils are coarse, thin, and characterized by low pH, cation exchange capacity,
and organic matter. The stream system is not gauged, so no budgets can be
determined for individual ions.
Total dissolved solids are low in all of the streams of the watershed,
typically around 20-30 mg/1. Silica is the major substance in solution,
ranging from 7 to 15 mg/1 as Si02. Bicarbonate is the dominant anion under
most conditions, with chloride and sulfate present in lower concentrations on
an equivalent basis. Magnesium is the major cation in solution with lower
concentrations (equivalents per liter) of calcium, sodium, and potassium.
Magnesium is a major constituent of the bedrock formations in the watershed,
particularly in the Catoctin formation; so its abundance in comparison to
other cations is not unexpected.
Unlike other streams in the survey area, pronounced changes occurred in
both Madison Run (Catoctin bedrock), and its major tributary, White Oak Run
(clastic bedrock) during the study period. From early June until mid-August,
pH increased in both streams, from 6.05 to 6.5 in Madison Run, and from 5.7 to
6.45 in White Oak Run. Alkalinity showed corresponding increases, from 70 to
150 (Jeq/1 and 18 to 80 (Jeq/1 for Madison and White Oak Runs, respectively. We
are uncertain of the cause of these changes, but changes in stream discharge/
groundwater flow and primary production by mosses and algae growing on bedrock
probably both play a role.
In the absence of precipitation, no significant diurnal changes occur in
any of the measured parameters on either Madison or White Oak Run. Alkalinity
and pH may undergo very small diurnal cycles on White Oak Run in response to
photosynthesis. During major precipitation events, however, substantial
changes occur, particularly on White Oak Run. On July 16, slightly less than
one inch of rain fell during a thunderstorm over the watershed, resulting in
major but brief changes in water chemistry on White Oak Run (Figure III-5).
Both pH and alkalinity fell substantially, with a corresponding increase in
acidity. Nitrate and sulfate levels showed a sharp increase, presumably
caused by precipitation input and washoff from leaf surfaces during the storm,
while silicon levels declined, probably from dilution of streamwater by pre-
cipitation rainwater. The concentrations returned to previous levels over
about an eight hour period.
A second storm, on August 11 and 12, was observed on Madison Run. About
2.5 inches of rain fell over a 24 hour period during this storm, resulting
from passage of a frontal system. Major ion analysis for this storm indicates
that its effects were minimal, despite the much larger amount of precipitation
(rainwater was not collected for chemical analysis during either storm).
Slight increases were observed in nitrate and sulfate concentration, and
silicon levels declined slightly during the storm, but no change was observed
in pH, alkalinity or acidity. Apparently the higher buffer capacity of the
stream, combined with more gradual input of acid (inferred from sulfate and
nitrate values) allowed the stream chemistry to remain relatively constant
throughout the storm.
38
-------
6.2-
6.1-
6. o-
5.9-
U1
QJ
801
60
40 -
20-
0
2.5
2.25 '
iI
\
a 2.0 -
1.75-
0
5.5
5.0
iH
| 4.5-
4.0-
0
.40
.30-
rH
\ .20-
g
.10-
Acidity
Alkalinity
0
T
1600
1200 1600 2000 0000 0400 0800 1200
July 16 July 17
Figure III-5. Changes in water chemistry in White Oak Run during the storm of
July 16, 1979. Rainfall occurred between 1300 and 1900 hours.
39
-------
A third storm, the remnant of Hurricane David, passed through the water-
shed on September 3, resulting in about 6 inches of rain over a twelve to
sixteen hour period. Although the pH of rain from this storm was rather high
(4.9), stream chemistry changed tremendously. Sequential samples on White Oak
Run showed a severe depression of pH to 5.1, alkalinity decreased to zero, and
acidity showed a major increase as bicarbonate was converted to carbonic acid.
Thirty-six hours after the storm, pH was still well below normal (5.6), and
alkalinity was only 12 [Jeq/1, about 20 percent of the pre-storm buffer capac-
ity. Madison Run was also sampled following the storm and it too showed much
lower pH and alkalinity (5.85 and 60 [Jeq/1) than normal.
Data on precipitation events is presented here only to indicate the type
of response one might observe in stream chemistry. Clearly, the specific
response will depend upon both the pre-storm chemistry of the stream and
groundwater in the watershed, and on the composition and magnitude of the
precipitation events. In retrospect, Madison Run was probably not an ideal
watershed to choose for concentrated study; the pH and alkalinity of Madison
and White Oak Run seem to show more seasonal variability than occur on other
streams. White Oak Run is clearly not typical of streams flowing from clastic
bedrock; by late summer its pH and alkalinity were well above those of other
streams originating in clastic rocks.
4. Summary of Intensive Study.
Within the Blue Ridge of central Virginia, the pH and buffer capacity of
streams varies greatly, with bedrock geology the major factor controlling
stream chemistry. With the exception of streams in areas with carbonate-
bearing bedrock, buffer capacity is generally low, and virtually nonexistent
in streams originating in clastic bedrock. Sensitivity to acidification is
presumably greatest in those streams with the lowest buffer capacity, but we
have no historical data available to determine whether significant changes in
the chemistry of these streams has occurred during recent years. Thus, these
data are valuable as baseline data to determine ongoing long-term changes in
stream chemistry.
Storm data from the Madison Run watershed clearly demonstrate short-term
effects of acid rain on stream chemistry, particularly on White Oak Run. Many
streams in the survey area have pH and buffer capacities significantly lower
than occur on White Oak Run; episodic changes in the chemistry of such streams
are probably more pronounced. The effects of such changes on stream fauna may
be substantial,43 but are beyond the scope of this study.
B. Extensive Surveys of Mountainous Areas of New Hampshire and North
Carolina
The states of New Hampshire and North Carolina, separated by over 800 km,
receive acid precipitation with volume weighted mean annual H+ concentrations
of approximately 114 peq/1 (pH 3.94) and 50 (Jeq/1 (pH 4.2).71 Both states
have variable topography with the White Mountains of central New Hampshire and
the Blue Ridge Mountains of western North Carolina providing terrain with
exposed bedrock, grading to regions of thin soil, then to deep soils. The
40
-------
MASSACHUSE ITS
Figure III-6.
Geologically-based sensitive areas map of New Hampshire. Highest
sensitivity (Type-I) is indicated by 1 on the map with 2, 3, and
A indicating progressively less sensitive areas. Dashed lines
are county boundaries.
41
-------
-p-
ho
Figure III-7. Geologically-based sensitive areas map of North Carolina. Highest sensitivity (Type I) is
indicated by 1 on the map with 2, 3 and 4 indicating progressively less sensitive areas.
Dashed lines are county boundaries.
-------
bedrock also is variable with respect to sensitivity, as can be seen in
Figures III-6 and III-7 with abundant outcrops of slow-weathering, igneous and
metamorphic rock. Thus these states were selected as ideal regions in which
to test the accuracy of the sensitive areas map. The test consisted of ident-
ifying specific sites which have low alkalinity (< 200 jjeq/1-1) from histor-
ical data, verifying these observations by new sampling, then comparing the
locations of these sensitive waters to the map of sensitive areas. The sens-
itive areas maps were not consulted when selecting sites for sampling, so that
the site selection, within the two regions expected to contain sensitive
areas, was not coupled to the map-making process.
1. Field Methods.
The first step in selecting field sites where samples were to be col-
lected, was to examine existing water quality records of the departments of
fisheries in each state so that new data could be compared to old data in an
analysis of temporal trends, as described below (Section IV). Stations which
had been previously sampled were then identified on county road maps. Wher-
ever possible, samples were collected at about the same time of year and at
the same sites used previously. Sites with alkalinity values greater than 200
peq/1-1 or pH greater than 7.0 were excluded. If recent cultural activities
such as farming, timber harvest, residential or recreational housing were
significant on site or upstream, the site was not included in the sampling.
At each location, two samples were taken in 250 ml acid washed polyethylene
bottles, rinsed three times in sample water before filling. One sample was
acidified for metal analysis to a pH of 2 by the addition of 1 ml of 6N HC1.
These samples were not acidified, but because suspended solids were low, and
because most calcium occurs in dissolved rather than particulate form,70 the
amount of calcium added in particulate form is believed to be negligible. The
other sample was placed in a cooler filled with ice and kept at or below 4°C
until pH and alkalinity measurements could be made (usually every two days).
Conductivity in microSiernens was measured with either a Beckman or Presto-tek
conductivity bridge. For pH and alkalinity determinations, a Beckman pH meter
with combination electrode was used. the samples were brought to room temper-
ature before pH was measured. Alkalinity was determined using Gran's plots.
Calcium analyses were performed on acidified samples by atomic absorption
spectrophotometry.
2. Results of Field Measurements.
The observed alkalinity values (Tables IV-1 and IV-2) in the ranges < 0
(jeq I-1, 0-50 peq I-1, 50-100 (jeq I-1 and 100-200 peq I-1 are shown for each
of the stations sampled in Figures III-8 and III-9 These can be compared to
Figure III-6 and Figure III-7, the respective state sensitive areas maps. In
New Hampshire all of these low alkalinity waters in the regions sampled fall
within regions predicted to have Types I and II sensitivity, and in North
Carolina all stations fall in Types I and II sensitive areas. Since stations
were selected for sampling only if historical data showed them to have low
alkalinities, they were expected to lie in regions underlain by sensitive
bedrock. The fact that the geologically based sensitive areas map confirms
this expectation constitutes a partial verification of this method of deter-
mining sensitive areas.
43
-------
-p-
feni
am
MEN HRMPSHIRE
RVG RLK(UEO/L)
- STflTIONS
SUMMER 79
LJ
a
en
72a33W
LONGITUDE
7044 H
LEGEND
x- LESS THRN 0
A - 0 TO 50
o- 50 TO 100
Figure III-8. Location and alkalinity of New Hampshire sites sampled during the summer of 1979.
-------
LJ
a
a:
K)
fO
NORTH CRROLHR - STRTIONS
RVG flLKdJEO/L) SUMMER 79
84^0 W
77 44 H
LEGEND
x- LESS THRN 0
A - 0 TO 50
o- 50 TO 100
75*32 W
LONGITUDE
Figure III-9. Location and alkalinity of North Carolina sites sampled during the summer of 1979
-------
C. Large-Scale Comparison
1. Acidification Chemistry Information Database. Data on Water Chem-
istry were collated from many Federal, state and local agency sources
(Appendix II). Information stored in computerized data bases (e.g. STORET),
as well as raw data from field reports gathered from these sources, were used
to establish thee Acidification Chemistry Information Database (ACID) at
Brookhaven National Laboratory. Two factors governed the selection of data.
Since alkalinity is the dominant factor of water chemistry which determines
sensitivity to acidification, all waters with alkalinity values in excess of
500 |Jeq/l were excluded. This exclusion selected for moderately to highly
sensitive waters and greatly reduced the amount of data to be included in
ACID. Second, the time scale for this project limited us to including about
half of the available and relevant data which we were able to locate.
A generalized data base management system, SYSTEM 2000 (propietary system
of MRI Systems Corp., Austin, Texas) was used for ACID. Analyses were done
using a combination of the CDC 6600 and CDC 7600 computers at Brookhaven
National Laboratory. Plotting of large scale maps was accomplished with a
36-inch Versatec electrostatic plotter. Some maps and tables were prepared
using a FR 80 COM device and many data tables were put into micro-fiche from
the FR 80. Plotting software was developed at BNL and is based on DISSPLA (a
plotting software proprietary package of Integrated Software Systems Corp.,
San Diego).
2. County-by-County Maps. As data are transferred from maps with a
scale as fine as 1:62,500 to maps at 1:2,500,000, fine resolution is com-
pletely lost. A problem arises in trying to preserve the identity of the
isolated sensitive areas. Preserving this information can be accomplished by
over-representing the size of the sensitive area. Data in ACID are used to
map the current alkalinity and pH on a county by county basis. Alkalinity
values were divided into 3 categories: a) < 100 peq/1; b) 100-200 [jeq/1; and
c) > 200 (Jeq/1 and values at each station for the period (1976-1979) were used
for this analysis. The stations included are shown in Figures 111-10 and
III-ll.
Two types of maps are presented:
1. An entire county is assigned the alkalinity category determined by the
lowest alkalinity value for the county and shaded according to that value
(Figures 111-12 and 111-13). These maps ensure that areas known to be
sensitive on the basis of current water chemistry are identified.
2. A county was assigned the average value of alkalinity, observed at all
stations in the county (Figures 111-14 and 111-15). The existence of low
alkalinity waters is frequently lost in these maps, but, when compared to
the minimum value maps, the true sensitivity of a whole county is more
realistically interpreted.
Maps for all states included in the ACID system are shown in the Appendix
(see Endnote, page 91).
46
-------
tin!
NEW HnMPSHIRE - SIRIIONS
RVG PH (1976-79)
72°33W 72°6W
LONGITUDE
7044 W
LEGEND
x- 5.1 TO 6.0
A - 6.1 TO 7.0
o = GREflTER THRN 7.0
Figure 111-10. Location of New Hampshire sites where pH and alkalinity data was available for the period
1976-1979.
