Content uploaded by Alan T Herlihy
Author content
All content in this area was uploaded by Alan T Herlihy on Jan 23, 2020
Content may be subject to copyright.
Environmental Pollution 77
(1992) 115-122
Sources of acidity in lakes and streams of the
United States
Philip R. Kaufmann, a Alan T. Herlihy ~ and Lawrence A. Baker b
a Utah State University and b University of Minnesota, c/o USEPA Environmental Research Laboratory,
200 SW 35th St, Corvallis, Oregon 97333, USA
Acidic (acid neutralizing capacity [ANC] < 0) surface waters in the United States
sampled in the National Surface Water Survey (NSWS) were classified into three
groups according to their probable sources of acidity: (1) organic-dominated
waters (organic anions >SO* + NO3); (2) watershed sulphate=dominated waters
(watershed sulphate sources > deposition sulphate sources); and (3) deposition-
dominated waters (anion chemistry dominated by inputs of sulphate and nitrate
derived from deposition). The classification approach is highly robust; therefore,
it is a useful tool in segregating surface waters into chemical categories. An
estimated 75% (881) of acidic lakes and 47% (2190) of acidic streams are
dominated by acid anions from deposition and are probably acidic due to acidic
deposition. In about a quarter of the acidic lakes and streams, organic acids were
the dominant source of acidity. In the remaining 26% of the acidic streams,
watershed sources of sulphate, mainly from acid mine drainage, were the
dominant source of acidity.
INTRODUCTION
Atmospheric deposition of acid anions derived from
fossil fuel combustion is a likely cause of chronically
acidic waters in several sensitive parts of North America
(Schindler, 1988). However, other anthropogenic activi-
ties, such as coal mining, are responsible for widespread
surface water acidification, and some surface waters are
known to be acidic because of organic acids or oxidation
of naturally exposed sulphide minerals (Huckabee
et al.,
1975). Other processes, including afforestation, natural
soil development, sulphate retention, neutral salt reten-
tion and natural hydrologic variations, may alter the
sensitivity of surface waters to acidification and may
postpone, ameliorate or exacerbate the pH depressions
that can result from anthropogenic loadings of acid
anions (e.g. Nilsson, 1982; Baker
et al.,
1988; Church
et al.,
1989).
Recent analyses of chemical data from surface water
surveys, combined with geochemical modelling and data
from other sources, have allowed interpretation of the
probable causes of acidification in streams (Herlihy
et
al.,
1990, 1991) and lakes (Marmorek
et al.,
1989) in
portions of the United States. This paper demonstrates
the utility of geochemical classification, in combination
Environ. Pollut.
0269-7491/92/$05.00 © 1992 Elsevier Science
Publishers Ltd, England. Printed in Great Britain
115
with survey data from the National Surface Water
Survey (NSWS) of the US Environmental Protection
Agency (EPA), in making a national-scale interpretation
of the probable sources of current acidity in acidic
surface waters of the United States. We present a
summary of the approach and a discussion of the
robustness of the classifications used in the US National
Acid Precipitation Assessment Program's (NAPAP)
State of Science/Technology report on the Current
Status of Surface Water Acid-Base Chemistry (Baker
et al.,
1990). This approach is one of the lines of
evidence used in NAPAP's final integrated assessment
of the role of acidic deposition in surface water
acidification.
SURVEY DESIGN AND CHEMICAL
CLASSIFICATION
National Surface Water Survey design
The NSWS (Fig. 1), conducted between 1984 and 1986,
delineated the distribution and numbers of acidic and
low-pH streams and lakes in acid-sensitive regions of the
United States. The NSWS employed a randomized
systematic sample of 500 stream reaches and 2300 lakes
to make population estimates of the chemistry in a target
116
I(ST
P. R. Kaufmann, A. T. Herlihy, L. A. Baker
PE|C[HT ANC ~: 0 ~eq/L
[--I < II
mn ~-sz
i
S-lOS
i ~o-2ol
inn > 201
/
/
/
/
fJ' ~
IIIU-A I LAII I1~ ,
HIGHLANDS
/
SOUTHEASTERN ,~4~
" '
A/
[|
LAND
lID-ATLANTIC
COASTAL PLAIN
FLORIDA
Fig. 1. Percentage of acidic surface waters in regions sampled by the NSWS. Lakes were sampled in all regions but the mid-Atlantic
coastal plain, the southern portion of the mid-Atlantic highlands, and the western portion of the south-eastern highlands. Streams
were not sampled in the west, upper mid-west, Adirondacks, or New England.
