United States Office of Water ERA 822-D-99-OD2
Environmental Protection 4304 November 1999
Agency
SERA Draft Ambient Water Quality
Criteria for Dissolved
Oxygen (Saltwater): Cape
Cod to Cape Hatteras
-------
-------
Ambient Water Quality Criteria for Dissolved Oxygen (Saltwater):
Cape Cod to Cape Hatteras
Introduction
Section 304 (a)(2) of the Clean Water Act calls for information on the conditions
necessary "to restore and maintain biological integrity of all.. . waters, for the protection
and propagation of shellfish, fish and wildlife, to allow recreational activities in and on
the water, and to measure and classify water quality." The Environmental Protection
Agency has not previously issued saltwater criteria for dissolved oxygen (DO) because
the available information on effects was insufficient. This document is the result of a re-
search effort to produce the required information to support the development of saltwater
DO criteria. The criteria presented herein represent the best estimates, based on the avail-
able data, of DO concentrations necessary to protect aquatic life and its uses.
The geographic scope of this document is limited to the Virginian Province of the
Atlantic coast of the United States (i.e., southern Cape Cod, MA, to Cape Hatteras, NC).
The document provides the information necessary for environmental planners and regu-
lators in the Virginian Province to decide whether the DO at a given site can protect
coastal or estuarine aquatic life. The approach can be used to evaluate existing localized
DO goals (e.g., Jordan, et al., 1992) or to establish new ones. This document does not ad-
dress direct behavioral responses (i.e., avoiding low DO) or the ecological consequences
of behavioral responses such as changes in predation rates or in community structures.
The document also does not address the issue of spatial extent of a DO problem. A given
site may have DO conditions expected to cause a significant effect on aquatic life, how-
ever, the environmental manager will have to judge whether the spatial extent of the low
DO area is sufficient to warrant concern. The approach presented here for deriving crite-
ria is expected to work for other regions. However, additional regionally specific data
may be required in order to amend the database for use in other regions. Animals may
have adapted to lower oxygen in locations where high temperatures have historically re-
duced concentrations, or in systems with natural high demands for oxygen. In addition,
effects of hypoxia1 may vary latitudinally, or site-specifically, particularly as reproduc-
tive seasons determine risks of exposure for sensitive early life stages.
As with the freshwater DO document (U.S. EPA, 1986), all data and criteria are
expressed in terms of the actual amount of DO available to aquatic organisms in milli-
grams per liter (mg/L). However, unlike the freshwater document, which provides limits
for DO in both warm and cold water, criteria are presented for only warm saltwater be-
cause hypoxia in Virginian Province coastal waters is primarily restricted to the warm
water of summer. Also, the freshwater criteria are based almost entirely on fish data even
though insects were often more sensitive than fish. The saltwater limits, on the other
hand, use data from fish and invertebrates.
1 Hypoxia is defined in this document as the reduction of DO concentrations below air saturation.
-------
The saltwater criteria described herein are intended to maintain and support
aquatic life and their designated uses. Criteria derived using the Guidelines2 are intended
to protect aquatic communities, but they rely primarily on data generated at the organism
level, and emphasize data for the most sensitive life stage. But a population of a given
species can potentially withstand some mortality to certain life stages without a signifi-
cant long-term effect on the population. Hence, an assessment of criteria should include
population-level considerations. One nuance of population-level assessment is the fact
that a population's sensitivity to hypoxia may depend on which stages have been ex-
posed. For example, many populations of marine organisms may be more impacted by
mortality occurring during the juvenile and adult stages than during the larval stage(s). In
this regard, a particular individual larva is not as important to the population as a par-
ticular individual juvenile or adult. With this in mind, the saltwater criteria for DO segre-
gate effects on juveniles and adults from those on larvae. The survival data on the sensi-
tivity of the former are handled in a traditional Guidelines manner. The cumulative ef-
fects of low DO on larval recruitment to the juvenile life stage, on the other hand, address
survival effects on larvae. The recommended DO approach uses a mathematical model to
evaluate the effect on larvae by tracking intensity and duration effects across the larval
recruitment season. Protection for larvae of all species is provided by using data for a
sensitive aquatic organism (the Say mud crab Dyspanopeus sayi in this case). This model
is used to generate a DO criterion for larval survival as a function of time.
For the reasons listed above, the approach recommended below to derive DO cri-
teria for saltwater animals deviates from EPA's traditional approach for toxic chemicals
outlined in the Guidelines. Where practical, however, data selection and analytical proce-
dures are consistent with the Guidelines. Therefore, some of the terminology and the cal-
culation procedures are the same. Thus, knowing the Guidelines are useful (but not es-
sential) for better understanding how the limits were derived. Terminology from the
Guidelines used here includes Species Mean Acute Value (SMAV), Genus Mean Acute
Value (GMAV), Final Acute Value (FAV), Genus Mean Chronic Value (GMCV) and
Final Chronic Value (FCV). Procedures from the Guidelines include those for calculating
FAVs, Criterion Maximum Concentration3 (CMC) and Criterion Continuous Concentra-
tion (CCC).
Overview of the Problem
The EPA's Environmental Monitoring and Assessment Program (EMAP) for the
estuaries in the Virginian Province has shown that 25% of its area is exposed to some de-
gree to DO concentrations less than 5 mg/L (Strobel et al., 1995). EMAP has also gener-
ated field observations that correlate biological degradation in many benthic areas with
low DO in the lower water column (Paul et al., 1997). The two reports serve to empha-
size that low DO is a major concern within the Virginian Province. Even though hypoxia
is a major concern, a strong technical basis for developing benchmarks for effects of low
DO has been lacking.
2 Guidelines for deriving Numerical National Water Quality Criteria for the Protection of Aquatic Organ-
isms and Their Uses (Stephen, et al., 1985—hereafter referred to as the Guidelines).
3 Although in the case of dissolved oxygen, CMC is more appropriately defined as the Criterion Minimum
Concentration.
-------
Hypoxia in the Virginian Province is essentially a warm-water phenomenon. In
the southern portions of the Province, such as the Chesapeake Bay and its tributaries, DO
may be reduced any time between May and ^October; in the more northern coastal and
estuarine waters, any time from late June into September. Hypoxic events may be sea-
sonal or diel. Seasonal hypoxia often develops as stratified water prevents the oxygenated
surface water from mixing downward. Low DO then appears hi the lower waters when
respiration in the water and sediment depletes oxygen faster than it can be replenished.
As summer progresses, the areas of hypoxia expand and intensify, then disappear as the
water cools in the fall. The cooler temperatures eliminate the stratification and allow the
surface and bottom waters to mix. Diel cycles of hypoxia often appear in unstratified
shallow habitats where nighttime respiration can temporarily deplete DO.
Although the primary fauna at risk from exposure to hypoxia in the Virginian
Province are summer inhabitants of subpycnocline4 (i.e., bottom) waters, hypoxia can
occur in other habitats as well. For example, upwelling may permit subpycnocline, oxy-
gen-poor water to intrude into shallow areas. Hypoxia also may appear in the upper water
of eutrophic water bodies on calm, cloudy days, when more oxygen is consumed than is
produced by photosynthesis and when atmospheric reaeration is limited. In spite of this
tendency, however, minima in DO are generally less severe above the pycnocline than
below it. Hypoxia above the pycnocline also tends to be more transient because it largely
depends on weather patterns.
Hypoxia may persist more or less continuously over a season (with or without a
cyclic component) or be episodic (i.e., of irregular occurrence and indefinite duration).
Continuous hypoxia without a cyclic component is exemplified in the subpycnocline
waters of western Long Island Sound and off the New Jersey coast (Armstrong, 1979).
Hypoxia in Long Island Sound may be interrupted temporarily by major storms, but re-
turns one or two weeks later, when the waters again become stratified (Welsh et al.,
1994).
Hypoxia may oscillate with tidal, diel or lunar frequencies. Tidal hypoxia is
common in subpycnocline waters of the mesohaline Chesapeake Bay main stem and the
mouth of the adjacent tributaries during summer (Sanford et al., 1990; Diaz et al., 1992).
In this case, DO concentrations oscillate as the tides alternately advect poorly oxygenated
subpycnocline water from the mid-bay trough or tributaries and better oxygenated water
from the lower bay. Diel cycles of hypoxia are found in small eutrophic embayments and
harbors all along the coast of the Virginian Province, where oxygen is depleted overnight
by respiration and replenished by photosynthesis after dawn. The Childs River is an ex-
ample of diel hypoxia (D'Avanzo and Kremer, 1994). Lunar cycles of oxygen may occur
in various systems but have been documented most clearly at the mouths of some Chesa-
peake Bay tributaries, where destratification from spring tides saturates the water with
oxygen and stratification afterward depletes the oxygen (Haas, 1977; Kuo et al., 1991;
Diaz et al., 1992).
4 The pycnocline is the region of density discontinuity in a stratified water column between surface and
bottom waters. The density difference between the two is primarily due to differences in temperature and
salinity.
-------
Episodic hypoxia has been noted in shoal waters of mid-Chesapeake Bay (Breit-
burg, 1990) and in adjacent tributaries (Sanford et al., 1990). Persistent winds tilt the
pycnocline laterally and displace low DO water onto the shoals or tributaries indefinitely.
As noted above, DO may also be reduced episodically in eutrophic surface waters, par-
ticularly during calm and cloudy weather, when photosynthesis is slow and daytime re-
oxygenation is reduced.
Biological Effects of Low Dissolved Oxygen
Oxygen is essential in aerobic organisms for the electron transport system of mi-
tochondria. Oxygen insufficiency at the mitochondria results in reduction in cellular en-
ergy and a subsequent loss of ion balance in cellular and circulatory fluids. If oxygen in-
sufficiency persists, death will ultimately occur, although some aerobic animals also pos-
sess anaerobic metabolic pathways, which can delay lethality for short time periods
(minutes to days). Anaerobiosis is well developed in some benthic animals, such as bi-
valve molluscs and polychaetes, but not in other groups, like fish and crustaceans (Ham-
men, 1976). There is no evidence that any free-living animal inhabiting coastal or estua-
rine waters can live without oxygen indefinitely.
Many aquatic animals have adapted to short periods of hypoxia and anaerobiosis
by taking up more oxygen and transporting it more effectively to cells and mitochondria,
i.e., by ventilating its respiratory surfaces more intensely and increasing its heart rate. If
these responses are insufficient to maintain the blood's pH, the oxygen carrying capacity
of the respiratory pigment will decrease. An early behavioral response might be moving
faster toward better-oxygenated water. However, if the hypoxia persists, the animal may
reduce its swimming and feeding, which will reduce its need for energy and hence oxy-
gen. Such reduce motor activity may make the animal more tolerant over the short term,
but will not solve its long-term problem. For example, even the modest reductions in lo-
comotion required by mild hypoxia may make the animal more vulnerable to predators,
and the reduced feeding may decrease its growth.
Compensatory adaptations are well developed in marine animals that commonly
experience hypoxia, e.g., intertidal and tide pool animals (McMahon, 1988), and bur-
rowing animals, which partly explains their reported high tolerance to low DO. In con-
trast, compensatory adaptations are poorly developed in animals that inhabit well-
oxygenated environments such as the upper water column. The animals most sensitive to
hypoxia are among this latter group. Details on compensatory adaptations to hypoxia are
provided in reviews for marine animals (Vernberg, 1972), aquatic invertebrates (Herreid,
1980) and fish (Holeton, 1980; Hughes, 1981; Kramer, 1987; Rombough, 1988a, and
Heath, 1995).
Overview of the Approach
The approach to determine the limits of DO that will protect saltwater animals
within the Virginian Province considers both continuous (i.e., persistent), and cyclic (e.g.,
diel) exposures to low DO. The continuous situation is covered first, and deals with expo-
sures longer than 24 hr. It is followed by sections on criteria for exposures of less than
10
-------
24 hr but that may be repeated for days. Both scenarios cover three areas of protection
(summarized here, and explained in more detail in the sections that follow):
1. Juvenile and adult survival—A lower limit is calculated for continuous exposures
by using Final Acute Value (FAV) calculation procedures outlined in the Guide-
lines (Stephan et al., 1985), but with data for only juvenile or adult stages. Limits
for cyclic exposures are derived from an appropriate time-to-death curve for ex-
posures less than 24 hr.
2. Growth effects—A threshold above which long-term, continuous exposures should
not cause unacceptable effects is derived from growth data (mostly from bioas-
says using larvae). This Final Chronic Value (FCV) is calculated in the same
manner as the FAV for juvenile and adult survival. This threshold limit as cur-
rently presented has no tune component (it can be applied to exposures of any du-
ration). Cyclic exposures are evaluated by comparing reductions in laboratory
growth from cyclic and continuous exposures.
3. Larval recruitment effects—A larval recruitment model was developed to project
cumulative loss caused by low DO. The effects depend on the intensity and the
duration of adverse exposures. The maximum acceptable reduction in seasonal re-
cruitment was set at 5%, which is equivalent to the protective limit for juvenile
and adult survival. The number of acceptable days of seasonal exposure to low
DO decreases as the severity of the hypoxic condition increases. The severity of
cyclic exposure is evaluated with a time-to-death model (as in the protective limit
for juveniles and adults).
Persistent Exposure to Low Dissolved Oxygen
Juvenile and Adult Survival
Data were used from tests with exposure ranging from 24 to 96 hr. This maxi-
mized the number of genera for the FAV calculation. Data for juveniles show that LC50
values calculated for 24 and 96 hr observations are very similar (Figure 1), therefore, all
values are applied as 24 hr data. The restriction of the data set to tests of 96 hr duration or
less was somewhat arbitrary; however, 96 hr is the duration used for most acute tests for
traditional water quality criteria (Stephan et al., 1985). In addition, there are insufficient
test data to compare 24 hr exposures versus those longer than 96 hr. Juvenile and adult
mortality data from exposures longer than 96 hr are compared to the final criterion in the
section on Other Laboratory Bioassay Data.
Data on the acute sensitivity of juvenile and adult saltwater animals to low DO is
available for 12 invertebrate and 11 fish species (almost all of the data are for juveniles).
The values are summarized in Table 1 and Appendix B. Overall Genus Mean Acute Val-
ues (GMAVs) range from <0.34 mg/L for the green crab Carcinus maenas to 1.63 mg/L
for the pipe fish Syngnathus fuscus; a factor greater than 4.8. Juvenile fish are somewhat
more sensitive than juvenile crustaceans (Table 1; Figure 2). In fact, the four most sensi-
tive genera are all fish, and the range of values for these is 1.32 to 1.63 mg/L; a ratio of
only 1.2.
As stated previously, the criterion for juveniles and adults exposed to continuous
low DO was calculated using the Guidelines procedures for derivation of an FAV
11
-------
2.0 .
1.8 -
1.6 .
_ 1'4-
| 1.2.
i" 1.0.
ll 0.8-I
x:
0.6 .
0.4.
0.2 .
Juveniles Only
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
24 hr LC50 (mg/L)
Figure 1. Relationship between 24 and 96 hr LC50 values for juvenile saltwater ani-
mals exposed to continuous low dissolved oxygen. Each point represents a paired set of
values calculated from the same test run. The line drawn represents a one-to-one rela-
tionship. Data for the plot are summarized by species in Appendix A. Appendix A also
contains data for test runs with larvae.
(Stephan, et al., 1985). However, the procedures outlined in the Guidelines were created
for toxicants. Since DO behaves in the opposite manner to toxicants (i.e., the greatest re-
sponse is associated with the lowest concentrations), DO concentration data were trans-
formed by using their inverse in the calculation. The FAV calculation is essentially a lin-
ear regression using the LC50 values for the four most sensitive genera and their respec-
tive percentile ranks. The final FAV is the value representing the 5th percentile genus5,
which for DO is 1.64 mg/L. This value is adjusted to a criterion of 2.27 mg DO/L by
multiplying by 1.38, the average LC5 to LC50 ratio6 for juveniles (Table 1). This value is
analogous to the CMC (Criterion Maximum Concentration) in traditional Water Quality
Criteria for toxicants.
5 Alternatively we could have modified the FAV calculation procedure to use untransformed data and es-
tablished the protective limit for the 95th percentile. However, the calculated results would be the same.
Since many researchers already have computer programs that calculate FAVs, we opted to remain consis-
tent with the Guidelines by using the inverse data.
6 The use of a ratio to adjust the FAV to a CMC is designed to estimate a negligible lethal effect concentra-
tion corresponding to the 5th percentile species. It may in fact represent an adverse effect concentration for
species more sensitive than the 5th percentile. The Guidelines use a factor of 2, however, there were suffi-
cient data available for low DO to use a factor specific to this stressor.
12
-------
J3
VD
f.
"
•8
I
I
I
> ^
g s
S (2
II
^^
^^] ^"
o
Is
o
»2 ^
Is
£>• a
§ u
f
so
S
i
Common na
I
B.
cs
Tl-
CO
o
V
ri
V
I
_s
1
*«
Carcinus maenus
ts ol
s
o '
CO «
o o
VO
1
2 *"•
o o
— ^o
1 1
Atlantic surfolara
Harris mud crab
j.
Spisula solidissima
Rithropanopeus harri,
o\
Q
o
00
o
^
o
9
oo
o
s
o
tt>
1
1^
a
1
§
Prionotus carolinus
00
^
o
in
o
«
1
flat mud crab
1
Eurypanopeus depres.
r-
vo
00
o
0
o
VO .
00
o
g
o
JD
1
I
Leiostomus xanthurus
S
0
*
g
o
?
«J.
s
0
J£
3
1
Tautoga onitis
n
Tf -
0) CM
0 C-
« o
.£ .2
" "
g.
0. 1
marsh grass shrin
daggerblade grasi
Palaemonetes vulgar!:
Palaemonetes pugio
*
CN
o
V
ON
o
V
J2
1
T3
O
f
Ampelisca abdita
CO CS
00 01
•* co
o o
CM CS
o —
ON ON
0 0
00 0»
^- co
0 0
oo ON
0 O
*3
1
«> *fl3
11
M *
c ^
windowpane flou
fourspine stickleb
Scopthalmus aquosus
Apeltes quadracus
.-
VO
*— (
ON
o
VO
VO
1— <
o\
o
a>
1
i
Homarus americanus
o ,j.
s
»— < '
f-
ON O
V
un
VO
t-
ON O
V
i
^3
1 ^
s- •?
sand shrimp
blue crab
c
Crangon septemspino.
Callinectes sapidus
00
CO
*
2
cn
«?
(S
o
1
s
Atlantic menhade
Brevoortia tyrannus
t>
>n
V
«n
V
j>
S
eastern oyster
Crassostrea virginica
VO
"n
0)
*n
o»
^3
3
1
Stenotomus chrysops
•n
vo
0
'
c~-
vo
0
o5
ID
1
S
Americamysis bahia
V
ON
^
O)
CO
ON
f-
0«
CO
o
1
summer flounder
Paralichthys dentatus
CO
0
vo
*
00
CO
0
Of
VO
oo
CO
—
a
winter flounder
a
Pleuronectes americai
cs —
CO t^
01 «
* »-•«
S? S
S £:
% °!
S S
V O
n
j ^
! -I
Morone saxatilis
Syngnathusfuscus
TJ- 00 t-
VO CO 0»
•^ « oi
13
-------
1.8 ,
I.O -.
1.64
1.6.
1.4.
_J
I 1.2-
| 1.0.
O
0>
|0.8.
1 0.6.
o
W
V)
° 0.4 .
0.2.
00
(
\
— i
)
N.Rsh
o Invertebrates
o.
0 .
0 0
O • o O
o
0 •
o
0
o
10 20 30 40 50 60 70 80 90 100
5o/0 % Rank of GMAV
Figure 2. Plot of low dissolved oxygen effect (Genus Mean Acute Values for
LCSOs) against percentile rank of each value in the data set Values for each genera
are listed in Table 1. Results from individual tests for each species are listed in Ap-
pendix E. The value highlighted on the y-axis is the calculated Final Acute Value
(FAV). This value is the LC50 that is higher than the values for 95% of the tested
genera. The line drawn through the four most sensitive genera is the line of best fit
for those four values. The LC50 values for the four most sensitive genera are the
only values used in the FAV calculation other than the total number ("n") of values.
Growth Effects
A threshold above which long-term, continuous exposures to low DO should not
cause unacceptable effects was calculated with growth data (mostly from bioassays using
larvae). Sub-lethal effects were evaluated with only growth data for two reasons. First,
growth is generally more sensitive to low DO than survival. There were only two excep-
tions where survival was more sensitive to low DO than growth. One test was with D.
sayi, however, growth was the more sensitive endpoint in eight other tests with this spe-
cies (Appendix C). The results from this one test were not included in Table 2. The other
exception was a 28-day early life stage test using the Atlantic silverside Menida menidia
(Appendix C). There was no effect at 4.8 mg/L DO, but there was 40% mortality and a
24% reduction in growth at a DO concentration of 3.9 mg/L. This 24% reduction in
14
-------
I
,
•§
H
II
B
b
I
re
•»•»
cc
e
&
03
t- r-
°5 *°.
—•" cs
A
00
c~i ci tn
V
ij- o vo •* „
o\ o o\ vo °°.
