United States
Environmental Protection
Agency
Office of Water
4304
EPA-822-R-00-012
November 2000
&EPA
Ambient Aquatic Life
Water Quality Criteria
for Dissolved Oxygen
(Saltwater): Cape Cod to
Cape Hatteras
-------
-------
Ambient Aquatic Life Water Quality Criteria
for Dissolved Oxygen (Saltwater):
Cape Cod to Cape Hatteras
November 2000
U.S. Environmental Protection Agency
Office of Water
Office of Science and Technology
Washington, DC
Office of Research and Development
National Health and Environmental Effects Research Laboratory
Atlantic Ecology Division
Narragansett, Rhode Island
-------
-------
Notices
This document has been reviewed by the Atlantic Ecology Division, Narragansett,
RI (Office of Research and Development) and the Office of Science and Technology
(Office of Water), U.S. Environmental Protection Agency, and approved for publication.
Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
Acknowledgments
This document was written by Glen Thursby, Don Miller, Sherry Poucher (Science
Applications International Corporation), Lauro Coiro, Wayne Munns, and Timothy
Gleason. Comments on two earlier versions of this document by Richard Batiuk (EPA's
Chesapeake Bay Program), Charles Delos, Keith Sappington (both from EPA's Office of
Water), and Walter Berry, Wayne Davis, and Diane Nacci (all from EPA's Atlantic
Ecology Division) improved the contents of the current version. The current version also
addresses comments by six peer reviewers. These include Larry Brooke, Daniel Call,
Gary Chapman, William Collins and Tyler Linton of the Great Lakes Environmental
Center (GLEC), Traverse City, MI, and Stephan Jordan from the Maryland Department of
Natural Resources. Useful discussions on several aspects of the final criteria also were
held with David J. Hansen of GLEC. Several individuals were involved with the
successful completion of many of the bioassays conducted at EPA's Atlantic Ecology
Division. These include Steven Rego, Kathy Simmonin, and Nan Hayden. Kenneth A.
Rahn provided valuable editorial comments for the final version.
in
-------
IV
-------
Executive Summary
This document recommends an approach to deriving the lower limits of dissolved
oxygen (DO) necessary to protect coastal and estuarine animals in the Virginian Province
(Cape Cod, MA, to Cape Hatteras, NC). The information on hypoxic effects used here
was obtained from studies conducted by the USEPA's Atlantic Ecology Division
specifically for this purpose, and from all other available reports applicable to hypoxic
issues of the Virginian Province. Hypoxia is defined here as concentrations of DO that
are below saturation. Literature on the effects of anoxia, while applicable to certain
ecological risk analyses, was not included in this document. This approach combines
features of traditional water quality criteria with a new biological framework that
integrates time (replacing the concept of an averaging period) and establishes separate
criteria for different life stages (larvae versus juveniles and adults). Where practical, data
were selected and analyzed in a manner consistent with the Guidelines for Deriving
Numerical National Water Quality Criteria for the Protection of Aquatic Organisms and
Their Uses (Stephan et al., 1985). This document considers how to protect three aspects
of biological health: survival of juveniles and adults, growth, and larval recruitment
(estimated with a generic model).
The recommended criteria described here apply to both continuous (persistent) and
cyclic (diel, tidal, or episodic) hypoxia. If the DO exceeds the chronic protective value
for growth (4.8 mg/L), the site meets objectives for protection. If the DO is below the
limit for juvenile and adult survival (2.3 mg/L), the site does not meet objectives for
protection. When the DO is between these values, the site requires evaluation of duration
and intensity of hypoxia to determine suitability of habitat for the larval recruitment
objective.
The limits identified are based entirely on laboratory findings but are supported in
part by field observations. For example, juvenile and adult animals showed field acute
effects at <2.0 mg/L, below the limit of 2.3 mg/L for juveniles and adults. Also,
behavioral effects were generally seen in the range of laboratory sublethal effects.
Unfortunately, however, no field observations are available for survival and growth of
larvae that are sensitive to hypoxia. This type of information is critical because two of
the three criteria are derived from laboratory responses of larvae.
Hypoxia as a stressor differs from chemical toxicants in that it can occur naturally
and because it is not controlled directly, whereas toxic chemicals are. Instead, hypoxia is
regulated primarily by controlling nutrients (largely nitrogen) and other oxygen-
demanding wastes. Criteria for DO may be used appropriately in a risk assessment
framework. The limits presented by the approach outlined here can be easily used to
compare the abilities of different areas to support aquatic life. Environmental managers
can determine which sites need the most attention, and how hypoxic problems vary in
time and space from one year to the next. Finally, environmental planners can make
better cost-benefit decisions by using this approach to evaluate how various management
scenarios will improve conditions.