population of 64 000 stream reaches (224 000 km) and
28 300 lakes. To make these quantitative estimates, it
was necessary to explicitly define a target population of
lakes and streams. Target lakes were those with surface
areas between 4 and 2000 ha in the east and between 1
and 2000 ha in the west. The NSWS target stream
population consisted of stream reach segments mapped
on 1:250 000-scale US Geological Survey maps with
drainage areas less than 155 km 2. Streams were sampled
during spring baseflow (between snowmelt and leaf out,
avoiding storm episodes). Lake samples were collected
just below the surface in the deepest part of the lake
during fall mixing. Details of the NSWS have been
presented elsewhere (Linthurst
et al.,
1986; Landers
et
al.,
1987; Kaufmann
et al.,
1988, 1991).
Acid source classification
We classified acidic NSWS streams and lakes according
to their dominant sources of strong acid anions in order
to identify the most likely cause of acidic conditions (e.g.
natural organic acids, acidic deposition, acid mine
drainage). Acidic NSWS waters were classified into three
groups: watershed sulphate-dominated, organic-domi-
nated, and deposition-dominated, as shown in the
flowchart in Fig. 2.
We considered surface waters to be dominated by
watershed sources of sulphate when the observed SO~-
concentration was more than twice as high as the
expected steady-state concentration, assuming evapo-
concentration of deposition (i.e. watershed sulphate >
deposition sulphate). The expected sulphate concentra-
tion ([SO2-kxp) was calculated for each NSWS site from
precipitation (P), run-off (R), precipitation sulphate
concentration ([SO]-]pr~), and dry sulphate deposition
(DR YDEP),
as shown in eqn 1"
[SO~-l¢,p
=
([SO~-]p~*P/R)
+ (DRYDEP/R)
(1)
In the eastern United States, values for these variables
were interpolated to specific NSWS site locations by
EPA's Direct/Delayed Response Project (Church
et al.,
1989). Precipitation and run-off were based on 30-year
(1951-80) annual averages. Precipitation sulphate con-
centrations were based on volume weighted average
1982-86 data. Dry deposition estimates were obtained
from the Regional Acid Deposition Model (Chang
et al.,
1987). In other regions, dry deposition inputs were
based upon literature values; for upper mid-west and
most Florida lakes,
P:R
ratios were estimated using
lake:precipitation chloride ratios (see Baker
et aL,
1990).
Streams and lakes in which the organic anion concen-
tration was greater than the sum of the SO* (sea salt
corrected or non-marine sulphate) and nitrate concen-
trations were classified as organic-dominated. Organic
anion concentration was calculated from dissolved
organic carbon (DOC) and pH using the approach of
Oliver
et al.
(1983). Chloride was not included in the
ratio since it enters watersheds as a neutral salt. If
organic anion concentrations exceed those for sulphate
and nitrate, surface water acidity is likely to be derived
from organic acids, rather than H ÷ associated with
additions of sulphate and nitrate in acidic deposition.
Sources of acidity in lakes and streams of the United States 117
NSWS Acidic Lakes and Streams ]
Table 1. Chemical characteristics (mean + SD) of acidic NSWS
streams and l=kes in different chemical dasses a
Acid source class b
Deposition Organic Watershed
dominated dominated sulphate
• exceeds2x
"
amount expected
from
qo
Yes
Acid
Source Class
~_1
Watershed Sulphate
--I Dominated Cass
"
exceed
Organic-
[SO4 .
+
NO3] / Dominated
Class
Streams
ANC -25 + 29
pH 4.8 + 0.3
SO* 152 + 69
NO3 7 + 11
C* 146 + 91
DOC 2.4 + 2-1
Org A- 18 + 17
Lakes
ANC -13 -+ 12
pH 5.0 _+ 0-3
SO* 102 + 42
NO3 2 + 3
C* 104 -+ 83
DOC 2.5 -+ 2-1
Org A- 19 + 16
-159 + 247 -280 -+ 330
4.5 + 0'6 4.4 + 0.4
40+39 2930+1960
7 -+ 16 38 + 62
290 -+ 162 2 490 + 1 620
50 -+ 61 3'5 +_ 7.1
247 + 237 91 + 128
-20_+ 19 -12_+ 10
4.7 + 0.3 5-1 + 0.3
33 "___ 22 554 + 338
2+4 7_+9
75 + 77 786 _+ 497
12 _+ 9 8-6 +_ 5.8
81 + 56 65 +_ 45
Fig. 2.
