ooro
SYNOPSIS OF WATER-EFFECT RATIOS
FOR HEAVY METALS AS DERIVED FOR
SITE-SPECIFIC WATER QUALITY CRITERIA
EPA Contract No. 68-CO-0070
March 1992
Prepared by:
William A. Brungs
Consultant
Todd S. Holderman
Work Assignment Leader
Mark T. Southerland
Project Manager
Submitted to:
Charles Delos
Health and Ecological Criteria Division
Office of Science and Technology
Office of Water
U.S. Environmental Protection Agency
Information Resources Center
US EPA (3404)
401 M Street, 3W
Washington, DC 20460
-------
INTRODUCTION
National water quality criteria, derived from a diverse data base and calculated on
the basis of numerous general assumptions (USEPA, 1985), have been used as the
principal requirement in developing state and federal water quality standards. In most
instances, the criteria are directly transcribed, without modification, into those standards.
Consequently, these standards are based on data for organisms that may or may not be
resident to the areas of regulatory concern, or the surface waters of the areas of
regulatory concern may not be comparable to the clean laboratory test waters used to
generate the toxicity data base from which the criteria or standards have been derived.
Once water quality standards are in place, monitoring programs are developed or
modified to incorporate regulatory oversight in the pursuit of violations of those
standards. State and federal monitoring programs have often observed excursions for
metals in watersheds that have been affected by the activities of man. These excursions
above the relevant water quality standards occur in a variety of uncontaminated and
contaminated watersheds.
Although apparent excursions of substantial frequency may not necessarily
indicate that water quality criteria or standards for metals are consistently overprotective,
these excursions cannot be ignored; investigations should be conducted to find the
ecological bases for such phenomena. These investigations may also lead to additional
lexicological or chemical research that will improve future criteria.
To develop a greater understanding of the phenomenon of excursions and to
provide opportunities for adjustments in water quality criteria, the Guidelines for
Deriving Numerical Aquatic Site-Specific Water Quality Criteria by Modifying National
Criteria were developed (Carlson et al., 1984): the Water Quality Standards Handbook
(USEPA, 1983) contained a brief synopsis of these guidelines. The guidelines state that
national criteria may be underprotective or overprotective because the species at the site
are more or less sensitive than those included in the national criteria data set, or because
the physical and/or chemical characteristics of the water at the site alter the biological
availability and/or toxicity of the material. Site-specific procedures have been developed
-------
to address each of these conditions separately, as well as the combination of the two.
For this report, only the indicator species procedure was studied because it results in a
water-effect ratio (WER) that accounts for differences in the biological availability
and/or toxicity of a material caused by physical and/or chemical characteristics of a site
water.
This indicator species procedure is principally employed to determine the acute
toxicity in site water and laboratory water using species resident at the site, or using
acceptable nonresident species as indicators or surrogates for species found at the site.
The difference in toxicity values, expressed as a WER, is used to convert the national
maximum concentration for a material to a site-specific maximum concentration from
which a site-specific Final Acute Value would be derived. The procedure also provides
three ways to obtain a site-specific Final Chronic Value. This latter alternative was
investigated in only a few of the retrieved studies and provided little information of value
to this report.
As will be detailed in a later section, the author of this report critically examined
the procedures used and data presented in numerous studies of the acute WER. The
author of this report, William A. Brungs, co-authored both the national guidelines
(USEPA, 1985) and the site-specific guidelines (Carlson et al., 1984); he used his
familiarity with the purposes, details, and nuances of those guidelines in examining the
available studies of the WER.
Charles Delos acted as Work Assignment Manager for this report, the final
deliverable for Work Assignment 1-07, EPA Contract No. 68-CO-0070. The focus of this
report is to obtain and analyze water-effect ratios and the ratio of dissolved metal to
total recoverable metal observed in the cited studies.
DATA PROCUREMENT
Few of the reports on water-effect ratios have been published in peer-reviewed
scientific journals. Many of the unpublished reports lack complete details on test
procedures, analytical endpoints, or proper data calculation procedures. Therefore, in
-------
some instances, the authors were contacted to obtain the necessary details or to assess
the validity of the studies and their conclusions. Telephone contact made with each
USEPA regional office water quality standards coordinator and/or regional laboratory
representative led to personal contact with several USEPA research laboratories, state
laboratories, industrial research laboratories, and others; most relevant reports were
obtained. A list of all contacts is presented in Table 1.
Only one study has been completed for a marine/estuarine system. In the final
report for that study, prepared by Larry Walker Associates (1991) for the City of
San Jose, California, copper, lead, and nickel WERs were determined for up to three
species in San Francisco Bay water. The State of California and EPA scientists are
conducting a final review of that report. This author has concluded that none of the
data in their present form meet all the guideline requirements because no confidence
limits were included with the toxicity data, (2) lead solubility problems existed, (3) results
were based on nominal concentrations when interpolation between measured metal
results could have been used, and (4) 7-day "chronic" tests were conducted. In the
following sections of this report, these results are discussed, and the reasons for rejection
are presented.
S.R. Hansen and Associates are currently conducting a second marine/estuarine
study for the San Francisco Bay area for copper and nickel with several species. Testing
began in September 1991. (The contact person is Cheryl Niemi, California Regional
Water Quality Control Board, tel. 415-464-1262.)
Two freshwater studies are currently in progress in Massachusetts. One, on the
Blackstone River, is testing copper, cadmium, nickel, and zinc in the test species
Ceriodaphnia dubia. A second study, on the Charles River, involves copper and the same
test species. Both studies are being conducted by independent contractors for
dischargers. (The contact person in Massachusetts is Jerry Szal, Massachusetts
Department of Environmental Protection, tel. 508-792-7480.)
-------
co
en
-S2
cq
U
CO
Q.