-------
VI
-------
Contents
Executive Summary v
Lists of Tables and Figures viii
Introduction 1
Overview of the Problem 3
Biological Effects of Low Dissolved Oxygen 4
Overview of the Approach 5
Persistent Exposure to Low Dissolved Oxygen 6
Juvenile and Adult Survival 6
Growth Effects 7
Larval Recruitment Effects 11
Application of Persistent Exposure Criteria 17
Less Than 24 Hr Episodic and Cyclic Exposure to Low Dissolved Oxygen 19
Cyclic Juvenile and Adult Survival 19
Cyclic Growth Effects 20
Cyclic Larval Recruitment Effects 25
Other Laboratory Bioassay Data 28
Laboratory Observed Behavioral Effects of Hypoxia 30
Observed Field Effects 32
DataNot Used 35
Virginian Province Criteria 36
Implementation 39
References 43
List of Appendices
Appendix A. Comparison of 24 Hr and 96 Hr Acute Sensitivity to
Low Dissolved Oxygen for Saltwater Animals A-l
Appendix B. Acute Sensitivity of Juvenile and Adult Saltwater
Animals to Low Dissolved Oxygen B-l
Appendix C. "Chronic" Sensitivity of Saltwater Animals to
Low Dissolved Oxygen C-l
Appendix D. Acute Sensitivity of Larval Saltwater Animals to Low
Dissolved Oxygen at 24 Hr and 96 Hr D-l
Appendix E. Explanation of Larval Recruitment Model and
How It Is Used E-l
Appendix F. Sensitivity Analysis of Larval Recruitment Model F-l
Appendix G. Time-to-Death Curves Used to Generate the Regressions
in Figures 9A and 9B G-l
Appendix H. Growth Data for Constant Versus Cyclic Exposure to
Low Dissolved Oxygen H-l
Appendix I. Comparison of American Lobster Growth Effects with
Other Saltwater Species 1-1
Appendix J. Other Data on the Sensitivity of Saltwater Animals to
Low Dissolved Oxygen J-l
vii
-------
List of Tables
Table 1. Acute sensitivity of juvenile and adult saltwater animals to low
dissolved oxygen 8
Table 2. Effects of low dissolved oxygen on growth of saltwater animals 10
Table 3. Dissolved oxygen and duration data from a hypothetical persistent
time series (Figure 8) 19
Table 4. Dissolved oxygen and duration data from a hypothetical cyclic time
series (Figure 13) 25
Table 5. Dissolved oxygen and duration data from the intervals selected from the
hypothetical cyclic time series in Figure 15 27
Table 6. Summary of Virginian Province saltwater dissolved oxygen criteria 37
List of Figures
Figure 1. Relationship between 24 and 96 hr LC50 values for juvenile
saltwater animals exposed to continuous low DO 6
Figure 2. Plot of low DO effect (GMAVs for LC50s) against percentile rank of each
value in the data set 9
Figure 3. Plot of low DO effect (GMCVs for growth) against percentile rank of each
value in the data set 12
Figure 4. Plot of the GMAV data from Figure 2 along with 24 hr and 96 hr
LC50 values for larval life stages of various saltwater animals 13
Figure 5. Twenty-four hr dose-response curves for nine genera used in the larval
recruitment model 15
Figure 6. Plot of model outputs that protect against greater than 5%
cumulative impairment of recruitment 16
Figure 7. Plot of the final criteria for saltwater animals continuously
exposed to low DO , 17
Figure 8. A hypothetical representative DO time series for one site 18
Figure 9. Slope (A) and intercept (B) versus low DO effect values at 24 hr from
time-to-death (TTD) curves 21
Figure 10. Criterion for juvenile saltwater animals exposed to low DO for
24 hr or less 22
Figure 11. Plot of test results from growth experiments pairing constant low
DO exposure with exposures to various cycles of low DO and
concentrations above the CCC 22
Figure 12. Plot of dose-response data for growth reduction in American lobster
(Homarus americanus) exposed to various continuous low DO
concentrations 24
Figure 13. • A hypothetical representative DO time series for one cycle..- 24
Figure 14. Time-to-death (TTD) curves generated for the Final Survival
Curve "genus" 26
Figure 15. The same hypothetical DO time series as Figure 13 26
Figure 16. The DO minima and the durations listed in Table 5 superimposed
on Figure 14 27
viii
-------
Figure 17. A plot that combines the information from Figures 5 and 6 into
a single cyclic translator.^ convert expected daily mortality from
cyclic exposures into allowable number of days of those cycles 28
Figure 18. A plot of the other juvenile/adult mortality data from Appendix J
along with the proposed DO criteria for juvenile/adult survival 29
Figure 19. A plot of the other larval survival data from Appendix J 31
IX
-------
-------
Introduction
This document provides guidance to States and Tribes authorized to establish water
quality standards under the Clean Water Act (CWA) concerning dissolved oxygen (DO)
values that protect aquatic life from acute and chronic effects. Under the CWA, States
and Tribes are to establish water quality criteria to protect designated uses. While this
document constitutes the U.S. Environmental Protection Agency's (EPA's) scientific
recommendations regarding ambient concentrations of dissolved oxygen that protect