Deposition-
Dominated Class
[
Flowchart of the classification scheme used to identify
probable sources of acidity in the NSWS.
SD = standard deviation; all units are in /zeq litre -~ except
for pH and DOC (mg litre-~); * = sea salt corrected; Org
A- = estimated organic anion concentration.
b See Fig. 2.
RESULTS
Sulphate concentrations in near-coastal systems
(within 200 km of the ocean) were corrected for marine
(neutral) sulphate inputs based on sea salt chloride
contributions (Clio) and the sulphate : chloride ratio of
seawater. C!~a was estimated by means of regression
relationships between distance from the ocean and
observed chloride concentrations, for sites with minimal
chloride contamination from watershed activities (e.g.
road salt), as given in eqns 2 and 3 (Baker et al.,
1990):
North-eastern lakes: Ln (Cl[eo)
-- 5.35 - 0.587*Dist + O.O003*Dist 2 (2)
Mid-Atlantic streams: Ln (CI~a)
-- 5.43 - O.180*Dist + O.O0004*Dist 2 (3)
with CI~, in/zeq litre-J and Dist equal to distance from
ocean (in miles for lakes, km for streams). Concentra-
tions for surface waters in Florida and the Pacific north-
west were corrected by assuming that CI~ equals
observed chloride.
Surface waters that are not organic-dominated or
watershed sulphate-dominated were classified as deposi-
tion-dominated. The dominant source of acid anions in
these waters is atmospheric deposition of sulphate and
nitrate. Acidic waters in the deposition-dominated class
are most likely to be currently acidic due to atmospheric
deposition.
Surface water chemistry
As would be expected, surface water chemistries were
distinctly different among the acid source classification
groups (Table 1). Mean sulphate concentrations in
streams and lakes dominated by watershed sources of
sulphate were an order of magnitude higher than those
in deposition-dominated sites. Similarly, organic anion
concentrations were much higher in organic-dominated
waters than in deposition-dominated waters. Sulphate
concentrations were much higher than nitrate concentra-
tions in all deposition-dominated lakes and streams
(Table 1). In waters dominated by acidic deposition, 88%
of the streams and 67% of the lakes had SO* concentra-
tions greater than base cation minus chloride concentra-
tions, indicating that H2SO4 inputs exceed the capacity
of the watershed to neutralize inputs by base cation
mobilization. This suggests that sulphate alone is suffi-
cient to cause acidic conditions in the majority of the
acidic deposition-dominated lakes and streams.
Sources of acidity in US surface waters
Overall, 8% of the streams and 4% of the lakes in the
NSWS are acidic (Baker et al., 1990). Acidic waters are
rare (<1%) in the west and south-eastern highlands and
most common in Florida (23%) and the Adirondacks
(14%, Fig. 1). Within the acidic NSWS lake and stream
118
P. R. Kaufmann, A. T. Herlihy, L. A. Baker
Table 2. Percentage of acidic NSWS waters in different classification groups ~
NSWS region Estimated
number of
acidic waters
Acidic waters (%)
Deposition Organic Watershed-S
dominated dominated dominated
Streams b
Mid-Atlantic highlands 2 414 56 -- 44
Mid-Atlantic coastal plain 1 334 44 54 --
Southeastern highlands 243 50 -- 50
Florida 677 21 79 --
All streams 4 668 47 27 26
Lakes
New England 173 79 21 --
Adirondacks 181 100 -- --
Mid-Atlantic highlands 88 100 -- --
South-eastern highlands ....
Florida 477 59 37 4
Upper mid-west 247 73 24 3
West 15 -- --
100
All lakes 1 181 75 22 3
a - no samples observed in this group.
b In the NSWS, streams were sampled at both the upper and lower ends of each reach. Data in this table
are based on upstream reach end chemistry. Estimates based on downstream reach end chemistry showed
fewer acidic systems but a similar pattern in sources of acidity, relative to upstream chemistry.
population, 75% of the lakes and 47% of the streams are
deposition-dominated (Table 2). About a quarter of
both the acidic lake and the acidic stream populations
are organic-dominated. Watershed sources of sulphate
are dominant in 26% of the acidic NSWS streams and in
only 3% of the acidic lakes.