~* en o oo |> oo
en en -^ en "S cs
•-« cs ^^ en
^ ¥ S S -o
cs m en n *—«
co
-------
growth, however, was not statistically significant. There was essentially no growth of
surviving M menidia at a DO concentration of 2.8 mg/L. Only the growth data were
summarized in Table 2.
The second reason for restricting sub-lethal effects to growth is that results are
available from only one saltwater test that measured reproductive effects. Data are pre-
sented in Appendix C from a 28-day life cycle test using the mysid Americamysis bahia.
Although growth was reduced 25% at 3.17 mg/L and was technically the most sensitive
endpoint in this test, the percentage reduction in growth was essentially the same at 2.76
and 2.17 mg/L as it was at 3.17 mg/L (20% and 27%, respectively). Reproduction was
reduced by 76% at 2.17 mg/L, the first treatment that resulted in a significant effect on
this endpoint. Although this test suggests that growth is more sensitive than reproduction,
there are insufficient data to confirm this conclusion for saltwater species. Data from two
standardized freshwater tests, however, indicate that growth is more sensitive than repro-
duction for both fathead minnows (Brungs, 1971) and Daphnia magna (Homer and Wal-
ler, 1983). Thus, DO limits that protect against growth effects also may be protective for
reproductive effects.
Data on the affects of hypoxia on growth are presented for four species offish and
seven species of invertebrates from a total of 36 tests. The sensitivity of growth to low
DO has been determined in only two standard 28-day tests which meet Guidelines re-
quirements; the above life cycle test with A. bahia and the above early life stage test with
M. menidia. Therefore, growth data from non-standard tests (i.e., not life cycle, partial
life cycle or early life stage tests) were used to augment the chronic database. These non-
standard tests ranged from 4 to 29 days long. Data from short duration tests were in-
cluded because effects of oxygen deprivation are assumed to be instantaneous. Oxygen is
required continuously for the efficient production of cellular energy. Therefore, even
modest reductions in DO may result in the redirection of energy use from growth to com-
pensatory mechanisms. In addition, data from larval growth of two bivalves (Morrison,
1971; Wang and Widdows, 1991) and several fish and crustaceans (Appendix C) show
that chronic values for DO do not change substantially for exposures ranging from a few
days to several weeks for most of the species tested. The Mercenaria mercenaria (Morri-
son, 1981) andMytilis edulis (Wang and Widdows, 1991) studies show that the effect on
larval bivalve growth within the same test run is the same over a series of days (13 days
forM mercenaria and 6 to 10 days forM edulis).
Overall Genus Mean Chronic Values (GMCVs) for effects on growth range from
> 1.97 for the sheepshead minnow Cyprinodon variegatus to 4.67 mg/L for the longnose
spider crab Labinia dubia; a ratio of < 2.4. Three of the most sensitive species were
crustaceans (Figure 3; Table 2). The range of chronic values for the four most sensitive
genera is 3.97 to 4.67 mg/L; a ratio of only 1.2. The Final Chronic Value (FCV) was cal-
culated in the same manner as the FAV (Stephan, et al., 1985). Because acutely resistant
taxa are under-represented hi the chronic database in Table 2, it could be argued that n,
the number of genera used in the calculation of the FCV, should be increased from 11 to
a higher value. We chose to increase n from 11 to 22 (the n for the FAV). This is the
same procedure that was used for the FCV in the ambient water quality.criteria for cad-
16
-------
4.8 ' 1
4.50 .
4.00 .
|> 3.50 .
0 3.00 .
e>
g 2.50 .
D)
8 2.00 .
13
§1
o 1.50.
in
° 1.00 -
0.50 .
0.00
(
V
c
)
^^ .Rsh
0 o Invertebrates
0 0
o 9
0
9
10 20 30 40 50 60 70 80 90 100
jo/o % Rank of GMAV
Figure 3. Plot of low dissolved oxygen effect (Genus Mean Chronic Values for
growth) against percentile rank of each value in the data set. Percentile rank was ad-
justed based on the total "n" from the acute data set (see text for explanation). Specific
values for each genus included are listed in Table 2. Results from individual tests for
each species are listed in Appendix C. The value highlighted on the y-axis is the calcu-
lated Final Chronic Value (FCV). This value is the chronic value that is higher than the
values for 95% of the species represented. The line drawn through the four most sensi-
tive values is the line of best fit for those four values. The chronic values for the four
most sensitive genera are the only values used in the FCV calculation other than the
total number ('n") of values.
mium7 (U.S. EPA, 1985). The final protective value for growth (the Criterion Continuous
Concentration or CCC) is 4.8 mg DO per liter, but would increase only to 5.0 mg/L if n
was kept at 11.
As presented here, the CCC is intended as a time-independent value. Areas where
the average minimum DO does not fall below 4.8 mg/L should have sufficient DO to
support the survival and growth of most aquatic species in the Virginian Province. Al-
though it is generally accepted that reduced growth means reduced overall fitness, there is
7 One assumption underlying the calculation procedure for FAVs and FCVs is that the sample of values
available is representative of the population of values in the community being protected. If the dataset is
too heavily weighted with values from the sensitive end of the distribution, then this skews the interpreta-
tion of the 5th percentile value that is calculated.
17
-------
1
^
^f
g
o
c
§>
1
10
o
"o
10
V)
b
3.5
3.0
2.5.
2.0.
1.5.
1.0.
0.5.
0.0
D. say/-megalopa
Larvae-96 hr
D
~ n ^
A u A • D. say/
Larvae-24 hr A ,
D /
A D /
A °
A y
0 ' A
*"°*o AAA
Juveniles * o 0 A A
o • • o 0
* •
• 5
0 • o
o
o
0 10 20 30 40 50 60 70 80 90 100
% Rank
Figure 4. Plot of the GMAV data from Figure 2 (circles) along with 24 hr (triangles)
and 96 hr (squares) LC50 values for larval life stages of various saltwater animals. The
open symbols are for invertebrates and the closed for fish. The open' square and triangle
for D. sayi represent the mean response for all larval life stages for this species. The
dashed line at top represents the LC50 for D. sayi exposed during the transition to
megalopa. The data for the juveniles are from Table 1. The data for the larvae are listed
in Appendix D.
little direct evidence for this in the field. In one study, Gleason and Bengtson (1996a,b)
found that for some estuarine fish bigger is not necessarily better. Bigger fish (as prey)
may be more susceptible to being eaten by predators. As an alternative to the growth
criterion, a criterion that addresses chronic stresses from long-term or short-term expo-
sures to low DC can be based on larval recruitment effects.
Larval Recruitment Effects
A generic model has been developed that evaluates the cumulative effects of acute
and chronic stresses on early life stages of aquatic organisms. Early life history informa-
tion and exposure-response relationships are integrated with duration and intensity of ex-
posure to provide an ecologically relevant measure of larval recruitment. There are ex-
isting recruitment models for marine organisms (e.g., Ricker, 1954; Beverton and Holt,
1957). However, these models address other processes such as parental stock size, popu-
lation fecundity, and density dependent processes such as cannibalism and intraspecific
competition. These existing models therefore are not appropriate for the-needs of the DO
document, which requires incorporation of abiotic stressor effects.
18
-------
Larvae are more acutely sensitive to low DO than juveniles (Figure 4); however,
the criteria are not being established to protect larvae and juveniles in the same manner.
A method is needed that estimates how many days a given DO concentration can be tol-
erated without causing unacceptable effects on total larval survival for the entire recruit-
ment season. This is accomplished with a generic larval recruitment8 model and applying
biological and hypoxic effects parameters for the Say mud crab (D. sayi). Parameters for
this larval crustacean are used for several reasons. Larval crustaceans are among the most
acutely and chronically sensitive larval saltwater animals, and the Say mud crab's late
larval to megalopa period is the most sensitive of the tested crustaceans (Table 2 and Fig-
ure 4). Among larvae at risk in estuaries, considerable information is available on Say
mud crab with respect to the biological parameters in the model. Laboratory responses of
D. sayi are indicative of a species the most at risk from hypoxia because it has a high DO
response threshold. In addition, these larvae are present in the lower water column coin-
cident with the expected hypoxia season present throughout the Virginian Province in
salinities >15 ppt, which strengthens the choice of this species for a Province-wide
model.
The model and the major assumptions used during its development are presented
in Appendix E. The life history parameters hi the model include only those mat relate
specifically to larvae: larval development time, larval season, attrition rate and vertical
distribution. The recruitment model assumes that the period of low DO occurs within the
larval season. The magnitude of effects on recruitment, defined as the cumulative number
of successful transitions to megalopae, is influenced by each of the four life history pa-
rameters. For instance, larval development time establishes the number of cohorts that
entirely or partially co-occur with the interval of low DO stress. The second parameter,
the length of the larval season, is a function of the spawning period, and also influences
the relative number of cohorts which fall within the window of hypoxic stress. The third
life history variable, natural attrition rate, gages the impact of slower growth and devel-
opment of the larvae in response to low DO by tracking the associated increase in natural
mortality (e.g., predation). The model assumes a constant rate of attrition, so increased
residence time in the water column due to delayed development translates directly to de-
creased recruitment. Finally, the vertical distribution of larvae in the water column de-
termines the percentage of larvae that would be exposed to reduced DO under stratified
conditions. Three exposure response curves that describe megalopa survival, zoea larval
survival, and molt delay versus DO concentration are used for estimating recruitment un-
der hypoxic conditions9. The model makes a simplifying assumption that hypoxic days
are contiguous. The model can be applied either to establish protective conditions or to
evaluate the severity of a given hypoxic condition.
The dose-response data used in the model in this document are presented in Fig-
ure 5. Figure 5A is a summary response curve for exposures that included a transition
from zoea to megalopa. These tests were necessarily longer (7 to 11 days) than other lar-
val tests to allow sufficient time for development to megalopa. Although some of the en-
Once the larvae are "recruited" into the juvenile life stage, the juvenile protective limit established above
is applied.
9 The model is designed to allow both biological and exposure-response data to be changed based on the
availability of appropriate data.
19
-------
hanced sensitivity in these tests may be due to the longer exposures to low DO, mortality
also appeared to be associated with the molt to megalopa1 . The model assumes a con-
stant rate of reproductive output per day, and a constant rate of development during the
larval season. Therefore, some larvae in the plankton would be molting to this stage
daily, and it is at this point that the crab larvae may be particularly sensitive to low DO.
The model assumes that the response of the late larvae in transition to megalopae could
occur following a single day of exposure (i.e., this response is independent of exposure
prior to the day of transition). Thus, the model applies this dose response as a 24 hr expo-
sure.
Figure 5B is a summary response curve for 24 hr exposures of zoea stage larvae.
Figure 5C shows data that suggest a delay in development time for D. sayi in going from
a stage 3 zoea to megalopa. However, the degree of developmental delay was difficult to
measure with sufficient resolution. Further, it was difficult to distinguish it from differ-
ential survival sensitivity among individuals within a replicate. Thus, the model has been
run with and without a delayed development effect. The results of these two runs are
shown in Figure 6. Points on the graph show which combinations of low DO concentra-
tion and exposure duration result in a seasonal reduction of recruitment that does not ex-
ceed 5%. Until further information is available, the output used to establish the criteria
for larval recruitment will be the one that assumes no delayed development (the solid line
in Figure 6).
The equation for the larval line (as well as the lines in Figure 5A and 5B) was de-
rived by an iterative process of fitting the best line through the points generated by the
output of the recruitment model. The equation is a standard mathematical expression for'
inhibited growth (logistic function—Bittinger and Morrel, 1993). This equation is:
P L
P(t) = r£ Equation 1
U P0+e-Lkt(L-P0)
For Figure 6, P(t) is the DO concentration at time t, PQ is the y-intercept, and L is the up-
per DO limit L was set as the DO concentration that allowed a 44-day exposure (the
maximum exposure period the model allowed using the current parameters—see appen-
dix page E-3 for further explanation). PO was first estimated by eye from the original plot,
and then adjusted higher or lower to minimize the residuals between the real recruitment
data and that estimated from the mathematical fit of the data. The rate constant, k, was
similarly empirically derived. For Figures 5A and 5B, the variables t and L represent DO
concentration and the upper limit for survival (100%), respectively.
10 Data for another crustacean, Cancer irroratus (rock crab), also lend some support for having separate
dose response curves for the zoea and megalopa larval life stages (Appendix F).
20
-------
Zoea Transition to Megalopa
V)
Po = 0.10
L = 100
k = 0.0208
1
100
90.
80 .
70
60
50
40 -I
30
20.
10.
0
B. Zoea Larvae
0.0
Po = 0.01
L = 100
k = 0.0489
0
1.4,
1.2
| S 1.0
S I 0.8
O K_
I = 0.6 .
2 S
> 1 0.4 .
^ 0.2.
C. Delayed Development
y=-0.0785x +1.3931
R? = 0.851
0 1 2 34 5 6
Dissolved Oxygen (mg/L)
Figure 5. Dose response curves for Say mud crab (Dyspanopeus sayi) used in
the larval recruitment model. Open symbols are the data from tests with con-
tinuous low dissolved oxygen exposures. Solid lines are the regression lines of
best fit. See text for explanation of Po, L and k. A: dose response curve for zoea
transition to megalopa. These data are from exposures durations greater than 24
hr but are applied as 24 hr exposures in the model (see text for explanation). B:
dose response curve for zoea larvae. Data are from 24 hr exposures. C: data for
delayed development of larvae to megalopae
21
-------
Application of Persistent Exposure Criteria
The final criteria for saltwater animals in the Virginian Province (Cape Cod to
Cape Hatteras) are indicated in Figure 7 for the case of continuous (i.e., persistent) expo-
sure to low dissolved oxygen. The most uncertainty with the application of these limits
usually will be when DO conditions are between the juvenile survival and larval growth
limits. Below the juvenile survival limit, DO conditions do not meet protective goals.
Above the growth limit, conditions are likely to be sufficient to protect most aquatic life
and its uses. Interpretation of acceptable hypoxic conditions when the DO values are
between the juvenile survival and larval growth limits depends on the characterization of
the duration of the hypoxia. To determine whether a given site has a low DO problem,
adequate monitoring data are required. The more frequently DO is measured the better
will be the estimate of biological effects.
Figure 8 is a hypothetical time series for average daily dissolved oxygen minima.
The portion of the data below the CCC is all that is considered. This area of the graph is
first divided into several intervals. We recommend using no finer than 0.5 mg/L DO in-
tervals because of limitations on most monitoring programs (see Implementation section).
However, larger intervals may be necessary if monitoring data are not taken frequently
enough. The resulting intervals in our example are (a) below 4.8 mg/L and above 4.3
mg/L, (b) below 4.3 and above 3.8, and so forth for intervals 'c' and 'd'. For each inter-
val, the number of days is recorded that the DO is between the interval's limits. For ex-
ample, in interval 'a' the DO is below 4.8 mg/L and above 4.3 mg/L from July 13th
through the 18th and again from July 23rd through the 25th, for a total of seven days. This
number of days is then expressed as a fraction of the total number of days that would be
allowed for the DO rninimum for each interval. For interval 'a', the allowed number of
days is 24 (using Figure 6 at 4.3 mg/L). Table 3 lists the information for all four intervals
from this hypothetical time series. The fractions of allowed days are totaled. If the sum is
greater than one, then the DO conditions do not meet the desired protective goal for larval
survival. If the sum is less than one (as is the case in our example), then the protective
goal has been met. This procedure uses a simplifying assumption that each interval is in-
dependent. That is, there is no increased risk to recruitment due to pre-exposure to hy-
poxia. This assumption is supported by the similarity of larval survival data for 24 and 96
hr exposures in Appendix A.
The current recruitment model is a first attempt at providing a method that incor-
porates duration of exposure in the derivation of DO criteria. A model that could inte-
grate gradual change in daily DO concentrations is desirable. However, the current model
may be adequate given the probable inaccuracies in assessments of DO conditions in
coastal waters (Summers, et al., 1997).
22
-------
I
5.5
5.0
4.5
4.0-,
3.5
3.0-
2.5-
2.0-
1.5
1.0
0.5-I
0.0
Recruitment Model Output at 5% Protection
with delayed development •
without delayed development
Po=2.15
L = 4.45
k = 0.036
5 10 15 20 25 30 35 40
Exposure Time (days)
45
Figure 6. Plot of model output that protects against greater than 5% cumulative im-
pairment of recruitment. Input parameters were the same for two runs of the model,
except for the inclusion of the delayed development response (Figure 5C), open sym-
bols, or the exclusion of molt delay, closed symbols, The solid line is the regression
line of best fit for the closed symbols. The area below the line represents conditions
of potential impairment. See text for explanation of Po, Land k.
55
5.0 .
4.5.
31 4.0 .
ra
S 3.5 .
g, 3.0 .
0 2.5.
•a
> 2.0 .
1 1
-------
01
c
§,
>,
£
•a
I
o
(0
tn
a
5.8 4v
5.3
4.8
4.3
3.8
3.3
2.8
2.3
ccc
o>
<
CM
<. <.
in o
T- CM
IO
CM
O)
^
O
CO
Figure 8. A hypothetical representative dissolved oxygen time series for one site. The
horizontal line represents the CCC of 4.8 mg/L. The portion of the curve below 4.8
mg/L is divided into four arbitrary intervals (a,b,c,d) to estimate effects on larval re-
cruitment. The dissolved oxygen minimum, and the duration for each interval are
Table 3. Dissolved oxygen and duration data from a hypothetical persistent time series (Figure 8). The
Below and Above columns show the range of D.O. covered by each interval. Number of Days Within
Range refers to the duration that the observed D.O. is between the range given. In the last column this
duration is expressed as a fraction of the number of days allowed by the recruitment model (Figure 6) for
the D.O. minimum of the interval. These fractions are totaled to evaluate whether the larval survival
protective goal has been met.
Range (mg/L)
Interval Below Above
a 4.8
b 4.3
c 3.8
d 3.3
4.3
3.8
3.3
2.8
No. Days
Within Range
7
3
1
1
No. Days
Allowed
24
13
7
4
Fraction of
Allowed
0.29
0.23
0.14
0.25
TOTAL
0.91
24
-------
Less Than 24 hr Episodic and Cyclic Exposure to Low Dissolved Oxygen
The criteria for continuous exposure to low dissolved oxygen do not cover expo-
sures times less than 24 hr. This section addresses this topic by describing the available
data and how they were used to evaluate the effect of low DO on exposure durations
lasting less than 24 hr. These included one-time episodic events, as well as either tidal- or
diel-influenced cycles where the DO concentrations cycle above and below the continu-
ous CCC. The approaches described for treatment of non-constant (e.g., cyclic) condi-
tions are intended to provide protective goals that are equivalent to those established for
persistent conditions. The data used come from two types of experiments. The first are
those which provide time-to-death (TTD) data and are used to derive TTD curves. The
second are experiments in which there were treatments consisting of a constant exposure
to a' given low DO concentration paired with a treatment in which the DO concentration
cycled between that low concentration and a concentration near saturation (or at least
well above concentrations that should cause significant effects). The data from both of
these experiments are discussed below.
Cyclic Juvenile and Adult Survival
The persistent hypoxic criterion for juveniles and adults is 2.3 mg/L. A conserva-
tive estimate of the safe DO concentration for exposures less than 24 hr would be to sim-
ply use 2.3 mg/L. However, time-to-death data indicate that this would be over protec-
tive. Data are available for two saltwater juvenile fish (Brevootia tyrannus and Leiosto-
mus xanthurus), one freshwater juvenile fish (Salvelinus fontinalis), and three larval salt-
water crustaceans (D. sayi, Palaemonetes vulgaris and Homarus americanus), providing
a total of 33 TTD curves (Appendix G). The curves represent a range of test conditions,
including acclimation to hypoxia with S. fontinalis, and a range of lethal endpoints. Two
general observations were made from this data. First, each curve can be modeled with the
same mathematical expression, a logarithmic regression, of the form:
Y=m(lnX)+b Equation 2.
where X=time, Y=DO concentration, m=slope and b=intercept where the line
crosses the Y-axis at X=l.
Second, the shape of the curve (i.e., the slope and intercept) was governed by the sensi-
tivity of the endpoint. This is true whether the sensitivity increase was due to interspecific
differences (including saltwater and freshwater species) or the use of different endpoint
(e.g., LC5 is more sensitive endpoint than LC50).
Figure 9 shows the relationship between sensitivity (i.e., 24 hr LC values) and the
slope (Figure 9A) and the intercept (Figure 9B) for all 33 TTD curves (Appendix G). The
DO value from each TTD curve at 24 hr was used as a measure of sensitivity. Plots using
other time intervals could have been used. The value at 24 hr was chosen in order to gen-
erate a curve for juveniles that meets the constant CMC at its 24 hr value (2.3 mg/L). The
slope and intercept for a time-to-CMC curve were calculated using Figure 9 equations
and the CMC 24 hr value of 2.3 mg/L. These were then used as the parameters in Equa-
tion 2 to generate a criterion for saltwater juvenile animals for exposures less than 24 hr
(Figure 10).
25
-------
0.6
0.5.