-------
Unl
oui
NORTH CRROLINR - SinilONS
RVG PH (1976-79)
82°8W
79^6 W
LONGITUDE
77"44 W
LEGEND
+ - LESS THRN 5. 1
x- 5.1 TO 6.0
A = 6.1 TO 7.0
o- GRERTER THRN 7.0
75a32W
Figure III-ll. Location of North Carolina sites where pH and alkalinity data was availble for the period
1976-1979.
-------
bni
am
NEW HRMPSHIRE - COUNTIES
MIN RLK (UEO/L) 1976-79
a
Z3
t
E
CE
72*53H
71a39W
LONGITUDE
DRTR RVRILRBLE
IGRERTER THRN 200
lioo TO 200
LESS THRN 100
Figure 111-12. Minimum alkalinity of counties in New Hampshire for the period 1976-1979
-------
NORTH CRROLINR - COUNTIES
MIN flLK lUEO/L) 1976-79
Ln
O
DRTfl RVRlLflBLE
6RERTER THRN 200
dlOO TO 200
Hi_ESS THRN 100
-------
NEN HRMPSHIRE - COUNTIES
RVG RLK (UEQ/L)1976-79
CD
t-n
DRTfl RVRILRBLE
GRERTER THRN 200
100 TO 200
LESS THRN 100
*^ 72%W 71°39W
LONGITUDE
Figure 111-14. Average alkalinity of counties in New Hampshire for the period 1976-1979.
-------
LJ
O
NORTH CRROLINn- COUNTIES
RVG RLK (UEQ/L)1976-79
NO DflTR RVRILRBLE
06RERTER THRN 200
0] 100 TO 200
THRN 100
19H
due
79*^6 M
LONGITUDE
7744W
75*32 H
Figure 111-15- Average alkalinity of counties in North Carolina for the period 1976-1979,
-------
Some reservations about these maps must be noted. First, it is most
unlikely that ACID contains information on all streams of any county; the
sample density must be judged by the site maps for each state (Appendix I),
recalling that only waters with alkalinities < 500 |Jeq/£ were included in
ACID. Second, for the average pH or alkalinity maps, a single low value may
be all that was available for a given county. Again, the site maps (Appendix
I) must be consulted. On the other hand, most state water quality data, and
especially data in STORE!, are biased in favor of downstream locations of
water withdrawal for irrigation or municipal use. The water chemistry data
from state fisheries agencies are of great value since we suspect that they
give a better representation of upstream waters. Inclusion of additional data
which we have located from state and local sources into ACID will improve
subsequent versions of these maps.
3. Comparing Geologically-Based Maps to Chemistry-Based Maps. The
accuracy of the geologically-based sensitive areas maps of North Carolina and
New Hampshire (Figures III-6 and III-7) was evaluated by comparing them to the
county-by-county maps of current minimum or average alkalinity (Figures 111-12
through 111-15), which were assumed to be the "correct" maps. (At this stage
the county-by-county maps which showed geology as a binary number, Figure
11-4, were not tested. One cannot distinguish 0 percent from 9 percent on
those maps. They are less precise than the parent maps such as Figure III-6).
The comparison consisted of first, making a prediction of "sensitivity" or
"non-sensitivity" for a particular county based on the bedrock geology alone.
A county was predicted to be sensitive if either Type I or Type II bedrock was
present, and non-sensitive if neither Type I nor Type II was present. The
prediction was then compared to the corresponding surface water alkalinity.
A correct prediction of sensitivity was scored if a county contained Type
I or Type II geological areas and the county alkalinity value was ^ 200 peq/1.
A correct prediction of non-sensitivity was scored if the county contained no
Type I or Type II geological areas and the county alkalinity value was > 200
[jeq/1. A false prediction of sensitivity was scored if the county had only
Type I or Type II geological areas but the alkalinity value was > 200 (Jeq/1.
A false prediction of non-sensitivity was scored if the county had no Type I
or Type II geological areas and the alkalninty value was ^ 200 peq/1.
Finally, an indeterminate prediction occurred when a county had Type I and/or
Type II plus Type III and/or Type IV geological areas and alkalinity values
> 200 [Jeq/1 were in ACID.
Percentages of these predictions were determined by adding the number of
each score class (i.e. in North Carolina correct predictions of sensitivity =
38) and dividing by the number of counties with the appropriate alkalinity
score (i.e. in North Carolina ^ 200 (Jeq/1 = 42), since the alkalinity score is
presumed to be "correct." Thus, in North Carolina,
O Q
correct prediction of sensitivity = 7 = 90%
Several factors confound the reliability of the geologically based map of
sensitivity. The false predicition of sensitivity may be due to insufficient
53
-------
data in ACID; there may be streams of < 200 |Jeq/l for which no data have been
located in counties where a false prediction of sensitivity was scored. The
false prediction of non-sensitivity may be due to lost resolution of the
geological map as contact surfaces were merged and small areas deleted in the
mapping process (see Section II of this report). Counties with both sensitive
(Types I and II) and non-sensitive (Types III and IV) bedrock may be expected
to have waters with alkalinities higher or lower than 200 |Jeq/l so absence of
low alkalinity data in this evaluation of the geologically based map of sensi-
tivity results in the indeterminate score.
The degree of agreement is rather good between the maps based on water
chemistry and the geologically based map. The results, shown in Table III-2
and III-3 indicate that the geologically based map predicted the possible
sensitivity in 90% of 42 counties which did indeed have sensitive waters in
North Carolina and all of such counties in New Hampshire. The "correct"
prediction of non-sensitivity was 17%, one of six counties in North Carolina.
These counties are located in the eastern portion of the state where soil
factors are more important than in the central and western parts of the state.
The other five of these six counties, where the lowest alkalinity in ACID was
> 200 |Jeq/l, were identified by the geologically based map as having Type III
or IV bedrock (in some cases, in addition to Type I and Type II). The false
prediction of non-sensitivity occurred in only 10% of 42 counties.
In New Hampshire, the geologically based map "correctly" predicted the
sensitivity of all 8 counties with minimum alkalinities of ^ 200 |Jeq/l, and
all 5 counties with average alkalinity values ^ 200 (Jeq/1. All 10 counties of
New Hampshire had Type I and Type II geological areas and 6 had Type III or
Type IV. Both counties with minimum alkalinity greater than 200 peq/l
included Type III or IV geology. Thus, two indeterminate predictions and no
false sensitivity predictions were scored.
Following this evaluation of the predictive capability of bedrock geol-
ogy, the utility of the less precise binary number geological map was assessed
by using it to also predict the sensitivity or non-sensitivity of the same
counties. A correct prediction of sensitivity was scored if the binary number
map indicated the presence of at least 10 percent of Type I or Type II rock in
a county and the county alkalinity value was ^ 200 |jeq/l. This resulted
because, as pointed out above, zero on the binary number map may actually
include up to 9 percent of the particular rock type. Predictions based on the
binary number map (Tables III-4 and III-5) agreed quite well with the contour
maps. The principal difference, because of the range represented by each
number, was an increase in the indeterminate category and a decrease in false-
sensitive .
D. Map Verification Summary
The ability of the geologically based map to accurately define areas
which may be sensitive to acidification has been verified in three ways. An
intensive study within the Shenandoah National Park found stream water pH and
alkalinity along the Blue Ridge to be controlled primarily by bedrock geology.
Data obtained during storm events demonstrate that pH and alkalinity in
54
-------
Table III-2. Comparison of the geologically-based sensitive areas map
(Figure III-7) to county-by-county maps of alkalinity (Figures
111-13 and 111-15), for 49 of 100 North Carolina counties. See
text for definitions of predictions.
Minimum Alkalinity Average Alkalinity
Basis Basis
Correct prediction 90% of 42 counties 93% of 15 counties
of sensitivity
Correct prediction 17% of 6 counties 15% of 33 counties
of non-sensitivity
False prediction of 17% of 6 counties 39% of 33 counties
sensitivity
False prediction of 10% of 42 counties 0% of 15 counties
non-sensitivity
Indeterminate 8% of 48 counties 33% of 48 counties
Table III-3. Comparison of the geologically-based sensitive areas map (Figure
III-6) to county-by-county maps of alkalinity (Figures 111-12
and 111-14) for all 10 counties in New Hampshire. See text for
definitions of predictions.
Minimum Alkalinity Average Alkalinity
Basis Basis
1 Correct prediction 100% of 8 counties 100% of 5 counties
of sensitivity
Correct prediction 0% of 2 counties 0% of 5 counties
of non-sensitivity
False prediction of 0% of 2 counties 60% of 5 counties
sensitivity
False prediction of 0% of 8 counties 0% of 5 counties
non-sensitivity
Indeterminate 20% of 10 counties 20% of 10 counties
55
-------
Table III-4.
Comparison of the geologically-based binary code sensitive areas
map (Figure II-4b) to county-by-county maps of alkalinity
(Figures 111-13 and 111-15), for 49 of 100 North Carolina
counties.
Minimum Alkalinity
Basis
Average Alkalinity
Basis
Correct prediction
of sensitivity
Correct prediction
of non-sensitivity
False prediction of
sensitivity
False prediction of
non-sensitivity
Indeterminate
86% of 42 counties
17% of 6 counties
83% of 6 counties
0% of 42 counties
13% of 48 counties
100% of 15 counties
21% of 33 counties
0% of 33 counties
0% of 15 counties
54% of 48 counties
Table III-5. Comparison of the geologically-based binary code sensitive areas
map (Figure II-4a) to county-by-county maps of alkalinity
(Figures 111-13 and 111-15) for all 10 counties in New
Hampshire.
Minimum Alkalinity
Basis
Average Alkalinity
Basis
Correct prediction
of sensitivity
Correct prediction
of non-sensitivity
False prediction of
sensitivity
False prediction of
non-sensitivity
Indeterminate
100% of 8 counties
0% of 2 counties
0% of 2 counties
0% of 8 counties
20% of 10 counties
100% of 5 counties
0% of 5 counties
60% of 5 counties
0% of 5 counties
20% of 10 counties
56
-------
streams lying on sensitive bedrock fall sharply, presumably in response to
inputs of acid precipitation.
Extensive surveys of surface waters with alkalinities ^ 200 |Jeq/l in
North Carolina and New Hampshire, conducted in 1979, found 83-100% of these
sites were located in Type I and Type II sensitive areas, as defined by the
sensitive areas map.
Information obtained from numerous sources of raw -data and from com-
puterized data bases was used to establish the Acidification Chemistry Infor-
mation Database (ACID). This was used to plot surface water alkalinity on a
county-by-county basis. This map, which for testing purposes was considered
to be the "correct" map, was compared to the geologically-based map defini-
tions of sensitive areas for each county in North Carolina and New Hampshire.
The geologically based sensitive areas map was -"correct" for 90% of the 42
counties in North Carolina, and for 100% of the 8 counties of New Hampshire,
which had alkalinities ^ 200 |jeq/l.
57
-------
IV. TEMPORAL TRENDS IN pH AND ALKALINITY:
VALIDATION OF THE SENSITIVE AREAS MAP
Historical records have been used to document temporal changes in the
chemistry of lakes brought about by acid precipitation.1'23'91'96'104 The
sensitive areas map, when combined with projections of trends into future air
pollution and rain chemistry, may become useful in predicting which regions
will be acidified as the deposition of anthropogenic atmospheric acids in-
creases. The fact that acidification has taken place in regions which are
both heavily loaded with anthropogenic acids, and are predicted to be sus-
ceptible to acidification by the sensitive areas map, is in itself a valida-
tion of the predictive value of the map.
Temporal trends in water chemistry were determined for low alkalinity
(< 200 peq/1) sites in New Hampshire and North Carolina by comparing histor-
ical records to data obtained in the 1979 extensive surveys described in
Section III of this report. Historical data from a variety of Federal, state
and local sources for most of the other states were also collected and
pre-1970 data were compared to post-1975 data.
A. North Carolina and New Hampshire: Data
All of the old data obtained for sites resampled in the 1979 field trip
to North Carolina came from the North Carolina Department of Inland Fisheries
river basin surveys of the early 1960s. Table IV-1 lists all of the sampling
sites from 1979 with the corresponding earlier data. For the old surveys,
alkalinity determinations were made using the methyl orange technique, and pH
was determined colorimetrically.
These methods are less precise and less accurate than methods currently
used which are potentiometric pH measurement for pH and multiple end-point
potentiometric alkalinity titration. The problems associated with comparing
data sets obtained by these different methods are discussed by Zimmerman and
Harvey.105 Determination of alkalinity using a fixed end-point titration to
pH 4.5 will require 30 to 40 |Jeq/l of titrant to remove the free H+ in solu-
tion at this pH, so alkalinity values are overestimated by this amount, com-
pared to the multiple end-point method (Gran's plot). The old alkalinity
values were corrected, therefore, by subtracting 40 peq/1. In some cases this
has resulted in obviously erroneous alkalinity values for the old (corrected)
data, e.g. Boylston Creek, with pH 6.9 and a corrected alkalinity value of -18
peq/1 (Table IV-1). Use of color-indicator solutions for pH in low-conduc-
tivity, poorly buffered waters will alter the water's pH and cause a spurious
reading. There is no way to systematically correct these pH errors mathemat-
ically.105
One might also question how accurately a single sample represents the
water quality of a lake or stream since pH and alkalinity of natural waters
58
-------
fluctuate daily and seasonally. It is assumed that daily and seasonal fluctu-
ation, when sampled during the day and at the same time of year as the pre-
vious sampling, are random and normally distributed. Thus, when reasonably
large populations of sample pairs obtained in synoptic surveys are compared,
the positive and negative errors should cancel each other out and regional,
temporal trends may be analyzed.