Watershed sulphate-dominated surface waters are
found principally in streams in the mid-Atlantic and
south-eastern highlands where they account for about
half the acidic stream population (Table 2). All of these
streams show evidence of mining impacts and are acidic
primarily because of H2SO4 inputs from acid mine
drainage (Herlihy
et al.,
1990). Watershed sulphate-
dominated acidic lakes are rare in the NSWS (3% of all
acidic lakes, Table 2). In Florida, acidic, watershed
sulphate-dominated lakes are found in the southern part
of the Peninsula, in the citrus producing area. The only
acidic lake in the west (representing a population of 15
lakes) is dominated by watershed sources of sulphate due
to a geothermal spring.
Organic-dominated streams account for about half
the acidic streams in the mid-Atlantic coastal plain,
primarily in the swampy lowland areas around Chesa-
peake Bay, and most (79%) of the acidic streams in
Florida (Table 2). Organic-dominated acidic lakes com-
prise 20--40% of the acidic lake population in Florida,
New England, and the upper mid-west.
All of the acidic lakes and 56% of the acidic streams in
the mid-Atlantic highlands are deposition-dominated, as
are all of the acidic lakes in the Adirondacks (Table 2).
Organic-dominated acidic lakes are present in the
Adirondacks, but are restricted mainly to small lakes less
than the 4 ha cut-off employed in the NSWS (Baker
et
a/., 1990; Sullivan
et al.,
1990). Deposition-dominated
acidic streams in the mid-Atlantic coastal plain are
found in the New Jersey Pine Barrens and on hilly
outcrops in the Pennsylvania Piedmont. Deposition-
dominated acidic streams in Florida are located in the
Panhandle. Most of the acidic lakes in Florida, New
England, and the upper mid-west are deposition-
dominated. Details on the distribution and character-
istics of these classes of acidic waters are described by
Herlihy
et al.
(1990, 1991) and Baker
et al.
(1990).
DISCUSSION
Although we classified waters into discrete categories,
the relative contributions of sources of acidity are con-
tinuously graded. Watershed sulphur sources, organic
acids and acidic deposition all contribute some degree of
acidity to every watershed, but in most cases one source
predominates. For example, a stream with 200/zeq
litre -~ of sulphate may receive 180/zeqlitre-' from
acidic deposition and 20/.~eq litre -1 from pyrite oxida-
tion in the watershed. Although watershed sources in
this example could supply 10% of the H2SOa inputs to
the system, we would still conclude that atmospheric
deposition is the dominant source of acid sulphate.
Similarly, almost all NSWS waters receive inputs of
organic anions in addition to sulphate from deposition.
Both acid inputs are neutralized to some extent by
processes in the watershed or water body. It was
impossible, in a regional survey such as the NSWS, to
Sources of acidity in lakes and streams of the United States
119
determine the neutralization rates for the different acid
inputs. For this reason, and because H ÷ enters surface
waters along with a mobile anion (Reuss & Johnson,
1986), our classification scheme is based on identifying
the dominant (>50%) source of acid anions in NSWS
surface waters. If organic anions are greater than
SO* + NOr, then the water is dominated by organic
acids. This does not mean that inorganic acids have
no effect, just that the effect is secondary. Similarly,
organic acids probably have some effect on deposition-
dominated systems. In the following sections, we
examine the robustness of our approach with respect
to these areas of uncertainty.
Robustness of watershed SO~4 - source classification
One indication of robustness is the clear separation in
sulphate histograms between deposition-dominated and
watershed sulphate-dominated systems. Figure 3 iUus-
5000
4000
P
E 3000
2000
5000
.s: 4000
(a)
ANC
(p.eqlL)
I
<:0 0-50
50-200
0 0 0
t i Y
0 0
0 0 0 0 0 0
o ~ o ,0 o o
I I I I I ! I
0 0 0 0
o ~ o o
8 o
5042-
( p.eq k -1 )
(b) ANC (l~eqlL)
I<:O
0 -50
IZ~ 50-200
P
E
~oo
" 20001
~ ~ [~;~J r-l,-1
"6
:~ 1000
0 0 0 0 0 0 0 0 0 0 0 0
0 I I I I I I I ^
= o g
oo o
o o o o o o
SO~-(p.eq L -1)
Fig.