0.4 .
y=0.191x- 0.064
R2 = 0.835
Dissolved Oxygen Concentration
Causing Effect Observed at 24 hr (mg /L)
2.0 -
1.8 .
1.6.
1.4.
1.2.
1.0 .
0.8.
0.6.
0.4.
0.2.
0.0
B
y = 0.392X + 0.204
R2 = 0.678
Dissolved Oxygen Concentration
Causing Effect Observed at 24 hr (mg /L)
Figure 9. Slope (A) and intercept (B) versus low D.O. effect values at 24 hr from
time-to-death (TTD) curves for two species of saltwater juvenile fish, one species of
juvenile freshwater fish and three species of saltwater larval crustaceans. Data used
mostly represent LT50 curves, but values for other mortality curves are included.
Species used and their associated TTD curves are presented in Appendix G. All TTD
curves were fit with a logarithmic regression.
26
-------
2.5,
2.0 .
«£ 1.5 J
0)
I
•a
I
o
10
to
5 0.5 J
1.0.
0.0
Time-to-"CMC"
= 0.370Ln(x) +1.095
8 12 16
Exposure Time (hr)
20
24
Figure 10. Criterion for juvenile saltwater animals exposed to low dissolved oxygen
for 24 hr or less. The line represents the same protective limit as the CMC for juve-
niles for continuous exposure. The line is a logarithmic expression with a slope and
intercept calculated from the regressions in Figure 9 at the dissolved oxygen con-
centration of 2.3 mg/L (the CMC).
Cyclic Growth Effects
The CCC for continuous exposure was derived based on growth effects data (Ta-
ble 2). The simplest way to determine effects from cyclic exposure to low DO is to com-
pare growth of organisms under cyclic conditions to those for the same species under
continuous conditions. Growth data are available from cyclic exposures to low DO for
three species of saltwater animals, D. sayi, P. vulgaris and Paralichthys dentatus (Coiro,
et al, 1999). These data are listed in Appendix H and summarized in Figure 11. Data are
from experiments in which a low DO treatment was paired with a treatment cycling be-
tween the same low DO concentration and one that was above the continuous CCC (usu-
ally saturation). All cyclic treatments had 12 hr of low DO within any one 24 hr period.
Most of the cycles consisted of 6 hr at the low concentration followed by 6 hr at the high
concentration. Only two tests (both with P. vulgaris) were conducted using a 12hr:12hr
cycle. There were a total of 20 paired treatments spread among the three species.
. As expected, at the end of each test, cyclic exposures generally resulted in more
growth than constant exposures to the minimum DO of the cycle (Figure 11). However, if
the effects of DO on growth were instantaneous (i.e., growth reduction begins as soon as
the DO concentration drops and growth rate returns to normal as soon as DO returns to
above CCC concentrations), then the cyclic exposures in the above experiments would
have been expected to cause one half of the growth reduction observed in the constant
treatment of each pair. (As noted above, the DO cycles had a total of 12 hr of low DO per
day.) If this were true, then the slope of the line in Figure 11 would be 0.5. However, the
slope of the line for the data (forced through the origin) is 0.778, a factor of 1.56 greater.
27
-------
Percentage Reduction in Growth Relative to Control
y = 0.778x
= 0.815
0
y = 0.5x
10 20 30 40 50 60 70 80 90
Constant Exposure
Figure 11. Plot of test results from growth experiments pairing constant low dis-
solved oxygen exposure with exposures to various cycles of low dissolved oxygen
and concentrations above the CCC. The dark line is a linear regression of the data
with the line forced through the origin. The lighter weight line is the "expected" rela-
tionship from a slope of 0.5 (see text for explanation). Species used and the experi-
mental conditions are listed in Appendix H.
Thus greater growth impairment occurs from cyclic exposures than expected. One hy-
pothesis for this discrepancy is that recovery from the low DO portion of the cycle is not
instantaneous, and the actual low DO effect period is then greater than 12 hr within each
day (by a factor of 1.56)11.
Figure 12 shows a dose-response for growth of larval Say mud crab (D. sayi) over
a range of constant DO concentrations12. The data are from ten tests (see Appendices C
and H) with durations ranging from 4 to 11 days. The percentage growth reduction is
relative to a control response. Growth reduction effects are considered instantaneous,
therefore the % reduction can be applied to any time period. Data for this mud crab are
emphasized, because it was the only sensitive species tested in cyclic exposures. In addi-
11 The data used to establish the relationship between cyclic and constant exposures (Figure 11) came from
experiments with a total low DO exposure of 12 hr per 24 hr period. We assume that as the total time of
exposure per 24 hr decreases the discrepancy between expected and observed should also decrease. Thus
the 12-hr data can be considered a worst case for any daily cycle of 12 hr or less exposure to low DO.
There is insufficient information for cycles with greater than 12 hr exposure periods per day. We recom-
mend assuming constant exposure conditions for these latter situations.
12 The relative sensitivity of Say mud crab growth to low DO versus other species tested is shown in Ap-
pendix I.
28
-------
c:
g
"o
T3
&
.c
1
0
SS
100 .. .
90 .
80 .
70 -
60 -
50.
40 -
30 -
20 .
10 -
0
• Constant Exposure
* y = -19.4x + 116.2
**»
• • t
jf 9 ™
"*'., •
•• v
• •
^
* ***.
v--.
•
0.0 0.5 1.0
1.5 2.0 2.5 3.0 3.5
Dissolvsd Oxygen (mg/L)
4.0 4.5 5.0
Figure 12. Plot of dose-response data for growth reduction in Say mud crab (Dys-
panopeus sayi) exposed to various continuous low dissolved oxygen concentrations.
Percentage growth reduction is relative to a control. The dashed line is a linear re-
gression through the data points.
tion, this species is used to represent larval crustaceans in the recruitment model for con-
stant exposures.
To evaluate a cycle for chronic growth effects, the above relationship between cy-
clic and constant exposure is needed as well as monitoring data from a representative, or
worst case, cycle of low DO for a given site. Figure 13 provides a hypothetical DO time
series. To estimate the expected growth reduction during this cycle the curve is divided
into three DO intervals13 for that portion of the cycle that falls below 4.8 mg/L (the
CCC). The DO mean, and the total duration that the cycle is within the interval's range of
DO, are determined for each interval. Data from this example are presented in Table 4.
Interval 'c' lasts a total of five hours. Interval 'b' lasts a total of three hours (bl before
plus b2 after interval 'c'). Similarly, interval 'a' lasts for a total of four and a half hours.
Each of these time intervals is multiplied by 1.56 to adjust for the cyclic effect.
A DO mean concentration for each interval is used with the equation from Figure
12 to estimate a daily growth reduction that is expected for larval crustaceans during con-
stant exposure to hypoxia. This value is then normalized for the interval's cyclic adjusted
duration. The normalized reductions for all intervals are added (growth effects are cu-
mulative) for an estimated growth reduction for the cycle. This reduction is compared to
the reduction estimated to occur at the CCC for constant exposures (23%, using the
13Any number of intervals can be chosen, even one. For simplicity, different DO ranges can be selected for
each interval so that each interval has approximately the same total time below the CCC. Alternatively, the
cycle can be divided by selecting a constant DO range (e.g., 0.5 mg/L), giving each interval a different time
value. Monitoring data, however, must be frequent enough to justify the chosen interval size.
29
-------
equation from Figure 12 at 4.8 mg/L DO). The percentage reduction in our example is
34%. This reduction is greater than the maximum allowed by the CCC, thus our hypo-
thetical cyclic hypoxic event does not meet the protective goal for growth.
"oi
^,
c
D)
o
•o
§
0
to
CO
Q
6.5
6.0 .
5.5.
5.0.
4.5.
4.0 .
3.5.
3.0.
2.5.
2.0
•
.
>—
a1
t
__
b1
4
^
*•.••[""
b2
c
^—•»
,
a2
.*•*.
•
m ccc
12:00 15:00 18:00 21:00 0:00 3:00 6:00 9:00 12:00 15:00 18:00
Time (hr)
Figure 13. A hypothetical representative dissolved oxygen time series for one cycle.
The horizontal line represents the CCC of 4.8 mg/L. The portion of the curve below
4.8 mg/L is divided into three arbitrary intervals (a,b,c) to estimate effects on growtii.
The range of dissolved oxygen, the mean dissolved oxygen and the duration for each
interval are listed in Table 4.
Table 4. Dissolved oxygen and duration data from a hypothetical cyclic time series (Figure 13).
These data are used to estimate the growth reduction occurring for the recruitment modeled species
during the cycle. Percentage reductions in growth for constant exposure are calculated with the
equation in Figure 12. These in turn are normalized for the cyclic adjusted duration.
Interval
al-a2
bl-b2
c
D.O. Range
(mg/L)
4.8-4.0
4.0-3.5
3.2-3.5
D.O. Mean
(mg/L)
4.40
3.75
3.35
% Daily
Reduction
in Growth
31
43
51
Actual
Duration (hr)
4.5
3
5
Cyclic Ad-
justed Dura-
tion (hr)
7.0
4.7
7.8
% Reduction
for Duration
9
8
17
Total % Reduction for Cycle
34
30
-------
Cyclic Larval Recruitment Effects
In order to evaluate cyclic" exposures for their potential impact on larval recruit-
ment to the juvenile life stage two pieces of information are needed. First, a set of larval
crustacean time-to-death curves to estimate the expected daily mortality for a given low
DO cyclic exposure. Second, a way to translate that predicted daily larval mortality into
allowable days for the given low DO cycle using the constant exposure recruitment
model output. Creation of the larval TTD curves is straightforward using the sensitivity
information (dose response curve) for the Say mud crab late larval to megalopa transition
period in Figure 5A14 and the sensitivity dependent relationships for TTD slopes and in-
tercepts in Figure 9. Creation of a series of larval TTD curves followed the same proce-
dure used to create the time-to-CMC curve for juveniles (Figure 10). Figure 14 shows the
results for nine calculated curves for mortalities ranging from 5 to 95%.
Estimating the daily mortality expected to occur with the model species also is
straightforward, and as with cyclic growth protection, requires representative or worst
case DO monitoring data. Figure 15 is a hypothetical monitoring data set for a single cy-
cle. As with growth, the portion of the cycle below the CCC is first divided into several
intervals. The DO minimum is determined for each interval. It should not matter how the
intervals are selected. All that is needed is a set of paired time and DO values. Table 5
lists the data for the intervals in this example. These data were plotted among the family
of larval TTD curves (Figure 16). In the example, the greatest effect datum lies closest to
the 10% mortality curve. Therefore, the hypothetical cycle of DO is expected to cause
10% daily mortality to the modeled larval crustacean. We are only concerned with the
greatest effect datum because survival effects are not cumulative (i.e., an individual can
only die once).
Now all that is needed is to translate the expected 10% mortality into the number
of allowable days for this hypothetical cycle to occur. This is accomplished using the fit-
ted curves in Figures 5A and 6. Figure 5A is the dose response curve for the Say mud
crab late larval transition to megalopa period used in the recruitment model. The infor-
mation in the figure is for percentage survival, but it can be converted easily into percent-
age mortality. Thus the information shows the expected cohort mortality to occur for a
given DO concentration. For the example, 10% mortality occurs at a DO concentration of
4.4 mg/L. From the equation used to fit the data in Figure 6, the 4.4 mg/L is allowed to
occur for up to 26 days without significant impairment to seasonal recruitment. Thus, the
cycle that resulted in an estimated 10% daily mortality to larval crustaceans can be re-
peated for up to 26 consecutive days without exceeding a 5% reduction in seasonal larval
recruitment to the megalopa life stage. All of the above can be simplified by merging the
information from Figures 5A and 6 into one cyclic translator figure using the DO axis
that is common between Figures 5A and 6. This is shown in Figure 17.
4T"he late larval to megalopa dose-response curve was selected because it is the most sensitive curve used
in the recruitment model.
31
-------
5.0 1
4.5.
4.0 .
3.5.
3.0.
2.5.
2.0 .
1.5.
1.0 .
0.5.
0.0
5
10
15
25
2 4
i 10 12 14 16 18 20 22
Time to Death (hr)
24
Figure 14. Time-to-death (TTD) curves generated for the recruitment model species.
Data to generate the curves were taken from Figures 5A, 9A and 9B. The numbers ad-
jacent to each TTD curve are the percentage mortality that each curve represents. The
dashed lines represent curves created with slopes and intercepts outside the range of the
original data used in Figure 9.
I
I
s
in
5
6.5,
6.0 .
5.5 .
4.5 .
4.0.
3.5 .
3.0 .
2.5.
2.0 .
12:00
ccc
15:00 18:00 21:00 0:00 3:00 6:00 9:00 12:00 15:00 18:00
Figure 15. The same hypothetical dissolved oxygen time series as Figure 13. This time
the portion of the curve below 4.8 mg/L is divided into several arbitrary intervals to es-
timate effects on mortality. The dissolved oxygen minimum and its duration for each
interval are listed in Table 5.
32
-------
Table 5. Dissolved oxygen and duration data from the intervals selected from the hypothetical
cyclic time series in Figure 15. These data are plotted in Figure 16 to estimate the expected
mortality occurring for recruitment modeled species during the cycle.
Interval
a
b
c
d
e
D.O. Minimum for Inter-
val (mg/L)
4.2
3.9
3.6
3.4
3.2
Duration of Interval
Off)
12
10
8
6
4
1
5.0
4.5
4.0
3.5
3.0 .
f
1 2.0
1
U)
5 1.5
1.0
0.5
0.0
5
10
15
25
0 24 6 8 10 12 14 16 18 20 22
Time to Death (hr)
24
Figure 16. The dissolved oxygen minima and the durations listed in Table 5 superim-
posed on Figure 14 (solid circles). The expected mortality from the cyclic exposure is
determined by the data point falling closest to a TTD curve of greatest effect, in this
case 10% mortality.
33
-------
Cyclic Translator
70
65
60
55
c
> 50
O
I ! I
UNACCEPTABLE
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
Daily Cohort Mortality (%)
Figure 17. A plot that combines the information from Figures 5A and 6 into a single
cyclic translator to convert expected daily mortality from cyclic exposures into allow-
able number of days of those cycles.
34
-------
Other Laboratory Bioassay Data
Additional available data on lethal and sublethal effects of hypoxia on saltwater
animals (Appendix J) do not indicate significantly greater sensitivity than indicated pre-
viously. The other data are divided into effects on juveniles and adults, and effects on
larvae. Figure 18 shows all of the juvenile mortality data from Appendix J plotted against
the criteria for juvenile and adult survival (limits for both persistent and cyclic exposures
are included). Most of the other survival data are well below the criteria. There are three
notable exceptions. The first is a single datum (LC50 of 1.9 mg/L) for the Atlantic men-
haden Brevoortia tyrannus at 6 hr (Voyer and Hennekey, 1972). However, several other
LC50 values (Burton et al., 1980) for Atlantic menhaden with durations ranging from 2 to
72 hr were much less (0.70 to 0.96 mg/L). The second is a single datum for the Atlantic
silverside M. menidia at 6 hr (also Voyer and Hennekey, 1972). There are no other data
for juvenile Atlantic silversides, but the unusually high sensitivities reported by Voyer
and Hennekey for the other species suggest that their exposure system might be a con-1
founding factor. In addition, the authors provided no information on control response for
either the Atlantic menhaden or the Atlantic silversides.
"S>
d
d
o
o
•s
£.
£
2.5
2.0 .
1.5 .
1.0 .
0.5 .
0.0
M. menidia
B. tyrannus ^ 0
E. affinis
o o
o w o
0.1
10 100
Time to Hfect (hr)
O O O
1000
Figure 18. A plot of the other juvenile/adult mortality data from Appendix J (open
symbols) along with the proposed dissolved oxygen criteria for juvenile/adult survival
(solid line).
35
-------
The third set of data above the criteria is a series of values at 0.5 hr for the cope-
pod Eurytemora affinis. Some are below the criteria, but many are above it (Vargo and
Sastry, 1978). However, the authors did not give any details on their experimental meth-
ods, including the number of replicates, the number of animals in each replicate, or on the
response in the control. Thus, it is difficult to adequately access the significance of these
results. However, in the absence of data to the contrary, it is worth noting the DO limit
for juveniles and adults may not be protective of copepods. Alternatively, one could con-
sider that short-lived species with high reproductive outputs (such as copepods) may be
more appropriately protected in a manner similar to larval recruitment. In this case all of
the E. affinis LC50 values would fall below the criterion provided by the larval recruit-
ment (see explanation for Figure 19A below).
Figures 19A and 19B present all of the lethality data from Appendix J for tests
using larval life stages. All of these data are from tests for effects on individuals, and the
criterion for larval survival acknowledges that some larval mortality is acceptable. Most
of the data for larvae are LC50 values for exposure durations other than 24 or 96 hr (these
two durations are used elsewhere in the document). The LC50 data are plotted in Figure
19A. The most appropriate protective limit to compare these values with is the time-to-
death (TTD) curve for 50% mortality for the Say mud crab (from Figure 14), because the
larval survival protective limit is based on data for this species. There are two series of
data points for LC50 values for larval rock crab Cancer irroratus for exposure durations
of two and four hours; each has some values above the 50% TTD curve (Vargo and Sas-
try, 1977). The more sensitive values in these sets are for tests run at 25°C, thus the ani-
mals were likely exposed to multiple stressors (temperature and low DO).
The rest of the other lethality data for larvae are plotted in Figure 19B. These data
are separated into three categories, LC5 toLC35, LC40 to LC65, and LC90 to LC100. As
with the LC50 values in Figure 19B, these values are plotted along with time-to-death
curves (10, 50 and 90% mortality) for late larval Say mud crab (From Figure 14). All of
the LC5 to LC35 values are at or below the 10% TTD curve. All of the LC40 to LC65
values are well below the 50% TTD curve. Finally, all but one of the LC90 to LCI00
values are below the 90% TTD curve. This one value is for 100% mortality of stripped
bass larvae, M. saxatilis that occurred after a 2 hr exposure to 1.90 mg/L DO. However,
there are two other stripped bass tests were 100% mortality of the larvae did not occur
until 24 hr of exposure to similar low DO.
The are fewer other data on sublethal effects than for lethality effects (Appendix
J). The sublethal effects included reduced feeding, growth, locomotion, and bivalve set-
tlement, as well as delays in hatching and molting. However, none of these values indi-
cate that the CCC would not be protective against these effects.
36
-------
6.0 ,
5.0.
4.0 .
en
E
O
Q,
6
J
13
I
3.0 .
2.0 .
1.0 .
0.0
0.01
0.10 1.00
Time to Effect (days)
10.00
100.00
6.0
5.0.
~ 4.0.
_j
^>
E
El. 3.0 .
o
O
•c
I
2.0.
1.0
0.0.
B
C. bosguianus
5% mortality
M. saxitalis
100% mortality
F. heteroclitus
10% embryo
mortality
0.010
10%TTD
50% TTD
90%TTD
0.100 1.000 10.000
Time to Effect (days)
100.000
Figure 19. A plot of the other larval survival data from Appendix J.
Figure 19A presents the available LC50 data (open circles) along
with the 50% time-to-death (TTD) curve for the Say mud crab. Fig-
ure 19B presents mortality data for other than 50%. Open circles
represent 5 to 35% mortality, open squares 40 to 65% mortality, and
closed circles 90 to 100% mortality. Figure 19B also includes the
10, 50 and 90% TTD curves for the Say mud crab.
37
-------
Laboratory Observed Behavioral Effects of Hypoxia
A number of laboratory studies report behavioral alterations following exposure
to hypoxia. The effects include low DO avoidance, changes in locomotion, burrowing
and feeding activity, and altered predator-prey behaviors. Most of the effects observed
occurred <2.3 mg/L, hence would be protected by even the 24 hr acute limit CMC. The
most hypoxia-sensitive behavioral effect occurs in red hake (Urophycis chuss). In red
hake, age 0+ fish leave their preferred bottom habitat and begin to swim continuously as
DO concentrations fall below 4.2 mg/L (Bejda et al., 1987). Food search time is also re-
duced as a consequence. Below 1.0 mg/L, most locomotor and other behavioral activity
ceases, and at 0.4 mg/L there is loss of equilibrium. Older red hake, (age 1+ and 2-3+),
did not exhibit these responses with low DO, except for loss of equilibrium at 0.6 mg/L.
The following effects are reported at less than the 2.3 mg/L protective limit. In the
red morph of green crabs (Carcinus maenas) the low DO avoidance EC25 was <2.3 mg/L
and the EC50 was 1.8 (Reid and Aldrich, 1989). The green morph was less sensitive. In
naked goby (Gobiosoma bosc) larvae, avoidance at 2.0 mg/L occurred with > one hr ex-
posure (Breitburg, 1994). No avoidance was observed at 3.0 mg/L. This same author re-
ported 100% avoidance in larval bay anchovy (Anchoa mitchilli) at 0.75 mg/L following
a one hr exposure. Reduced locomotor activity occurred in daggerblade grass shrimp (P.
pugid) at 1.8 mg/L (Hutcheson et al., 1985). Burrowing in the northern quahog (M. mer-
cenarid) was reduced 1.4 to 2 fold when exposed to 1.8 to 0.8 mg/L and slowed 4 fold in
Atlantic surfclam (Spisula solidissimd) at 1.4 mg/L (Savage, 1976). The polychaete,
Nereis virens, EC25 for emergence from the sediment was 0.9 mg/L (Vismann, 1990).