saltwater aquatic life in the Virginian Province, this document does not substitute for the
CWA or EPA's regulations, nor is it a regulation itself. Thus, it cannot impose legally
binding requirements on EPA, States, Tribes, or the regulated community, and may not
apply to a particular situation based upon the circumstances. State and Tribal
decisionmakers retain the discretion to adopt approaches on a case-by-case basis that
differ from this guidance when appropriate. EPA may change this guidance in the future.
Section 304 (a)(2) of the CWA 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." EPA has not previously issued
saltwater criteria for DO because the available information on effects was insufficient.
This document is the result of a 10-year research effort to produce the required
information to support the development of saltwater DO criteria. During that effort there
were several technical work group meetings involving stakeholders and external scientists
that helped to guide the process. The criteria presented herein represent the best
estimates, based on the available data, of DO concentrations necessary to protect aquatic
life and its uses.
These water quality criteria recommendations apply to coastal waters (waters within
territorial seas, defined as within 3 miles from shore under Section 502(8) of the CWA)
of the Virginian Province (southern Cape Cod to Cape Hatteras). However, with
appropriate modification, they may be applied to other coastal regions of the United
States. The document provides the information necessary for environmental planners and
regulators 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
address 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, however; 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 criteria 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 reduced concentrations, or in systems with natural high demands for oxygen.
-------
In addition, effects of hypoxia1 may vary latitudinally, or site-specifically, particularly as
reproductive 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
milligrams per liter (mg/L). Unlike the freshwater document, which provides limits for
DO in both warm and cold water, criteria are presented for warm saltwater only because
hypoxia in Virginian Province coastal waters is restricted primarily to the warm water of
summer. However, these warm-water limits can be considered protective for colder times
of the year. 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.
The saltwater DO criteria described herein were derived using the Guidelines2 and
are intended to maintain and support aquatic life communities and their designated uses.
Although the criteria are intended to protect aquatic communities, 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 significant long-term effect on the population. Hence, an assessment
of criteria should preferably 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 exposed. 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 particular individual juvenile or adult. With this in mind,
the saltwater criteria for DO segregate effects on juveniles and adults from those on
larvae. The survival data on the sensitivity of the former are handled in a traditional
Guidelines manner. The cumulative effects of low DO on larval recruitment to the
juvenile life stage, on the other hand, address survival effects on larvae. The DO
approach presented here uses a mathematical model to evaluate the effect on larvae by
tracking intensity and duration effects across the larval recruitment season. The model is
used to generate a DO criterion for larval survival as a function of time. It is
recommended that the parameters for this model be evaluated and adjusted where
necessary to meet site-specific conditions, especially those for length of recruitment
season and larval development time.
For the reasons listed above, the approach recommended in this document to derive
DO criteria for saltwater animals deviates from EPA's traditional approach for toxic
chemicals outlined in the Guidelines. Where practical, however, data selection and
analytical procedures are consistent with the Guidelines. Therefore, some of the
'Hypoxia is defined in this document as the reduction of DO concentrations below air saturation.
Guidelines for Deriving Numerical National Water Quality Criteria for the Protection of Aquatic
Organisms and Their Uses (Stephan et al., 1985—hereafter referred to as the Guidelines).
2
-------
terminology and the calculation procedures are the same. Thus, knowing the Guidelines
is useful (but not essential) 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
concentration (CCC).
Overview of the Problem
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
degree to DO concentrations less than 5 mg/L (Strobel et al., 1995). EMAP has also
generated 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
emphasize 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 have been lacking.