3. Frequency distribution of sulphate concentration in
acidic and Iow-ANC streams in the mid-Atlantic highlands
region of the NSWS: (a) watershed sulphate-dominated
streams and (b) deposition-dominated streams. Data are based
on upstream reach end chemistry.
trates this pattern for the mid-Atlantic highlands, which
have the highest sulphate concentrations of any NSWS
region. In this region, there are few waters with
intermediate sulphate concentrations between 250 and
500/~eq litre -~. More importantly, there is no overlap in
concentration among acidic streams in the deposition-
dominated and watershed sulphate-dominated cate-
gories. All deposition-dominated acidic streams had
SO 2- concentrations <250 p.eq litre -t, whereas all wa-
tershed sulphate-dominated acidic streams had SO~-
concentrations >450/zeq litre -l. Thus, in this case, our
classification system yields populations that are dis-
tinctly different.
Another measure of robustness is the sensitivity
of the classification to the watershed sulphur source
criteria. To evaluate this, we compared the number of
systems placed in the watershed sulphate-dominated
category using 1.5 times [SO2-]~p as the cut-off concen-
tration, rather than 2-0 times [SO2-]cxp. Out of the 881
deposition-dominated acidic lakes, only 85 were
classified as watershed sulphate-dominated with the
more restrictive sulphate cut-off. No acidic streams
changed categories with the new classification rule.
All deposition-dominated acidic streams had observed
sulphate concentrations < 1.4 times [SO~-]oxp.
Robustness of organic anion estimates
A minor amount of uncertainty enters our classification
as a result of estimating organic anion concentrations
Table 3. Percentage of acidic NSWS waters with organic
dominance using three different organic anion estimates
Region Organic anion estimate
Oliveff C s -CA b Driscoll c
Streams d
Mid-Atlantic highlands 0 0 c
Mid-Atlantic coastal plain 54 36 c
Florida 79 79 c
All streams 27 22 c
Lakes
Adirondacks 0 0 0
Mid-Atlantic highlands 0 0 0
New England 21 5 5
Florida 35 18 38
Upper mid-west 24 27 23
All lakes 22 14 21
a Organic anion concentration calculated from model devel-
oped by Oliver
et al.
(1983).
b Organic anion concentration calculated from anion deficit: X
cations (Ca 2+, Mg 2÷, Na ÷, K +, H +, NH~, Mn 2+, AP +) - X
anions (SO~-, NOL HCO~, CO]-, OL F-, OH-).
c Organic anion concentration calculated from Driscoll
et al.
(1989) modification of Oliver model. The Driscoll modification
was developed for NSWS lakes but not for streams.
a Stream estimates are based on upstream reach and chemistry,
as in Table 2.
120
P. R. Kaufmann, A. T. Herlihy, L. A. Baker
from measured DOC and pH. We used Oliver's model
(Oliver
et al.,
1983) to estimate organic anions, but the
classification was similar when organic anions were
calculated from anion deficits (sum of cations minus
inorganic anions) or from a region-specific modification
(Driscoll
et al.,
1989) of the Oliver method (Table 3). In
the mid-Atlantic highlands and the Adirondacks, none
of the acidic NSWS streams and lakes are dominated by
organics, regardless of how organic anions were esti-
mated. Overall, the Oliver model gave slightly higher
estimates of organic anions. Thus, for acidic lakes, 22%
were found to be organic-dominated using the Oliver
model estimates of organic anions, compared with 14%
using the anion deficit approach or 21% using the
Driscoll modification (Table 3). Thus, we may be slightly
overemphasizing the importance of organic anions
(Baker
et al.,
1990). We judged the Oliver method to be
preferable to anion deficit estimates for two reasons: (1)
the determination of anion deficits aggregates analytical
errors to substantial levels (for duplicate samples from
streams in the NSWS, the standard deviation of anion
deficit measurements was 27 #eq litre -~ or 20% of the
mean value); and (2) uncertainty in speciation of metals
(Fe, Mn, AI) affects the calculated anion deficits,
particularly for the acidic waters of concern here.
Organic influence in deposition-dominated waters
New England (L).
Adirondacks (L)-
M.A. Highlands (L).
M.A. Highlands (S).
M.A. Coastal (S)-
Florida (L)-
Florida (S).