The shelter guarding and nest guarding behavior by adult male naked goby (G. bosc) was
not altered at 0.7 mg/L, but they abandoned shelters at 0.38 mg/L and nests at 0.3 mg/L.
Death occurred in these animals at 0.26-0.24 mg/L (Breitburg, 1992).
The following low DO effects on feeding are reported in a bivalve and four poly-
chaetes. In eastern oyster (Crassostrea virginicd) early post-settlement stage (436 um
mean shell height), exposure to 1.9 mg/L for 6 hr resulted in 54 to 61% reduction in
feeding rate; at <0.4 mg/L for the same period, 86 to 99 % reduction occurred (Baker and
Mann, 1994b). In older post-settlement animals (651 um mean shell height), feeding rate
was not altered with 1.9 mg/L exposure for 6 hr, but at < 0.4 mg/L it was reduced 97 to
99%. In the polychaetes, feeding stopped in Nereis diversicolor at 1.2 mg/L and in N. vi-
rens at 0.9 mg/L (Vismann, 1990). In adult Loimia medusa, feeding stopped at 1.0 mg/L
during <20 hr exposure, then resumed in 42 to 113 hr in 42% of the animals (Llanso and
Diaz, 1994). At 0.5 mg/L, there was no resumption of feeding after initially ceasing dur-
ing the same initial exposure period. Following exposure in Streblospio benedicti adults,
the initial response to 1.0 mg/L was cessation of feeding, but it resumed in 3.5 days; with
0.5 mg/L exposure, the initial response was the same, with feeding resuming in 4.5 days
(Llanso, 1991).
Changes were observed in predator-prey activities in two fishes in low DO In na-
ked goby (G. bosc) larvae, avoidance of the sea nettle (Chyrsaora quinquecirrhd) preda-
tor was reduced 60% following 3 hr exposure to 2.0 mg/L. In striped bass (M. saxatilis)
juveniles, predation on naked goby larvae was reduced 50% following 1-hr 35 min expo-
sure to 2.0 mg/L (Breitburg et al., 1994).
38
-------
Observed Field Effects
Field reports of the biological consequences of hypoxia could be used to derive
DO criteria if they include information to describe the exposure conditions. Yet sufficient
data are rarely available. In most cases, DO conditions prior to observed effects are un-
known, making it difficult to predict an exposure threshold for the observed effect. A
field report of hypoxic effects must, at a minimum, provide a description of the concur-
rent DO exposure conditions if it is to be useful in deriving criteria. Ten studies in the
Virginian Province have provided concurrent DO measurements. The DO observations
often are only point measurements, not continuous records, and they rarely provide in-
formation on DO conditions prior to the observed effects. The biological effects reported
include alterations in the following: presence of fish and crustaceans, diel vertical migra-
tion of copepods, recruitment and population density of an oyster reef fish (naked goby),
recruitment and growth of eastern oyster spat, and macrobenthic community parameters.
Effects were usually not observed above 2 mg/L. Exceptions are the Long Island Sound
trawl studies, where effects were reported in the 2.0 to 3.7 mg/L range.
The relationship between low DO and presence of fish and shellfish in Long Is-
land Sound was examined in two trawl studies. Howell and Simpson (1994) reported
marked declines in abundance and diversity in 15 of 18 study species when DO was be-
low 2 mg/L. When DO was between 2 and 3 mg/L, there were significantly reduced
abundances of three species: winter flounder, windowpane flounder and butterfish. In a
subsequent three year study, the aggregate data for 23 species of demersal finfish showed
a decline for two community indices, total biomass and species richness, with declining
DO (Simpson et al., 1995). The DO concentration that corresponded with a 5% decline
below a response asymptote was 3.7 mg/L for total biomass and 3.5 mg/L for species
richness. Dissolved oxygen declines below these concentrations resulted in further exclu-
sion of these animals, which has implications for the secondary productivity of these wa-
ters. Reduced species number implies reduction of community resilience, should this
condition persist. The consequences of habitat crowding on animals occurring in adjacent
waters is unknown.
Hypoxia-induced changes in the distribution of fish and crustaceans have also
been reported in the lower York River, located in the Virginian portion of Chesapeake
Bay (Pihl et al., 1991). Subpycnocline DO <2 mg/L developed during neap tide periods
and the study species (spot, croaker, hogchoker, blue crab, and mantis shrimp) migrated
to shallower and better oxygenated habitats. The degree and order of vertical movement
was believed to be a function of the water column DO concentration and species sensi-
tivity to hypoxia, i.e. croaker > spot = blue crab > hogchoker « mantis shrimp. Water
column destratification and reaeration occurred with spring tide or strong winds and all
species except the burrowing mantis shrimp returned to the deeper strata, indicating a
preference for the deeper habitats.
Diel vertical migration of copepods Acartia tonsa and Oifhona colcarva is dis-
rupted by hypoxia (Roman et al., 1993). In mid-Chesapeake Bay during the summer,
these copepods typically occur near the bottom during the day and migrate to the surface
waters at night. However, when DO concentrations fell below 1 mg/L in subpycnocline
waters, the copepods were displaced to the pycnocline, where the highest numbers were
39
-------
found both day and night. When mixing occurred during the summer, the bottom waters
were reaerated, and the copepods once again were found at depth during the day. Vertical
migration is believed adaptive in that it places the copepods in the chlorophyll maximum
at night to maximize food intake, yet it provides day-time avoidance of the surface wa-
ters, protecting the copepods from visual feeding bay anchovy.
The consequences of hypoxia on recruitment were examined for two species at a
mid-Chesapeake Bay site: the naked goby G. bosc, a benthic oyster reef fish, (Breitburg,
1992), and Eastern oyster C. virginica (Osman and Abbe, 1994). In the naked goby study,
low DO episodes were short-lived, but extreme (<0.5 mg/L), the result of movement of
deep, oxygen-depleted bottom water into the near shore reef habitat. Following each se-
vere intrusion, the naked goby population density fell dramatically at the deeper stations,
which experienced the lowest DO (0.4 mg/L). Small, newly recruited, juveniles were ab-
sent, presumably due to extremely high mortality. There is evidence, based on observed
densities, that older juveniles and adults survived these events by temporarily moving to
inshore portions of the reef where DO was not as low, then return during the weeks fol-
lowing the event. Embryonic development was also affected. Males abandoned egg-
containing tubes placed at deeper sites, and the majority to all of the embryos were dead.
In addition, the youngest embryos collected from the shallower, less hypoxia-stressed site
developed abnormalities following laboratory incubation. The severe intrusions occurred
during peak periods of recruitment, with the lowest DO occurring on portions of the reef
where recruitment was expected to be highest. These adverse effects were not observed at
sites experiencing low DO >0.7 mg/L.
In the study with the eastern oyster C. virginica (Osman and Abbe, 1994), mor-
tality was observed in newly-set (2 to 4 days old) animals during periods of prolonged
intrusions of low DO water (<1 mg/L 40% of the time in bottom water during the first
two weeks of two experiments). Mortality was proportional with depth, which corre-
sponded to severity of hypoxia. Growth rate of surviving spat decreased after 1, 2, and 4
weeks following deployment, with a greater effect also occurring at the deeper stations.
Survival and growth of juvenile oysters were unaffected following simultaneous deploy-
ment at the same stations, indicating greater tolerance of the older animals. The authors
concluded hypoxia to be a plausible causative factor, acting directly or indirectly, al-
though other causative factors also are possible.
Responses of the macrobenthic community to DO < 2 mg/L are reported for the
lower Chesapeake Bay and tributaries (Dauer and Ranasinghe,1992; Diaz et al., 1992;
Llanso,1992; Pihl, et al.,1991, 1992). Two community effects are reduced species num-
ber and abundance, with these effects increasing spatially and temporally with increasing
severity and duration of hypoxia. There also is a shift with hypoxia from dominance of
longer-lived, deeper burrowing species of a mature community to short-lived, shallow
burrowing opportunistic species. The response of benthic species, and their subsequent
recoveries following hypoxia, depends on species tolerance, the timing of the hypoxic
event relative to larval availability and settlement, and life history strategy. Some infau-
nal organisms migrate towards the sediment surface with hypoxia, beginning around 2
mg/L (Diaz et al.,1992). Animals that migrate to the surface are exposed to predation by
hypoxia-tolerant fish and crustaceans (Pihl et al., 1992). Defaunation may only occur
40
-------
below 1 mg/L. These studies support 2 mg/L as the hypoxic effect threshold for the mac-
robenthos, which is consistent with the global literature (Diaz and Rosenberg, 1995).
To summarize, demersal finfish community biomass has been observed to dimin-
ish at DO <3.7 mg/L, and species richness to diminish at <3.5. These effects become in-
creasingly pronounced with further DO decline. Below 2.0 mg/L, migration of the infau-
nal species to the sediment surface and movement of epifaunal species to better aerated
water were observed. All effects reported at <1 mg/L DO concern hypoxia-tolerant spe-
cies and life stages (i.e. disruption of die! vertical migration in copepods, reduced growth
and survival of newly settled oysters, and lethality in larval goby) as demonstrated in par-
allel laboratory studies (Breitburg, 1992, Roman et al., 1993) or by other workers (Baker
and Mann, 1992 and 1994a).
Data not used
Data from a variety of published literature were not used. The literature on effects
of anoxia was not used, as it provides negligible information on threshold requirements of
aerobic animals. Information on anoxic effects may be found in a recent symposium (Ty-
son and Pearson, 1991) and a review (Diaz and Rosenberg, 1995) of this subject. Results
of hypoxia effects studies were not cited for species which do not commonly occur in
coastal and estuarine waters between southern Cape Cod, MA and Cape Hatteras, NC
during the spring to autumn period which brackets the occurrence of hypoxia. Reports for
occasional visitor species that occur in these waters during a favorably warm or cold
summer were excluded.
Data were not cited if the test temperature was outside the temperature range of
Virginian Province waters during the hypoxic season, e.g. American lobsters tested at 5
°C (McLeese, 1956). Data were not used if they are probably not reliable. Examples in-
clude indications that the test animals may have been stressed, e.g. American lobster
tested at 25 °C which were not fed during a 8-10 week acclimation period (McLeese,
1956); excessive control mortality (> 10% for juveniles or adults and > 20% for early life
stages); the DO exposure concentration was uncertain, whether due to questionable DO
measurements or failure to directly measure test chamber DO conditions (e.g. Reish,
1966); or if test animals were removed and handled during the test to make other meas-
urements, e.g. for an energetics study (Das and Stickle, 1993). Literature on physiological
responses of animals to hypoxia was reviewed, but was not found useful to determine low
DO effect thresholds. See Herreid (1980) for a discussion of difficulties in using oxygen
consumption results to describe DO requirements of invertebrates. Rombough (1988b)
has developed an approach to identify the DO requirements for fish embryos and larvae,
but this approach has not been employed with species applicable to Virginian Province
saltwaters.
Some data are not used for juvenile blue crabs, C. sapidus (Stickle, 1988; Stickle
et al., 1989). Effect concentrations for this species from this laboratory are an order of
magnitude higher than values from an earlier study using adult C. sapidus (Carpenter and
Cargo, 1957). In addition, these effect concentrations for juvenile blue crabs are almost
all higher than values for larvae of all tested species. Another study (DeFur et al., 1990)
showed that adult C. sapidus make respiratory adjustments that allow them to tolerate
41
-------
long-term (25 days at 22 °C) exposure to 2.6 to 2.8 mg DO/L. These data for juvenile
blue crabs are considered outliers until further testing shows otherwise.
Just prior to final completion of this document, a paper appeared (Secor and Gun-
derson, 1998) describing the effects of hypoxia and temperature on juvenile Atlantic
sturgeon, Acipenser oxyrinchus. There was 22% mortality at 19 °C and an average within
tank DO concentration of 2.7 mg/L (within tank data provided by author). This sensitivity
is not that different from that of stripped bass. However, a combination of low DO (ca.
3.5 mg/L) and high temperature (26 °C) resulted in 100% mortality of A. oxyrinchus
within approximately 24 hr. Because the greatest sensitivity was associated with the high
temperature the data were not included in this document. In addition, the salinity during
the experiments only ranged between one and three ppt, therefore it is likely that this data
is more appropriately associated with freshwater criteria which are much higher than
those for saltwater (see Implementation section).
National Criteria
The national criteria for ambient dissolved oxygen for the protection of saltwater
aquatic life from Cape Cod to Cape Hatteras are summarized in Table 6 and presented
graphically on Figure 20 (for persistent exposure) and Figure 21 (for episodic and cyclic
exposure). These criteria are briefly described below:
(1) Protection of Juvenile and Adult Survival from Persistent Exposure
This limit is derived following the Guidelines procedures and is analogous the criterion
maximum concentration (CMC), except that a protective DO concentration limit is ex-
pressed as a minimum as opposed to a maximum, as would be the case for a toxicant.
This limit represents the floor below which dissolved oxygen conditions (for periods of >
24 hours) must not occur. Shorter durations of acceptable exposure to conditions less
than the CMC have been derived from laboratory studies, as described in (4) below.
Please refer to Table 1 for a detailed explanation of the derivation of this limit.
(2) Protection of Growth Effects from Persistent Exposure
This limit is derived following the Guidelines procedures and is analogous to the criterion
continuous concentration (CCC) for a toxicant. This limit represents the ceiling above
which dissolved oxygen conditions should support both the survival and growth of most
aquatic species from Cape Cod to Cape Hatteras. Please refer to Table 2 for a detailed
explanation of the derivation of this limit. This limit may be replaced with a limit derived
in (3) as described below, when exposure data are adequate to derive an allowable num-
ber of days of persistent exposure.
(3) Protection of Larval Recruitment Effects from Persistent Exposure
This limit is derived from a generic larval recruitment model using data for the Say mud
crab, a sensitive species native to the waters from Cape Cod to Cape Hatteras. It provides
a degree of protection equivalent to the CCC described above in (2). The limit represents
allowable dissolved oxygen conditions below the CCC, provided the exposure duration
does not exceed a corresponding allowable number of days that assure adequate recruit-
ment during the recruitment season. The cumulative effects of all exposure interval dura-
42
-------
tions at a given DO below the CCC can be accounted for by totaling the fractions of the
actual (or projected) exposure duration (in days) divided by the allowable exposure dura-
tion for each interval of a specific DO concentration. Please refer to Table 3 and Figure 6
of this document for a detailed explanation of the derivation of this limit.
(4) Protection of Juvenile and Adult Survival from Episodic or Cyclic Exposure
This time dependent limit was derived to represent the responses of the most sensitive ju-
veniles tested in the laboratory. It provides an equivalent degree of protection as the
CMC, but for shorter exposure durations than a day. It is assumed that adults are no more
sensitive than juveniles. This limit represents the minimum dissolved oxygen conditions
that must be maintained on an hourly basis (e.g., one-hour minimum, two-hour minimum,
etc.). The limit applies to conditions occurring on a single given day; even if this limit is
met, recurring exposure patterns still must be checked for agreement with the larval re-
cruitment limit described in (6) below. Please refer to Figure 10 of this document for a
detailed explanation of the derivation of this limit.
(5) Protection of Growth Effects from Episodic or Cyclic Exposure
This limit is derived from the dose-response relationship for DO vs. growth reduction for
the Say mud crab, and comparisons of the effects of cyclic exposure versus constant ex-
posure on growth for a variety of species. It provides an equivalent degree of protection
as the CCC, but for shorter exposure durations than a day. The limit represents the DO
conditions that maintains a daily percent growth reduction in Say mud crab not greater
than the level provided at the CCC for whole day exposures (23%). The cumulative ef-
fects of all exposure interval durations at a given DO below the CCC are accounted for
by summing the percent reductions for time intervals at representative D.O. concentra-
tions. An adjustment factor of 1.56 was derived to estimate time-variable effects from
intermittent exposure tests that indicated residual, or delayed recovery effects from vari-
ous growth-inhibiting conditions. The limit applies to conditions that may occur as a re-
curring pattern throughout the year without adverse growth effects at the CCC level of
protection. However, a recurring pattern of exposure may be limited for a certain number
of days based on the larval recruitment limit (6). Recurring patterns of DO conditions that
do not meet the growth limit may be allowed for a limited number of days in a recruit-
ment season, provided the larval recruitment limit is met according to (6). Please refer to
Table 4 and Figure 12 of this document for a detailed explanation of the derivation of this
growth limit. The larval recruitment limit can be substituted in whole for the growth
limit.
(6) Protection of Larval Recruitment Effects from Episodic or Cyclic Exposure
This limit is derived from the modeled relationships between daily cohort mortality for
the Say mud crab and the allowable number of days at a given maximum daily cohort
mortality that protects against greater than 5% cumulative impairment of recruitment
over a season. It provides an equivalent degree of protection as the limits described in (3)
above, but for recurring patterns of low DO as opposed to continuous low DO conditions.
Figure 16 of this document illustrates how to determine the maximum daily cohort mor-
tality from duration intervals of DO minima. Figure 17 of this document illustrates how
to determine the allowable number of days of cyclic exposure for a given maximum daily
43
-------
cohort mortality. This limit provides additional information that should be used in con-
junction with the limits described in (4) and (5) above. The limit determines the number
of days that recurring episodic or cyclic conditions may occur, including whether the
pattern may occur for an unlimited number of days. For example, a cyclic pattern that in-
cludes a DO minimum of 3.6 mg/L for 8 hours results in a daily cohort mortality of 10%
(see Figure 16). Assuming this represents the maximum daily cohort mortality for the cy-
clic pattern, the allowable number of days for the cyclic exposure is 26 (see Figure 17).
Please refer to pages 31-34 of this document for a detailed explanation of the derivation
of this limit.
In summary, limits (1) and (4) establish one day and hourly minimum conditions that
should be maintained for persistent and cyclic exposures, respectively; limits (3) and (6)
establish conditions that may occur for a limited number of days for persistent and cyclic
exposures, respectively; and limits (2) and (5) establish long term conditions that should
be maintained for the remaining number of days for persistent and cyclic exposures, re-
spectively.
Implementation
Dissolved oxygen criteria should be implemented differently from those of toxicants, but
not for reasons associated with biological effects or exposure. Uncertainties associated with
aquatic effects of DO, such as behavior, synergistic relationships with temperature, salinity, or
toxics, apply to toxics as well. Dissolved oxygen also does not differ from toxics for reasons as-
sociated with exposure. Dissolved oxygen can vary greatly in the environment, but so can toxics.
Effluents and their receiving waters can vary daily, even hourly, in their toxicity to aquatic life.
Toxicity of saltwater receiving waters also can vary with the tide and the depth of water. It may
be mistakenly perceived that DO varies more in concentration simply because it can be measured
easily and nearly continuously.
From the standpoint of environmental management, DO differs from toxic compounds
primary because it is not regulated directly. Hypoxia is a symptom of a problem; not a direct
problem. Dissolved oxygen is regulated primarily by controlling discharges of nutrients (in the
marine environment, most commonly nitrogen). Dissolved oxygen also differs from most toxic
compounds because hypoxia can have a large natural component. Therefore, criteria for hypoxia
should not automatically be applied in the same way as limits for toxicants are.
This document provides the information necessary for environmental planners and regu-
lators in the Virginian Province to address the question of whether DO at a given site is sufficient
to protect coastal or estuarine aquatic life. The document does not address how compensatory
mechanisms such as avoidance can influence the response of local populations to seemingly ad-
verse DO conditions. The document also does not address the issue of spatial extent of a DO
problem. In other words, even if the DO at a site is low enough to significantly affect aquatic
life, the environmental manager will have to judge whether the hypoxia is widespread enough for
concern.
44
-------
Table 6. Summary of Saltwater Dissolved Oxygen Criteria.
Persistent Exposure
(24 hour or greater continuous low DO
conditions)
Episodic and Cyclic Exposure
(less that 24 hour duration of low DO
conditions)
Juvenile and Adult
Survival
(minimum
allowable
conditions)
(1) a limit for continuous exposure
DO = 2.3 (mg/L)
(criterion minimum concentration,
CMC)
(4) a limit based on the hourly duration of
exposure.
DO = 0.37*ln(t) +1.095
-where:
DO = allowable concentration (mg/L)
t = exposure duration (hours)
Growth Effects
(maximum
conditions required)
(2) a limit for continuous exposure
DO = 4.8 (mg/L)
(criterion continuous concentration,
CCC)
(5) a limit based on the intensity and hourly
duration of exposure.
Cumulative cyclic adjusted percent daily
reduction hi growth must not exceed 23%.
ti*\56*Gredi
24
< 23%
and
Gredi = -19.4*DOi + 116.2
-where:
Gred; = growth reduction (%)
DO; = allowable concentration (mg/L)
t; = exposure interval duration (hours)
i = exposure interval
Larval Recruitment
Effects1
(specific allowable
conditions)
(3) a limit based on the number of days
a continuous exposure can occur
Cumulative fraction of allowable days
above a given daily mean DO must not
exceed 1.0
n .(actual)
Z*.