CO
2
O
Em e
1*11
|| |<3
5 ^5 §
£ C V K
££•?*;>
' ' o 5 3
00 Ov *- 3 J
c c e Q N
.2 .2 .2 IS
QJQ So 5b j
v w o Q*
QM C£ fi£ U
CJ
fr c
c «
.£ X
J^ qj
II
ce
o
N
' • ** i
8 a =
« I I
> ^5
t «
a Z
JZ Jj
CO J t/5
U UJ UJ
i
P
55
.—
c c
*— M
?I"
Vi j/5 • —
c
'«
o
o
u'
CQ
C
N ^
»
a
t
£
CL
u
w:
o
•8
O
2
u.
c "2
o
Q
•s £
i£5
" S'=P
u
^ «£-
E §i^
— = Ui 2 IT •— « ~~ -» •:
UUI^?^ I
—
n
3
CO
X
« ** Cu
IllS^iSg
fr> O
C ^
c
O
'5b
eN
.2 ._ ._ ._
Ofi Ofi DO OO
O 1> 1> 4>
o
C
o
c
C
sO r**
e e
.2 .2
'ob'ab
c
c
I
o
<
To co
N .e
»tl
to
u ^ ^
CO
'C
C9
^
CO
c =
o -^
181
«c^
«- _
o i •-
J= !C
sso
s s
i a. E
x ° —
B-TJ o
6> U
Q «
e u =
ITS »
, X ^J CO
UJ to CO
.2
u
TJ
.2: c .S
COG
D Z D
|
to
-------
DATA INTERPRETATION
The detailed site-specific guidelines (Carlson et al., 1984) contain
recommendations, precautions, and additional information not contained in the brief
version in the Water Quality Standards Handbook (USEPA, 1983). The latter was the
reference usually cited in early studies funded by the EPA's Office of Water Regulations
and Standards, which funded 17 freshwater site studies distributed throughout the 10
Regions; of these, 11 involved one or more metals. One of these site studies, for the
Walkill River in New Jersey, was prematurely terminated after a variety of major
problems occurred (pers. comm. Jeff Reading). No report on the Flint River in
Michigan could be found even after a diligent search. For the nine remaining site
studies, the resulting reports contained data obtained under a variety of conditions, some
contrary to the detailed guidelines. Errors noted included the following:
Site and laboratory waters were significantly dissimilar
* Site waters contained point source discharges
Flow-through and static test results were compared
Site and laboratory tests were not conducted concurrently with organisms
from the same population
Mixtures of metals were tested
Control mortality was excessive
LCSO and EC50 values did not include confidence limits
As a result, most of the data presented are hot considered to be valid. This
author is using the same judgmental factors specified in the national and site-specific
guidelines for determining the acceptability of data. Invalid or unacceptable data are
presented for completeness and qualified accordingly.
The issue of significantly dissimilar test waters requires further explanation.
Several studies, including some not supported by the initial EPA funding, used site and
-------
laboratory waters considered to be noncomparable in terms of hardness, alkalinity, and
pH. Consequently, water-effect ratios incorporated the effects of those variables as well
as the effects of organic solutes, inorganic and organic colloids, and suspended solids.
These latter varieties are the major factors of interest in using the indicator
species/water-effect ratio procedure because hardness and pH have been used previously
to derive water quality dependent criteria. Therefore, for the purposes of evaluating the
utility of WERs, studies confounded by water difference in hardness, alkalinity, or pH
were deemed to be unacceptable.
In this diversity of studies, varied procedures were used to obtain laboratory
waters. Reconstituted water was commonly used in the initial field studies. Water
supplies in toxicology laboratories came from a variety of sources, including Lake
Superior water, dechlorinated tap water, and well water. All were considered to be
acceptable when their water quality characteristics were similar to the clean site water.
Each metal will be discussed separately because there are more consistent
patterns within studies of a single metal than there are between metals. More than half
of the acceptable data were from two sources: the EPA's Environmental Research
Laboratory in Duluth, Minnesota, and the Lake Superior Research Institute at the
University of Wisconsin-Superior. In addition, most of the data presented on the
dissolved to total metal ratios were from the studies reported by those two laboratories.
WATER-EFFECT RATIOS FOR COPPER
Final reports were obtained for 10 studies in which copper was investigated to
determine WERs (Tables 2 and 3). One of these studies was marine/estuarine (Larry
Walker Associates, 1991). Several freshwater studies on the South Fork of the Crow
River (JRB Associates, 1983a), the Cuyahoga River (De Graeve et al., 1990), the
Spokane River (Bailey and Saltes, 1983), Mulberry Creek, NC (Department of Natural
Resources, 1982), and Prickly Pear Creek, MT (Miller et al., 1983) resulted in toxicity
data that were judged to be unacceptable based on reasons discussed in the Data
A
Interpretation section and specified in Table 3.
-------
.1
*rf
CO
i-
u
'w
13*
| £
= i
5/3 r
^
O
1
00
Jg
J
• -^
o
1
CO
J<
J
w
u
c
f
*"•
2
3
C
i/i
r^
u
«£i S pj. v
u. ^ " "" ""
CB
»
JtJ o
« U ^~
a. MO,
8 o E
< 91
> • O\ 30
d d
(S
4>
^
f" 1
Soecies
g
fathead min
tx:
b
hT
§»
5
<«^
ca
So
«
Z
5J -O
*tn —
w r-
0 0
d d
53
^
•^
a
Ceriodaphni
'175 _
Si
d d
|
g
fathead min
Z
if
>
5
3
3
«
«Ml
'35 — "55 -2 '55 —
oo r^i ^o ^* f*} rs
<«•> i— ' r»i o tso
d d do d d
«
^5 Sc
I 1 -a
1 1 1
= f |
2 Q «
•""" "ffl
ON a
<*J O
d d
amphipod
i
uT
Si
£
f
Z
•^^ ^c
oo ^
d d
amphipod
^^
u
c2
(9
E
&
3
BL.
•5 —
3*
^
3>
•£
c
ao
c
•5
s
^
t-
iS
^
^>
ot significant
c
S
1
-------
u
.S
"5
a
S
od
1
rt i_
^** fll
u i-
ss
c « .2
ft?
« .tr
•g.a
.2 S
=5 2
c «2
£ 35
fl> »-.
«M v
'•5 'c
a u-
•§ '" .2
III
•o fc «
s-ai
I
C/3
V
u
Q.
a
u
3
I
a
t/5
u
O.
u
"i •* oc
ir
cs
O' W W V W o
•« .^s .^i .^ -Ci s?
1/3 '3; '35 '35 m
u ^ a>
•—
-------
c
«
£
4>
o
TJ
•O £
M
> c ^
a £3
! 11
I «
^ «« _
-§ '".2
111
"2 « S3
S -c E
* 00 C
•S "-S 2
O.NM' C
I/)
i
•o
s
:s a
c I-sl S
•* fS. 2S ^ W
5 -o-S
CQ gj
11
£ S
«i w C* w .-<
S1^ lii|
I J5 I o
8>
CH
3 *!
3
8
V3
U
O.
I
TS
4J
reconst
•<*
A
•o
^ (/j —
« o V
*•• m "-1
.r «
o
0
r
oo
^* •••^ ^^ • •*• -™ «
S «5 S S
o o o o o
^ r-;
fl O
V
O C5
O Tf
So
o d
vC
£i
•sC
(N
I,
c/:
<3
!>
c
"8
•a
'S.
CO
o.