Distributions of apparent pH and alkalinity changed in the fifteen to
nineteen years between samples (Figures IV-1 to IV-6, uncorrected data). Mean
hydrogen ion concentration changed from pH 6.77 in the early 1960's to pH 6.51
in 1979, an 82% increase in hydrogen ion concentration. The pH decreased at
79% of the sampling sites. The differences between the two sets of values was
found to be significant at the 0.02 probability level. Mean alkalinity values
have also experienced a significant decline since the early 60s from 116 |Jeq/l
(corrected) to 80 (Jeq/1 in 1979, with 71% of the 1979 values being lower than
the old (corrected) values.
For New Hampshire, extensive sets of old data, collected in the late
1930s, were found. No other data for small streams and lakes were located
except for some scattered records from the 1970s done by the New Hampshire
Water Pollution Control Commission. Unfortunately, this older survey con-
tained only pH data for small streams; alkalinity titrations were performed
only on lakes. It is not known exactly what methods were used to determine pH
and alkalinity for these old surveys, but it was almost certainly some form of
colorimetric method for both. The old and new data for New Hampshire are
listed in Table IV-2.
A comparison of Figures IV-7 to IV-9 reveals that in New Hampshire there
has been an apparent decrease in stream pH values since the 1930s. The mean
hydrogen ion concentration for the old data is pH 6.66 and in 1979 it is 6.12,
a 247% increase in free hydrogen ion. The pH values at 90% of the sampling
sites decreased between the 1930s and 1979. Again, the differences between
the two sets of data are significant at the 0.02 probability level. Of the
five sites where old alkalinity data were available, four of the new samples
had lower alkalinity.
59
-------
'labie IV-i. A comparison of old and new data for headwater
40 [ieq/1.
streams in North Carolina. Old alkalinity values were corrected by
subtracting
Stream
Roaring Fork
Little Elkin Creek
Pike Creek
Gardner Creek
Rich Mountain Creek
Bullhead Creek
Fall Creek
North Harper Creek
Webb Prong
Wilson Creek
Boone Fork
Laurel Fork
Cold Prong
Moody1 s Mill Creek
Beech Creek
Buckeye Creek
Blevins Creek
Beetree Creek
K lessee Creek
Cathey ' s Creek
Cove Creek
Davidson River
Looking Glass Creek
Cooper Creek
Deep Creek
Dick ' s Creek
Culowhee Creek
Tanessee Creek
Carson Creek
Williamson Creek
Turkey Creek
South Fork Mill River
Boylston Creek
North Mill River
South Fork Toe River
Middle Creek
Locust Creek
South Fork Toe River
Date
08-03-61
08-24-61
08-24-60
07-25-61
08-25-61
07-02-61
08-14-61
07-19-63
07-31-63
07-31-63
06-06-63
06-06-63
06-10-63
06-12-63
06-01-63
06-20-63
06-25-63
06-14-62
06-14-62
06-12-62
06-12-62
06-19-62
06-12-62
07-23-61
06-28-61
06-27-61
07-19-61
07-24-61
06-19-62
06-19-62
06-29-62
07-17-62
07-09-62
07-17-62
06-02-64
06-08-64
06-09-64
06-02-64
pH
7
6
7
6
6
7
7
6
6
6
6
6
6
6
6
6
6
6
6
6
6
7
6
6
6
7
6.
7
6
6
6.
6.
6.
6.
6.
6.
6.
.4
.9
.4
.7
.9
.0
.2
.9
.9
.9
.9
.7
.9
.7
.9
.9
.7
.0
.8
.8
.9
.0
.6
.7
.8
.2
.6
.6
.9
.9
.8
.9
.9
.8
.8
.8
.8
Old
Alkalinity
(as reported)
(|Jeq/l)
200
180
180
200
180
160
200
42
176
200
200
180
140
180
200
160
200
36
40
100
180
120
140
40
22
120
160
80
140
160
140
140
22
180
180
200
180
200
Alkalinity
(corrected)
(Meq/1)
160
140
140
160
140
120
140
2
136
160
160
140
100
140
160
120
160
-4
0
60
140
80
100
0
-18
80
120
40
100
120
100
100
-18
140
140
160
140
160
Date
06-29-79
06-29-79
06-29-79
06-29-79
06-29-79
06-29-79
06-30-79
06-30-79
06-30-79
07-01-79
07-01-79
07-01-79
07-01-79
07-01-79
07-01-79
07-01-79
07-01-79
07-02-79
07-02-79
07-02-79
07-02-79
07-02-79
07-02-79
07-03-79
07-03-79
07-03-79
07-03-79
07-03-79
07-04-79
07-04-79
07-04-79
07-04-79
07-04-79
07-04-79
07-04-79
07-04-79
07-04-79
07-04-79
pH
6.
7.
6.
6.
6.
6.
.6.
6.
6.
6.
6.
6.
6.
6.
6.
6
6.
6.
6
6
6
6
6
6
6
6
6
6
6
6
6.
6.
6
6
5
6
6
.7
.1
.8
9
92
89
.8
.5
.78
.5
.52
.62
.3
.7
.8
.59
.7
.33
.4
.5
.4
.34
.2
. 4
.41
.8
.4
.4
.9
.7
.5
.63
.6
.2
.99
.4
.7
New
Alkalinity
(Meq/1)
96
206
96
110
128
89
80
46
90
23
73
116
49
86
73
63
107
68
-+-J
60
77
61
61
60
70
77
192
64
65
109
67
57
145
67
23
16
56
84
Specific
Conduc tance
(^Siemens )
21
35
18
19
22
15
15
10
14
18
16
32
14
19
41
26
50
12
10
10
15
14
14
12
16
19
29
13
15
18
12
12
23
13
10
10
13
17
C a 1 c i um
(Meq/1)
41
103
33
35
57
26
35
32
44
88
77
106
41
71
108
76
147
23
24
22
44
38
40
26
48
48
111
34
38
43
29
27
74
31
24
21
38
48
-------
Table IV-2. A Lompai ibou ot ultl .ind new data fni Nru u.impsii i i i' sLir.ims ,innt r.it i nn
lintty calculated by subt rac-t i rig new n I k.i 1 i n i I y I rom i-sL i m.i led old ,i I k,i I i u i I y .
ksi
fi i iii-b wt-i'f LUI' i L-ul L'd uy biiuLi di. Ling 40 peq/ 1 -
s model51 and the est i ma ted rharige in alka-
Stream or Lake
Kimball Creek, Coos County
Millsfield Pond Bk. , Coos Co.
Horne Brook, Coos County
Upper Ammonoosuc, Coos County
Jericho Brook, Coos County
Wild River, Coos County
Imp Brook, Coos County
Peabody River, Coos County
Nineteen Mile Brook, Coos Co.
Ellis River, Carroll County
Wildcat Brook, Carroll County
Great Brook, Carroll County
Slippery Brook, Carroll Co.
Mountain Pond Outlet,
Carroll County
Paugus Brook, Carroll County
Swift River, Carroll County
Wonalancet Brook, Carroll Co.
Whiteface River, Carroll Co.
Cold River, Carroll County
Atwood Brook, Carroll County
Bearcamp Brook, Carroll Co.
White Lake, Carroll County
Slater Brook, Carroll County
Chocorua River, Carroll Co.
Cream Brook, Carroll County
Moat Brook, Carroll County
Mason Brook, Carroll County
Weeks Brook, Carroll County
Upper Kirnbal Lake, Carroll Co.
Allard Brook, Carroll County
Hobbs Brook, Carroll County
Steam Mill Brook, Carroll Co.
Oliverian Brook, Carroll Co.
White Brook, Carroll County
Downes Brook, Carroll County
Douglass Brook, Carroll Co.
Razor Brnf.k, Cirroll County
Albany Brook, Carroll County
Sawyer River, Carroll County
Nancy Brook, Carroll County
GJbbb Brook, Coos County
'
Date
08-01-38
07-13-37
07-08-37
08-03-37
08-08-37
07-07-37
07-06-37
07-06-37
07-06-37
06-28-37
06-24-37
06-24-37
06-24-37
07-15-37
06-15-37
06-15-37
06-15-37
06-11-37
06-20-37
06-17-37
06-19-37
06-21-37
06-12-37
06-15-37
06-29-37
06-23-37
06-21-37
06-26-37
.07-23-37
06-29-37
06-29-37
06-30-37
06-29-37
06-29-37
06-29-37
06-29-37
06-.:R-1 7
06-29-37
06-25-37
06-25-37
07-26-38
pl-1
6.
7.
7.
7.
7.
6.
6.
6.
7.
7.
7.
7.
7.
6.
7.
7.
6.
6.
7.
6.
6.
6.
6.
7.
7.
6.
6.
7.
6.
7.
6.
6.
6.
6.
6.
7.
7 .
7.
6.
6.
6.
old
A 1 k.i I i n i ty
(as reported)
(|lei(/ 1 )
6
0
0
0
0
8
8
8
0
2
0
2
0
1
0
0
8
4
0
8
6
5 100
2
0
2
7
4
0
0 1 40
2
8
4
4
2
8
0
0
0
8
4
8
A 1 k n I i u i I y
{ i o r roc t cd )
(I'fl/U
(17
07
07
07
07
07
07
07
07
07
07
07
07
07
07
07
0 7
07
07
07
07
60 07
07
07
07
07
07
07
D.il
-26
-26
-27
-27
-27
-27
-27
-27
-27
-27
-27
-27
-27
-27
-29
-29
-29
-29
-29
-29
-29
-30
-30
-30
-30
-30
-30
-30
e
-79
-79
-79
-79
-79
-79
-79
-79
-79
-79
-79
-79
-79
-79
-79
-79
- 79
-79
-79
-79
-79
-79
-79
-79
-79
-79
-79
-79
100 07-30-79
07
07
07
07
07
07
07
07
07
-30
-30
-30
-30
-30
-30
-30
-31
- 31
07-31
07
07
-31
-31
-79
-79
-79
-79
-79
-79
-79
-79
-/')
-79
-79
-79
pll
6.
6 .
6.
7.
b .
6.
6.
6.
b .
6.
6.
6.
6.
5.
6.
6.
6.
6.
6.
6.
6.
5.
b.
6.
6.
6.
6.
6.
6.
6.
6.
6.
6.
6.
5.
6.
6.
(j.
5.
5.
5.
6
9
5
0
4
4
5
0
4
1
2
1
4
6
2
8
/,
3
4
3
3
8
3
3
3
3
4
6
2
5
2
0
I
0
6
6
1
1
8
8
1
A 1 ka 1 i r
(l-if'l/l
196
1 24
151
189
1 )9
39
65
8
it )
85
77
85
78
25
53
155
')'(
91
b6
127
127
4
119
89
189
138
138
79
42
1 05
27
50
40
09
-10
134
103
1 ! .'
32
18
-15
i n j l y
Spe
Cond
(I'SL
New
c i f i c
uctaiice
eni(jns )
37
28
25
37
23
9
16
44
19
14
20
14
13
14
16
32
9U
25
21
26
26
12
21
20
52
26
21
23
16
21
20
18
23
23
16
18
id
.:')
24
18
17
Calcium
(Meq/l)
84
155
154
173
55
20
68
76
52
70
73
81
61
100
157
147
118
107
195
197
58
129
129
188
162
162
1 18
91
95
98
79
95
109
57
135
14 i
i 5 7
101
79
52
Es t i ma ted
old Alk.
from Ca
87
188
187
214
46
-4
65
76
42
67
72
83
55
110
191
177
136
120
245
248
50
151
151
235
198
194
136
97
106
107
80
103
123
49
ihO
1 71
I'M
111
80
42
Est ima ted
AAlkalinity
+ 109
-64
-36
-25
-8
+67
-57
-13
+43
+ 10
+ 13
-5
-30
-57
-36
-83
-45
-54
-118
-121
-46
-32
-62
-46
-60
-60
-57
-55
-1
-80
-30
-63
-54
-59
-26
-68
-b4
-79
-62
-57
-------
Table ltf-2. (continued)
ON
Stream or Lake Date pH
Mt. Pleasant Bk., Carrol] Co.
Clay Brook, Coos County
Jefferson Brook, Coos County
South Br. Israel R., Coos Co.
Deception Brook, Coos County
Zeland Brook, Grafton County
Little River, Coos County
Tucker Brook, Grafton County
Reel Brook, Grafton County
Ham Branch, Grafton County
Black Brook, Grafton County
Lost River, Grafton County
Walker Brook, Grafton County
Cascade Brook, Grafton County
Hancock Branch, Grafton Co.