7
i */
.(allowed)
and
DO, =
9.57
(2.15 + 2.3eai6ti)
•where:
DO; = allowable concentration (mg/L)
t; = exposure interval duration (days)
i = exposure interval
(6) a limit based on the number of days an
intensity and hourly duration pattern of
exposure can occur.
Maximum daily cohort mortality for any
hourly duration interval of a DO minimum
must not exceed a corresponding allowable
days of occurrence.
•where:
Allowable number of days is a function of
maximum daily cohort mortality (%).
Maximum daily cohort mortality (%) is a
function of DO minimum for any exposure
interval (mg/L) and the duration of the
interval (hours).
1 model integrating growth and survival effects to maintain a minimally impaired Say mud crab larval population
45
-------
2 .
»»«•«•>•«*«»•»»»»*»»>••••« »•«•«!»»>•••»»>»•
. _ . Larval Recruitment
—»—Juvenile Survival
10 20 30 40
Exposure Time (days)
Figure 20. Summary of Criteria for Persistent Exposure. The larval recruitment line represents the minimum DO
concentration that may persist for a given exposure interval duration (number of days). The cumulative effect of
multiple intervals during a season must be accounted for as described in (3) above and in the equation provided on
Table 6.
Growth
_ .Larval Recruitment
+ Juvenile Survival
4 8 12 16 20 24
Exposure Time (hours)
Figure 21. Summary of Criteria for Episodic and Cyclic Exposure. The growth line represents the minimum DO
concentration that may persist for a given exposure interval duration (i.e., the exposure duration/DO concentration
that results in a 23% daily growth reduction). The cumulative effect of multiple intervals during the course of a day
must be accounted for as described in (5) above and in the equation provided on Table 6. The larval recruitment
line represents the hourly exposure duration/DO concentration intervals that may recur for an unlimited number of
days (corresponds to the 8% daily cohort mortality as shown on Figure 17).
46
-------
Finally, as with all criteria, this document does not address changes in sensitivity to low
DO that accompany other stresses such as high temperature, extremes of salinity, or toxi-
cants. Chief among these concerns would be high temperature because high temperature
and low DO often appear together. Low DO will be more lethal at water temperatures
approaching the upper thermal limit for species. This effect has been seen for freshwater
species (U.S. EPA, 1986; Secor and Gunderson, 1998), and saltwater species (e.g., C. ir-
roratus and E. affinis). The limits provided here should be sufficient under most condi-
tions where aquatic organisms are not otherwise unduly stressed.
Many programs that monitor coastal DO with electronic equipment cannot meas-
ure DO to better than 0.5 mg/L due to limitations of instrument accuracy and resolution
(e.g., Strobel, et al., 1995; Strobel and Heltshe, 1999) or sampling design (Summers, et
al., 1997). Attempts to refine the limits presented here or to apply these limits in assess-
ing field DO conditions should take this into account. Criteria for DO can be used in a
risk assessment framework. The approach outlined in this document can be easily used to
compare among areas the DO conditions that are adequate to support aquatic life. Envi-
ronmental managers can determine which sites need the most attention, and evaluate the
spatial and temporal extent of hypoxic problems from one year to the next for sites of
major concern.
Environmental managers who wish to use the protective approach presented here
will have to decide several questions about how the limits will be used, four of which are
described below.
1. Accuracy of monitoring data—The most important decision is to determine
how accurate the monitoring data are—the better that hypoxia is character-
ized, the more reliably it can be decided whether it meets the criteria. Data
from existing monitoring programs may not always be accurate enough to
take full advantage of the approach provided here. For example, a recent as-
sessment of conventional sampling procedures along the Atlantic and Gulf
coasts has suggested that hypoxia in estuarine waters is substantially more
widespread that previously believed (Summers, et al., 1997). Deciding what
data can adequately characterize hypoxia is a matter of risk management. Cy-
clic conditions may require measurements every 30 min for several days,
whereas persistent hypoxia may need only several measurement a week. Deci-
sions also have to be made about the number and locations of sampling sites
to properly represent a given area.
2. Biological effects—Potential biological effects are most difficult to predict
when DO lies between the limits for juvenile and adult survival and larval
growth. Concentrations below the juvenile and adult limit do not protect; con-
centrations above the limit for growth probably protect most aquatic life and
its uses15. Deciding whether concentrations between the limits are acceptable
will depend on the duration of hypoxia and on the acceptable impairment of
larval recruitment. The acceptable impairment can be a risk-management de-
15 The larval growth protection limit is based on statistically significant differences that result in chronic
values similar to EC25s for growth of many organisms. EC25 values are listed as a part of Appendix C for
four species of crustaceans and two species offish. The geometric mean of these values (by species) corre-
lates with the geometric mean of the chronic values.
47
-------
cision. The 5% impairment level was selected to be consistent with the pro-
tection provided to juvenile and adult life stages. In addition, a model that in-
tegrates gradual change in daily DO conditions may more accurately predict
recruitment effects than the current simplified model and its application.
3. Spatial extent—After environmental managers have found a hypoxic area,
they must decide whether it is small enough relative to nearby unaffected .ar-
eas to allow the coastal region as a whole to meet the criteria.
4. Freshwater versus saltwater—It is not trivial to decide whether the DO in
certain parts of estuaries should be judged by freshwater criteria or saltwater
criteria, particularly where the tides vary the salinity between near fresh and a
few parts per thousand. This decision is important because the criteria for
freshwater can be up to twice as great as the saltwater limits developed here,
depending on water temperature and the life stage being protected (U.S. EPA,
1986). A reasonable way to start is by considering an estuary's biological
communities. If they are more like freshwater organisms, freshwater criteria
should be applied. If they are more like saltwater, then saltwater criteria apply.
5. Threatened or endangered species—In cases where a threatened or endan-
gered species occurs at a site, and sufficient data exists to suggest that it is
more sensitive at concentrations below the criteria, it is appropriate to con-
sider development of a site-specific criterion.
48
-------
References
Armstrong, R.S. 1979. Bottom oxygen and stratification in 1976 and previous years, pp.
137-148. (in) Swanson, R.L. and CJ. Sindermann (eds). Oxygen Depletion and Asso-
ciated Benthic Mortalities in New York Bight, 1976. NOAA Professional Paper 11.
U.S. Dept. of Commerce, Washington, D.C.
Baker, S.M. and R. Mann. 1992. Effects of hypoxia and anoxia on larval settlement, ju-
venile growth, and juvenile survival of the oyster Crassostrea virginica. Biol. Bull.
182:265-269.
Baker, S.M. and R. Mann. 1994a. Description of metamorphic phases in the oyster Cras-
sostrea virginica and effects of hypoxia on metamorphosis. Mar. Ecol. Prog. Ser.
104:91-99.
Baker, S.M. and R. Manri. 1994b. Feeding ability during settlement and metamorphosis
in the oyster Crassostrea virginica (Gmelin, 1791) and the effects of hypoxia on post-
settlement ingestion rates. J. Exp. Mar. Biol. Ecol. 181:239-253.
Bejda, A.J., A.L. Studholme and B.L. Olla. 1987. Behavorial responses of red hake, Uro-
phycis chuss, to decreasing concentrations of dissolved oxygen. Environ. Biol. Fishes.
19:261-268.
Beverton, RJ.H. and S.J. Holt. 1957. On the dynamics of exploited fish populations. U.K.
Min. Agric. Fish., Fish. Invest. (Ser 2) 19:533 pp.
Bittinger, MX. and B.B. Morrel. 1993. Applied Calculus. 3rd ed. Addison-Wesley Pub.
Reading, MA. 818 pp.
Breitburg, D.L. 1990. Near-shore hypoxia in the Chesapeake Bay: Patterns and relation-
ships among physical factors. Esutarine, Coastal and Shelf Sci. 30:593-609.
Breitburg, D.L. 1992. Episodic hypoxia in Chesapeake Bay: Interacting effects of re-
cruitment, behavior, and physical disturbance. Ecol. Monogr. 62:525-546.
Breitburg, D.L. 1994. Behavioral response offish larvae to low dissolved oxygen con-
centrations in a stratified water column. Mar. Biol. 120:615-625.
Breitburg, D.L., N. Steinberg, S. DuBeau, C. Cooksey and E.D. Houde. 1994. Effects of
low dissolved oxygen on predation on estuarine fish larvae. Mar. Ecol. Prog. Ser.
104:235-246.
Brungs, W.A. 1971. Chronic effects of low dissolved oxygen concentrations on fathead
minnow (Pimephalespromelas). J. Fish. Res. Bd. Canada. 28:1119-1123.
Burton, D.T., L.B. Richardson and CJ. Moore. 1980. Effect of oxygen reduction rate and
constant low dissolved oxygen concentrations on two estuarine fish. Trans. Amer.
Fish. Soc. 109:552-557.
Carpenter, J.H. and D.G. Cargo. 1957. Oxygen requirement and mortality of the blue crab
in the Chesapeake Bay. Technical Report XHI. Chesapeake Bay Institute, The Johns
Hopkins University.
Chesney, E.J. and E.D. Houde. 1989. Laboratory studies on the effect of hypoxic waters
on the survival of eggs and yolk-sac larvae of the bay anchovy, Anchoa mitchilli.
Chapter 9. pp. 184-191. (in). E.D. Houde, E.J. Chesney, T.A. Newberger, A.V. Vaz-
49
-------
quez, C.E. Zastrow, L.G. Morin, H.R. Harvey and J.W. Gooch. Population Biology of
Bay Anchocy in Mid-Chesapeake Bay. Maryland Sea Grant Final Report.
Coiro, L.L., S.L. Poucher, and D.C. Miller. 1999. Hypoxic effects on growth ofPalae-
monetes vulgaris larvae: Using constant exposure data to estimate cyclic exposure re-
sponse. Memorandum to Glen Thursby on draft document. AED contribution number
2066. January 20.
Das, T. and W.B. Stickle. 1993. Sensitivity of crabs Callinectes sapidus and C. similis
and the gastropod Stramonita haemastoma to hypoxia and anoxia. Mar. Ecol. Prog.
Ser. 98:263-274.
Dauer, D.M. and J.A. Ranasinghe. 1992. Effects of low dissolved oxygen events on the
macrobenthos of the lower Chesapeake Bay. Estuaries. 15:384-391.
D'Avanzo, C. and J.N. Kremer. 1994. Diel oxygen dynmaics and anoxic events in an
eutrophic estuary of Waquoit Bay, Massachusetts. Estuaries 17:131-139.
Davis, R.M. and B.P. Bradley. 1990. Potential for adaptation of the estuarine copepod
Eurytemora affinis to chlorine-produced oxidant residuals, high temperature, and low
oxygen, (in) R. L. Jolley et al., (eds) Water Chlorination: Chemistry, Environmental
Impact and Health Effects. Vol. 6. pp. 453-461. Lewis, Boca Raton, FL.
DeFur, P.L., C.P. Mangum and J.E. Reese. 1990. Respiratory responses of the blue crab
Callinectes sapidus to long-term hypoxia. Biol. Bull. 178:46-54.
De Silva, C.D. and P. Tytler. 1973. The influence of reduced environmental oxygen on
the metabolism and survival of herring and plaice larvae. Netherlands J. Sea Res.
7:345-362.
Diaz, R.J., R.J. Neubauer, L.C. Schaffher, L. Pihl and S.P. Baden. 1992. Continuous
monitoring of dissolved oxygen in an estuary experiencing periodic hypoxia and the
effect of hypoxia on macrobenthos and fish. Sci. Total Environ. Supplement. 1992.
pp. 1055-1068.
Diaz, RJ. and R. Rosenberg. 1995. Marine benthic hypoxia: A review of its ecological
effects and the behavioural responses of benthic macrofauna. Oceanography and Ma-
rine Biology: an Annual Review. 33:245-303.
Gleason, T.R. and D.A. Bengtson. 1996a. Growth, survival and size-selective predation
mortality of larval and juvenile inland silversides, Menidia beryllina (Pisces: Ather-
inidae). J. Exp. Mar. Biol. Ecol. 199:165-177.
Gleason, T.R. and D.A. Bengtson. 1996b. Size-selective mortality of inland silversides:
Evidence from otolith microstructure. Trans. Am. Fish. Soc. 125:860-873.
Gleason, T. and W. Munns. 1997. Response to questions concerning the recruitment
model developed for the dissolved oxygen report. Memorandum to Don Miller, At-
lantic Ecology Division, U.S. Environmental Protection Agency, Narragansett, Rhode
Island 02882. April 4.
Haas, L.W. 1977. The effect of spring-neap tidal cycle on the vertical salinity structure of
the James, York, and Rappahannock rivers, Virginia, USA. Estuarine, Coastal Shelf
Sci. 5:485-496.
50
-------
Hammen, C.S. 1976. Respiratory adaptations: Invertebrates, pp, 347-355. (in) M. Wiley
(ed). Estuarine Processes. Vol. 1. Uses, Stresses, and Adaptations to the Estuary.
Academic Press. NY, NY.
Health, A.G. 1995. Water Pollution and Fish Physiology. 2nd ed. Lewis Publishers. 359
PP-
Herreid, C.F., II. 1980. Hypoxia in invertebrates. Comp. Biochem. Physiol. 67A:311-320.
Hillman, N.S. 1964. Studies on the distribution and abundance of decapod larvae in Nar-
ragansett Bay, Rhode Island, with consideration of morphology and mortality. MS
Thesis. University of Rhode Island. 74 pp.
Holeton, G.F. 1980. Oxygen as an environmental factor of fishes, pp. 7-32. (in) M.A. Ali
(ed). Environmental Physiology of Fishes. Plenum Press.
Homer, D.H. and W.E. Waller. 1983. Chronic effects of reduced dissolved oxygen on
Daphnia magna. Water, Air and Soil Pollut. 20:23-28.
Howell, P., and D. Simpson. 1994. Abundance of marine resources in relation to dis-
solved oxygen in Long Island Sound. Estuaries 17:394-402.
Hughes, G.M. 1981. Effects of low oxygen and pollution on the respiratory systems of
fish. pp. 121-146. (in) A.D. Pickering (ed). Stress and Fish. Academic Press. NY,NY.
Hunnington, K.M. and D.C. Miller. 1989. Effects of suspended sediment, hypoxia, and
hyperoxia on larval Mercenaria mercenaria (Linnaues, 1758). J. Shellfish Research
8:37-42.
Hutcheson, M, D.C. Miller and A.Q. White. 1985. Respiratory and behavioral responses
of the grass shrimp Palaemonetes pugio to cadmium and reduced dissolved oxygen.
Mar. Biol. 8:59-66.
Johnson, D.A. and B.L. Welsh. 1985. Detrimental effects of Ulva lactuca (L.) exudates
and low oxygen on estuarine crab larvae. /. Exp. Mar. Biol. Ecol. 86:73-83.
Johnson, D.F. 1985. The distribution of brachyuran crustacean megalopae in the waters
of the York River, lower Chesapeake Bay and adjacent shelf: Implications for re-
cruitment. Estuarine, Coastal and Shelf Sci. 20:693-705.
Jones, M.B. and C.E. Epifanio. 1995. Settlement of brachyuran megalopae in Delaware
Bay: an analysis of time series data. Mar. Ecol. Prog. Ser. 125:67-76.
Jordan, S., C. Stenger, M. Olsen, R. Batiuk, and K. Mountford. 1992. Chesapeake Bay
dissolved oxygen goal for restoration of living resource habitats. Reevaluation Report
#7c. CBP/TRS 88/93. Chesapeake Bay Program Office. Annapolis, Md.
Kramer, D.L. 1987. Dissolved oxygen and fish behavior. Environ. Biol. Fishes. 18:81-92.
Kuo, A.Y., K. Park and M.Z. Moustafa. 1991. Spatial and temporal variabilities of hy-
poxia in the Rappahannock River, Virginia. Estuaries 14:113-121.
Llanso, R.J. 1991. Tolerance of low dissolved oxygen and hydrogen sulfide by the poly-
chaete Streblospio benedicti (Webster). J. Exp. Mar. Biol. Ecol. 153:165-178.
Llanso, RJ. 1992. Effects of hypoxia on estuarine benthos: the Lower Rappahannock
River (Chesapeake Bay), a case study. Estuarine, Coastal and Shelf Sci. 35:491-515.
51
-------
Llanso, RJ. and R.J. Diaz. 1994. Tolerance to low dissolved oxygen by the tubicolous
polychaete Loimia medusa. J. Mar. Biol. Assoc. U.K. 74:143-148.
Lutz, R.V., N.H. Marcus and J.P. Chanton. 1992. Effects of low oxygen concentrations
on the hatching and viability of eggs of marine calanoid copepods. Mar. Biol.
114:241-247.
Lutz, R.V., N.H. Marcus and J.P. Chanton. 1994. Hatching and viability of copepod eggs
at two stages of embryological development: anoxic/hypoxic effect. Mar. Biol.
119:199-204.
McLeese, D.W. 1956. Effects of temperature, salinity and oxygen on the survival of the
American lobster. J. Fish. Res. Bd. Canada. 13:247-272.
McMahon, B.R. 1988. Physiological responses to oxygen depletion in intertidal animals.
Amer. Zool. 28:39-53.
Miller, D.C. and K.M. Huntington. 1988. Larval hard clam mortality under high sus-
pended sediment and low dissolved oxygen concentration. Final Report. April 1988.
Prepared for: Dept. Natural Resources and Environmental Control, State of Delaware.
College of Marine Studies, University of Delaware, Lewes, DE.
Morrison, G. 1971. Dissolved oxygen requirements for embryonic and larval develop-
ment of the hardshell clam, Mercenaria mercenaria. J. Fish. Res. Bd. Canada.
28:379-381.'
Osman, R.W. and G.R. Abbe. 1994. Post-settlement factors affecting oyster recruitment
in the Chesapeake Bay, USA. pp. 335-340. (in) Dyer, K.R. and RJ. Orth (eds).
Changes in Fluxes in Estuaries. Olsen and Olsen, Denmark.
Paul, J.F., J.H. Gentile, KJ. Scott, S.C. Schimmel, D.E. Campbell and R.W. Latimer.
1997. EMAP-Virginian Province Four-Year Assessment Report (1990-93). EPA
600/R-97/XXX. U.S. Environmental Protection Agency, Atlantic Ecology Division,
Narragansett, Rhode Island.
Pihl, L., S.P. Baden and RJ. Diaz. 1991. Effects of periodic hypoxia on distribution of
demersal fish and crustaceans. Mar. Biol. 108:349-360.
Pihl, L., S.P. Baden, RJ. Diaz and L.C. Schaffher. 1992. Hypoxia-induced structural
changes in the diet of bottom-feeding fish and Crustacea. Mar. Biol. 112:349-361.
Poucher, S. 1988a. Effects of low dissolved oxygen onMysidopsis bdhia in two modified
chronic tests. Memorandum to David J. Hansen. U.S. Environmental Protection
Agency, Atlantic Ecology Division, Narragansett, Rhode Island 02882.
Poucher, S. 1988b. Chronic effects of low dissolved oxygen onMenidia menidia. Memo-
randum to David J. Hansen. U.S. Environmental Protection Agency, Atlantic Ecology
Division, Narragansett, Rhode Island 02882.
Poucher, S. and L. Coiro. 1997. Test Reports: Effects of low dissolved oxygen on salt-
water animals. Memorandum to D.C. Miller. U.S. Environmental Protection Agency,
Atlantic Ecology Division, Narragansett, Rhode Island 02882. July 1997.
Poucher, S. and L. Coiro. 1999. Data print out of ICp values for effects of dissolved oxy-
gen on growth of saltwater species. Memorandum to G.B. Thursby. U.S. Environ-
52
-------
mental Protectoin Agency, Atlantic Ecology Division, Narragansett, Rhode Island
02882.
Reid, D.G. and J.C. Aldrich. 1989. Variations in response to environmental hypoxia of
different colour forms of the shore crab, Carcinus maenas. Comp. Biochem. Physiol.
92A:535-539.
Reish, D.J. 1966. Relationship of polychaetes to varying dissolved oxygen concentra-
tions. Section III. Paper 10. Third International Conference on Water Pollution Re-
search. Munich, Germany.
Ricker, W.E. 1954. Stock and recruitment. J. Fish. Res. Bd. Canada. 11:559-623.
Roman, M.R., A.L. Gauzens, W.K. Rhinehart, and J.R. White. 1993. Effects of low oxy-
gen waters on Chesapeake Bay zooplankton. Limnol. Oceanogr. 38:1603-1614.
Rombough, P.J. 1988a. Respiratory gas exchange, aerobic metabolism, and effects of hy-
poxia during early life. pp. 59-161. (in) W.S. Hoar and D.J. Randall. Fish Physiology.
Vol. XI: The Physiology of Developing Fish. Part A. Eggs and Larvae. Academic
Press, NY, NY.
Rombough, P.J. 1988b. Growth, aerobic metabolism and dissolved oxygen requirements
of embryos and alevins of steelthead Salmo gairdneri. Can. J. Zool. 66:651-660.