Upper Midwest (L)-
0.0
To evaluate the importance of organic acids in deposi-
tion-dominated waters, we examined A-: (SO~4 + NO;)
ratios (Fig. 4). Organic acids probably exert very little
influence on most of the deposition-dominated acidic
streams in the mid-Atlantic highlands, where organic
anion concentrations are less than 8% of (SO~4 + NO~-)
concentrations. The organic contribution to deposition-
dominated lakes in the upper mid-west is much more
substantial, where organic anion concentrations are 20-
45% of the (SO~4 + NO~-) sum (Fig. 4). Organic influence
is more moderate in deposition-dominated acidic waters
I i i i
I I I
I I I
I I
Koy
25%
Meal
75%
r]
I I
I I
I I I
0'.1 0'.2 0:a o'4 0s
Org A-/(SO 4"
+ NO3)
Fig.
4. Estimated organic anion concentration as a propor-
tion
of
SO* plus NOr concentration in deposition-dominated
acidic NSWS lakes (L) and streams (S).
in the other NSWS regions, where ratios of median
organic anion to (SO$4 + NO;) are between 0-1 and
0.2.
Neutral salt H ÷ exchange
Except in unusual circumstances, chloride loadings to
watersheds in acid-sensitive regions of the United States
do not occur as hydrochloric acid, but as inputs of
neutral salts from marine origin or application of road
de-icing compounds. However, neutral salt exchange has
been hypothesized as a mechanism of surface water
acidification in coastal surface waters. Rosenqvist (1978)
and Krug & Frink (1983) described a mechanism in
which base cations in the neutral salt solution replace H +
ions in the soil. The resultant drainage water is rendered
richer in H ÷ and thus more acidic than the solution
added to the soil.
Episodic acidification due to neutral sea salt inputs
has been observed in coastal streams (Wright
et al.,
1988;
Langan, 1989). If sea salt displacement of H ÷ were
causing chronic ANC depressions, we would expect to
see evidence that watersheds are retaining Na + and Mg 2+,
the principal base cations in sea salt. Accordingly,
surface water CI- should be enriched relative to Na ÷ and
Mg 2÷, when compared with the sea-salt molar ratios of
these ions. There were no indications of Na ÷ or Mg 2+
retention in acidic New England lakes (Sullivan
et al.,
1988) or streams in the mid-Atlantic coastal plain
(Morgan & Good, 1988; Herlihy
et al.,
1991). Neutral
salt exchange, however, may be a source of H + to some
of the acidic waters in Florida, where there is evidence of
base cation retention (Baker
et al.,
1988; Herlihy
et al.,
1991). The neutral salt acidification mechanism is not
relevant to acidic surface waters in the Adirondacks,
mid-Atlantic highlands, and upper mid-west, where C1-
concentrations are very low.
Afforestation
It is recognized that aggrading forests may exacerbate the
acidification of surface waters (Nilsson
et al.,
1982). The
exacerbating effect of forest growth on surface water
acidification stems from mechanisms such as base cation
and ammonium uptake by tree roots, or from increased
scavenging of atmospherically derived acid anions. How-
ever, while soils may be acidified by forest growth, it does
not appear that afforestation results in
acidic
surface
waters unless there are mobile anions (such as sulphate or
organic anions) to transport the soil H + into the surface
water (Miller, 1989). Thus, acid anion source remains the
major factor explaining surface water acidity.
Temporal variability
We classified acidic waters using chemical data from the
NSWS, which represents autumn mixing conditions for
Sources of acidity in lakes and streams of the United States 121
lakes and spring baseflow for streams. Acid anion
composition in other seasons may be somewhat dif-
ferent. In streams draining wetlands in the mid-Atlantic
coastal plain, there is some evidence of a seasonal shift in
acid anion dominance from sulphate in the spring to
organics in the summer (Eshleman & Kaufmann, 1989).
These shifts generally coincide with summer increases in
ANC, so they do not greatly alter conclusions regarding
the sources of acidity during the spring, when stream pH
is the lowest and the most acid-sensitive developmental
stages of fish are usually present. To our knowledge,
seasonal shifts in anion dominance have not been
reported for other regions.
SUMMARY
We classified acidic waters in the US National Surface
Water Survey into groups according to their acid anion
composition and the likely origin of their dominant
anions. Using the mobile carrier anion mechanism
(Reuss & Johnson, 1986) as a working assumption, we
interpret these classes as reflections of the probable
source of acidity in these waters. Because the groupings
were quite distinct, our major conclusions are robust and
relatively insensitive to differences in the criteria for
assessing the relative contributions of watershed sulphur
sources and estimating the concentrations of organic
anions.