CO
I
.8
I,
i
I
•b
V
tf
1
i
u
a
3
4
-------
Carlson et al. (1986) studied the toxicity relationship of copper in site water from
the Naugatuck River in Connecticut and laboratory water from Lake Superior. Of the
tests conducted with larval fathead minnows (Pimephates promelas) and Ceriodaphnia
dubia, a cladoceran, only those conducted with Naugatuck River water collected on
September 5, 1983, and shipped to the ERL-Duluth were considered to be acceptable.
Concurrent tests with the same populations were performed at the Duluth laboratory.
The WERs based on measured concentrations for both species were 1.1 (Table 2): there
was no difference in copper toxicity between the two waters because the confidence
limits on these pairs of tests overlapped. Additional testing during this study provided a
large amount of data (Table 3) that did not meet the guideline requirements. For the
three species tested, there was a pattern of increasing WERs from upstream to
downstream in the Naugatuck River. This could be due to increasing amounts of
municipal effluents that contain many of the chemicals known to reduce copper toxicity
and/or bioavailability. Caution should be used when attempting to discern patterns
among data that were considered to be unacceptable.
Hammermeister et al. (1983) conducted a comprehensive study using four species
and St. Louis River, Minnesota, site water and dechlorinated Superior, Wisconsin, tap
water from wells under Lake Superior. Static 96-hour tests were conducted with fathead
minnows, rainbow trout (Salmo gairdneri), and scuds (Gammarus pseudolimnaeus). Static
48-hour tests were conducted with the cladoceran, Daphnia magna. All test results were
, acceptable and based on measured concentrations. Total hardness and pH were very
similar for the two waters in the concurrent tests. The WERs were 3.2 for the rainbow
trout, 6.3 for the fathead minnow, 9.2 for the amphipod, and 153 for Daphnia magna
(Table 2). There was no discussion concerning the cause of the reduced copper toxicity
in the St. Louis River water, although total residue values were from 1.6 to 2.4 times
greater in that site water.
In a quarterly report (Call et al., 1982) on the U.S. EPA Cooperative Agreement
No. CR 809234-01-0, the Center for Lake Superior Environmental Studies (currently the
Lake Superior Research Institute) of the University of Wisconsin-Superior conducted
12
-------
studies with two different site waters and laboratory water. Static tests (96-hour EC50)
with measured concentrations were used for copper with the amphipod, Hyalella azteca.
The WER for the Nemadji River in Wisconsin and laboratory water from Lake Superior
was 4.3 (Table 2). For the Little Pokegama River in Wisconsin and laboratory water
from Lake Superior, the copper WER was 3.1 (Table 2). Additional tests were
conducted with three more species, but the hardness and pH of the two site waters were
quite different from the laboratory water at that time. Those WERs ranged from 7.0 to
44, but because hardness effects were the dominant factor, the data are presented in
Table 3.
DeGraeve et al. (1990) conducted 7-day LC50 tests using reconstituted water,
Cuyahoga River water, and effluent. Since hardness was significantly different in these
waters, the authors normalized their results based on the copper criteria. WERs were
higher in the mixture of effluent and clean site water (Table 3), which indicates that the
effluent contained chemicals that reduced the toxicity and/or bioavailability of copper.
The LC50 confidence limits were not available for all the data; however, because the
WERs were relatively high, the results may be useful in defining water-effect ratio
patterns.
The results of the study on San Francisco Bay (Larry Walker Associates, 1991)
have limited potential for use (Table 3). However, both species were tested monthly
and, although the individual results are questionable, the internal pattern over time may
be meaningful. There appears to be an approximate range of WERs having a factor of 2
for both species.
In summary, only 4 of the 10 copper studies provided acceptable WER data;
values ranged from 1.1 to 153. There appeared to be a significant species effect because
the WERs for four species in the same waters ranged from 32 to 153 (Hammermeister
et al., 1983). With this degree of variation due to species differences, the requirement in
the site-specific guidelines for only two species for the indicator species procedure may
be inadequate. A similar major variation with cadmium and five species will be
discussed later.
13
-------
WATER-EFFECT RATIOS FOR ZINC
Ten studies were conducted with zinc to determine the effect of site water
differences on toxicity. Studies on the Straight River (Carlson and Roush, 1985),
Spokane River (Bailey and Saltes, 1983), Mingo Creek (JRB Associates, 1983b), and
Prickly Pear Creek (Miller et al., 1983) resulted in toxicity data that were judged to be
unacceptable as discussed in the Data Interpretation section and Table 4. Studies were
also conducted on the Trinity, Sabine, and Red Rivers in Texas (Parkerton et al., 1989).
While the hardness of the three site waters differed considerably from the hardness of
the laboratory water, the toxicities could not be used to calculate acceptable WERs.
Therefore the authors used the zinc hardness-dependent criterion to adjust their toxicity
results accordingly (Table 4). They concluded that when adjusted for site water
hardness, the national zinc criteria accurately predicted differences in zinc toxicity
observed in the three site waters. Their predicted WERs adjusted for hardness agreed
well with the observed ratios. This indicated that hardness alone would explain the
differences in observed zinc toxicity between the three site waters and laboratory waters.
Thus, the WERs for the three site waters would have approximated 1.0 if hardness had
been the same. They tested both the fathead minnow and a cladoceran, Daphnia pulex.
The Norwalk River in Connecticut was studied by Dunbar and Pizzuto (1982)
using rainbow trout and Daphnia magna. Site water was transported to the EG&G
Bionomics Laboratory in Wareham, Massachusetts where concurrent tests were run with
Norwalk River water and reconstituted water as the laboratory comparison. Both waters
were the same hardness and pH. The WERs for the two species were 1.5 and 2.3
(Table 5). In an attempt to explain the cause of this difference, the authors presented
data indicating a concentration of total organic carbon in the site water of 5 mg/1; none
was found in the reconstituted water.
During the study on Boggy Creek in Oklahoma (JRB Associates, 1983c), site
water tests with the red shiner (Notropis lutrensis) were conducted using flow-through
conditions; the laboratory reconstituted water was used in static conditions. The inherent
variability of the two test methods probably masked any effect of water source; the data
14
-------
are included in Table 4. Tests with a caddisfly (Cheumatopsyche sp.) provided a WER of
\'.2 indicating no effect of the site water on the toxicity of zinc (Table 5).
Hammermeister et al. (1983) tested four species in St. Louis River water and
dechlorinated tap water from Lake Superior under similar conditions of water quality.
As is typical with zinc, the WERs were close to 1.0 (Table 5). Tests conducted with the
rainbow trout, fathead minnow, and an amphipod (Gammarus pseudolininaeus)
demonstrated no significant effect of the site water on toxicity. The WER for Daphnia
magna was 2.9.