Saksena, V.P. and E.B. Joseph. 1972. Dissolved oxygen requirements of newly-hatched
larvae of the striped blenny (Chasmodes bosquianus), the naked goby (Gobiosoma
bosci), and the skilletfish (Gobiesox strumosus). Chesapeake Sci. 13:23-28.
Sandifer, P.A. 1973. Distribution and abundance of decapod crustacean larvae in the
York River estuary and adjacent lower Chesapeake Bay, Virginia, 1968-1969.
Chesapeake Science. 14:235-257.
Sandifer, P. A. 1975. The role of pelagic larvae in recruitment to populations of adult de-
capod crustaceans in the York River estuary and adjacent lower Chesapeake Bay,
Virginia. Estuarine and Coastal Marine Science. 3:269-279.
Sanford, L.P. K.R. Sellner and D.L. Breitburg. 1990. Covariability of dissolved oxygen
with physical processes in the summertime Chesapeake Bay. J. Mar. Res. 48:567-
590.
Savage, N.B. 1976. Burrowing activity in Mercenaria mercenaria (L.) and Spisula
solidissima (Dillwyn) as a function of temperature and dissolved oxygen. Mar. Be-
hav. Physiol. 3:221-234.
Secor, D.H. and T.E. Gunderson. 1998. Effects of hypoxia and temperature on survival,
growth, and respiration of juvenile Atlantic sturgeon, Acipenser oxyrinchus. Fishery
Bulletin 96:603-613.
Shepard, M.P. 1955. Resistance and tolerance of young speckled trout (Salvelinus fonti-
nalis) to oxygen lack, with special reference to low oxygen acclimation. /. Fish. Res.
Bd. Canada. 12:387-446.
Shumway, S.E. and T.M. Scott. 1983. The effects of anoxia and hydrogen sulfide on sur-
vival, activity and metabolic rate in the coot clam, Mulinea lateralis (Say). J. Exp.
Mar. Biol. Ecol. 71:135-146.
53
-------
Simpson, D.G., M.W. Johnson and K. Gottschall. 1995. A study of marine recreational
fisheries in Connecticut. Cooperative Interagency Resource Assessment, pp. 87-114.
(in) Final Report to U.S. Fish and Wildlife Service, Project F54R. Study of Marine
Fisheries in Connecticut. Fisheries Div., Bur. Natural Resources, CT Dept. Environ-
mental Protection, Hartford, CT.
Stephan, C.E., D.I. Mount, DJ. Hansen, J.H. Gentile, G.A. Chapman and W.A. Brungs.
1985. Guidelines for deriving numerical national water quality criteria for the protec-
tion of aquatic organisms and their uses. NTIS Publication No.: PB85-227049.
Stickle, W.B., M.A. Kapper, L. Liu, E. Gnaiger and S.Y. Wang. 1989. Metabolic adapta-
tions of several species of crustaceans and molluscs to hypoxia: Tolerance and micro-
calorimetric studies. Biol. Bull 177:303-312.
Stickle, W.B. 1988. Tables for 96-hour and 28-day survival for seven species of marine
animals. Memorandum dated October 6 to Don Miller. U.S. Environmental Protection
Agency, Atlantic Ecology Division, Narragansett, RI 02882.
Strobel, C.J., H.W. Buffum, S.J. Benyi, E.A. Petrocelli, D.R. Reifsteck and DJ. Keith.
1995. Statistical Summary: EMAP-Estuaries Virginian Province - 1990-1993. U.S.
Environmental Protection Agency, National Health and Environmental Effects Re-
search Laboratory, Atlantic Ecology Division, Narragansett, RI. EPA/620/R-94/026.
Strobel, CJ. and J. Heltshe. 1999. Application of indicator evaluation guidelines to dis-
solved oxygen concentration as an indicator of the spatial extent of hypoxia in estua-
rine waters. Chapter 2 (in). L. Jackson, J. Kurtz and William Fisher (eds). Evaluation
Guidelines for Ecological Indicators. U. S. Environmental Protection Agency. Office
of Research and Development, (in press).
Summer, J.K., S.B. Weisberg, A.F. Holland, J. Kou, V.D. Engle, D.L. Breitberg, and R.J.
Diaz. 1997. Characterizing dissolved oxygen conditions in estuarine environments.
Environ. Monitoring and Assessment 45:319-328.
Theede, H., A. Ponat, K. Hiroki and C. Schlieper. 1969. Studies on the resistance of ma-
rine bottom invertebrates to oxygen-deficiency and hydrogen sulphide. Mar. Biol.
2:325-337.
Tyson, R.V. and T.H. Peason. 1991. Modern and Ancient Continental Shelf Anoxia.
Geological Society Special Publication No. 58.
U.S. EPA. 1985. Ambient Water Quality Criteria for Cadmium - 1984. U.S. Environ-
mental Protection Agency. Office of Water Regulations and Standards. Criteria and
Standards Division. Washington, D.C. EPA 440/5-84-032.
U.S. EPA. 1986. Ambient Water Quality Criteria for Dissolved Oxygen. U.S. Environ-
mental Protection Agency. Office of Water Regulations and Standards. Criteria and
Standards Division. Washingtion, D.C. EPA 440/5-86-003.
U.S. EPA. 1994. Interim Guidance on Determination and Use of Water-Effect Ratios for
Metals. U.S. Environmental Protection Agency. Office of Water. Office of Science
and Technology. EPA-823-B-94-001.
van Montfrans, J., C.A. Peery, and RJ. Orth. 1990. Daily, monthly and annual settlement
patterns by Callinectes sapidus andNeopanope sayi megalopae on artificial collectors
deployed in the York River, Virginia: 1985-1988. Bill. Mar. Sci. 46:214-229.
54
-------
Vargo, S.L. and A.N. Sastry. 1977. Acute temperature and low dissolved oxygen toler-
ances of Brachyuran crab (Cancer irroratus) larvae. Mar. Biol. 40:165-171.
Vargo, S.L. and A.N. Sastry. 1978. Interspecific differences in tolerance of Eurytemora
afftnis and Acartia tonsa from an estuarine anoxic basin to low dissolved oxygen and
hydrogen sulfide. pp. 219-226. (in) D.S. McLusky and A.J. Berry (eds). Physiology
and Behaviour of Marine Organisms. Proceeding of the 12th European Symposium
on Marine Biology, Stirling, Scotland, September 1977. Pergamon Press.
Vernberg, FJ. 1972. Dissolved gasses: Animals, pp. 1491-1526. (in) O. Kinne. Marine
Ecology: A Comprehensive, Integrated Treatise on Life in Oceans and Coastal Wa-
ters. Vol. I, Part 3: Environmental Factors. Wiley-Interscience, NY, NY.
Vismann, B. 1990. Sulfide detoxification and tolerance in Nereis (Hediste) diversicolor
and Nereis (Neanthes) virens (Annelida: Polychaeta). Mar. Ecol. Prog. Ser. 59:229-
238.
Voyer, R.A. and RJ. Hennekey. 1972. Effects of dissolved oxygen on two life stages of
the mummichog. Prog. Fish. Cult. 34:222-225.
Wang, W.X. and J. Widdows. 1991. Physiological responses of mussel larvae Mytilus
edulis to environmental hypoxia and anoxia. Mar. Ecol. Prog. Ser. 70:223-236.
Welsh, B.L., RJ. Welsh and M.L. DiGiacomo-Cohen. 1994. Quantifying hypoxia and
anoxia in Long Island Sound, pp.131-137. (in) K.R. Dyer and RJ. Orth. Changes in
Fluxes in Estuaries: Implications from Science to Management. Olsen and Olsen,
Fredensborg, Denmark.
55
-------
This page intentionally left blank.
-------
Appendix A. Comparison of 24 hr and 96 hr acute sensitivity to low dissolved oxygen for saltwater animals.
Each pair is from the same test run.
bpecies
Common name
24 hr LC50 96 hr LC50 Reference
Americamysis bahia
Americamysis bahia
Apeltes quadracus
Brevoortia tyrannus
Brevoortia tyrannus
Crangon septemspinosa
Leiostomus xanthurus
Morone saxatilis
Morone saxatilis
Palaemonetes pugio
Palaemonetes vulgaris
Paralichthys dentatus
Pleuronectes americanus
Pleuronectes americanus
Prionotus carolinus
Tautoga onitis
Tautoga onitis
Juveniles
mysid shrimp 1.22 1.29
mysid shrimp 1.20 1.25
fourspine stickleback 0.92 0.91
Atlantic menhaden 1.14 1.21
Atlantic menhaden 0.88 1.04
sand shrimp 0.77 0.97
spot 0.67 0.70
striped bass 1.50 1.53
striped bass 1.62 1.63
daggerblade grass shrimp O.55 0.72
marsh grass shrimp 0.84 1.02
summer flounder 1.10 1.10
winter flounder 1.44 1.45
winter flounder 1.28- 1.30
northern sea robin . 0.55 0.55
tautog 0.82 0.82
tautog 0.80 0.82
Larvae
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Burton, etal. 1980
Poucher and Coiro, 1997
Burton, etal. 1980
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Cancer irroratus
Cancer irroratus
Cancer irroratus
Cancer irroratus
Cancer irroratus
Dyspanopeus sayi
Dyspanopeus sayi
Dyspanopeus sayi
Dyspanopeus sayi
Dyspanopeus sayi
Dyspanopeus sayi
Dyspanopeus sayi
Eurypanopeus depressus
Homarus americanus
Homarus americanus
Homarus americanus
Homarus americanus
Homarus americanus
Homarus americanus
Homarus americanus
Libinia dubia
Menidia beryllina
Morone saxatilis
rock crab
rock crab
rock crab
rock crab
rock crab
Say mud crab
Say mud crab
Say mud crab
Say mud crab
Say mud crab
Say mud crab
Say mud crab
flat mud crab
American lobster
American lobster
American lobster
American lobster
American lobster
American lobster
American lobster
longnose spider crab
inland silverside
striped bass
2.20 3.09 Poucher and Coiro, 1997
2.14 2.80 Poucher and Coiro, 1997
<1.72 2.17 Poucher and Coiro, 1997
<1.75 2.22 Poucher and Coiro, 1997
1.85 2.20 Poucher and Coiro, 1997
1.66 2.50 Poucher and Coiro, 1997
<1.18 1.73 Poucher and Coiro, 1997
1.61 1.73 Poucher and Coiro, 1997
1.88 2.13 Poucher and Coiro, 1997
1 -95 1.97 Poucher and Coiro, 1997
<1.55 1.57 Poucher and Coiro, 1997
<1.83 2.40 Poucher and Coiro, 1997
2.09 2.10 Poucher and Coiro, 1997
3.31 3.43 Poucher and Coiro, 1997
2.66 3.21 Poucher and Coiro, 1997
2.46 2.82 Poucher and Coiro, 1997
2.27 2.27 Poucher and Coiro, 1997
2.14 3.08 Poucher and Coiro, 1997
2.44 2.83 • Poucher and Coiro, 1997
<2.32 3.19 Poucher and Coiro, 1997
1.83 2.71 ' Poucher and Coiro, 1997
1-43 1.44 Poucher and Coiro, 1997
1.96 1.96 Poucher and Coiro, 1997
-------
Appendix A. Continued
Palaemonetes pugio
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
daggerblade grass shrimp 1.24 1.58
marsh grass shrimp 0.84 1.02
marsh grass shrimp 1.50 2.18
marsh grass shrimp <2.05 2.16
marsh grass shrimp O.48 0.98
marsh grass shrimp <1.56 >1.92
marsh grass shrimp <1.59 2.05
marsh grass shrimp 1.77 1.87
marsh grass shrimp . 1.70 1.72
marsh grass shrimp 1.66 2.15
marsh grass shrimp 1.95 2.10
marsh grass shrimp <1.79 <1.79
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Poucher and Coiro, 1997
Appendix A: page A-2
-------
.
0?
•a
s*
«2
0
•*•*
"o
.2
•S
_«
i
.3
03
g
'S
03
••-D
o
•w
'j>
'-4mt
°5
S
a>
en
-2
S
O
•<
CQ
'•3
a.
c.
Reference
B§
"""
si
J.E,
||
G"
)
&
S
u
^
t^
M
•8 £*
u ^
Q ^,
tS
T3
O
V
s
s
G
u
,J
u
«
5
S
o
U
S
I
ON O\
00 OO
O\ O\
r- i- t- r- t- ^ "^^
2^^^^'O\ _H o\ o\ o a> o\ o\ o\ c\ o\ os o\
&* v\ O\ G\ O\ OX ON(V1"" O\ O\OSO\OXONO\
2 2" 2" 2" 2" g g1 S S g S « o 2" o 2" 2" 2" £ 2" 2"
ooooo^!Oo~u I^^^^^SsSjSSStSS
1 1 1 1 ^ 1' i ^ * § 1 i i 5 5 ? i • * § i 1 1 1 1
1 1 1 1 1 1 ! 1 1 1 1 1 1 1 1 J I I i 1 1 III
_P O O O O ^ ™ O • *G O ® "-C "-J3 ^ ^ ^ O^OOOOOO
^o o)r^o\ m ovoi— tr^fOf-oooN
CN C^ » «, voSOa^-*«0^
" ___ _ _cjcq____^
9 3's 1 3 S s-g g i 2 1 g s.s si |;2 S g § 2 2
V - V V V V
SoJc^^S <^ • ~ cso— «trj|r)»n
AiAo^o\r5 *^2o\^j^ ' !0 ^ M 'rS ' ' ' * o "^
s ' cs cs ^^ *— • ^~* 2 cj 7i Pi ^ ^ ^
^£pi0£^0£ 0 pj^So^pJS^
nmrn(r^^V^r*^m'~'c*ie^Cri^(N'*^'rriO^^^'~' ^
fr) rn n c^i CN
SSSSSISSSSSSSSSSSSSSSSISSIS
L^UnfcitlHtl-ipI-iCOtI-C/3pUKJCOC/3COWCOtl -g § §
-S w ca-'K'K2 — «^— w^^^rto
| | .1- -| -S -S | 1 § 1 | | 1 | | | | * | | J> s s
J5 "w "o, <— c3 w o o o *o ^ -w « o u w o ^_i S S^i 60 wj £ ™
3 £ ro<2<
-------
s
I
a.
•a
I
J
c,
H
"o
*o
c
ra
u
1
VO
VO
^3;
f?
CN
2
en
-
«-«
|
i
y
1
i.
>
i
I
S5
2
3
•a
1
£
en
CN
VO
"
o
en
o
CN
0\
O
en
CJ\
CN
Tj-
a
"E
•o
S
o
1
^
•o
c
o
."s
c
Pleitronectes america
S
y
3
"O
c
ro
1
1
^
oo
o
U"»
o
ej\
2
en
-
1
o
1
_c
J3
rn sea ro
i
Prionolus carolinus
c\
oo
o\
*«
o
to
cxT
oo
o\
JJ"
55
VJ
o
o
d
o
o
en
*
S
00
o
g
Tfs
o
•a
i
en
i
s
:~
Rithropanopetis harrl
£
a
3
•o
1
00
Tf
*"^
00
d
o
CN
O
o
en
CN
S
u.
0
I
(_
•o
c
o
c?
1
1-
0
1
Scopthallmis agiiosus
§>
2
8
"O
§
Poucher
en
VO
0
tj-
O
1
CN
CN
O
en
Tf
S
0
1
E
1
CO
1
Spisiila solidissima
S
2
1
1
O
cu
cs
o
cs
o
CO
~
s
tu
s
I
cu
1
Stenotomus chrysops
o\
g
3
•a
c
ea
1
1
CN
-------
•o
I
ca
S
a
C5
I.
S
•a
•s
- I
w
o S
IS
g
b« u
•— S,
~ E
£
s
'5
OH
I
C/l
P
2
Cfl
P
O
tD
00
TO
O
.*
•3
S
a,
.s
00"
O O O" O"
. O «—i
CO CN CN *-3
*
CD
o"
g
t-.
5
f-
c^ Tf ^ o\
»n r-
-------
M
C3
0,
Appendix C:
•
Reference
f •«, ff
lof
5 "• •="
^ »-.
lit
*£"
f
=>
— "^
L> 2
3 §.
"o §
0 -E
=8 B ^
1 w "a
| §S
"* S.
• 5
Q ""
" a
* S
* u
" "*
1 E
s
J)
s
4
;
:
~
0
1
C
3
;
i
)
i
u
n
This report;
Poucher & Coiro,
1999
tn
en
CO
^
o o"J|
eS ^ 0
oo ^- o
NO OS — «
«-« OO OO
vi en en
en c^i rn
O
m .
CO ' S 2
o. c«i rn
1> O o Q)
60 *C8 M e« ' 00 ,„
S M So. So.
CO W CO ^O CO O
1 1 11 1 1
S m S E .2 1
XI J3 JS
§ 0 1
•a -o -a
Ell
s5* ^ ^*
CO CO CO
1 1 I •
g 3 g
^. o_ D
o o o^
I i I
tt
o
&.
15
t
1
CO CO CO
£H ?* OO
VO 00 ON
0 0" 0" C5" g
r-- ob es o\ o
vo vo f- os *--
(Q »
CO
1
g
1 -
1
This report;
Poucher & Coiro,
1999
*>
ON
00
CO
o" o" cf ^
vo o\ */i o
Os" of vf oo
m m cs *~*
c- ,3-
C--
-------
Q_J E3
§5
_ 73 8B
6>-&
a
•** "^
i!
18
2 "B-
•± 5^
Q. s-
t!
§
<» I
ON O
O O"«f
>? vP vP
>*• 0s O^
o\ —
cs' cs
•a- —i o
vd
-------
a
ta
0.
6"
Appendix
i Reference
1
| §1
O ^^
§ J fe^
** ^* e
5>~
V
t^
3
* 1
1 S
"b ^
o &
* « ~
2 w la
19s
O 5£ *^^
o *
If
3 S.
c-s-
i £i
?i
- La
S 6J
11
H
tl
Bfl
B
u
5
s
±
£
i
3
3
5
•j
|
CO
•e
.52
'
§
r~"7
CO C/3
^? f^
CN r>*
000"
gvf g
M >o t~
CNTJ--0
•^l- en co
en (S ~
^
vd
oo
^
n
3
J
g
X
|
S
a.
3
|
3i
i
S
«o
1
I
3
u
g
5
3
a.
2"
This report;
Poucher & Coi
1999
en
«s
v>
oo
^ t
VO
O OOO"
— CN Tf VO
t-ri «'
s"°
vo «ri
^
S
en
•a
o
JS
VO
V
O.
Ig
en
a
1
E
t,
i
g
•S
—
O ir> O
CS CS VO
c«i cs ~
35
vd
CO
VO
S
i
en
a
•o
o
£
VO
"v
n.
E .
~
CO
S
U)
CO
E
.03
*.
"g
•Si
§
s
J
Q
a.
This report
•
c\
CO
OT
«£
a,
o"
S
— '
»-* oo en oo e^]
vd en en oi ^
r-
o
en
S
en
d.
en
oi)
S
•— en
£2
ea ^ l
1
cn
en
on
2
CO
E
^
I
§
8
1
1
<§
g"
This report;
Poucher & Coil
1999
S
"w
O
CU
1
-C
en
en
oh
1
E
,co
^>
3
S
§
5
J
Q
0,
This report
•
9
°i
O
en
VO
tn
— •
— O 0
oo 10 vt
vo en CN
^
VO
^
en
0
en
„_
5
.2
•z
o
o.
o.
J3
CO
CO
13]
1
S
feo
£!
"g
S
CO
1 -
1
^1
2
«
H
g
OD
O O O O c/3
so vp so ^p vO
5^ ^ 5v 5^ £v
oo o r^ vo vo
-
o r-;
-------
10
6
l
§
O-l
en
o
en
o
o
"to
"S?
o
"c
CO
•S
'§
-S
"5
51
t:
o
f
CO
£
CO
O4
v
"co
*! «
I s .
0 0 -d
° 0>
4_t 4-14=!
O « co
sS 5 "
o So
° ^ 1 g
Z O OO 0
en o <— « «^ — co
r- «o -^- en en o-*
^_
> E
11
•s
c
o
CS
!_.
0
E
CO
s
• K
1
1"
"s
s
t:
O
.«
00
o\
in
OO O"
f^ •* O
•^
£-
c
a
O
^*^
"5
rrl
I^J
"S
1
CA
-O
9
u
w
o
_o
t!
§
c_>
C
O
O
•o
53
•a
fe
cn
0
o
^
U
0
,?
03
&
at the effect
•S
CO
a
CO
1
.22
-xB
CO
j-
J_(
M
™
«
o
c
o
o
3
>
*3
_C3
"3
C
_o
o ^_;
3 C
*T3 CO
a) o
it 5fJ
la
"S OT
o3 "co
.« to
o '§
cS "
CL) o
O
Appendix
-------
This page intentionally left blank.