Watershed sulphate sources were identified as the
probable cause of acidic conditions in 26% of the acidic
streams and 3% of the acidic lakes in the survey area.
The vast majority of the watershed sulphate-dominated
streams were acidic due to acid mine drainage and were
located in the mid-Atlantic highlands. For 22% of the
acidic lakes and 27% of the acidic streams in the NSWS,
natural organic acids were the dominant source of
acidity; in coastal lowlands, more than half of the acidic
streams were organic-dominated. An estimated 75%
(881) of the acidic lakes and 47% (2190) of the acidic
streams in the NSWS had anion chemistry dominated
by sulphate and nitrate derived from atmospheric
deposition.
ACKNOWLEDGEMENTS
The research described in this article was funded by the
US Environmental Protection Agency (EPA). This
document was prepared at the EPA Environmental
Research Laboratory in Corvallis, Oregon, through
cooperative agreements CR815168 with Utah State
University and CR813999 with the University of Minne-
sota. The manuscript has been subjected to the Agency's
peer and administrative review and approved for pub-
lication. Mention of trade names or commercial prod-
ucts does not constitute endorsement or reeommenda-
tion for use. Completion of the EPA's National Surface
Water Survey, upon which this article is based, depended
upon several years of dedicated work by many in-
dividuals. We thank M. Mitch for data analysis and J.
Mello for word processing. R. Church and S. Christie
pro-vided insightful comments on an earlier draft.
REFERENCES
Baker, L. A., Pollman, C. D. & Eilers, J. M. (1988). Alkalinity
regulation in softwater Florida lakes. Water Resour. Res.,
24, 1069-82.
Baker, L. A., Kaufmann, P. R., Herlihy, A. T. and Eilers, J. M.
(1990). Current Status of Surface Water Acid-Base Chem-
istry. NAPAP Report 9, Acidic Deposition: State of Science
and Technology. National Acid Precipitation Assessment
Program, 722 Jackson P1, Washington, D.C.
Chang, J. S., Brost, R. A., Isaksen, I. S. A., Modronich, S.,
Middleton, P., Stockwell, W. R. & Walcek, C. J. (1987).
A three-dimensional Eulerian acid deposition model:
physical concepts and formulation. J. Geophys. Res., 92,
14681-700.
Church, M. R., Thornton, K. W., Shaffer, P. W., Stevens, D.
L., Rochelle, B. P., Holdren, G. R., Johnson, M. G., Lee, J.
J., Turner, R. S., Cassell, D. L., Lammers, D. A., Campbell,
W. G., Lift, C. I., Brandt, C. C., Liegel, L. H., Bishop, G. D.,
Mortenson, D. C., Pierson, S. M. & Schmoyer, D. D. (1989).
Direct~Delayed Response Project: Future Effects of Long-
term Sulfur Deposition on Surface Water Chemistry in the
Northeast and Southern Blue Ridge Province. EPA/600/3-89/
061a-d US Environmental Protection Agency, Washington,
D.C., 887 pp.
Driscoll, C. T., Fuller, R. D. & Schecher, W. D. (1989). The
role of organic acids in the acidification of surface waters in
the eastern US. Water Air Soil Pollut., 43, 21~10.
Eshleman, K. N. & Kaufmann, P. R. (1989). Unpublished data
on sources of acidity and controls on stream acidification in
the mid-Atlantic Coastal Plain Region, United States. Paper
presented at the American Geophysical Union Chapman
Conference on Hydrogeochemical Responses of Forested
Catchments, Bar Harbor, Maine, USA. 18-21 Sept., 1989.
Herlihy, A. T., Kaufmann, P. R., Mitch, M. E. & Brown, D.
D. (1990). Regional estimates of acid mine drainage impact
on streams in the mid-Atlantic and Southeastern United
States. Water Air Soil Pollut., 50, 91-107.
Herlihy, A. T., Kaufmann, P. R. & Mitch, M. E. (1991).
Chemical characteristics of streams in the Eastern United
States: II. Sources of acidity in acidic and low ANC streams.
Water Resour. Res., 27, 629--42.
Huckabee, J. W., Goodyear, C. P. & Jones, R. D. (1975). Acid
rock in the Great Smokies: unanticipated impact on aquatic
biota of road construction in regions of sulfide mineral-
ization. Trans. Am. Fish. Soc., 104, 677-84.