Cedar Brook, Grafton County
E. Branch Pemigewasset,
Grafton County
Baker River, Grafton County
Merrill Brook, Grafton County-
Still Brook
Stinson Lake, Grafton County 08-06-38 6.8 70
Ellsworth Outlet, Grafton Co. 08-10-38 6.8 100
Drakes Brook, Graftou County
Osceola Brook, Grafton County
Eastman Brook, Grafton County
Little East Pond Brook,
Grafton County
Russell Pond, Grafton County 07-24-38 6.6 90
07-25-38
07-25-38
07-25-38
07-13-38
07-26-38
07-26-38
07-26-38
07-14-38
07-14-38
07-14-38
07-12-38
08-27-39
08-27-39
08-27-39
08-29-39
08-29-39
08-28-39
08-25-39
08-26 3?
07-19-38
08-06-38
08-10-38
07-21-39
07-24-39
07-24-38
07-24-38
6.9
6.9
6.9
7.0
6.8
6.8
6.4
6.8
6.8
6.8
6.8
6.5
6.7
6.1
6.9
6.2
6.9
6.9
6.8
7.0
6.8
6.8
6.8
6.8
7.0
7.0
AlkaI ini ty Alkal ini ly
(as ropurleil) ^ corr eel ed J
30
60
50
07-31-79
07-31-79
07-31-79
07-31-79
07-31-79
07-31-79
07-31-79
07-31-79
08-01-79
08-01-79
08-01-79
08-01-79
08-01-79
08-01-79
08-01-79
08-01-79
08-01-79
08-02-79
08-02-79
08-02-79
08-02-79
08-02-79
08-02-79
08-02-79
08-02-79
08-02-79
pll
6.0
6.2
6.4
6.3
6.3
6.4
6.4
6.4
5.9
6.2
6.3
6.1
6.4
6.5
6.0
5.9
6.4
6.4
u. 7
6.5
6.2
6.1
6.2
6.3
6.6
6.6
08-02-79 6.1
A 1 ka 1 i n i t y
6
29
18
63
150
64
52
123
52
150
11J
43
76
116
65
63
51
33
173
111
33
57
46
74
90
90
49
New
H p e c i f i c
Conduct auce
(|jSi emens )
15
]2
12
19
32
25
16
222
6
37
29
124
21
64
36
28
24
18
38
25
20
19
24
38
25
25
Calcium
(Ueq/l)
79
40
42
103
155
119
94
449
98
164
107
105
75
150
116
97
107
57
148
101
88
106
103
117
103
89
Estimated
old Alk.
from Ca
80
25
28
114
188
137
101
607
107
201
120
117
75
181
133
106
120
49
178
111
93
119
114
134
114
94
Estimated
AAlkalinity
-74
+4
-10
-51
-38
-73
-49
-484
-55
-51
-9
-74
+1
-65
-68
-43
-69
-16
-5
0
-60
-62
-68
-60
-24
-4
20
111
-62
-------
14
CO
bJ , ^
_J 2
Q_
iio
C/)
DO
0
NORTH CAROLINA FISHERIES DATA (JUNE,JULY
or AUGUST 1960-1963)
0 20 40 60 80 100 120 140 160 180 200 220 240
ALKALINITY (/ieq./l)
Figure IV-1. Distribution of alkalinity in samples from North Carolina fisheries data, 1960-1963.
-------
00
CL
2
<
CO
10
QQ
§4
0
CAROLINA FIELD TRIP (6-29-79-7-4-79)
0 20 40 60 80 100 120 140 160 180 200220240
ALKALINITY (/ieq./l )
Figure IV-2. Distribution of alkalinity in samples from North Carolina field trip, 1979.
-------
NORTH CAROLINA FISHERIES DATA (JUNE,JULY
or AUGUST 1960-1963)
1 T
C/}
UJ
14
o 8
m c
QQ 6
0
1 1 1 T
6.0 6.2 6.4 6.6 6.8 7.0 7.2 7.4 7.6
PH
Figure IV-3. Distribution of pH in samples from North Carolina fisheries data, 1960-1963.
-------
NORTH CAROLINA FIELD TRIP (6-29-79-7-4-79)
10
14
QQ
^
\ /i
z> 4
n
-
i i i
i
i
i
i i i i i -I 1
6.0 6.2 6.4 6.6 6.8 7.0 7.2 7.4 7.6
PH
Figure IV-4. Distributions of pH in samples from North Carolina field trip, 1979.
-------
150-
cr
-------
pH vs. pH FOR
NORTH CAROLINA STREAMS
5.8 6.0
6.5 7.0 7.5
pH,1960-64
Figure IV6. Comparative plot of pH for North Carolina streams, 19601964 versus 1979.
-------
20
CO
IR
o i 5
a:
UJ
99 10
0
NEW HAMPSHIRE FISHERIES DATA
(SUMMERS of 1937,1938,1939)
I I I I I
1 I
i I I I I I I I i i r r
n i i i i i i r
\ r
5.2 5.4 5.6 5.8 6.0 6.2 6.4 6.6 6.8 7.0 7.2 7.4 7.6
PH
Figure IV-7. Distribution of pH in samples from New Hampshire fisheries data 1937-1939.
-------
NEW HAMPSHIRE FIELD TRIP (7-26-79-8-2-79)
30
CO
CL
20
LL_ j r-
o I 5
cr
LU
GO ltx
0
I I I I I I I I I I I I I I I I I I I I I I T
i i r
5.2 5.4 5.6 5.8 6.0 6.2 6.4 6.6 6.8 7.0 72 7.4 7.6
PH
Figure IV-8. Distribution of pH in samples from New Hampshire field trip, 1979.
-------
cr>
7.2 F
7.0
6.8
6.6
6-4
x 6.2
Q.
6.0
5.8
5.6
5.4
I I I
pH vs. pH FOR NEW HAMPSHIRE
STREAMS & LAKES
I I I I I I I I I I
I I I -
I I I I I I I I I I i
5.4 5.6 5.8 6.0 6.2 6.4 6.6 6.8 7.0 7.2 7.4 7.6
pH, 1937-39
Figure IV-9. Comparative plot of pH for New Hampshire streams, 1937-1939 versus 1979.
-------
B. North Carolina and New Hampshire: Discussion
A method for estimating the expected alkalinity value from the observed
Ca concentration has been derived by Henriksen,51 based on water chemistry
observations obtained in acidified and non-acidified regions underlain by
chemically similar bedrock (described further in Section VI of this report).
These calculated values, Alkalinity (|jeq/l) = 1.42 Ca (peq/1) - 32 for the New
Hampshire data are presented in Table IV-2. Acidification, defined as loss of
alkalinity51 was estimated by substracting the expected alkalinity from
current measured alkalinity. In this data set, 59 stations had alkalinity
values in 1979 which were lower, by an average of 57 |Jeq/l, than expected.
Regional acidification was thus calculated as 57 |Jeq/l for New Hampshire.
This method of calculating acidification depends on an approximately
constant relationship between Ca and Mg. We found that it could not be
applied to the North Carolina data.
In 1955-56 the Blue Ridge area of North Carolina was receiving precipita-
tion with a weighted annual average hydrogen ion concentration of pH 4.7-
5.2.69 By 1972-73, the same area was receiving precipitation with a weighted
annual average of about pH 4.2.69 This decline in pH appears to have caused a
significant decline in the pH and alkalinity of sensitive streams in the Blue
Ridge area of North Carolina. If the pH of rainfall continues to decline in
the southeast these streams may experience a continued decline in pH and
alkalinity.
The 1979 data for New Hampshire and North Carolina are plotted with
Henriksen's52 empirically derived curve (Figure IV-10) of pH versus [Ca] ,
which separates acidified and non-acidified lakes (for a more detailed discus-
sion, see Section VI). Those points lying above the empirically drawn curve
are presumed to have been significantly acidified by precipitation and those
below presumably have not. None of the samples from North Carolina lie above
the curve, while there has been a significant decrease in pH and alkalinity.
Henriksen's curve assumes a constant Ca/Mg ratio, which may not be valid for
the North Carolina waters, which are underlain by sedimentary and metamorphic
rocks. Approximately 54% of the data points for New Hampshire lie on or above
the curve, indicating that many of the sensitive streams and lakes have become
significantly acidified. The average pH of samples in New Hampshire is lower,
and their calcium values are higher, than in North Carolina.
72
-------
CALCIUM vs. pH FOR HEADWATER
STREAMS IN N.C. & N.H.
4.0-
i i i i i i i i i T i i i i i i i r i
NORTH CAROLINA
* NEW HAMPSHIRE
I I I I I I I I I I I I I I I I I I I I
0
Figure IV-10. Calcium versus pH for headwater streams in North Carolina and New Hampshire.
The curve is Henriksen's line.
-------
V. IMPACTS OF ACIDIFICATION ON AQUATIC BIOTA
Acidification of surface waters in sensitive areas causes a variety of
changes in the aquatic flora and fauna. The extent of this impact ranges from
changing population strengths to complete elimination of many species. The
purpose of this section is to provide a review and description of the kinds of
biological impacts which have been observed and which may be anticipated in
sensitive waters which are altered by acid precipitation. Impacts on organ-
isms other than fish are stressed here. Impacts on fish are discussed in
Section VI. This section is a collation of materials previously presented,
and includes references to affects of acid mine drainage on biota.
Although the literature concerning effects of acid mine drainage (AMD) on
freshwater ecosystems is extensive, it is not directly applicable to the acid
precipitation problem. Often the uncontaminated waters of coal mining areas
have higher alkalinity and hardness than the very soft waters acidified by
acid precipitation. AMD water usually has a heavy load of iron and other
heavy metals, and frequently depresses the oxygen concentration of receiving
waters. High turbidity and the presence of chemical floe are also common and
greatly alter aquatic habitats. These factors make it very difficult to
extrapolate observations from AMD situations to those in the Laurentian
Shield, for example.
The variety of species of both plants and invertebrate animals occurring
in fresh waters is enormous and speciation differs markedly from one locale to
another even though water chemistry may be similar. It is at best difficult
and probably futile to try to interpret ecosystem damages at lower trophic
levels by comparing lists of species. On the other hand, changes in major
processes such as primary and secondary production, and decomposition, can be
broadly described and compared. Effects of stress on the major functional
guilds may be compared from place to place. Finally a few groups of organisms
seem to be remarkably insensitive to strong mineral acidity and are common to
many acid environments, while some other groups are clearly intolerant of pH
levels below 6.0 to 5.5.
A. Effects on Microbiota
The production of fish and other animal life in a lake is ultimately
dependent upon the availability of organic food resources, primarily plant
materials. The sources of organic materials may be divided into two major
categories: autochthonous, originating by primary production in the lake, and
allochthonous, transported into the lake by inflowing water, airborne litter,
or dissolved in rain. The relative importance of each of these sources varies
greatly from lake to lake. One principal route for both autochthonous and
allochthonous organic matter into the trophic system of a lake is via the
detritus.
74
-------
Bacterial consumption and mineralization of dietrital organic matter,
both particulate (POM) and dissolved (DOM), allows a cycling of carbon which
dominates the structure and the functioning of the system and provides what
Wetzel98 has called a fundamental stability to the system. In the deep, open
water of the pelagic zone, where phytoplankton production normally provides a
substantial portion of the nonrefractory organic matter, bacteria rapidly
assimilate dissolved labile organic substances (PDOM) derived from photosyn-
thesis29'46 and convert it into bacterial biomass. Particulate organic matter
(POM) from phytoplankton is assimilated at a somewhat slower rate. Only a
small portion of the PDOM is refractory material likely to survive longer than
24 hours. The bacterial mineralization rate appears to be rather slow, a
few percent per day,17'99 so that this new biomass is actually available to
other trophic levels. Not only do the bacteria conserve the energy stored in
labile PDOM, which otherwise would be lost from the system, but they also
convert (at a slower rate) some of the refractory DOM into a usable form.
Fungal and bacterial communities render other POC into forms which are useful
for detritivores.13 The significance of these activities to ecosystem ener-
getics can be better appreciated when one considers that on the order of 90%
of the organic carbon in the water column is DOM and that detrital POM is many
times larger than the total living carbon biomass.
There are other sources of detritus in the pelagic zone. In some lakes,
particularly smaller and/or shallow lakes, macrophytes and benthic algae are
important sources of autochthonous organic carbon. Material from these plants
may contribute significantly to detritus in the pelagic zone. In deciduous
forest lakes, leaf litter falling or blown onto the surface annually has been
found to be 200 to 500 g dry leaves per meter of wooded shore line.35'60 This
forest litter, plus that which is added by stream inputs, contributes to both
DOM (after leaching) and POM.
Water column detritus generated from all of these sources has three
possible fates. It can be transformed biologically, it can sink to the sedi-
ments where it accumulates and/or is transformed biologically, or it is lost
from the system by outflow. In the first two cases, microbial activity plays
a key role in removing detritus.