-------
VO
•a
s
CU
IS
"O
CU
£
o
C3
.S
'a
CS
cu
i
en
a
C>
Q
f)
8;
ON
2
r and Co
1
CN
^^
V
CN
^^
™"
V
O
CN
3
CN
CN
ON
ON
a
•o
O
1
2
CN
ON
0
en
ON
a
r and Co
1
S
V
CN
0
CN
-
CN
r>-
CN
2
•o
g
O
1
S'
V
CN
CS
C")
e^i
en
o\
o\
8
rand Co
o
•g
1
oo
CN
ON
CN
en
0
en
ON
ON
P
a
1
•g
1
ON
oo
V
CN
•S
CN
CN
ON
%•
i
CO
1
ca
on
o
CN
O
VI
CN
0
s
*
*5
CN
r--
2
P
o
1
o
a,
«
V
CN
en
0
en
r-
ON
ON
P
a
•o
g
u
1
oo
V
CN
o
CN
CN
en
o
en
5
ON
O
a
i
ca
•§
1
vo
eN
CN
O
en
OO
CN
r—
ON
ON
P
O
U
ca
1
OO
oo
«
s
m
en
1
p
r and Coi
1
1
oo
V
vo
s
CN
o
r^
2
P
' and Coi
o
1
vo
-
2
-
o
en
1
P
a
•a
g
1
1
ON
O
CN
ON
O
CN
ON
O
CN
CN
en
0
en
2
nd Henn
o
CN
V
o
CN
V
O
CN
V
CN
O
en
CN
ON
•0.
•— »
T3
c
CO
g
0>
_<
^^
CN
0
CN
— i
CN
2
a.
•o
c
CO
c
o
_
o
en
*-4
O
m
s
0
CN
CN
0
ON
• O
and Coi
I
1
C7\
CN
V
CN
CN
V
*
CN
P!
°
1
p
C
C3
1
1
CN
en
V
CN
CN
ON
ON
P
a
•o
g
JC
o
VO
VO
ON
oo
en
ON
CN
r-
2
o
and Coi
1
1
^.
CN
—
O
CN
en
o
en
ON
ON
P
a
•a
i
1
1
VO
CN
ON
oo
S
O
en
ON
P
a
T3
ca
o
en
en
o
ON
en
°
r-
ON
O
a
1
•§
1
r*
o
2
n,
Q
a
'o
|
en en en
*« ea ec
§•
•3
oo
g>
"O T3 ^ ^
(O 03
-H -a t
eg ea cd cq g
£ £
|
• "
— CN CN CN
I1
M DO W) OO
"Eo tn "oo tn
£ £
•
cu
a
a,
•§-
B
^
u
2
S
t>
•g
O
JD
1
2
•§ •§
S S
•a -a
2 2
_n
^
o
S
>>
c
c
o
2
•g
-B*
1
o
•o
3
!>•
•1
u
•o
1
s-
00
•s -§ •§ •§ •§ -g
o S S o o t>
•§ •§ 1 •§.•§ •§
£ E E E E 1
C/3 C/3 CO C/3
E S
I S
a a
S .8
o o
.8
I I
o> u
_ l_^ %_l \J l^l (J (J (J
g g
•S •§
1 1
§ §
11
a a
§3§g|'§>&e5-e>&^&-i<
llll|s'ssssa_>
-fe .fe .fe -fe I I I I I I I I §•
S •§
od
fel§llllllllll
llllllllllfl&-
av3v3v5S^Q>Q1Q>^Q-Q't§cS
-
(J
o
"O
1
Q
1
•«i
<3
s
^
B
Si
Q
S
R
1
a
s
8
S
s
0
Jo
t5
s
^
-3
«^
g
'g
5
S
<>}
§
K
%
£2
s;
S
&
K;
a
&a
|
1
a
j~
g
"S
5
S
to
i
5
t§
*3
s:
.§
x. -
Cli
£
5
eo
1
5
i
S3
a
§
s
R:
Ct
f^
§
K:
!§
-------
2
I
c.
.
•a
a
g
Rcfercn
m
5
O
|
CO
3"
"a
E,
6
Ci
e
u
S JlP
Q x~"
K
O
ft»
Common nam
|
CO
r-
$
y
a
1
1
OH
?
CS
0
CS
en
o
en
en
a
1
e
»
u
American lobsl
S3
§
|>
martis amei
!§
^
§:
H
a
1
1
1
vo
CO
cs
^3.
cs
cs
cs
CO
CO
CtO
s
T5
0
American lobsl
60
g
^
martis amei
!§
ON
ON
2
"S
O
•o
g
1
1
s?
V*
cs
CO
CO
OO
a
CO
CO
0
American lobsl
g
.§
g
!§
o\
cy»
2
a
1
1
o
1
oo
o
Cj)
oo
CO
a
f
CO
"3
^
a
g
American lobsl
i
.8
I3UID SIIJOIU
t§
^
o\
0\
g
a
i
C3
5
|
o
PH
«0
O
CS*
o
oi
CO
^
cs
o
cs
CO
*
oo
s
V)
To
^
s
o
L-
longnose spide:
nnia dubia
$
t —
Os
g*
"o
O
•a
O3
1
1
ON
to
CN
TI-
CS
en
en
o.
o
oo
u
CS
.O
2
o
longnose spide:
linia dubia
3
£
2
1
C
C3
Poucher
o
?
(
a.
o
re
oo
u
*«
—
03
«
longnose spide
linia dubia
$
Cf\
o\
2
a
•a
03
Poucher
o
o
-H*
o
o
~*
V
0
^
V
o
T3
o
I
long fm squid
1
"3
Os
2
a
•o
g
Poucher
en
— ;
V
en
^
V
?!
V
to
CS
"
"S
6
!
CO
inland silversid
§
1
^
ON
2
a
•a
Poucher
9
o
en
"o
cs
I
0
inland silversid
.§
.g
:f
o\
g
a
•o
g
fc-.
1
1
cs
Os
cs
oo
cs
Os
cs
1
s
f
0
IH
•o
"c
.1
'5
a
s
"o
o
•o
g
1
1
0
1— 1
rs
CO
1
03
!
«L>
•o
^>
"to
•o
g
.E
.1
-Ci
$
Os
OO
h
1
•o
2
.£
1
(
o
v'
o
V
§
V
cs
cs
s>
.2?
1
•o
o
>,
•*
hardshell clam
.0
1
^
rcenaria nit
:S
«-
OS
1
i
0
V
V)
cs
oo
cs
o
1
hardshell clam
a
1
rcenaria im
^
i
2
o
O
•a
g
1
1
ON
en
cs"
o
en
cs"
vo
ON
_^
CS
o
cs
vo
o
D.
striped bass
.£
rone saxati
:f
a
g
o
O
"O
c
CO
Poucher
!2
o
f^.
£
1
ex
striped bass
.s
|
1
a
2
o
u
•a
Poucher
cs
cs
o\
oo
to
1
1
o.
striped bass
.a
rowe saxati,
1
a
2
o
O
•o
g
Poucher
^
cs'
n-
to
cs
s
cs
VO
^f.
cs
cs
0
CO
•g
CO
0
&
1
Burry's octopus
'opus bwry,
o
ON
ON
g
O
O
•o
c
CO
Poucher
CT\
en
V
cs
_^
cs
to
cs
^-
cs
en
0
en
I
"O
o
1
o.
B
JZ
CO
1
daggerblade gr;
.a
a
'aemonetes t
eg
c?
2
o
U
•a
%
i,
_^
V
§
V
o
m
en
0
en
1
CO
a.
S
marsh grass shi
I
x
'aemonetes
eg
a
g
o
O
•o
C
CO
Poucher
oo
to
CS
o
cs
en
2
a
s
VI
1
c.
marsh grass shi
!
'aemonetes
s,
ON
2
a
1
Poucher
S
V
vo
CS
en
0
en
g
*O
o
vo
"v
o.
£
marsh grass shi
.=0
"1)
a
'aemonetes
£
a
2
o
O
1
1
1
0
V}
v>
cs
,-J.
cs
CO
0
CO
03
03
"O
0
1
o.
E
marsh grass shi
60
1
~
'aemonetes
eg
a
2
o
U
n
03
Poucher
V)
O
V
vo
CS
,-]•
cs
CO
0
CO
1
CO
"O
0
VO
v"
0,
E
marsh grass shi
.60
3
'aemonetes
eg
ON
ON
2
a
c
C3
Poucher
VO
V
to
CS
cs
en
CO
0
£
0.
E
marsh grass shi
I
«
'aemonetes
eg
ON
ON
g
a
1
Poucher
to
V
o
CO
Os
CS
cs
0
CO
8
03
*o
o
1
o.
E
marsh grass shi
60
1
~
'aemonetes
-------
o.
a,
S
co
o
£
%
Q-
m
2" 2" 2"
a a a
•o -o -o
o o
O CJ
0>
I I
BH Pa
I .
I I
r- o eo \o
t^ t-^ Ti; r-^
•— —' o —
Vt OO OO
CM
— en
r» " o ^
ON
2
a
•o
CO
ss
•g
3
&
"^
CN
r~
CN
OS
0
en
CN
O
(S
CM
on
0
en
ON ON ON
ON ON ON
222
o o o
O O 0
•O *O T3
i § i
S3 S3 S3
•g •§ -g
333
o, P< a.
0 CN r-
CM CM CM
CM CM CM
ON CO O
r*~i rO t^i
en cM en
•a
g
o
U
•o
S3
•g
3
£
0
n
^-«
on
oo
en
ON
CM
ON
ON
2
a
•o
CO
S3
"o
^ on
o o
bo bo
CO CO
"co "co
^g J2
*CO "°
t3 o
2 2
«n
en
%
s
1
s
o
•S
2
en
1
to
1
JO
t
o
2
1
O
CO
S?
E
t
O
2
S
—
*o
"o
c^
o
^
'i
1
1
1
2
js
oo
CM
O
C
1
1
0.
1
CO
i
"co
t
C3
£>
s
5
•a
3
g.
i
eo
t
CO
ca
0
t3
I
on
"co
CO
CO
.0
eo
0
s
£
on
i
eo
CO
CO
•s
b
-a
3
S
CO
on
CO
CO
CO
s
o
•a
3
s
g-
CO
on -<3- CM
S 3 2
CO CO CO
111
Bj S3 3
|||
111
E E E
g-. g> «
Ct3 Co TO
OO OT 53
en
I
eo
CO
CO
•s
1
i
"S
i
CO
•ca
CO
53
eo
•§
C
ca
o
I
3
CO
ca
CO
*J
o
o
1
a>
M
CO
CO
CO
*3
o
g
o
i
CN
a>
f
CO
75
CO
£2
o
s
o
a>
1
-------
t
Q
o
8P
a.
Q
«
c
K
|
5
|
CO
2"
a
0
1,
>
CO
O
LO
t
t
OT
ggggsgjg;^
2 2" 2 2 2" 2" 2" 2"
88888888
'c"c"S"c"^"a"a*a
c3c3rocJeac3c3ea
| || || | | |
f2<2<2
V
O TT t-
cs ~ —
V
ssssss;?;:
V
*J3 c^c?c?r?'«r?
4 ~* CN ~H CN s
^ O CN CM — < CN
• 3
03
O
E -a
m co m — • — i ^J S.'ra
j> !>!>!>!> & •§ |
ti_ in « to t5 ~ -. ^
iiiiiiil
*-, i_ o b o ^
j | .| .| i i |
a a 8 K g ~ o
'C 'C ° ° p •« "^
c £ S^ S^ 2P co J2
< •< £ £ £ is <<
a ^
11 Q
• "£ •£ j| •=!
S 5553 £> '5
53 ^ *S •§ *^ "S S
§ a '~ 'S -S «: S
5 5 ~ Jj ^ "5 '5
^ t§ ^ ^ ^ ^ ^
Sas^gSSg;?;^
22222 S 8222
8888888888
•o-o-a-a-o-a-o-a-o-o
wrteortracsecwnjco
aJO4>a3a3wa3a3oo
•5-g-g-g-g-g-g-g-g-g
Illlllllll
^ ^:
V
m oo «o
•^J- tn co
ts « «
V
VOCO^VOOOOOC*VO0
•-^c^oirn^^cs-^escNcN
V
4 1 -1 1 -1
j3S88j3jSe3j8(3t3
"s9'i:i'i^>^>^)^>J1
.2" •& .S* .S" hn ^^ Ja *-• *-* ^
•tS-faiSil^pcepc
w wa co w ~ C C C B C
.a -a -a -a -a
•§> J> J> I? J> j?
•a -a -a -a ^ "3 "S "» "S "a
^coaSSSSSS
^s5^S3Si22SiS;SJ|i
ccScSSSSSS
SS2S§§ac§§
lill^i§^(S(g^
a s a s a
.§.§§§ 2"
6"" 'o 'o 'o 'o
o o o o
•O *O T3 "O T3
1 1 1 1 1
1 1 1 1 1
VI VO t- CN 00
CN — i — . — . 0
c^ ut c^ 2
CN cK CN t/^ CO
m m en m
M §
i^ ^ & & _
O. O. O. O. O.
.1 .§ .1 .1 .i
C *k« • *JU *C *E
to « « w "w
g c§ g g g
j? ^ ^ j? ^
M 2 *2 £2 ^
S S 2 2 2
c c c c c
-S -S3 .%. .S3 .S3
bo fc$ tx) t>o bo
? § § g 1
S S S S S
w
'§
a.
CO
&^
o
«
•s
OT
S
CO
CO
fi
c«
B
o
'S
CD
S
O
O
-------
Appendix E. Explanation of the larval recruitment model and how it is used.
The model is available as a Microsoft® Excel file (APPEND-E.XLS). The input
parameter fields and the model output are on the sheet labeled "I-O". A sample of the I-O
sheet is shown in Figure E-l. The input parameters are divided into three categories: biology,
bioassay, and exposure. The model is run by inputting the necessary biological and bioassay
parameters, selecting the dissolved oxygen (D.O.) concentration (interval) to model, and then
iteratively assessing various exposure days until the "% Recruitment Impairment" is at or
near 5% (Figure E-l). Alternately, one can assess the expected impairment for a given site by
inputting the D.O. interval that represents the minimum for than site, and inputting the
number of days that the site is experiencing D.O. concentrations less than the CCC (4.8
mg/L). The second sheet in the Excel file is labeled "Model" and contains the calculation
fields for recruitment -without hypoxia and with hypoxia (sample outputs are shown in Tables
E-l and E-2, respectively).
Model Input Parameters
The recruitment model is a discrete time, density-independent model. The input
parameters for the model include: the length of the recruitment season (R), the duration of
larval development (L), the cohort initial abundance (N0), percentage of the initial cohort
exposed to low dissolved oxygen (p), the attrition rate without hypoxia (a), the duration of
the hypoxic event in days (E), and the minimum D.O. experienced during that event (DOmin).
Three exposure response models describe the following: late larval to megalopa survival vs.
D.O. concentration, larval survival vs. D.O. concentration, and delayed development vs. D.O.
concentration. These exposure response models are used for estimating recruitment under
hypoxic conditions. The general model presented for assessment of ecological risk, while
developed for use with hypoxia, is applicable to any type of time-variable environmental
stress. In addition, life history parameters of the model can be redefined to reflect site-
specific qualities or to describe another species of concern.
The cumulative impact of low D.O. on recruitment is expressed as a proportion of the
potential annual recruitment for the species of concern (Say mud crab, Dyspanopeus sayi, in
the current run of the model). The recruitment season (R) takes into account information hi
the literature from various Virginian Province locations. Consideration was given to capture
the period of predominant recruitment, rather than observance of the first and last dates for
zoeal presence in the water column. Peak larval abundance between June and September is
typical of brachyurana crustaceans in the Virginian Province (Hillman, 1964; Sandifer, 1973;
Dittel and Epifanio, 1982; Johnson, 1985; Jones and Epifanio, 1995). Settlement of D. sayi in
the megalopal stage is relatively continuous, and unrelated to lunar periods (van Montfrans,
et al., 1990). The larval season, or period of presence in the water column, chosen for the
running of the model for D. sayi is 66 days. This value is derived from a representative
hatching season of 45 days and a larval development time of 21 days. The development time
Appendix E: page E-l
-------
of 21 days was estimated from field data (Hillman, 1964), as well as from laboratory
observations made during EPA's D.O. testing with D. sayi.
Only one data set (Hillman, 1964) was available to represent natural attrition. It was
generated from a full season of weekly collections in Narragansett Bay, Rhode Island.
Mortality per day was estimated by applying the assumption that the observed densities of
each zoeal life stage represented the relative survivorship of each stage, and that the total
number of zoeal development days was 21. The rate of attrition, 7.8% loss per day, is the
exponential loss constant based on the best fit to these data.
The model assumes that only 75% of the available mud crabs are exposed to low D.O.
on any given day (i.e., the other 25% remain above the pycnocline). This assumption is based
on observations of water column position of these larvae and the recognition of the
importance of observed vertical migration for estuarine retention of these larvae (Hillman,
1964; Sandifer 1973,1975). The choice to apply the 75% lower water column distribution to
all stages is a conservative assumption, which particularly emphasizes risk hi the more
sensitive later stages. A general assumption regarding vertical (and horizontal) distribution is
that zoea do not successfully avoid hypoxia.
For each individual run of the model, the exposure input parameters are limited to one
exposure duration and one D.O. concentration. The conservative approach to deriving the
exposure parameters used by the model is to treat each D.O. time series as a number of
intervals of D.O. less than the CCC (see main text for a more detailed explanation). This
approach defines the duration as the total number of sequential days of hypoxia for each
interval. The D.O. value for the model is the minimum D.O. concentration that occurs during
that duration of time.
Bioassay input parameters are presented and discussed in the main body of the D.O.
document, however, specific values used in the model are presented in Table E-3. The final
protective limit for larval survival that is presented in the D.O. document was derived
assuming the there was no delayed development (i.e., a value of 1.0 was used for each D.O.
exposure interval).
Model Assumptions
The creation of any model necessitates the use of simplifying assumptions that
introduce some limitations to the application of the model. A complete understanding of the
utility of the model output for a given set of circumstances requires an understanding of these
underlying assumptions. The model divides the recruitment period into 24 hr time periods.
The model assumes that a new cohort of larvae (those released within a 24 hr period) are
available each day of the recruitment period. This is a reasonable assumption for larval mud
crabs. However, if the model is adopted for use with species for which daily cohorts are not
available, then the model may be overprotective1. Under these conditions the model may
'The model applies a daily effect on each cohort that is included in the sum of effects for the total number of
days of exposure. If on a given day there is no new cohort, then there is no effect registered for that day (i.e., if
Appendix E: page E-2
-------
need to be modified for a different time period (e.g., weekly or lunar cycle). The model
assumes that there is no change in sensitivity for an individual zoeal larva (the "early life
stage" in the model) exposed to low D.O. for multiple days. In other words, the same 24 hr
dose response relationship is used for each day of exposure without any consideration as to
whether or not an available individual was exposed on the previous day. The model also
assumes that once a zoeal larva has made the development transition to megalopa, then there
is no further low D.O. effect (the model only applies the late larval to megalopa dose-
response curve for one 24 hr tune period).
The recruitment model has two assumptions with respect to duration of exposure to
hypoxia. First, the model assumes that exposure to low D.O. will not occur over the entire
recruitment season (R). The maximum number of days that low D.O. can exist in the model
is R-(L+1), or 44 days in the current run. Any exposure longer than 44 days gives the same
output as 44 days. This is not a serious limitation for the current D.O. protective limit for
larval survival because the protective limit essentially reaches an asymptote at around 30
days (Figure 7, main text). Second, the model was developed with the maximum number of
exposure days equal to the length of the development period of the modeled species (21 days
in the current run). This only affects the zoeal life stage portion of the model. If exposure
exceeds 21 days, then the model behaves as if there were delayed development and the output
is not as accurate and is slightly over protective. The inaccuracy associated with exposures
longer than 21 days does not show up under conditions that allow for such long exposures to
low D.O. (i.e., which keep the percentage impairment at or below 5%). This is demonstrated
in Figure E-2.
An implicit assumption that the model makes is that the various over- and under-
protection issues more or less cancel each other out. The assumptions were necessary to
construct a reasonably simple and tractable model.
Model Equations
Recruitment under non-hypoxic conditions
In the non-hypoxic example, the number of recruits from each daily cohort is
expressed by the following equation:
Eq.l
NR represents the number of surviving recruits from the initial cohort, all other parameters are
as described in the above section on Model Input Parameters. The total number of recruits
for non-hypoxic conditions is then determined by summing NR for all daily cohorts.
there are no recruits, then there is nothing to have an effect on for that day). The model cannot apply a "zero':
effect to a cohort that does not exist.
Appendix E: page E-3
-------
Recruitment under hypoxic conditions
In the hypoxia example, the above equation is modified to account for D.O. effects on
megalopae, larvae, duration of larval period, and percentage of larvae exposed. These
modifications are performed using several intermediate calculations, but the overall equation
is:
Eq.2
The variable Sy represents the number of unexposed individuals from a cohort that survive to
recruitment during the hypoxic event. Note that this equation is Eq. 1 (the recruitment model
for non-exposed cohorts) multiplied by the proportion of the population that does not
experience hypoxic conditions.