In a study on the Naugatuck River in Connecticut (Carlson et al., 1986),
laboratory water from Lake Superior was shipped to the site to conduct concurrent tests
with the clean site water. Static 48-hour tests were conducted with a cladoceran
(Ceriodaphnia dubia), and the fathead minnow was tested for 96 hours. The results of
this pair of tests were not significantly different (Table 5). As with copper (Table 3), the
testing on the Naugatuck River included stations downstream of domestic and industrial
discharges. While these latter data (Table 4) do not meet the guideline criterion of
clean site water, they do have some internal merit. There appears to be a slight general
increase in WERs with distance downstream (stations 4A to 7) from 0.8 to 1.5.
The LC5Q values between stations, however, are not significantly different. The authors
concluded that this small difference, if real, would have been expected based on the
hardness correction factor for zinc toxicity used in the zinc national criterion. This, they
also concluded, indicates that the domestic and industrial effluents within this part of the
Naugatuck River had little or no effect on the toxicity and/or bioavailability of zinc other
than their contribution to hardness. The daphnid data (Table 4) are inconclusive in this
regard.
DeGraeve et al. (1990), in their study on the Cuyahoga River in Ohio, obtained a
somewhat similar relationship between zinc toxicity and municipal/industrial effluent
(Table 4). They normalized their WERs to account for hardness using the zinc national
15
-------
Reasons for Reject
<« W.
12
is
£§»
i5"e
•o c
r- g
s-8
a*
1 *
* 58
* I
in ^
* &
- <••»
S
I
o2
S s
* 8
g-g
S
d
Al
wa
-ass
3 J=
Sfe
U •—
•= 1?
il
•o 2
II
.
X! ">
•- « 4J
re
we
89
w
ts
— -5 «
sts
su
" *t3 •§ W V
*«"• _ ^.i? w w
3 3 —
•a o -
-
iS f
Is
ws V
rf E
w; O
V M
c .
"S c
5 -
•^ (if V *• fc.
— > C l_ 3
.s " « « §
S/5 « f * U
A
U
"I
Je
V
v
j>
Ou
c
D
Tf
3
1
8
u
1
CO
<9 a «
00 O (N
•a ^2
^^ O 1> U
.— '35 '35 '33
^ ^
00 O
•* in \o aq w-i
o o d d o
1/9 1/9 CA
(N r-;
M' o
.•= «
3a«
J
vn
w-i
^ o <~
-a
3
^^
I
V
00 00
o d
A A
«* "35
.2
x
o c —
odd
A A
I
10
l^^ ^T *^ ^5
•*G iri r* in
V V
tfl !/5 i«
>C -" 9^ £
V V
V
•5
C
"8
c
X.
o
4>
U
o
CO
a
&o
n
Z
1
CO
a
CO
-------
_
H*
° W W
111
V
tfl
u
F i— u -
4) ® Si
«•* f^ UM U
* ll I
T3 ^ o C
3 fe ~ .2
8 « 2 S
|| fl
O C
2-a
*r c
u- a
^J; CD C3 *-*
a '5 c i
^^ ^ CM fli
82^1
r» *> -c
SJ ^ «• a
«« .5
.2 -S
•§3
2 5
CO SO
'C C
§.c.
Ei/;
v
Sao
S3
o
i
_c
"^
I
I
-o
«<*
a
c
2 2
8
8
O
W
000000
o o
c/;
3
•o
CO
V
u
-S
1
o
I
*s
2
«rf
5/5
U
I
fi
U
i_
CO
4)
S.
-------
3
.S
o
u
^_if
> •»
i—
3 -8 a>
JD 5 C -O
c
t
.i
"I
• 2
£
u-
C
5
«
^
^
*o ^
n
11
r- > = 4>
3^5 wj3
Jsl — •" 13
«> E S) E
J- g — O
S» t^. w C
IS v ^
"* « ** e -=
| "S « ^.
« ^ -^ £
.s _c w v
Sll|
15=5 o 2J
0. 0.
2CQ
^
"O 'TS
« ^ «
CQ ^f ffl
•S 3 .£
o 'S o
1 § 1
^ Si -o
*d
§3 *> 9 «
53 *-' N o
o
,0
• lfc>
X
un —
ri U
•"• 1/3
> o'
•5 ""'
•»>
"P 'S x">
« 0 ^
5 « TJ
•o
4>
«rf
2
(/:
1
OS
o
Du
2
1 1
•i .1
O "S
1 §
-S a
ON '<«•
od c4
o
^
c
6
2
0
. U
^
V
c
—
2
S -S S
S ea
.ti an
as
.s-s
J
erf Os
rt ^
00 —i
ON ^O
r*- '-* o
^
0)
Xi
M
P
o:
3,
U
00
I
M
'S
o
«A
t>
I
o
c
UJ
-------
V
n
*S£
8 1
1|
^
B
•o
3
o?
1
^
•^
V
4—1
*^
*S
S
8
v
ha
•^
4)
3
_*^
8
k«
2
T>
*^
.5
'C
_0
o
•Tl
o.
«
•o
u
c
_o
U
N
£
.2
(S
.s-s
« JS
•a ^2
^
CO
I
««
o
0^
wi
S J
ii
1 °?
,— . vM4
CZ5 ^!!!!^
O CJ
OC
r4 o o c'o
p
'53
5
a
•*»#
•<
a
c
c
•a
c
s?
£
J<
o
z.
U
z
1
eg
*rf
"8
«^
CO
"C
^ .
S
Os
r4
a JB.
•35 ^.
!? S
C3 O
1
i
.2
-1
g
n
"S .£
4_> 'C
c a.
'C 3
0 V3
E «
^
ea «
« t^
0 0
SJ X> i £i
'7> . J2 *M —
ao CO QS in
c^ f*"i f^ *n
0 C5 00
• 1
1 1
f U
C9 **-
t
iJ'
.£
^
3
S)
3
1-
_c
1
^
V
3
«J
cc
o
.•= -^
Uf! -^
\O OC
1
•2
1
a
i
1
2
JS
i/5
1
SJ
tf.
V
.E
3
5b
x:
4prf
C
•-
S£,
.E
"S
S
<
c
'•5
1
'E
SO
'35
C
1
4A
^3
rs
0
U
>
in
y
-------
criterion. Site water mixed with effluent and clean site water alone produced WERs that
were not significantly different. This again indicates no detectable effect of domestic/
industrial effluents on zinc toxicity and/or bioavailability when water hardness is taken
into account. These tests with the fathead minnow and Ceriodaphnia dubia were
conducted over 7 days. Chronic WERs were also determined during these 7-day tests,
but the report contained insufficient details to fully evaluate the results.
Studies on the Straight River, Mingo Creek, Prickly Pear Creek, and Spokane
River varied under the guidelines and were too confounded to attempt any
generalization (Table 4).
The Trinity, Sabine, and Red Rivers in Texas were studied by Parkerton et al.