The inhibition of microbial decomposition can have profound effects
throughout an aquatic ecosystem. Detritus removal, conservation of energy,
nutrient recycling, primary production, detritivore production and thus
production at higher trophic levels, can all be affected by changes in microb-
ial activity. Several investigations have indicated that microbial decomposi-
tion is greatly inhibited in waters affected by acid precipitation.
An abnormal accumulation of coarse organic detritus has been observed on
the bottom of six Swedish lakes where the pH decreased by 1.4 to 1.7 units in
the past three to four decades.37 Bacterial activity apparently decreased,
while in some of the lakes the sediment surfaces over large areas were made up
of dense felts of fungal hyphae. In one of the lakes, GSrdsjb'n, 85% of the
bottom in the 0 to 2 m depth zone was covered with a thick felt of algae and
organic debris, plus fungal hyphae. Lime treatment caused a rapid decomposi-
tion of the organic litter as well as great reductions of the felt,4 indi-
cating that an inhibition of bacterial activities had taken place at low pH.
Similar neutralizations of acidified lakes in Canada resulted in a significant
75
-------
increase in aerobic heterotrophic bacteria in the water column.89 Results from
field and laboratory experiments with litter bags in Norway49 also indicate
reduced weight loss from leaves in acidic waters. Dissolved organic carbon
(DOC) in the inflowing water was found to contribute approximately 50% of
allochthonous inputs and 8% of all organic carbon in Mirror Lake, while fine
particulate organic carbon (FPOC) was negligible.60 The extent to which this
DOC input is converted to bacterial biomass or otherwise enters into the
energetics of a lake is not known. Observations of abnormal accumulations of
organic debris have also been made in AMD waters in South Africa44 and West
Virginia (J. DeCosta pers. comm.).
In laboratory experiments the decomposition rates of peptone by micro-
biota decreased with pH and the oxidation of ammonia ceased below pH 5.
Bacterial cell counts and the species number of ciliates also decreased.12
Numerous other studies indicate that the microbial decomposition of organic
materials is markedly reduced at pH levels commonly encountered in lakes
affected by acid precipitation.49
Accumulations of organic debris and extensive mats of fungal hyphae, as
observed in the Swedish lakes,37 both seal off the mineral sediments from
interactions with the overlying water and retain organically bound nutrients
which would otherwise have become available if normal decomposition had
occurred. The reduction in nutrient availability can be expected to have a
negative feedback effect on both plants and bacteria, further inhibiting their
activities. The reduction of nutrient supplies to the water column from the
sediments, because of the physical covering and from reduced mineralization of
organic materials in the water itself, will lead to reduced phytoplankton
productivity. These ideas have been formulated into the hypothesis of "self-
accelerating oligotrophication" by Grahn et aJL.37 Qualititative observations
support this hypothesis but a quantitative evaluation is lacking.
Reduction of microdecomposer activities may have a direct effect upon the
invertebrates. Although certain benthic invertebrates appear to feed directly
on the allochthonous detritus material, it seems that "coditioned" (colonized
by microorganisms) material is preferred, and that the nutritional value of
the detritus is highly increased by conditioning.13 Bacteria may also be a
food source to be remoed by the filtering apparatus of organisms such as
Calanoida. An inhibition of the microbiota or a reduction in microbial
decomposition processes would therefore have a direct impact on the lakes'
animal communities.
B. Effects on Benthic Plants
In waters affected by acid precipitation major changes occur within plant
communities. Most of the available data are qualitative and descriptive
although some experimentation has been done. Intact lake sediment cores which
included the rooted macrophyte Lobelia dortmana were incubated at three pH
levels (4.0, 4.3 to 5.5, 6.0) at Tovdal in southern Norway. The growth and
productivity of the plant (02 production) were reduced by 75% at pH 4 compared
to the control (pH 4.3 to 5.5) and the period of flowering was delayed ten
days at the low pH.66
76
-------
In five lakes of the Swedish west coast, a region severely affected by
acid precipitation, in the past three to five decades the macrophyte commun-
ities dominated by Spaghnum have expanded. In the sheltered and shaded
locality of Lake Orvattnet, in the 0 to 2 m depth zone, the bottom area
covered by Sphagnum increased from 8 to 63% between 1967 and 1974. In the 4
to 6 m depth zone, the increase was from 4 to 30%. At the same time, pH in
Orvattnet decreased 0.8 units to approximately 4.8.38 Similar growth of
Sphagnum occurred in other Swedish lakes, in Norwegian lakes, and in AMD water
as well.42'44'45'49 At the pH of these acid waters, essentially all of the
available inorganic carbon is in the form of CC>2 or H^COs- Conditions are
more favorable for Sphagnum, an acidophile which appears to simply outgrow the
flowering plants under acid conditions.
In developing their hypothesis on oligotrophication, Grahn et a^l.37 have
stressed two biologically important consequences of this Sphagnum expansion.
First, Sphagnum has an ion-exchange capacity which results in the withdrawal
of base ions such as Ca from solution, thus reducing their availability to
other organisms. Secondly, dense growths of Sphagnum form a distinct biotope
which is unsuitable for many members of the bottom fauna.
Under some acid conditions, unusual accumulations of both epiphytic and
epilythic algae may occur. In the Swedish lakes Mougeotia and Batrachospermum
become important components of the benthos.37 In Lake Oggevatn (pH 4.6), a
clear-water lake in southern Norway, not only is Sphagnum beginning to choke
out Lobelia dortmana and Isoetes lacustris, but these macrophytes have been
observed to be festooned with filamentous algae.49
A floristic survey, conducted in Lake Golden (elevation 842 m) in 1932,
does not mention Sphagnum, but describes a macrophyte community typical of
clear oligotrophic lakes of the Adirondack Mountains (Eriocaulon, Lobelia,
Myriophyllum, Isoetes, Utricularia, etc.). On 6/11/58 Golden's surface pH was
5.8. In mid-July, 1979, pH was 4.9 and the flora from the shore to 0.5 m
depth (about 2-5 m from shore) around most of the lake perimeter was dominated
by extremely dense, uniform stands of Sphagnum pylaesii Brid. (390 g/m2). In
some areas Utricularia alone or with Sphagnum also formed dense stands which
were covered over by flocculent growth of algae in which Fragilaria
virescence, Tabellaria fenestrata, Eunotia triodon and Mougeotia sp. were
abundant. Prominent mats composed primarily of blue-green algae, fungal
hyphae and plant detritus were spreading over the dense Utricularia stands (G.
H. Hendrey and F. Vertucci, unpublished data). Golden, once famous for its
trout fishing, is now too acidic to support fish life. These observations,
the first of their kind in North America, parallel findings in some acidified
Swedish and Norwegian lakes. The question of why some acidified lakes show
these extreme floristic changes, while many others do not, remains a mystery.
Heavy growths of filamentous algae and mosses occur not only in acidified
lakes but have also been reported in streams in Norway affected by acidifica-
tion. In experiments in artificial stream channels using water and the natur-
ally seeded algae from an acidified brook (pH 4.3 to 5.5), an increase in
acidity to pH 4.0 by addition of sulfuric acid led to an increased accumula-
tion of algae compared to an unmodified control.48 The flora was dominated by
Binuclearia tatrana, Mougeotia spp., Eunotia lunaris, Tabellaria flocculosa,
and Dinobyron spp., each accounting for at least 20% of the flora at one time
77
-------
or another. The rate of radioactive carbon uptake per unit of chlorophyll in
the channels, measured on just two occasions, was found to be lower in the
acid channel by about 30%, suggesting greater algal accumulation at low pH
despite lower productivity.
Several factors may contribute to these unusual accumulations of certain
algae. The intolerance of various species to low pH or to consequent chemical
changes76 will allow just a few algal species to utilize the nutrients avail-
able in these predominantly oligotrophic waters. Many species of inverte-
brates are absent at low pH, and removal of algae by grazing is probably
diminished. Microbial decomposition is inhibited, as was previously noted,
also reducing the removal of algal mass.
C. Effects on Phytoplankton
There is no consistency among various investigations as to which taxa are
likely to be dominant under conditions of acidification. The Pyrrophyta may
be more common (e.g., species of Peridinium and Gymnodinium) than others in
lakes near 4.0. With decreasing pH in the range of 6.0 to 4.0, many speices
of the Chlorophyta are eliminated although a few tolerant forms are found in
the acid range. In the survey of 155 Swedish west coast lakes, blue-green
algae became less important with decreasing pH,1 but Kwiatkowski and Roff65
found the opposite to be true in lakes of the Sudbury, Ontario, region.
There are, however, conspicuous decreases in phytoplankton species
number, species diversity, biomass, and production per unit volume (mg/m3)
with decreasing pH. Lake clarity and the compensation depth increase with
lake acidification, so that areal primary production (mg/m2), although lower
in acid than nonacid lakes, is not as severely depressed as is production per
unit volume.1'58'65 The low phytoplankton biomass (< 1 mg/1) has been corre-
lated to the concentration of available phosphorus, which generally decreases
with lower lake pH.1 Low availability of inorganic carbon has also been
suggested as a factor limiting primary production in acidic lakes.58'63
Kwiatkowski and Roff65 have carried out a highly quantitative study of
phytoplankton in six lakes of the Sudbury region, which are impacted by atmos-
pheric effluents from a huge nickel-copper smelter complex. Curvilinear
equations are presented for phytoplankton diversity, secchi transparency
depth, chl a, and productivity.
D. Effects on Invertebrates
Zooplankton analyzed from net samples collected from 84 lakes in Sweden
showed that acidification caused the elimination of many species and led to
simplification of zooplankton communities.1 Crustacean zooplankton were
sampled in 57 lakes during a Norwegian lake survey in 1974,47 and the number
of species was found to decrease with pH. The distributions and associations
of crustacean zooplankton in 47 lakes of a region of Ontario affected by acid
precipitation were strongly related to pH and to the number of fish species
present in the lakes. However, fish and zooplankton were each correlated with
the same limnological variables, especially pH.94 Zooplankton communities
78
-------
become less complex with fewer species present as acidity increases. Food
supply, feeding habits, and grazing of zooplankton will probably be altered
following acidification, as a consequence of decreased biomass and species
composition of planktonic algae and bacteria. In streams continuously pol-
luted by AMD, the number of zooplankton species was low compared to the number
of individuals in less polluted conditions downstream.84
Surveys at many sites receiving acid precipitation in Norway, Sweden, and
North America4'14'18'47 have shown that waters affected by acid precipitation
have fewer species of benthic invertebrates than localities which are less
acid. In 832 Norwegian lakes, J. 0kland81 found no snails at pH values below
5.2; snails were rare in the pH range 5.2 to 5.8 and occurred less frequently
in the pH range 5.8 to 6.6 than in more neutral or alkaline waters. The
amphipod Gammarus lacustris, an important element in the diet of trout in
Norwegian lakes where it occurs, is not found in lakes with pH less than
6.O.82 Experimental investigations have shown that the adults of this species
cannot tolerate 24 to 48 hours of exposure to pH 5.O.15
In the River Duddon in England, pH is the overriding factor which
prevents permanent colonization by a number of species of benthic inverte-
brates, primarily herbivores, of the upper acidified reaches of the river.95
In the more acid tributaries (pH < 5.7), the fauna consisted of an impover-
ished Plecopteran community. Ephemeroptera, Trichoptera, Ancylus (Gastro-
poda), and Gammarus (Amphipoda) were absent. The epiphytic algal flora was
reduced (in contrast to increases noted in Norway), and leaf litter decompo-
sition was retarded. The food supply of the herbivores was apparently
decreased, and this may have played a role in the simplification of the
benthic fauna. Quantitative data concerning the effects of low pH on the
benthic fauna are also available for some acid Norwegian lakes,49 where
notably low standing crops have been observed.
Many studies of invertebrate communities in streams receiving AMD have
been conducted. Comparisons are usually made between affected and unaffected
zones or tributaries, and experimental acidification has been performed.54
The numbers of species, species diversity, and biomass are usually greatly
reduced. Generally, in AMD waters Chironomidae (midges) and Sialis (alderfly)
are the most tolerant macroinvertebrates. The order Trichoptera has more
tolerant species than does Plectoptera (stone flies), and Plecoptera has more
tolerant species than does Ephemeroptera (mayflies).25'26'44'45'84'102
This order of tolerance is essentially the same as seen in waters acidi-
fied by acid precipitation. However, the Hemiptera, Notonectidae (back-
swimmer), Corixidae (water boatman), and Gerridae (waterstrider) are often
abundant in acidified soft waters at pH as low as 4.0. This may, in part, be
due to lack of fish predation.
Benthic plant communities in lakes may be greatly altered as a conse-
quence of lake or stream acidification (as discussed above). Under these
conditions, benthic invertebrate populations may be affected by starvation,
evacuation, or extinction due to the loss of preferred habitat. Chironomids83
and other benthic invertebrates,20 present in many of the poorly buffered
northeastern lakes, have diverse feeding habits and habitats. These inverte-
brates, in many situations, will be affected by altered decomposition cycles
and variations in available foods caused by increased acidification.