Eq. 3
Su = (1 - p 1 100) * N0 (1 - a 1 100)L
The variable SH represents the number individuals from a cohort exposed to low D.O. that
survive to recruitment during the hypoxic event.
Eq.4
N0]*(l-a/10Q-)L'}
SDQ represents the total number of individuals from a daily cohort that survive the hypoxic
event. The equation describing SDO will follow below. L' represents the duration of larval
development modified by the developmental delay due to D.O. exposure. The equation
describing L' will follow below. All other variables have been described previously.
Eq.5
The variable ER surv represents the proportion of the cohort surviving at the selected D.O.
concentration using the laboratory exposure-response data (Table E2). Two possible exposure
response models can be used in the calculation of the survival of low D.O. conditions and
their selection is based on whether the individuals of a particular cohort experience the
hypoxic condition during the larval stage or the megalopa stage (Table El). All other
variables have been described previously.
Eq. 6
The variable ER delay represents the molt delay observed at the selected D.O.
concentration using the laboratory exposure-response data (Table E3). All other variables
used in this equation have been described previously.
Appendix E: page E-4
-------
As described for the non-hypoxic condition, the total number of recruits for hypoxic
conditions is determined by summing NR's for all daily cohorts. The percent recruitment
impairment due to hypoxic conditions is calculated as follows:
Eq. 7
O/T • i 1 — JVn(hypoxic) 4.1nr.
%lmpairment = — * 100
JY^J(non - hypoxic)
Appendix E: page E-5
-------
Table E-l. Calculation field from the recruitment model showing
summed to estimate the seasonal recruitment -without hypoxia. Only
(Table E-3).
Cohort
(W
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Days Exposed
(E)
0
0
1
2
3
4
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Molt During E?
N
N
Y
Y
Y
Y
Y
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N.
N
the results for
the biological
each cohort that are then
input parameters are used
No. Surviving D.O. Realized L No. Surviving »
Exposure (days) Attrition
100
100
100
• 100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17.
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
"18.17
18.17
18.17
Appendix E: page E-6
-------
Table E-2. Calculation field from the recruitment model showing the results for each cohort that are then
summed to estimate the seasonal recruitment -with hypoxia. The input parameters for this run are shown in
Table E-3.
Cohort
W
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Days Exposed
(E)
0
0
1
2
3
4
5
5
5
5
5
5
.5
5
5
5
5
5
5
5
5
5
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Molt During E?
N
N
Y
Y,
Y
Y
Y
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
No. Surviving D.O.
Exposure
100
100
50.425
50.425
50.425
50.425
50.425
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7
99.7-
99.7
99.7
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Realized L
(days)
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
21
No. Surviving
Attrition
18.17
18.17
9.16
9.16
9.16
9.16 '
9.16
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12
18.12 •
18.12
18.12
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
18.17
Appendix E: page E-7
-------
Table E-3. Bioassay input parameters used in the current run of the larval recruitment model. The final
protective limits for larval survival assumed no delayed development (i.e., Delayed Development set to 1.0 for
each dissolved oxygen interval).
D.O. Cone.
Interval
D.O. Minimum
(mg/L)
Survival—Larvae1
Survival—
Megalopae1
Delayed
Development
(fraction of
control)1
1
2
3
4
5
6
7
8
2.0
2.1
2.2
2.3
2.4
2.6
2.8
3.0
Data Set I
0.639
0.745
0.825
0.885
0.926
0.971
0.989
0.996
Data Set H
0.060
0.073
0.089
0.107
0.128
0.183
0.253
0.339
1.24
1.23
1.22
1.21
1.20
1.19
1.17
1.16
1
2
3
4
5
6
7
8
3.2
3.4
3.6
3.8
4.0
4.2
4.4
4.6
0.998
0.999
1.000
1.000
1.000
1.000
1.000
1.000
0.438
0.541
0.641
0.731
0.804
0.862
0.904
0.935
1.14
1.13
1.11
1.09
1.08
1.06
1.05
1.03
'From regression equations for Figure 5.
Appendix E: page E-8
-------
linput Parameters
J2
Q.
sn
&
(0
re
o
m
T3
CD
>i
CD
CD
Q
"co
'E
^s
CO
I
>
CO
o
Q
"CD
3
O
Q
-
-*-*
c:
t^_
g °^
c: c "o
Q. o c:
° "•§ c
CD ~ <">
Q
ZJ
.2
CO
05
us
_>>
CD
UJ
E
i
[c
« *
£ O
0)
D)
JS
CO
g,
CO
CO
--.
^
o
o
'
o
CO
o
0
"Jt
o>
•
>, CD
CD 2-
^-- *;
0 ®
8 E
CD a.
n* r^
Biological Inputs
Length of recruitment sc
Duration of larval deveic
o
o
'i
<<. * >
00
,°°.
; ~
-s
"CO~O> Tft
£. co o
O O T"'
o
"*
^^
O
^
»o* in
't**
*i
•
V
f1
CM
|
if
O
Initial cohort abundance
^\CO
o
^^
CM
co
h-»
T3
CD
8
Percentage of cohort ex
o
2~
u>
00
^OQ
'o
«">
w.
•*
CO
II
CD
?
5
I
c:
.0
<
o o
°. ^
S
-
«>
o <=>.
05O
"
^
•4
CM €0^ U> - *
* ^
8 j;
o> o
0/0
^~
"t
II
UJ
"t/r
>^
1 Exposure Inputs
Duration of hypoxia (da\
O
-
eo
o>
0
oo
CM
t*-
eo
II
Q
1
6
Q
o
> ^
~~ ^ T*
ftO
at
Ja>
Q
^
„
o
eo
*""
co
I
00
I
Q
S
CO
52
4-1
Q.
's
O
0
TS
0
o>
CO
h-
II
Ul
CO
"x
o
a.
s^
%
•*->
§
1
CD
"CD
o
CO
co
CD
CO
CO
CO
CO
II
•z
.CD
O
a.
^*
^_,
o
i
•4-'
0
'S
b
"CD
o
co
CD
CO
IO
IO
a>
|
'«.
Q.
E
a>
|
2
O
s-
o
IS
I
1
w
-------
80
70
60.
C? 50.
I 40 J
-------
Appendix F. Justification for treating transition to megalopa as a more
sensitive early life period than zoea life stages. Data are summarized for
both the Say mud crab Dyspanopeus sayi and the rock crab
Cancer irroratus.
Appendix F: page F-l
-------
Say mud crab (Dyspanopeus say/)
10Ch
80-
60-
CO
40-I
20-
zoea
zoea 3-megalopa
0
0.0
1.0 2.0 3.0 4.0
Dissolved Oxygen (mg/L)
5.0
Figure F-l. Comparison of effects of low dissolved oxygen on zoea and megalopa of the Say mud
crab Dyspanopeus sayi. Zoea data are from five tests with a four day duration. Megalopa data are from
2 tests with durations of 8 and 10 days. The test began with stage 3 zoea and the longer exposure times
were required for test animals to molt to megalopae. Observations made during the tests suggested that
most of the mortality occurred during the transition to megalopa. For the zoea data, the point is the
mean, the box the standard deviation and the line the range. Data are from this study.
Appendix F: page F-2
-------
Atlantic rock crab (Cancer irroratus)
100
80-
60H
ca
40-
20-
zoea
zoea 5-megalopa
0
O.O 1.0 2.0 3.0 4.0 5.0 6.
Dissolved Oxygen (mg/L)
Figure F-2. Comparison of effects of low dissolved oxygen on zoea and megalopa of the rock crab
Cancer irroratus. Data are -from four zoea tests each with a four day duration, one zoea 5 to megalopa
test of four day duration, and one megalopa test of seven day duration. The longer durations were
necessary to allow sufficient time for individuals to molt to megalopae. Observations made during the
tests suggested that most of the mortality occurred during the transition to megalopa For the zoea
tests, the point is the mean and the line is the range. Data are from this study.
Appendix F: page F-3
-------
This page intentionally left blank.
-------
Appendix G. Time-to-death curves used to generate the regressions in
Figures 9A and 9B.
Appendix G: page G-l
-------
Dyspanopeus sayi
2.5.
2.0.
a
f 1.5.
I
o
to
CO
s
1.0.
0.5.
0.0
LT10: y = 0.511 Ln(x) + 0.892
r2=0.98
LT25: y = 0.493Ln(x) + 0.672
r2=0.99
750: y= 0.291 Ln(x)+0.568
LT90: y=0.378Ln(x)+0.329
r2=0.79
10
Time (hr)
15
20
Figure G-l. Time-to-death curves for LT10, LT25, LT50 and LT90 for larvae of the Say mud crab
Dyspanopeus sayi exposed to low dissolved oxygen. Data are from this study. Solid lines are
logarithmic regressions of the four data sets. Regressions were calculated using Microsoft® Excel 5.0.
Appendix G: page G-2
-------
-J
"S
o
D)
I
13
>
O
CD
CO
2.5
2.0
1.5.
1.0.
0.5.
0.0
Palaemonetes vulgaris
LT10: y = 0.449Ln(x) + 0.856
r2=0.94
LT25: y = 0.299Ln(x)+0.881
r2=0.87
LT50: y = 0.287Ln(x) + 0.819
LT90: y = 0.268Ln(x) + 0.750 r2=0.89
r2=0.97
10
15
20
25
Time (hr)
Figure G-2. Time-to-death curves for LT10, LT25, LT50 and LT90 for larvae of the marsh grass
shrimp Palaemonetes vulgaris exposed to low dissolved oxygen. Data are this study. Solid lines are
logarithmic regressions of the four data sets. Regressions were calculated using Microsoft® Excel 5.0.
Appendix G: page G-3
-------
Homarus americanus
4.0
3.5
~ 3.0.
1 2.5.
g
O!
•O
| U
o
w
S 1.0.
0.5.
0.0
LT10: y = 0.487Ln(x) +1.827
r2=0.71
LT50: y = 0.363Ln(x) + 1.413
r2=0.91
LT90: y = 0.255Ln(x) +1.340
r2=0.83
10 15
Time (hr)
20
25
Figure G-3. Time-to-death curves for LT10, LT25, LT50 and LT90 for larvae of the American lobster
Homarus americanus exposed to low dissolved oxygen. Data are from this study. Solid lines are
logarithmic regressions of the four data sets. Regressions were calculated using Microsoft® Excel 5.0.
Appendix G: page G-4
-------
1.6
1.4
J 1.2.
0)
f 1.0J
0)
I 0.8 J
O 1
| 0.6.
"5
i8 0,
5
0.2.
0.0
O)
e
CD
O>
0
0.9.
0.8.
0.7.
0.6.
0.5
0.4.
§ 0.3.
1 0.2.]
Q
0.1 J
0.0
B
BreVoortia tyrannus
LT5: y=0.156Ln(x)+0.804
R2=0.91
LT50: y=0.084Ln(x)+0.618
R2 = 0.96
LT95: y=l).043Ln(x)+0.469
R2 = 0.90
20
40
60
80
100
Time (hr)
Leiostomus xanthurus
LT5: y=0.058Ln(x)+0.542
R2 = 0.96
LC95: y = 0.043Ln(x) + 0.421
R2 = 0.95
20 40 60 80 100
Time (hr)
Figure G-4. Time-to-death curves for LT5, LT50 and LT90 for juveniles of the saltwater fish Atlantic
menhaden Brevoortia tyrannus (A) and spot Leiostomus xanthurus (B) exposed to low dissolved
oxygen. Data are from Burton et al., 1980. Solid lines are logarithmic regressions of the four data sets.
Regressions were calculated using Microsoft® Excel 5.0.
Appendix G: page G-5
-------
Salvelinus fontinalis -small fingerlings
§
B>
I
•D
>
O
CO
CO
5
1.6-
1.4.
1.2.
1.0 .
0.8.
0.6.
0.4.
0.2.
0.0
y = 0.204Ln(x) + 0.97
R2 = 0.97
y= 0.21 OLn(x)+0.844
= 0.91
1 "~A
y= 0.150Ln(x)+0.884
R2 = 0.97
x
y = 0.111Ln(x)+0.907
R2 = 0.91
y = 0.083Ln(x) + 0.880
R2 = 0.85
10
15
Time (hr)
20
25
30
Salvelinus fontinalis - fry
y = 0.235Ln(x) +1.177
R2 = 0.95
•y=0.207Ln(x)-H.013
R2 = 0.87
Figure G-5. Time-to-death curves for LT50s of small fingerlings (A) and fry (B) of the freshwater
brook trout Salvelinus fontinalis acclimated to different concentrations of low dissolved oxygen and
then exposed to different concentrations of low D.O. Data are from Shepard, 1955. Solid lines are
logarithmic regressions of the four data sets. Regressions were calculated using Microsoft® Excel 5.0.
Appendix G: page G-6
-------
Salvelinus fontinalis - large fingerlings
o>
§
c
o
O)
I
"5
2.0
1.8
1.6
1.4.
1.2.
1.0.
0.8.
0.6.
0.4.
0.2.
0.0
y = 0.345Ln(x) + 0.773
R2 = 0.990 O
y = 0.260Ln(x) + 0.709
R2 = 0.945
y = 0.200Ln(x) + 0.645
/ R2 =0.984
y = 0.180Ln(x) + 0.606
R2 = 0.963
y = 0.165Ln(x) +0.616
1^ = 0.913
10 15 20
Time (hr)
25
30
I
o
•o
o
"5
-------
This page intentionally left blank.
-------
Appendix H. Growth data for constant versus cyclic exposure to low dissolved oxygen (Coiro, et al., 1999).
Species
Dyspanopeus sayi
Dyspanopeits sayi
Dyspanopeus sayi
Dyspanopeus sayi
Dyspanopeus sayi
Dyspanopeus sayi
Dyspanopeus sayi
Dyspanopeus sayi
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Paralichthys dentatus
Paralichthys dentatus
Paralichthys dentatus
Paralichthys dentatus
Paralichthys dentatus
Paralichthys dentatus
Lifestage
larva!
larval
larval
larval
larval
larval
larval
larval
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
newly hatched
juvenile
juvenile
juvenile
juvenile
juvenile
juvenile
juvenile
juvenile
juvenile
juvenile
juvenile
juvenile
Cycle (mg/L)
4.5-sat.
3.6-sat.
2.6-sat.
1.5-sat
4.2
3.4
2.4
1.6
1.9-sat.
1.6-sat.
1.9
1.6
2.2-sat.
1.7-sat.
2.3
1.9
3.0-sat.
2.2-sat.
3.2
2.3
2.8-sat
2.6
3.2-sat.
2.1-sat.
1.8-sat.
3.4
2.3
1.8
3.7-sat.
2.5-sat.
1.5-sat.
3.5
2.5
1-5
1.8-4.4
1.8
2.2-7.2
1.8-7.2
2.3
1.8
Cycle Duration Test Duration
(hr) (days)
6 low/6 hi
6 low/6 hi
6 low/6 hi
6 low/6 hi
constant
constant
constant
constant
6 low/6 hi
6 low/6 hi
constant
constant
6 low/6 hi
6 Iow/6 hi
constant
constant
12 low/12 hi
12 low/12 hi
constant
constant
6 low/6 hi
constant
12 low/12 hi
12 low/12 hi
12 low/12 hi
constant
constant
constant
6 low/6 hi
6 low/6 hi
6 low/6 hi
constant
constant
constant
6 low/6 hi
constant
6 low/6 hi
6 low/6 hi
constant
constant
7
7
7
7
7
7
7
7
4
4'
4
4
8
8
8
8
8
8
8
8
7
7
8
8
8
8
8
8
14
14
14
14
14
14
10
10
14
14
14
14
D.O.
Minimum
(mg/L)
4.5
3.6
2.6
1.5
42
3.4
2.4
1.6
1.9
1.6
1.9
1.6
2.3
1.7
2.3
1.9
3
2.2
3.2
2.3
2.8
2.6
3.3
2.2
1.8
3.4
2.3
1.8
3.7
2.5
1.5
3.5
2.5
1.5
1.8
1.8
2.2
1.8
2.3
1.8
% Reduction
in Growth
6
30
49
89
33
51
53
90
36
59
67
78
36
56
46
66
25
41
28
60
35
51
15
5l'
69
21
56
75
0
13
55
3
13
64
35
45
18
31
33
47
Appendix H: page H-l
-------
This page intentionally left blank.
-------
Appendix I. Comparison of Say mud crab growth effects with other saltwater
species. Data are from this study.
Appendix I: page 1-1
-------
g
I
I—I
5
I
O
So
03
&
P-l
100 -
90 •
80-
70-
60-
50-
40-
30-
20-
10-
^
\
\
\
o>
0.0 1.0 2.0 3.0
Dissolved Oxygen (mg/L.)
• \
4.0
5.0
6.0
Figure 1-1. Plot of growth (percentage impairment realtive to control) for several species of
saltwater animals. The Say mud crab (Dyspanopeus sayi—bold solid line) is among the most
sensitive tested. Experimental conditions are listed in Table 1-1.
Appendix I: page 1-2
-------
a
S
.S
I
I
•I
•I
•3
e
g
ex
W
S
£
2 S
•3 S3 S
§^
0.
S
e
o
O
«s
e
o
U
u
K
O.
cc
""» O C>OO(=>
sgjsaoS'
OO
OO
'
CN
OOOO
vocsooo
CN«NCN«N-<— •
O
$
o\
$
1
w-> ^
cS
00
U
to
0,
W VW WW I I \_^
•a «> g> - « -a
t f ^f m to » „ m
CO 00 IU "S ~S bO too toD > > tj)
grocncScoCN ^)l* S)
o. o. o, a.
•O T3 -O 13
"I f f "
11
II I i i
i
1 1 1
'g
&, fx CL, a, a, &< a,
s e s s s s s
1~~''~~'c!aSG w w cnc/3 w
^^0>2SS2S;H
S S 2 w>cuoej)W)ooej)
s sll ttttttdtt
&Q CQ &3 C^ , C*3 Co ^3 Co
11111111
OO^JOOOOO
""««
1 1 1 1 1 I I I I
EO GO EO c»a • c*a co 03
-------
This page intentionally left blank.
-------
vae
to low olved oxygen. Data are segregated into juvenile/adult and
with the different protection limits.
iss
on
X
g »>
S 'C
*e ra
El-
tn «
CM EfS)
CO CO CO
§
1
a
ou
— ^ -? 5->
t— t- f- r- r-
r-1
S ' '
SSSS2S2SS2SSSS2SS22
0.0 O-Q Q
S 8 8 3 3 3 -
•o •§
& a
3333333333
OUOOVU4JOOO42OOOQJOOVOQJ
•>'-'"'''*'*'"''
..
S -S' -S -S .2 -S -S -S -S -S -S -S -S .2 .2 .2 -S -S -S -S ^ ^ J3 §
-------
a
ra
o,
•a
•»
*e3 "3 •§•
HI
E E 5
£ SS 8
3 O O
5 * £
S CN CS
i i i
*- *- «
"3 3 3
3 -O *O
« cs ca
a "ra "ca
3 o o
L> O O
§ § g
"S "5 "S
III
*O «O »3
U >
•s ^
11
S -a
S g
ob g
a
1
a §•
I I
R co"
§ i
•S OO
1 1
o o
r*I
ON
ON
•"• *
C
1
Baker and
•n
o
0.&
H I
ts
?3
JO
'c
I
fe
1
§
1
.1
*5n
•~
Q
1
1
ON
ON
•••«
C
1
Baker and
V)
_C
^
^o
1
•o
S 00
cs
?3
0
*c
.2,
u.
Ho
1
1
1
Kn
•ft
§
S
1
6
.0
ON
ON
i— *
C
i
Baker and
ON
c
n
.2 *2
11
11
j=
^r
cs
o
eN
tN
"o
CO
1
c
(U
e _^
ss-
II
0
I
1
U
•§
S;
*^
ts
I
1
o
2
?*>
—
£
"O
c
C3
1
VO
o
1
•*
C-l
00 00 S) 00
£) § 0 £ g
O 0 O O 0
«.*«««
o o o o o
2 2 — 2 12
000
1 1 1 1 1
& XI 13 !2 £3
•a-8-s**
§* O C3 S" ooooo«oooooooo
oooooSooouuoooo
**«««««««*««*««
o'o'ooooo'ooooooo'o*
°°°°°°
S *
o o
o o
0
S -a
<8 E
of vT
3 a
•o -o
o o
o o
0. 0.
So
u
•£2 -S2
s: JC
B g
0 0
1 1
b b
-------
4*
C
1
iii
<3 s
u
<2
e
^o
CB
S
O
n-
u
S
3
1/3
S
S
3
4»
[IIBU UOUIU
a
en
u
u
en
CS CS
ON ON
§ii
& s s
S3 S S
™ M kH
co EC ffi
•0-0*0
O IM t-.
"> g. £.
a f? &
> > >
VO tf VO
O O O
o o o
§£§
u*
J3 .c
^ VO VO
£ "H "3
1 u u
fc. C" «?
!£.•§•§
> O5 ffl
ON vo ON
OO VN *3-
0 0 0
O V> O
>n o ^j
I 1
3 3
V> ""
0 <=>
|