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
seasonal or die!. Seasonal hypoxia often develops as stratified water prevents the
oxygenated surface water from mixing downward. Low DO then appears in 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,
oxygen-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
Although in the case of dissolved oxygen, CMC is more appropriately defined as the criterion
minimum concentration.
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.
-------
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
returns 1 or 2 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
example 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
Chesapeake 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 etal., 1992).
Episodic hypoxia has been noted in shoal waters of mid-Chesapeake Bay (Breitburg,
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,
particularly during calm and cloudy weather, when photosynthesis is slow and daytime
reoxygenation is reduced.
Biological Effects of Low Dissolved Oxygen
Oxygen is essential in aerobic organisms for the electron transport system of
mitochondria. Oxygen insufficiency at the mitochondria results in reduction in cellular
energy and a subsequent loss of ion balance in cellular and circulatory fluids. If oxygen
insufficiency persists, death will ultimately occur, although some aerobic animals also
possess anaerobic metabolic pathways, which can delay lethality for short time periods
(minutes to days). Anaerobiosis is well developed in some benthic animals, such as
bivalve molluscs and polychaetes, but not in other groups, like fish and crustaceans
(Hammen,. 1976). There is no evidence that any free-living animal inhabiting coastal or
estuarine 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, that
is, 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
oxygen. Such reduced 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 locomotion 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, for example, intertidal and tide pool animals (McMahon, 1988) and
burrowing animals, which partly explains their reported high tolerance to low DO. la
contrast, 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; 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
exposures longer than 24 hr. It is followed by sections on criteria for exposures of less
than 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 FAV calculation procedures outlined in the Guidelines
(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
exposures 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
bioassays using larvae). This FCV is calculated in the same manner as the
FAV for juvenile and adult survival. This threshold limit as currently
presented has no time component (it can be applied to exposures of any
duration). 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 recruitment was set at 5% (although other percentages also may be
-------
appropriate on a site-specific basis), 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 maximized
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, Other Laboratory Bioassay Data.
Juveniles Only
1.4 1.6 1.8 2.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
24hrLC50(mg/L)
Figure 1. Relationship between 24 and 96 hr LC50 values for juvenile saltwater animals exposed to
continuous low DO. Each point represents a paired set of values calculated from the same test run. The
line drawn represents a one-to-one relationship. Data for the plot are summarized by species in Appendix
A. Appendix A also contains data for test runs with larvae.
-------
Data on the acute sensitivity of juvenile and adult saltwater animals to low DO are
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 GMAVs range from
<0.34 mg/L for the green crab, Carcinus maenas, to 1.63 mg/L for the pipe fish,
Syngnaihusfuscus, a factor greater than 4.8. Juvenile fish are somewhat more sensitive
than juvenile crustaceans (Table 1; Figure 2). In fact, the four most sensitive 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
(Stephan et al., 1985). However, the procedures outlined in the Guidelines were created
for toxicants. Since DO behaves in a manner opposite to that of toxicants (i.e., the
greatest response is associated with the lowest concentrations), the calculation is reversed.
The FAV calculation is essentially a linear regression using the LC50 values for the four
most sensitive genera and their respective percentile ranks. The final FAV is the value
representing the 95th percentile genus,5 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 in traditional Water
Quality Criteria for toxicants.
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). Sublethal effects were evaluated with only growth data for two reasons. First,
growth is generally more sensitive than survival to low DO. There were only two
exceptions where survival was more sensitive to low DO than growth. One test was with
Dyspanopeus sayi; however, growth was the more sensitive endpoint in eight other tests
with this species (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 were 40%
mortality and a 24% reduction in growth at a DO concentration of 3.9 mg/L. This 24%
reduction in 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 standard calculation for toxicants in the Guidelines uses the fifth percentile. The 95th
percentile is used here because, unlike toxicants, DO effects decrease as the concentration of DO increases.
6The use of a ratio to adjust the FAV to a CMC is designed to estimate a negligible lethal effect
concentration 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 sufficient data available for low DO to use a factor specific to this stressor. There was not a
significant relationship between genus sensitivity and the LC5/LC50 ratio; therefore, all ratios were
included in the calculation of the final ratio.
7
-------
.
js
\o
o
o5
S
Ǥ
•o
o
oo
c
8
CO
.2
1
a
•o
1
V)
a
w
c
o
&
o
•a
->
3
£
<*
0
^
"
IBs
OT 1-3 O
V
. .