79
-------
The tolerance of aquatic invertebrates to low pH varies over their life
cycles, and the emergence of adult insects seems to be a period particularly
sensitive to low pH levels. Bell11 and Moss,76 im similar studies with
Trichopter and Ephemeroptera, found emergence patterns to be affected at pH
levels which were higher than the 30-day survival limits. Many species of
aquatic insects emerge early in the spring, even through cracks in the ice and
snow cover. Because of the contamination of spring meltwaters by atmospheric
pollutants, including heavy metals,41'50'55'56 the early emergers must, in
many cases, be exposed to the least desirable water conditions.
Damage to invertebrate communities will influence other components of the
food chains. Benthic invertebrates assist with the essential function of
removing dead organic material. In litter bag experiments, the effects of
invertebrates on leaf composition were much more evident at higher pH than at
low pH.49 A reduction of grazing by benthic invertebrates may also contribute
to the accumulation of attached algae in acidified lake and streams.
A short reach of Norris Brook, a tributary to Hubbard Brook in New
Hampshire, was acidified to pH 4 in the spring-summer of 1977, to evaluate the
effects of acidification on a stream ecosystem. Excessive accumulations of
algae occurred, bacterial biomass and heterotrophic activity per unit of
organic matter were reduced, and both invertebrate diversity and biomass
decreased.43
In unstressed lake ecosystems there tends to be a continuous emergence of
different insect species available to predators from spring to autumn. In
acid-stressed ecosystems the variety of prey is reduced and periods may be
expected to occur in which the amount of prey available to fish (and water-
fowl) is diminished.
E. Effects on Vertebrates
Pough86 has described effects of acid precipitation on spotted sala-
manders (Ambystoma jeffersonianum and A. maculatum), which breed in temporary
rain pools. Below pH 5 and 7, respectively, these species suffered high
mortality during hatching in laboratory tests. This mortality was associated
with distinctive embryonic malformations. The development of salamander eggs
in five ponds near Ithaca, New York, ranging from pH 4.5 to 7.0, were
observed. An abrupt transition from low to high mortality occurred below pH
6. Although a synergistic effect of several stresses may have been possible,
the studies suggested that pH was the critical variable. Pough cites studies
which indicate a deline in British frog populations.86
In Tranevatten, a lake acidified by acid precipitation, near Gothenburg,
Sweden, the lake pH has declined to 4.0 to 4.5, and all fish have been elim-
inated. The frog species Rana temporaria is being eliminated as well.
Currently, only adults 8 to 10 years old are found. While many egg masses
were observed in 1974, few were found in 1977 The few larvae observed in
1977 subsequently died. Another species, Bufo bufo, is also being eliminated
from this lake.42
80
-------
Frogs and salamanders are important predators on invertebrates, including
mosquitoes and other pests, in lakes and puddles or pools. In turn, they are
themselves important prey for higher trophic levels in an ecosystem.
F. Biological Effects Summary
Acid precipitation, by causing increased acidity in lakes, streams,
pools, and puddles, can cause slight-to-severe alteration in communities of
aquatic organisms. The effects are similar to those observed in waters
receiving acid mine drainage (AMD), but the toxicology and chemistry are not
as complicated by the presence of high concentrations of heavy metals, chem-
ical floes, turbidity, etc., such as are found in AMD.
The kinds of effects that are likely to result from increasing acidifica-
tion are shown in Table V-l. In order to provide step-functions for damages,
which may be of use in modeling ecosystem acidification, a summary of damages
to aquatic organisms as functions of pH is presented in Table V-2.
81
-------
Table V-l. Damages to aquatic biota likely to occur with inceasing acidity.
1. Bacterial decomposition is reduced and fungi dominate saprotrophic com-
munities. Organic debris accumulates rapidly.
2. The ciliate fauna is greatly inhibited.
3. Nutrient salts are taken up by plants tolerant of low pH (mosses, fila-
mentous algae) and by fungi. Thick mats of these materials may develop
which inhibit sediment-to-water nutrient exchange and choke out other
aquatic plants.
4. 'Phytoplankton species diversity, biomass, and production are reduced.
5. Zooplankton and benthic invertebrate species diversity and biomass are
reduced. Remaining benthic fauna consists of tubificids and Chironomus
(midge) larvae in the sediments. Some tolerant species of stone flies
and mayflies persist as does the alderfly. Air-breathing bugs (water-
boatman, backswimmer, water strider) may become abundant.
6. Fish populations are reduced or eliminated (see Section VI).
Table V-2. Summary of damages to aquatic organisms with decreasing pH.
pH range Effects
8.0-6.0 Long-term changes of less than 0.5 pH units likely to alter the
biotic composition of freshwaters to some degree. The significance
of these slight changes is however, not great.
A decrease of 0.5 to 1.0 pH units in the range 8.0 to 6.0 may cause
detectable alterations in community composition. Productivity of
competing organisms will vary. Some species will be eliminated.
5.5-5.0 Many species wil be eliminated, and species numbers and diversity
indices will be reduced. Crustacean zooplankton, phytoplankton,
molluscs, amphipods, most mayfly species, and some stone fly species
will begin to drop out. In contrast, several pH-tolerant inverte-
brates will become abundant, especially the air-breathing forms
(e.g., Gyrinidae, Notonectidae, Corixidae), those with tough cuti-
cles which prevent ion losses (i.e. Sialis lutaria), and some forms
which live within the sediments (Oligochaeta, Chironomidae, and
Tubificidae). Overall, invertebrate biomass wil be greatly reduced.
5.0-4.5 Decomposition of organic detritus will be severely impaired.
Autochthonous and allochthonous debris will accumulate rapidly.
Most fish species will be eliminated.
< 4.5 All of the above changes will be greatly exacerbated, and all fish
will be eliminated.
82
-------
VI. REGIONAL ASSESSMENT Of ACIDIFICATION IMPACTS OF FISH POPULATION
A. Introduction
Acidification of surface waters in geologically sensitive regions by acid
precipitation may be viewed as a continuous process of water quality change,
analogous to the acidimetric titration of a bicarbonate solution. The "titra-
tion curve" can be segmented into three stages, which define both the extent
and nature of water quality change and the biological impacts. The initial
stage of acidification is characterized by decreased alkalinity, but pH levels
remain greater than 6.0 and bicarbonate buffering is maintained. No signifi-
cant impacts on fish population have been observed at this level of acidifica-
tion. Loss of HC03 buffering, and resulting severe temporal pH fluctuations
occur in the second stage of acidification. Stress, reproductive inhibition,
and episodic mortalities may initiate recruitment failure and eventual extinc-
tion of fish populations during this stage of acidification. The final stage
of acidification is characterized by chronically depressed pH and elevated
toxic metal levels. Fish are generally absent from waters at this level of
acidification. This general view of acidification impact levels on fish
populations is supported by extensive, carefully documented data from sensi
tive regions in Norway103 and eastern North America.9'91
More recently, Henriksen,52'53 has developed models of the acidification
process which quantitatively relate the stages of acidification described
above to precipitation acidity levels in sensitive regions. These models
provide a basis for predicting the regional extent and levels of acidification
which have occurred in sensitive areas where historial data are lacking. The
models also permit evaluation of expected fish population status on a regional
level. Considering the paucity of carefully documented fish population data
for waters in sensitive areas of the eastern United States and the time
constraints for obtaining and interpreting new data, it was decided to utilize
the predictive models to obtain a regional perspective of acidification
impacts on fish populations. The approach taken was to first calibrate the
Norwegian model using North American data and verify the predictions obtained
by comparison with known water quality and fish population data from the
Adirondack region (New York) and Experimental Lakes Area (Ontario, Canada).
The next step will be to apply the models to water quality data from other
sensitive regions in the United States to evaluate acidification and popula-
tion status.
B. Procedure
1. Data sources. Water quality and fish population data obtained from
the following sources were utilized in model calculation and testing.
83
-------
1. Water chemistry and fish survey data for Adirondack Mountain lakes (New
York): Schofield (1976 and file data); New York State Biological Surveys
of the Upper Hudson, Raquette, Mohawk-Hudson, and Black River Watersheds
(1931-1934, NYS Dept. Conservation).
2. Stream water quality and fish population status in selected areas of West
Virginia: Personal Communication and file data from Donald Gasper, West
Virginia Dept. of Natural Resources.
3. Water quality data for lakes in the BWCA, N.E. Minnesota: G. E. Glass,
EPA Research Laboratory (Duluth), Draft report on Impacts of Air Pollu-
tants on Wilderness Areas of N. Minnesota.
4. Water chemistry and fish population data for Connecticut lakes.24'79
5. Water chemistry and fish population data for Rhode Island lakes.39'40
6. Water chemistry for Massachusetts lakes.21
7. Water chemistry and fish population for ELA lakes.9
8. Water chemistry and fish population data for the North Carolina bay
lakes.30'31
9. Water chemistry for coastal plain streams in South Carolina and
Georgia.10
C. Model Calibration and Verification
1. Acidification. Acidification is defined as the decrease in alka-
linity, which results from an equivalent increase in strong acid input. Three
independent approaches to estimating regional acidification were applied to
the Adirondack lake data base. The first approach developed by Henriksen51
utilized the empirical relationships between precipitation pH (volume weighted
annual mean) and estimated regional acidification of lakes in southern Norway.
Regional acidification, a decrease in alkalinity associated with excess
sulfate (corrected for seasalt) in precipitation (Ac-S, |jeq/l), was calculated
by regression as:
_ (lQ-pH x 106) - 17
AC S ~ 0.38
The volume weighted, mean annual hydrogen ion concentration for precipitation
collected in the western Adirondack region (Hinckley, New York; USGS network)
in 1975 was pH 4.26. Substituting this value in Equation (1) yields an esti-
mated acidification level of 99.9 MecL/l as the average regional acidification
of the western Adirondacks.
The second approach to estimating acidification was also as given by
Henriksen51: Acidification = preacidification alkalinity - present day alka-
84
-------
Unity. In this case, preacidification alkalinities (ALKp) for the Adirondack
lakes were estimated from 1975 measurements of calcium concentrations and the
empirical relationships between calcium concentration and alkalinity observed
in the Experimental Lakes Area (Ontario) waters:
ALKp (peq/1) =1.42 (Ca peq/1) - 32 (2)
The assumptions made in employing this method of estimating preacidification
alkalinities are as follows:
a) Alkalinity in pristine (unacidified) lakes (e.g., the ELA region) is due
primarily to HCO 3, which is accompanied by approximately equivalent
amounts of calcium and magnesium, and the Ca/Mg ratio is relatively
constant. This relationship (Equation 2) is of a general nature and can
be applied to other regions with similar geological settings. This
assumption is supported by the similarity of the calcium-alkalinity
relationship for pristine lakes in nothern Norway and the ELA.52
b) Calcium levels do not increase or decrease in response to acidification.
Historical calcium data are not available for the Adirondack region, but
estimates of calcium from the 1930s alkalinity data and Equation (2) were
not significantly different (p = 0.05) from calcium levels measured in
1975 for 36 Adirondack lakes. Additional support for this assumption
comes from studies in Sweden73 which demonstrated that no significant
changes in calcium concentration were detected in acidified lakes of
south-central Sweden, where significant increases in H and S042 did
occur Also, Wiklander100 reported that base exchange efficiency is very
low in acidic podzols, where the exchange complex is typically dominated
by H and Al 3 ions, rather than Ca 2. Watt (1979) compared the pH and
ionic composition of lakes in Nova Scotia sampled in 1955 and 1977.
Significant decreases in pH levels were observed but no significant
change in Ca++ was found.96
Acidification (Ac-Ca) was then calculated as the difference between
preacidification alkalinity (Equation 2) and current alkalinity measurements
(ALK*):
Ac-Ca peq/1 = ALKp - ALK" (3)
The estimated mean acidification for 215 Adirondack lakes sampled in 1975 was
Ac-Ca = 99.6 peq/1 (SD = 30.6), which compares well with the first estimate of
acidification (Ac-S = 99.9 peq/1). These estimates of lake acidification are
not correlated with current pH levels (Figure VI-1), which is indicative of
the regional uniformity of acidification. Estimated acidification levels are
significantly correlated with sulfate levels (Figure VI-2), indicating that
atmospheric strong acid (H2S04) inputs are the primary source of acidifi-
cation.
The third approach to estimating acidification was to directly compare
historical (1930s data) and current (1975) alkalinity measurements. Alka-
linity measurements obtained in the 1930s for Adirondack lakes employed methyl
orange indicator for end point determination in the titration of water samples
85
-------
CO
ffl
z
O
150
O
Lu
O
O
100
50
0
**>*
-'
** -
*..
.
V
O
»
Figure VI-1.
4.0 4.5 5.0 5.5 6.0 6.5 7.0
pH
Acidification of Adirondack lakes in relation to pH measured in 1975 (215 lakes)
Mean acidification = 99.6 ueq/1, SD = 30.6, where acidification (ueq/1) = 1.42
(Ca ueq/1) - 32 (alls, yeq/1) .