*
(1989). The only variation from the guidelines was that the hardness of the site waters
was not consistent with that of the laboratory water (Table 4). Since the WERs were
normalized for hardness as discussed earlier, this study complements those by Carlson et
al. (1986) and De Graeve et al. (1990). All three studies involved hardness as a variable,
and all three concluded that only hardness, not industrial or municipal waste
components, caused the observed differences in WERs.
WATER-EFFECT RATIOS FOR CADMIUM
Seven studies investigated the effects of site water on the toxicity of cadmium to
freshwater organisms. Of these, four contained data determined to be acceptable
according to the guideline requirements (Table 6). The most comprehensive and
enlightening, in terms of understanding the toxicity of cadmium, is the study by Spehar
and Carlson (1984). Unacceptable data are listed in Table 7, and some tentative
observations can be drawn from these data.
In their investigations, Spehar and Carlson (1984) tested five species in St. Louis
River site water and reconstituted laboratory water. A cladoceran (Simocephalus
semilatus), the fathead minnow, the rainbow trout and brown trout, and an amphipod
(Gammams pseudolimnaeus) provided WERs from 0.8 for the amphipod to 10.8 for the
brown trout (Table 6). This large variation can be attributed to species differences
20
-------
\
alone, since there was no difference between the two waters. This interspecific variation
is similar to that demonstrated for copper (Call et al., 1982). In addition to these
studies, Spehar and Carlson (1984) conducted a unique series of tests with larval fathead
minnows in which site water was collected monthly for 8 months and tested concurrently
with reconstituted water or Lake Superior water (Table 6). These WERs varied by a
factor of 3 and increased with increasing suspended solids, total organic carbon, turbidity,
and dissolved solids concentration. Correlations calculated for linear regressions of acute
toxicity and these parameters were 0.58, 0.60, 0.68, and 0.77, respectively (Spehar and
Carlson, 1984). The authors further concluded that "the large degree of binding or
complexing of cadmium that occurred during times (high river flow) when concentrations
of particulates in this water were highest was the apparent cause of reduced cadmium
toxicity."
In a separate study on the Nemadji River, WI, Call et al. (1982) determined
WERs of 1.9, 6.2, and 2.8, for two amphipods and Daphnia magna respectively (Table 6).
Although the LC50 values for the one amphipod (WER of 1.9) differed by a factor of 2,
the confidence limits overlapped and the difference was not significant. As with the
study by Spehar and Carlson (1984), the total residue and turbidity of the site water were
appreciably higher than in the laboratory water. At the same time, Call et al. (1982)
conducted parallel tests with the same species using site water from the Little Pokegama
River. WERs were 2.2, 2.6, and 1.8 (Table 6). Again, LC50 confidence limits used to
calculate two of those WERs overlapped resulting in nonsignificant differences. Total
residue and turbidity values were also higher in the site water.
JRB Associates (1983d) conducted site water tests on Selser's Creek with
laboratory reconstituted water. The grass shrimp (Palaemonetes kadiakensis) had a WER
of 3.0 (Table 6). An additional pair of tests with the pygmy sunfish (Elassoma zonatum)
produced a WER of 0.74, but the test was confounded by the use of different methods.
The site water test used flow-through conditions, and the laboratory water test was
conducted using static conditions (Table 7).
21
-------
sj
3 -2
£.3
s
o
1
+-
***
S
I
s
•o
2
9
•o
2
**
i
§
1
•O
«*
_**
OIM
— • o
Qp
10 -sr
en"* «">— ^»oo oo o o» — • ift — i
INM — o ««'« ^rsi «»>-•« 'O " •
— o <=> o p p p p pp pp
00 OO OO OO O O OO
— p
o o
SD
V
I
o - «
1 i 8
,2
I
I
2
1
1
S3
I
•fe-
ll
I'
Bl
-------
II
•o
v
2
+•*
1
VJ
JS
I
f/1
,o
te
o>
3
V
' |
' Vi
.2
I
« «
.2
I
.2
I
0>
ca
w
CS
oo
s<0 243
V, S l» - .- ^ •- « •- Jg
O
U
O
o o
o q
o o
— q
o o
OO C4
00
q q
o o
— o
O Cd ^) ^f
n no «N
(s) q q q
d d d d
q q
d d
a
a
I
JS
a
Q.
a
/
v
•ii
wa
-------
* a
o g
o *i
t/5 -Jl
.2
I
2
CO
O
u
a
OO
I
1
o
w
43
1
«•*
I
,2
&c
el
o o
TT 00
o o
53
o o
I
a
rt
•o
V
a
A
o
V
I
2
o
G
i
1
1
€
o
oo
B
8
oo
.SP
v
§
g^ J
I "2 j-
§§ g
|'i o
CU O ">
«J3 fe
£
to
i
I
-------
3
|
1
c: c
A
CO
1
V
U
3
I
•o
u
O
o
V
S.
•2
1
<+•«
•o
8
Studies were
4>
C
en
i
o
w
J5.
ffl
1
Si
fll
guidelines w<
35
s
tT
M
1
e
TJ
JS
i
c
E
B
U
'•5
"i
* as
comparisons
ages, no WE
i_
1
t/a
C
0
1
*•*
1
8
No cadmium
•o
3
i/l
1
U
^
—
1-1
^^
!
•^
detectable in
3
^•rf
t/i
•D
C
«
E
5
effluent was 1
*^3
( r
.52
"S.
c
4>
**
«
U
wastewater ti
V
CO
I/]
c
4>
£
S
CQ
cadmium me
questionable
p
^
B
J=
•*-
2
°3
C
f
s/i
a
^
t I
.0
u
1
e
D
-— CO
.a
E
(U Al
3 **
S *
(S
sq sq
r4 «*i
o o o
c
1
•a
u
."°
1
u
BO
•5
o
"i
.£•
Q.
0>
4>
U
•£
o
ya
"<5
C/3
O
cs:
00
5
tf
g
n
O
0.
C/3
X
j*"
u
2
U
c
o
,2
I
t/:
H>
"rt
B
LU
-------
The data from De Graeve et al. (1990) were normalized to account for different
hardness levels in the three test waters. The highest WERs for the fathead minnow (4.3)
and Ceriodaphia dubia (> >3.8) were observed in the mixture of effluent and clean site
water (Table 7). These data were considered unacceptable because the duration of the
tests was 7 days instead of the usual 96 hours for fish and 48 hours for cladocerans.
However, patterns within such a study can be evaluated even if the data cannot be
compared among studies. Chronic WERs were also determined from these 7-day tests,
but the details in the report were insufficient to fully evaluate the results.
No. useful data resulted from the studies on the Spokane River and Leon Creek
(Table?).