Kaufmann, P. R., Herlihy, A. T., Elwood, J. W., Mitch, M. E.,
Overton, W. S., Sale, M. J., Messer, J. J., Cougan, K. A.,
Peck, D. V., Reckhow, K. H., Kinney, A. J., Christie, S. J.,
Brown, D. D., Hagley, C. A. & Jager, H. I. (1988). Chemical
Characteristics of Streams in the Mid-Atlantic and South-
eastern United States. Volume I." Population Descriptions and
Physico- Chemical Relationships. EPA/600/3-88/021 a, US En-
vironmental Protection Agency, Washington, D.C., 397 pp.
Kaufmann, P. R., Herlihy, A. T., Mitch, M. E., Messer, J. J. &
Overton, W. S. (1991). Chemical characteristics of streams in
the Eastern United States: I. Synoptic survey design, acid-
base status, and regional patterns. Water Resour. Res., 27,
611-27.
122 P. R. Kaufmann, A. T. Herlihy, L. A. Baker
Krug, E. C. & Frink, C. R., (1983). Acid rain on acid soil: a
new perspective. Science, 221, 520-5.
Langan, S. J. (1989). Sea-salt induced strcamwater acidi-
fication. Hydrol. Processes, 3, 25-41.
Landers, D. H., Eilers, J. M., Brakke, D. F., Overton, W. S.,
Kellar, P. E., Silverstein, M. E., Sehonbrod, R. D., Crow,
R. E., Linthurst, R. A., Omernik, J. M., Teague, S. A. &
Meier, E. P. (1987). Characteristics of Lakes in the Western
United States. Volume I: Population Descriptions and
Physico-Chemical Relationships. EPA/600/3-86/054a, US En-
vironmental Protection Agency, Washington, D.C.,
176 pp.
Linthurst, R. A., Landers, D. H., Eilers, J. M., Brakke, D. F.,
Overton, W. S., Meier, E. P. & Crowe, R. E. (1986).
Characteristics of Lakes in the Eastern United States. Volume
L" Population Descriptions and Physico-Chemical Relation-
ships. EPA/600/4-86/007a. US Environmental Protection
Agency, Washington, D.C., 136 pp.
Marmorek, D. R., Bernard, D. P., Wedeles, C., Sutherland, G.,
Malanchuk, J. A. & Fallon, W. E. (1989). A protocol for
determining lake acidification pathways. Water Air Soil
Pollut., 44, 235-57.
Miller, H. (1989). Forests and acidification. In Acidification in
Scotland 1988. Scottish Development Department, Edin-
burgh.
Morgan, M. D. & Good, R. E. (1988). Stream chemistry in the
New Jersey Pinelands: the influence of precipitation and
watershed disturbance. Water Resour. Res., 24, 1091-100.
Nilsson,
S. I., Miller, H. G. & Miller, J. D. (1982).
Forest
growth as a possible cause of soil and water acidification: an
examination of the concepts. Oikos, 39, 40-9.
Oliver, B. G., Thurman, E. M. & Malcolm, R. K. (1983). The
contribution of humic substances to the acidity of colored
natural waters. Geochim. Cosmochim. Acta, 47, 2031-5.
Reuss, J. O. & Johnson, D. W. (1986). Acid Deposition and the
Acidification of Soils and Waters. Springer-Verlag, New
York.
Rosenqvist, I. (1978). Acid precipitation and other possible
sources for acidification of rivers and lakes. Sci. Total
Environ., 10, 271-2.
Schindler, D. W. (1988). Effects of acid rain on freshwater
ecosystems. Science, 239, 149-57.
Sullivan, T. J., Driscoll, C T., Eilers, J. M. & Landers, D. H.
(1988). Evaluation of the role of sea salt inputs in the long-
term acidification of coastal New England lakes. Environ.
Sci. Technol., 22, 185-90.
Sullivan, T. J., Kugler, D. L., Small, M. J., Johnson, C. B.,
Landers, D. H., Rosenbaum, B. J., Overton, W. S., Kretser,
W. A. & Gallagher, J. (1990). Variation in Adirondack, New
York, lakewater chemistry as a function of surface area.
Water Resour. Bull., 26, 1-10.
Wright, R. F., Norton, S. A., Brakke, D. F. & Frogner, T.
(1988). Experimental verification of episodic acidification of
freshwaters by sea salts. Nature, 334, 422-4.