-------
CO
-------
for total alkalinity measurement. This method yields an apparent end point pH
of approximately 4.4, which overestimates the true alkalinity (as measured by
Gran plot location of the titration curve inflection for the 1975 data) by the
amount of free H+ ion present in solution at this pH level (40 (Jeq/1 H+). The
1930s alkalinity measurements were corrected to correspond to true alkalinity
levels by subtracting 40 |Jeq/l from each value. Acidification was then esti-
mated as the difference between corrected 1930s alkalinity levels and 1975
alkalinity measurements. The mean acidification for 36 Adirondack lakes,
where data were available from both periods, was 104.7 |Jeq/l (SD 50.9).
Estimated acidification for the same lakes, using the calcium-alkalinity model
(Equation 2), was 105.7 (Jeq/1 (SD 31.8) which is not significantly different
from the previous estimate (p 0.01).
2. Lake classification. Lakes were classified into three categories,
which define acidification and fish population status as discussed previously,
utilizing regional acidification estimates and lake calcium concentrations.
Henriksen53 constructed a nomogram by grouping data from 719 lakes in southern
Norway into discrete pH classes and then determined the least squares regres-
sions of calcium on acidification for each class. The resulting nomograph
(Figure VI-3) can then be used to predict lake pH (and fish population status)
from lake calcium levels and either regional acidification or precipitation
pH. The two regression lines derived from the Norwegian lake data define the
boundaries for the three classifications:
HC03 lake boundary Ca (|Jeq/l = 49.9 (0.28 + 0.0192 Ac-S) (4)
Acid lake boundary Ca (peq/l = 49.9 (0.07 + 0.0139 Ac-S)
The characteristics, espected pH values, and boundary calcium levels for this
classification, as applied to the Adirondack data, are given in Table VI-1.
The observed pH levels for lakes classified as described below are illus-
trated in Figure VI-3. For the Adirondack data, 70% of the lakes were
correctly classified and 98% of the ELA lakes were corrected designated as
HC03 lakes. These results are similar to those obtained by Henriksen53 for
classification of lakes in southern Norway. An additional test of the
validity of this model for predicting lake pH changes resulting from acidifi-
cation is illustrated in Figure VI-4. Increase in H+ concentration between
the 1930s and 1975, as determined from differences in pH, were compared to
calcium concentrations measured in 1975 for 36 Adirondack lakes. The break in
the plot of H+ against calcium occurs precisely at the calcium level predicted
from Equation (4). The results of these model predictions for Adirondack and
ELA lake data appear to be quite adequate for determining acidification
status.
Fish population status in the Adirondack lakes is presented in Table
VI-2. The frequencies of lakes not supporting fish life in each of the three
lake classifications are comparable to those reported by Henriksen53 for lakes
in southern Norway. The lakes of th ELA are still predominantly pristine,
HC03 type waters and the fish populations are unaffected.9
88
-------
50%
50%
FREQUENCY
CP
£
3.5
3.0
2.5
2.0
1.5
1.0
0,5
0
n ELA
I;. HC03 3";
;!;LAKES /
/
TRANSITION
LAKES
ACID LAKES
6.0 5.0 4.8 4.6 4.4 4.2
pH PRECIPITATION
4.0
pH
>6.0
5.3-6.0
4.7-5.3
<4.7
o
Q
o:
£ VECTORS
UJ
ACIDIFICATION
(LONG TERM)
Figure VI-3. Nomograph for predicting lake pH and fish population status from
lake Ca levels and either regional acidification or precipita-
tion pH. Regression lines for the boundary between HCO-3 and
transition lakes, and for the acid lakes boundary with transi-
tion lakes, are given in the text as Equations (4) and (5)
respectively.
89
-------
0)
O
i
in
en
25
c_ %*/
20
0
5
0
1
1
0
0.5
JO L5
Co
2.0 T 2.5
'/*. '
3.0 3.5
Figure VI-4.
Changes in the H+ ion concentrations in 36 Adirondack lakes from the 1930s to 1975, in
relation to Ca levels measured in 1975. The arrow designates the critical Ca level for
loss of HC03 buffering as predicted from Henriksen's model53 for a region exposed to
;ion witti a weighted mean pH of 4.26.
-------
Table VI-1. Lake classification for Adirondack Mountain lakes, based on the
nomograph of Henriksen.53
Lake Class
Expected pH
Fish Population
Boundary Ca peq/1
(Adirondack)
HC03
Transitional
Acid
> 5.3
4.7 - 5.3
< 4.7
Good
Sparse
Barren
> 110
73 - 100
< 73
Table VI-2. Fish population status in 214 Adirondack Mountain lakes, class-
ified according to the nomograph of Figure VI-3.
Lake Type
Expected pH
Number of lakes
Number in pH range
% correctly classified
% without fish
HC03
>5.3
69
61
88.4
2.9
Trans
4.7-
97
65
67.
54.
5.3
0
6
Acid
<4.
48
27
56.
91.
.7
,3
,7
ENDNOTE
The 94 pages of Appendix I and II are available separately from National
Technical Information Service (NTIS) as EPA-600/3-80-024b under the same title
as this volume (EPA-600/3-80-024a). The NTIS address is Springfield, VA
22161.
Appendix I contains four maps for each of 22 states, depicting minimum
and average alkalinity of surface waters on a county basis, and indicating the
location of pH and alkalinity sampling stations with th measured data range
for 1976-1979. The states included are: Connecticut, Delaware, Florida,
Georgia, Kentucky, Maine, Maryland, Massachusetts, Michigan, Minnesota, New
Hampshire, New Jersey, New York, North Carolina, Pennsylvania, Rhode Island,
South Carolina, Tennessee, Vermont, Virginia, West Virginia, and Wisconsin.
Appendix II is a three page list of additional sources for water chemis-
try data which have been used in this report.
91
-------
REFERENCES
1. Aimer, B., W. Dickson, C. Ekstrom, E. Hb'rnstrb'm, and U. Miller. 1974.
Effects of acidification of Swedish Lakes. Ambio 3:330-336.
2. Aimer, B., W. Dickson, C. Ekstrom, and E. Hornstrom. 1978. I_n Sulfur in
the Environment, Part 2 (ed. Nriagu, J. 0.) 272-311 (Wiley, New York).
3. American Public Health Association. 1971. Standard Methods for the
Examination of Water and Wastewater. Thirteenth ed. APHA. Washington,
B.C.
4. Andersson, I., 0. Grahn, H. Hultberg, and L. Landner. 1974. Jamforande
undersokning av olika tekniker for alterstallande av forsurade sjoar.
Institute for Water and Air Research, Stockholm, STU Report 73-3651.
5. Anonymous. 1979. Soil Survey of Augusta County, Virginia. USDA Soil
Conserv. Ser and For. Ser.
6. Barnes, R. S., D. E. Spyridakis, and P. B. Birch. 1979. Atmospheric
input of trace metals to lake sediments in western Washington State
(abs.). Am. Chem. Soc. Ann. Meet., Environ. Div.
7. Bartholomew, M. J. 1978. Geology of the Greenfield and Sherando Quad-
rangles, Virginia. Virginia Dept. of Mineral Resour. Pub.
No. 4.
8. Beamish, R. J. and H. H. Harvey. 1972. J. Fish Res. B. Can. 29:1131
1143.
9. Beamish, R. S., L. M. Bloues, and G. A. McFarlane. 1976. Tech. Rep. No.
607, Dept. of Environ., Fish and Marine Serv. Can.
10. Beck, K. C., J. A. Router, and E. M. Perdue. 1974. Geochem. et.
Cosmochem. 38:344.
11. Bell, H. L. 1971. Effect of low pH on the survival and emergence of
aquatic insects. Water Res. 5:313-319.
12. Bick, H. and E. F. Drews. 1973. Selbstreinigung und Ciliatenbesiedlung
in saurem Milieu (Modellversuche). Hydrobiologia 42:393-404.
13. Boling, R. H. , Jr, E. D. Goodman, J. A. VanSickel, J. 0. Zimmer, K. W.
Cummins, R. C. Petersen, and S. R. Reice. 1975. Toward a model of
detritus processing in a woodland stream. Ecology 56:141-151.
93
-------
14. Borgstrom, R. , J. Brittain, and A. Lillehammer. 1976. Evertebrater og
surt vann: oversikt over innsamlingslokaliteter. SNSF IR 21/76.
SNSF-Project, 1432 Aas-NLH, Norway.
15. Borgstrom, R. and G. R. Hendrey. 1976. pH tolerance of the first larval
stages of Lepidurus articus (Palas) and adult Gammarus lacustris G. 0.
Sars. SNSF IR 22/76. SNSF-Project, 1432 Aas-NLH, Norway.
16. Cogbill, C. V. and G. E. Likens. 1974. Precipitation in the north-
eastern United States. Water Resour. Res. 10:1133-1137.
17. Cole, J. J. and G. E. Likens. 1978. Measurement of mineralization of
phytoplankton detritus in an oligotrophic lake. (Manuscript in press).
18. Conroy, N., K. Hawley, W. Keller, and C. LaFrance. 1975. Influences of
the atmosphere on lakes in the Sudbury area. J. Great Lake Res. 2,
Suppl. 1(1976):146-165.
19. Cragin, J. H., M. M. Herron, and C. C. Langway, Jr. 1975. The chemistry
of 700 years of precipitation at Dye 3, Greeland. CRREL Res. Report No.
341, 18 pp.
20. Cummins, K. W. 1973. Trophic relations of aquatic insects. Ann. Rev.
Entomol. 18:183-206.
21. Daly, L. M. 1965. DJ Completion Report F-20-R-2, Mass. Fish and Game
Dept.
22. Davis, R. B. , S. A. Norton and D. F. Brakke. 1979. Heavy metal deposi-
tion in bottom sediments and acidification of New England lakes from
atmospheric inputs, and the effect on lake biota (abs.). Conf. on Great
Lakes Res.
23. Davis, R. 0., M. 0. Smith, J. H. Bailey, and S. A. Norton. 1978. Acid-
ification of Maine lakes by acid precipitation. Verh. Intern. Verein.
Limnol., (XX Congress, Copenhagen, 1977). Vol. 20:532-537.
24. Deevey, E. S. and J. S. Bishop. 1942. Bull. No. 63, Conn. State Bd.
Fish and Game.
25. Dills, G. arid D. T. Rogers. 1974. Macroinvertebrate community structure
as an indicator of acid mine pollution. Environ. Pollut. 6:239-262.
26. Dinsmore, B. H. 1968. The aquatic ecology of Toms Run, Clarion County,
Pennsylvania, preceding watershed reclamation. Pennsylvania Dept. of
Health, Bureau of Sanitary Engineering, Div. of Water Qual. Pub. No. 21.
27. Dochinger, L. S. and T. A. Seliga (eds.). 1976. Proc. of the First Int.
Symp. on Acid Precip. and the For. Ecosystem. USDA For. Ser. Gen. Tech.
Rep. NE-23.
28. Edgerton, E. Personal communication.
94
-------
29. Fogg, G. E. 1977. Excretion of organic matter by phytoplankton.
Limnol. Oceanogr. 22:576-577-
30. Frey, D. G. 1949. J. Elisha Mitch. Soc. 65:1.
31. Frey, D. G. 1951. J. Elisha Mitch. Soc. 67:1.
32. Galloway, J. N. and G. E. Likens. 1976. Calibration of collection
procedures for the determination of precipitation chemistry. Water, Air,
and Soil Pollut. 6:241.
33. Galloway, J. N. and E. B. Cowling. 1978. The effects of precipitation
on aquatic and terrestrial ecosystems: a proposed precipitation chemistry
network. APCA J. 28:229.
34. Galloway, J. N. 1979. Unpublished data.
35. Gasith, A. and A. D. Hasler. 1976. Airborne litterfall as a source of
organic matter in lakes. Limnol. Oceanogr. 21:253-258.
36. Gathright, T. M. II. 1976. Geology of the Shenandoah National Park.
Virginia Dept. of Mineral Resour. Bull. 86.
37. Grahn, 0., H. Hultberg and L. Landner. 1974. Oligotrophication--a
self-accelerating process in lakes subjected to excessive supply of acid
substances. Ambio 3:93-94.
38. Grahn, 0. 1975. Macrophyte succession in Swedish lakes caused by depo-
sition of airborne acid substances. In Proc. of the First Int. Symp. on
Acid Precip. and the For. Ecosystem, Ohio State Univ., 12-15 May 1975.
USDA Forest Service Gen. Tech. Report NE-23, 1976, p. 519-530.
39. Guthrie, R. C., J. A. Stolgitis, and W. L. Bridges. 1973. Fish Rep. No.
1, Rhode Island Dept. Nat. Res.
40. Guthrie, R. C. and H. A. Stolgitis. 1977. Fish Rep. No. 3, Rhode Island
Dept. Nat. Res.
41. Hagen, A. and A. Langeland. 1973. Polluted snow in southern Norway and
the effect of the meltwater on freshwater and aquatic organisms.
Environ. Pollut. 5:45-57-
42. Hagstrom, T. 1977. Grodornas Forsvinnande i en Fb'rsurad sjo (The
disappearance of frogs in an acidified lake). Sveriges Natur 11:367-369.