Of the metals reviewed in this report, cadmium provided more data from which to
observe patterns than any other metal. Like copper, cadmium provided data (Spehar
and Carlson, 1984) to demonstrate a large interspecific variability; this variability
invalidates attempts to use WERs when only two species are tested with cadmium. That
same study demonstrated significant variability in WERs over an 8-month period due to
seasonal flow conditions affecting the concentration of suspended solids or completing
agents. The data from Call et al. (1982) suggested a similar phenomenon. Other studies
with cadmium indicated low WERs.
WATER-EFFECT RATIOS FOR CHROMIUM, LEAD AND NICKEL
Since there were insufficient data for chromium, lead, and nickel from which to
summarize or draw conclusions, these metals are addressed together in this section and
in Tables 8 and 9. Three studies addressed site water effects on chromium toxicity, and
all were supported by the initial funding by EPA's Office of Water Regulations and
Standards. Only one, Mingo Creek (JRB Associates, 1983b), resulted in no acceptable
data.
A caddisfly (Cheumatopsyche sp.) was tested in Boggy Creek site water and
reconstituted laboratory water (JRB Associates, 1983c); this produced a WER for
chromium of 1.2 (Table 8). An additional test with the red shiner was considered to be
26
-------
unacceptable because the site water test was conducted under flow-through conditions,
and the lab water test used static conditions (Table 9). The Leon Creek study
(Melancon et al., 1983) tested an amphipod (Hyalella azteca) and derived a chromium
WER of 2.7 (Table 8). A second set of paired tests with the fathead minnow resulted in
no mortalities at concentrations as high as 8 mg/1.
Five studies were conducted with lead, but only one resulted in acceptable data.
The marine/estuarine study in San Francisco Bay (Larry Walker Associates, 1991) and
three freshwater studies in the Norwalk River (Dunbar and Pizzuto, 1982), Selser's Creek
(JRB Associates, 1983d), and the Spokane River (Bailey and Saltes, 1983) were rejected
primarily owing to excessive precipitation of lead during the toxicity tests. Lead is
relatively nontoxic compared to other metals, and it has low aqueous solubility. In some
instances, other problems necessitated the elimination of the resultant data from this
review. One excellent study (Hammermeister et al., 1983) compared St.. Louis River site
water and dechlorinated Superior, Wisconsin, tap water drawn from wells under Lake
Superior. These investigators tested the fathead minnow, rainbow trout, Daphnia magna,
and an amphipod (Gammarus psuedolimnaeus), and produced WERs of 1.9, 3.4, 5.7, and
4.1, respectively (Table 8) Again, these results indicate a significant interspecific effect.
Nickel was successfully tested in Selser's Creek, Louisiana (JRB Associates,
1983d), and the WER for the grass shrimp was 1.6 (Table 8). The LC50 values,
however, were not significantly different. No data were presented in the Spokane River
study by Bailey and Saltes (1983) (Table 9).
While the data from the San Francisco Bay study by Larry Walker Associates
(1991) were considered unacceptable for a variety of reasons (Table 9), a comparison of
data within this study was made. There appeared to be no difference between the
WERs for the two species, although the fish (Menidia beryllina) was about 20 times more
resistant to nickel. The tests were conducted during each of four consecutive months
and showed no variation in WERs over time; the eight WERs were close to 1.0,
indicating no difference in the toxicity of nickel in site and laboratory water. It must be
remembered that the laboratory water was not fully characterized.
27
-------
RATIOS OF DISSOLVED TO TOTAL METAL CONCENTRATIONS
\
As directed in the work assignment, this report attempted to evaluate these
site-specific studies in terms of the ratio of dissolved, metal to total recoverable metal.
As the numerous reports were reviewed, it became apparent that there is a great
inconsistency in the terminology used to describe the analytical endpoints characterizing
metal concentrations. The author believes that the principal reason for this lack of
detail is the primary focus of the senior author on the biological or lexicological aspects
of the study. In this author's experience, sufficient detail is present only when a chemist
«•
coauthors the report and writes the analytical section. In an attempt to provide clarity,
a list of definitions (Table 10) was obtained from the Lake Superior Research Institute,
which conducted some of the most complete studies on this subject (Call et al., 1982;
Hammermeister et al., 1983).
Most of the reports did not provide a comparison of dissolved to total recoverable
metal, and some did not provide details sufficient to indicate which analytical endpoints
were being discussed. It is interesting, and fortunate, to note that the studies that
provided the best lexicological data also provided the best analytical results. These were
the studies conducted by the EPA's Environmental Research Laboratory-Duluth and the
Lake Superior Research Institute at the University of Wisconsin-Superior.
Dissolved to total recoverable metal ratios in this report (Table 11) were selected
on the basis of two considerations. First, the analytical endpoints had to be described in
clear detail and, second, the metal concentrations had to be at or close to the reported
LC50 and EC50 values. This latter condition is important because precipitation is
common at higher concentrations of metals. In addition, because the maximum criterion
concentrations are based on LC50 and EC50 values, it was felt that data selection would
provide the best and most meaningful ratios.
There are two major patterns or conclusions that can be drawn from the data in
Table 11. First, the dissolved concentration LC50 and EC50 values are typically 80 to
100 percent of the total concentration values. Second, there appears to be no differences
28
-------
<*- £•
O o u
*•» JBI
8 11
If
•o
• «
*rf
3
•econstil
0.
ra
•o
u
_c
hp>
O
1
f
T3
A% '
eg
.2 " '
j=
o.
2
•o
_c
o
4S
' 13
a.
eg
*o
4U
re
°C
£
o=
•o
sir
, "S E -
.«J C 0. -0 ?
j 1 f 1 I"
T a i s B
•S3 J2 -5 ^2 ' 53 ^2
O t~- O «N
(NO O O r-^ rf
1 -t
I "8 1
!| "i" *
1" t 2
Cj n OD
o
t/:
C*
UJ
-C
c
•o
1
1
o
c
'•g
8
u
C
kl
•5
o*
-------
e
t
CL
a
Reasons foi
u
C9
«
•o
3
*•»
conditions and si
continuous flow.
E °° •
E <«
3 «
'p JS
|.2P
if
No mortality at c
concentrations a;
mg/1
VI
_o
fa
**
1/3
U
U
•o
3
**
e/3
U
^_j
£
2
I
M
V
V
conditions and si
continuous flow.
t/3
j^
1
.i
'35
u
*•*
1/3
C
1
^
gljj
1
M
continuous flow ;
CL.
J
. _f* ^*2
« <«
ffi 2
R) C
•2. 5
00 |
^1
.a 8
g.s
«« w
!-s
> *-
c
CO
.3
concentration un
.=
^ &
» 2
^^
wei
er, i
.« tt)
able, no tests
h clean ab wa
•S «
w
W
— C
3 «5 5 -
^ £ n J
C O
l-c.