43. Hall, R. J., G. E. Likens, S. Fiance, and G. Hendrey. 1980. Experi-
mental stream acidification in the Hubbard Brook Experimental Forest.
Accepted for publication in Ecology.
44. Harrison, A. D. 1958. The effects of sulphuric acid pollution on the
biology of streams in the Transvaal, South Africa. Verh. Internat. Ver.
Limnol. 13:603-610.
95
-------
45. Harrison, A. D. and J. D. Agnew. 1962. The distrubtion of invertebrates
endemic to acid streams in the western and southern Cape Province. Ann.
Cape Prov. Mus. 2:273-291.
46. Hellebust, J. A. 1974. Extracellular products. In W. D. Steward (ed.)
Algal physiology and biochemistry. Univ. Calif.
47. Hendrey, G. R. and R. F. Wright. 1975. Acid precipitation in Norway:
Effects on aquatic fauna. In Proc. First Specialty Symp. Atmos. Contrib.
Chem. Lake Waters, Int. Assoc. Great Lake Res., Orillia, Ontario, 28
Sept-1 Oct 1975. J. Great Lakes Res. 2 Suppl. 1:192-207-
48. Hendrey, G. R. 1976. Effects of low pH on the growth of periphytic
algae in artificial stream channels. SNSF-Project, 1432 Aas-NLH, Norway.
49. Hendrey, G. R. , K. Baalsrud, T. Traaen, M. Laake and G. Raddum. 1976.
Acid precipitation: Some hydrobiological changes. Ambio 5:224-227-
50. Henriksen, A. and R. F. Wright. 1976. Concentrations of heavy metals in
small Norwegian lakes. SNSF-Project Report. SNSF-Project, 1432 Aas-NLH,
Norway.
51. Henriksen, A. 1979. A simple approach for identifying and measuring
acidification on freshwater. Nature Vol. 278.
52. Henriksen, A. 1979a. Nature 278:542.
53. Henriksen, A. Personal communication. (Manuscript in preparation).
54. Herricks, E. E. and J. Cairns, Jr. 1974. Rehabilitation of streams
receiving acid mine drainage. Bull. 66, Virginia Water Resour. Res.
Center, VPI, Blacksburgh, Virginia 24061.
55. Hultberg, H. 1975. Thermally stratified acid water in late wintera
key factor inducing self-accelerating process which increase acidifica-
tion. In Proc. First Int. Symp. Acid Precip. and the For. Ecosystem,
Ohio State Univ., 12-15 May, 1975. USDA For. Ser. Gen. Tech. Rep. NE-23,
1976.
56. Johannessen, M. , T. Dale, E. Gjessing, A. Henriksen, and R. F. Wright.
1975. Acid precipitation in Norway: The regional distribution of
contaminants in snow and the chemical concentration processes during
snowmelt. Proc. Int. Symp. Isotopes and Impurities in Show and Ice, Int.
Assoc. Hydrol. Sci., Grenoble, France, Aug 1975. SNSF FR 4/75.
SNSF-Project, 1432 Aas-NLH, Norway.
57 Johnson, A. J. 1979. Acidification of headwater streams in the New
Jersey Pine Barrens. J. of Environ. Qual., Vol. 3. (In press).
58. Johnson, M. G. , M. F. P. Michalski, and A. E. Christie. 1970. Effects
of acid mine waste on phytoplankton communities of two northern Ontario
lakes. J. Fish. Res. Bd. Can. 27:425-444.
96
-------
59. Johnson, N. M. 1979. Acid rain: neutralization within the Hubbard
Brook ecosystem and regional implications. Science 204:497.
60. Jordan, M. and G. E. Likens. 1975. An organic carbon budget for oligo-
trophic lakes in New Hampshire, USA. Verh. Int. Verein. Limnol.
19:994-1003.
61. Keroher, G. C., et al. 1966. Lexicon of geologic names of the United
States for 1936-196oT~ Parts 1, 2, and 3: U.S. Geol. Sur. Bull. 1200.
62. Keroher, G. C., e_t al. 1970. Lexicon of geologic names of the United
States for 1961-1967-. U.S. Geol. Sur. Bull. 1350.
63. King, D. L. 1970. The role of carbon in eutrophication. J. Water
Pollut. Control Fed. 42:2035-2051.
64. Kramer, J. R. 1976. Geochemical and lithological factors in acid
precipitation. In Proc. of the First Int. Sym. on Acid Precip. and the
For. Ecosystem. U.S. Dept. Agric. For. Ser. Gen. Tech. Rep. NE-23, p.
611-618.
65. Kwiatkowski, R. E. and J. C. Roff. 1976. Effects of acidity on the
phytoplankton and primary productivity of selected northern Ontario
lakes. Can. J. Bot. 54:2546-2561.
66. Laake, M. 1976. Effekter av lav pH piproduksjon, nedbrytning og
stoffkretslop i litteralsoner. SNSF-Project, 1432 Aas-NLH, Norway.
67. Lazrus, A. L., E. Lorange, and J. P. Lodge, Jr. 1970. Lead and other
metal ions in United States precipitation. Environ. Sci. and Tech.
4:55-58.
68. Leivestad, H., G. Hendrey, I. P. Muniz, and E. Snekvik. 1976. p. 87-111
In Impact of Acid Precipitation on Forest and Freshwater Ecosystems in
Norway (Summary Report, Phase I of SNSF-Project).
69. Likens, G. E. 1976. Acid precipitation. Chem. Eng. News 54:29-44.
70. Likens, G. E., F. H. Bormann, R. S. Pierce, J. S. Eaton and N. M.
Johnson. 1977. Biogeochemistry of a Forested Ecosystem. Springer-
Verlag, New York.
71. Likens, G. E., R. F. Wright, J. N. Galloway, and T. J. Butler. 1979.
Acid rain. Sci. Am. 241:43-51.
72. Liljestrand, H. M. and J. J. Morgan. 1979. Acidic precipitation in
southern California (abs.). Am. Chem. Soc. Ann. Meet., Environ. Div.
73. Malmer, N. 1975. SNV Res. Rep. No. PM 402E, Univ. Lund, Sweden.
74. McFee, W. W. 1980. Sensitivity of soil regions to long-term acid
precipitation. U.S. Environmental Protection Agency Rep. EPA-600/3-80-
013.
97
-------
75. Mills, A. 1979. Dept. of Environ. Sci., Univ. of Vir., Charlottesville,
Virginia. Personal communication.
76. Moss, B. 1973. The influence of environmental factors on the distribu-
tion of freshwater algae: an experimental study, II. The role of pH and
the carbon dioxide bicarbonate system. J. Ecol. 61:157-177.
77. National Academy of Sciences. 1978. Sulfur Oxides.
78. National Atmospheric Deposition Program. 1979. NADP First data report:
July 1978 through February 1979. Nat. Resour. Ecol. Lab., Colo. State
Univ.
79. Norvell, W. A. and C. R. Frink. 1975. Bull. No. 759, Conn. Agric. Exp.
Stn., New Haven.
80. Oden, S. 1976. The acidity probleman outline of concept. WASP 6:137-
166.
81. 0kland, J. 1969. Distribution and ecology of the freshwater snails
(Gastropoda) of Norway. Malacologia 9:143-151.
82. 0kland, K. A. 1969. On the distribution and ecology of Gammarus
lacustris G. 0. Sars in Norway, with notes on its morphology and biology.
Nor. J. Zool. 17:111.
83. Oliver, D. R. 1971. Life history of the Chironomidae. Ann. Rev.
Entomol. 16:211-229.
84. Parsons, J. D. 1968. The effects of acid strip-mine effluents on the
ecology of a stream. Arch. Hydrobiol. 65:25-50.
85. Parsons, J. D. 1977. Effects of acid mine wastes on aquatic ecosystems.
W.A.S.P. 7:333-354.
86. Pough, F. H. 1976. Acid precipitation and embryonic mortality of
spotted salamanders, Ambystoma maculatum. Science 192:68-70.
87. Rockingham County, Virginia, Soil Conservation Service. 1979. (Unpub-
lished data).
88. Saunders, G. W. and T. A. Storch. 1971. Coupled oscillatory control
mechanism in a planktonic system. Nature 230:58-60.
89. Scheider, W., J. Adamski, and M. Paylor. 1975. Reclamation of acidified
lakes near Sudbury, Ontario. Ont. Min. of the Environ., Texdale,
Ontario, Canada.
90. Schofield, C. L. 1975. p. 477 In Proc. First Int. Symp. on Acid Precip.
and the For. Ecosystem. USDA For. Ser. Gen. Tech. Rep. NE-23.
91. Schofield, C. L. 1976. Acid precipitation: effects on fish. Ambio.
5:228.
98
-------
92. Schofield, C. L. 1979. Acidification impacts on fish populations
(abs.). Am. Chem. Soc. Ann. Meet., Environ. Div.
93. Shaffer and Galloway. Unpublished data, to be presented as an M.S.
thesis. Dept. of Environ. Sci., Univ. of Virginia.
94. Sprules, G. W. 1975. Midsummer crustacean zooplankton communities in
acid-stressed lakes. J. Fish. Res. Bd. Can. 32:389-395.
95. Sutcliffe, D. W. and T. R. Carrick. 1973. Studies on mountain streams
in the English lake district, I. pH, calcium and the distribution of
invertebrates in the River Duddon. Freshwater Biol. 3:437-462.
96. Watt, D. W. , D. Scott, and S. Ray. 1979. Acidification and other chem-
ical changes in Halifax County lakes after 21 years. Limnol. Oceanogr.
24:1154-1161.
97. Werner, J. J. 1966. Geology of the Vesuvius Quadrangle, Virginia. Vir.
Dept. of Mineral Resour. Rep. of Investigations No. 7.
98. Wetzel, R. G. 1975. Limnology. W. B. Saunders Co. 743 pp.
99. Wiebe, W. J. and D. F. Smith. 1977. Direct measurement of dissolved
organic carbon release by phytoplankton and incorporation by microhetero-
trophs. Marine Biol. 42:213-223.
100. Wiklander, L. 1979. EPRI Rep. No. SOA77-403.
101. Winkler, E. M. 1976. Natural dust and acid rain. p. 209-218 In: Proc.
First Int. Symp. on Acid Precip. and the For. Ecosystem. USDA For. Ser.
Gen. Tech. Rep. NE-23.
102. Wojcik, B. and L. Butler. 1977. Aquatic insects as indicators of stream
environmental quality in northern West Virginia. Bull. 653T. West Vir
Agric. and For. Exp. Stn., Morgantown, West Virginia.
103. Wright, R. F. and E. Snekvik. 1979. Verh. I. Verein. Limnol. 20:765.
104. Wright, R. F. and E. T. Gjessing. 1976. Changes in the chemical compo-
sition of lakes. Ambio 5:219.
105. Zimmerman, A. P. and H. H. Harvey. 1979. Sensitivity to acidification
of waters of Ontario and neighboring states. Final rep. for Ontario
Hydro. Univ. Toronto, Ontario. 48 pp.
99
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-80-024
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Geological and Hydrochemical Sensitivity to the Eastern
United States to Acid Precipitation
5. REPORT DATE
January 1980 issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
G. R. Hendrey, J. N. Galloway, S. A. Norton, and
C. L. Schofield
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Brookhaven National Laboratory
Upton, New York
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
IA 79-D-X0672
12. SPONSORING AGENCY NAME AND ADDRESS
Corvallis Environmental Research Laboratory
U. S. Environmental Protection Agency
Corvallis, OR 97330
13. TYPE OF REPORT AND PERIOD COVERED
Final, May-Oct. 1979
14. SPONSORING AGENCY CODE
EPA/600/02
15. SUPPLEMENTARY NOTES
16. ABSTRACT
State geologic maps were used to produce a detailed map of the bedrock geology of
the states east of the Mississippi River, on the premise that impact of acidic
precipitation to aquatic ecosystems is largely a function of the buffering capacity
of the underlying bedrock. Available alkalinity data for surface waters were simul-
taneously compiled, on a county-by-county basis in each of the 27 eastern states,
and compared with the bedrock to verify the geologic definition of sensitivity.
Intensive studies in the Blue Ridge Mountains in Virginia demonstrated the relation-
ship; pH and buffer capacity of streams there varied greatly, with bedrock geology
the major factor controlling stream chemistry. In North Carolina and New Hampshire,
the geological map was evaluated by comparing it to county-by-county maps of current
minimum alkalinity. In North Carolina, predictions were 83 percent correct, and in
New Hampshire, 100 percent. Impacts of acidification on aquatic biota are discussed.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
cos AT I Field/Group
Acidification, precipitation, geological
maps, geology, water chemistry, Aquatic
biology
Acid precipitation
Acid rain
Atmospheric deposition
08-G
08-H
18. DISTRIBUTION STATEMENT
Release to public
19. SECURITY CLASS (This Report)
unclassified
20. SECURITY CLASS (This page)
unclassified
21. NO. OF PAGES
110
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
PREVIOUS EDITION IS OBSOLETE
100
------- |