* t«
M .v
- 00
o w
l« ^
i_ C
CQ O
Q. «-
1^
8^5
•a "o
c u
I
1
.o
II
§
•a
*+
3
8
4>
1/3
I
u
8
15
«
«^
a.
O O
t/3
1/9
1
C
I
•o
a
42
a.
O
a
2
U
U
eg
j*
g.
-------
v
C
K
C
&
8.1
f-» •ted
C " .S 3
• . « ->\.
jj ,»T «*
IPS-!
_ •=3.a
"^ 3 "S 52 -
Q 3 w g c
II
3 _ « -
^Sll*
| § 12-s
§ E -° E g
Q 6 — ^ 5
" I "I-8 S
v 8 S S 5
J C is a. a.
o ' co O ~ 3 ™*
o . C wi *; «
£ w O t J -Q w
=
-------
c
*£
.1
C
tt
**>
(/
c
V
n
4,
tt
"O
t>
u
3
"O
1
V
l_
Studies we
2
1
i
^5
•a
'I
£
tests we
o
before gui<
available, i
*•*
V)
*•*
C
U
*"•
1
CQ
MM
c
"w
4-^
f
"5
I
c
compariso
c«
S
^
o
c
4>
DO
different a
e
0
+.*
1
B
^
I
O
Z
T3
2
3
unmeas
-------
Table 10. METAL ANALYSES CONDUCTED BY THE LAKE SUPERIOR
RESEARCH INSTITUTE.
Acid-exchangeable metal - Samples are collected, immediately acidified to a pH 2 and
analyzed directly by atomic absorption spectrophotometry (AA), The sample will
typically have properties similar to drinking water and be paniculate-free.
Frequently a comparison will be made between acid exchangeable metals and a more
stringent digestion to determine matrix interferences from more polluted waters. If
there is a good agreement between methods, the digestion could be dropped.
Dissolved metals - EPA, 1979, Metals 4.1.1). An unacidified portion of the sample is filtered
through a conditioned 0.45 //m membrane filter. The filtrate is acidified arid analyzed by
AA without further treatment.
Acid extractable metals - (APHA, 1985, meth. 302B) - Metals that are lightly absorbed on
paniculate matter. An acidified sample (nitric acid, hydrochloric acid) is heated for 15
minutes on a steam bath, then filtered through a 0.45 pm membrane filter. The volume
is adjusted and the sample is analyzed by AA.
Total recoverable metals - (EPA, 1979, Metals 4.1.3). The sample is treated with nitric and
hydrochloric acids and reduced in volume on a steam bath. The volume is restored and
analyzed by AA.
Total metals - (EPA, 1979, Metals 4.1.3). A more stringent digestion to free inorganically
and organically bound metals. An acidified sample is digested and refluxed with nitric
acid followed by treatment by hydrochloric acid. The volume is restored and the sample
is analyzed by AA.
Acid soluble metals - (EPA, 1985, p. 1). The sample is acidified to a specified pH (pH
1.5-2.0) and is filtered through a 0.45 jum membrane filter. The sample is then analyzed
byAA.
Ambient Water Quality Criteria for Lead -1984. EPA, 1985. p. 1-3.
Standard Methods for the Examination of Water and Wastewater. APHA. 16th ed. 1985.
Methods for Chemical Analysis of Water and Wastewater. EPA. March, 1979.
33
-------
8
"
£ o-o
3 Si
» B- ~
111
ta-i jjj o
— ^ «
C/3 -J C
1
1.
•*
8
i
u
3
2
I
s
2
•E I
«•»
•o3
(9
00
o
7i ^ Q.Q. 0.0.0.' L.
g1* «« eosaco «
ill il ^ iilll iiilll s.'i
*1 CQ CQ CO OM OH Qb ^W CO CO Q!i Q£ Q£ mtoC *U
CL ^*0 "CC CCC O Q,
3 55 ^ ^ ••••«•«• L^ u» fc* "S 'S *S3 ^ -"^
<* ooTt- ^-ir-,
C^ ^^ 00 ^^ ^n^ ^^ ^2 ^^-
s
2
j
TJ
CO
U
a:
3
a**
2
2
_
I
i/a
1
55'
c3
00
-------
« s
ll
10 S
•* * —
WOT?
4) 4) 4>
ex ex
pa eg
** «•* i- k.
T3 T3 g |>
4) 4) i> >
UU 3
tn y> 'S
Al U CM
s
8
- --
(U W *J
3
1
•2 i
oo
e" S)
8 "S
a. •}
.00
C/3
I
a
8
o
«j
i/:
z
I
8
1
3
A
-------
in percent dissolved concentrations between clean laboratory water and a variety of
surface waters containing variable amounts of dissolved and suspended organic and
inorganic materials.
36
-------
SUMMARY AND CONCLUSIONS
The purpose of the site-specific criteria studies reviewed in this report was the use
the indicator species procedures to calculate WERs for concurrent acute toxicity tests.
To derive site-specific criteria, these results would be used to provide a site-specific Final
Acute Value from which, using acute/chronic ratios in the criteria documents, a Final
Chronic Value or equation would be derived. Too few studies conducted chronic tests in
site and laboratory waters to warrant discussion in this report.
This survey of water-effect ratios was an attempt to understand or explain the
excursions above chronic criteria observed in the monitoring of waters for metals.
The results of this review do provide some insight into potential causes of these
excursions, but a complete understanding does not appear to be possible at this time.
The data for copper suggest a potential ability to adjust criteria for site water
effects, but the data base is insufficient to do so. In addition, one test with four species
(Hammermeister et al., 1983) demonstrated a factor of 5 variation in WERs due to
differences in species alone. This pattern was also demonstrated for cadmium (Spehar
and Carlson, 1984) and for lead (Hammermeister et al., 1983). The site-specific
guidelines presently require only two species for testing the indicator species approach;
the author believes this is too few considering the results with copper, cadmium, and
lead.
Little variation of any kind was noted in the WERs for zinc, and almost no
difference was seen in the sensitivity of organisms between site and laboratory waters.
Even when four species were tested (Hammermeister et al., 1983), there was little
difference. Other than hardness, no major water parameter appears to affect zinc
toxicity.
The cadmium results present the greatest potential for modifying national criteria.
Consistently high WERs were found in most site waters, and several of the studies
suggested effects due to suspended solids. The eight monthly WERs for fathead
minnows (Spehar and Carlson, 1984) demonstrated this effect quite well. These results
suggest that it may be beneficial to correlate cadmium excursions with suspended solids.
37
-------
The few data available for chromium, lead, and nickel do not permit any firm
generalizations. However, the consistent problem with lead solubility during testing and
the relative insensitivity of test organisms to lead indicate that lead excursions may not
be indicative of an ecological problem.
^s»-'
The problems of defining analytical endpoints for total recoverable and dissolved
metal concentrations have been discussed. Several slightly different procedures were
used to determine total metal, and those studies did not investigate the potential
variability due to methodology. If this inconsistency in methods occurs in a monitoring
program, it will also cause the monitoring results to be less precise. However, the
consistent similarity of-percent dissolved values to total metal values for all the metals
(Table 11) indicates that this methodological problem should have had a large effect on
the results reviewed in this report, especially since site waters and clean laboratory
waters did not differ.
38
-------
REFERENCES
Bailey, G.C. and J. Saltes. 1983. The Development of Some Metal Criteria for the
Protection of Spokane River Rainbow Trout. Project Completion Report from
Washington State University to the Washington State Department of Ecology.
Call, DX; L. Brooke and P.P. Vaishnav. 1982. Aquatic Pollutant Hazard Assessments
and Development of a Hazard Prediction Technology by Quantitative Structure-Activity
Relationships. Fourth Quarterly Report, USEPA Cooperative Agreement No. CR
809234-01-0.
Carlson, A.R. and T.H. Roush. 1985. Site-Specific Water Quality Studies of the Straight
River, Minnesota: Complex Effluent Toxicity, Zinc Toxicity, and Biological Survey
Relationships, EPA/600/3-85/005.
Carlson, A.R. et al. 1984. Guidelines for Deriving Numerical Aquatic Site-Specific
Water Quality Criteria by Modifying National Criteria. EPA 600/3-84/099.
Carlson A.R. et al. 1986. Development and Validation of Site-Specific Water Quality
Criteria for Copper. Environmental Toxicology and Chemistry. 5:997-1012.
Carlson, A.R. et al. 1986. Evaluation of Site-Specific Criteria for Copper and Zinc: An
Integration of Metal Addition Toxicity, Effluent and Receiving Water Toxicity, and
Ecological Survey Data. EPA/600/3-86/026.
DeGraeve, G.M. et al. 1990. A Field and Laboratory Site-Specific Evaluation of
Cadmium, Copper, and Zinc for the Cuyahoga River in the Vicinity of the Northeast
Ohio Sewer District's Southerly Effluent Discharge. Battelle Columbus Division.
Department of Natural Resources and Community Development. 1982. Development of
Site-Specific Criteria, Unnamed Tributary to Mulberry Creek, Wilkes County, North
Carolina.
Dunbar, L.E. and E. Pizzuto, Jr. 1982. Derivation of Site Specific Water Quality
Criteria: Norwalk River at Georgetown, CT. State of Connecticut, Department of
Environmental Protection.
Hammermeister et al. 1983. Comparison of Copper, Lead and Zinc Toxicity to Four
Animal Species in Laboratory in St. Louis River Water. Internal Report, Center for
Lake Superior Environmental Studies. University of Wisconsin-Superior.
JRB Associates. 1983a. Demonstration of the Site-Specific Criteria Modification
Process: South Fork of the Crow River, Hutchinson, Minnesota. EPA Contract
68-01-6388.
39
-------
JRB Associates. 1983b. Demonstration of the Site-Specific Criteria Modification
Process: Mingo Creek, Tulsa, Oklahoma. EPA Contract 68-01-6388.
JRB Associates. 1983c. Demonstration of the Site-Specific Criteria Modification
Process: Boggy and Skeleton Creeks, Enid, Oklahoma. EPA Contract 68-01-6388.
JRB Associates. 1983d. Demonstration of the Site-Specific Criteria Modification
Process: Selser's Creek, Ponchatoula, Louisiana. EPA Contract 68-01-6388.
Larry Walker Associates and Kinnetic Laboratories. 1991. Site-Specific Water Quality
Objectives for South San Francisco Bay.
Melancon, S.M.S. et al. 1983. The Toxicity of Cadmium and Chromium in Leon Creek,
Texas. EPA 600/X-83-036.
Miller, T.G. et al. 1983. Site Specific Water Quality Assessment: Prickly Pear Creek,
MT. Environmental Monitoring Systems Laboratory, Las Vegas, Nevada.
Spehar, R.L. and A.R. Carlson. 1984. Derivation of site-specific water quality criteria
for cadmium and the St. Louis River Basin, Duluth, Minnesota. Environ. Toxicol. Chem.
3:651-665.
U.S. Environmental Protection Agency. 1983. U.S. EPA Water Quality Standards
Handbook. Office of Water Regulations and Standards. Washington, DC: U.S. EPA.
U.S. Environmental Protection Agency. 1985. Guidelines for Deriving National Water
Quality Criteria for the Protection of Aquatic Organisms and their uses. NTIS
PB85-227049.
40
-------
-------
u
.E
Reasons for Rejection
K
st results could be determ
V
O
Z
r^*>i
V*J
O
c
JO
B
£5
•*_»•
L.
V
+•»
«o
~
(A
4>
+*
§
"35
V]
1)
"5 "r:
SLg.
un Q,
•S 8
.8*
sl
"33 .-s
-C TJ
J^
>. U
•s -s
3 I
CO O
s^
J *
es were conducted before
55
V3
»/)•••*••
t? S
lines were available, no te
run with clean lab water, t
as
IJ
«•*
arisons were with differen
no WERs.
— •
E S
8 tf
£'«
U u
«^ ,,2
C4 U
5 js
u .r
•35 J
. «T to
C^^
T3 W
r- '53
£^
v .S
S3 E
? c
« «
w JJ
w ^
__ <«
^ i
o s
«= i
water mixed with effluent,
confidence limits for site '
^ S
|Q
k.
^^
WERs were normalized ft
ess.
-""2
en S
s£-
WJ
4>
"S
nfidence limits on LC50 vi
. 8
0
Z
w-S
asured concentrations wer
o calculate LC50 values, 1:
was not described.
« ** «-
C T3 «
1^5
335
I
e
>H
j
c
n
§
.2
«
I
V
1
"S
econs
•o
V
3
a
8
V
|
^
C
3
unknow
unkn
8
(N
OO
m
U
T3
'E
q\ ^^
o
"35
.§
E
'
.•= CH
m ao »n
ao •* —
r<-> r« ^
Sac ao
c^ r^
r^ >—' O
O O C5
A
r-i
-------
a
a
I
4.
tt
~ - "1
c
c
8
V
c
3
(S
C
3
.r at
2 3
0 G5
0 O
o o
A
.t: eg
2 —
f^l fn
c o
A
c/]
4J
"u
g
1/9
V
I
"O
CO
«
£
T3
•«
-*
*>
«>
2 O
So
------- |