"s
o ^
M§
B si
W C
o u
t*M >
"*1 3
Common Name
green crab
Species
Cardnus maentis -
en in
vq tr
o o
r^ oo
0 0
en ^^ in
d d 0
en in
vq •*
o o
p": oq
o o
en — * in
rt in in
000
000
III
333
Atlantic surfclam
Harris mud crab
northern sea robin
Spisula solidissima •
Rithropanopells harrisii
Prionotus carolinus
V") vo r^
\o o
S 2
0 «
t~- o exi
in c^ oo
o' o' o
vo o
oo -^
t^ o cs
in r-- oo
0 0 O
aj o o
1 1 1
333
flat mud crab
spot
iautog
Eurypanopeus depressus
Leiostomus xantlmrus
Tautoga onitis
oo
in
CN
,_;
VO
o'
t~- en
en in
^ -:
Sr^
r-
— o
0 u
1 1
"a.
E
marsh grass shrimp
daggerblade grass shri
Palaemonetes vulgaris
Palaemonetes pugio
Ov °
oo
^
C~J
Ov °
0 0
V
OO
s
o\ oo
0 0
V
u n
S S
~
in
S N
(U 13
1 1
s s
•o
g- -^
Stenotomus chrysops
Americamysis bahia
CT\ O
^
f-H *"*
oo en
R ^
S °i
oo en
u u
4) §
3 3
striped bass
pipe fish
Morone saxatilis
Syngnathus fuscus
° —: «
• -* ~H ni
-------
1.8-1
1.64
1.6-
1.4-
I '
£. 1.2-
1 1.0-
1 '
|0.8-
j! 0.6-
0
»
0 0.4-
0.2-
0 0 •
i
•
0 *
Q
0 ° 1
o • • o 1 '
•°| T
A O
o
o
o
Y o Invertebrates
0 10 20 30 40 50 60 70 80 90
m
1
100
% Rank of GM AV
95%
Figure 2. Plot of low DO effect (GMAVs 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 Appendix B. The value
highlighted on the y-axis is the calculated FAV. This value is the LC50 that is
higher than the values for 95% of the tested genera. 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. Arrows refer to those values that are less thans.
The second reason for restricting sublethal effects to growth is that results are
available from only one saltwater test that measured reproductive effects. Data are
presented 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 reproduction for both fathead minnows (Brungs, 1971) and
Daphnia magna (Homer and Waller, 1983). Thus, DO limits that protect against growth
effects also may be protective for reproductive effects.
-------
U
§
I*
.a
a
Q
en
C3
•s
c
o
o
o
•o
.«
•B
z
I
CM" .
V
S
-• o d »n en «S rt u-j
oit^c^^cN^voxnvdc^cnmc^'^c^^Ti-r^^v^inr^Ti'int^for^^ocnt^^i-vdtoininr^in
II
O O
II
33338 &""> "i 323333333
3 2 V V .2,JS
6£MCV
-------
Data on the effects of hypoxia on growth are^ presented for 4 species offish and 7
species of invertebrates from a total of 36 tests. Sensitivity of growth to low DO has been
determined in only two standard 28-day tests that meet Guidelines requirements; the
above life cycle test with A. bahia and the above early life stage test with M. menidia.
Therefore, growth data from nonstandard tests (i.e., not life cycle, partial life cycle, or
early life stage tests) were used to augment the chronic database. These nonstandard tests
ranged from 4 to 29 days long. Data from short duration tests were included 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
compensatory 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 (Morrison, 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 for M. mercenaria and 6 to 10 days for M. edulis).
Overall 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
species in the Virginian Province.7 The consequences of reduced growth in the field,
however, are uncertain.
Larval Recruitment Effects
A generic model has been developed that evaluates the cumulative effects of stresses
on early life stages of aquatic organisms. Early life history information and
exposure-response relationships are integrated with duration and intensity of exposure to
provide an ecologically relevant measure of larval recruitment. There are existing
recruitment models for marine organisms (e.g., Ricker, 1954; Beverton and Holt, 1957).
However, these models address other processes such as parental stock size, population
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.
Larvae are more acutely sensitive to low DO than juveniles (Figure 4). A method is
provided that estimates how many days a given DO concentration can be tolerated
7However, the CCC represents the potential for an approximate 25% reduction in growth. The
CCC for growth is based on statistically significant differences that result in chronic values similar to IC25s
for growth of many organisms. IC25 values are listed as a part of Appendix C for four species of
crustaceans and two species of fish. The geometric mean of these values (by species) correlates with the
geometric mean of the chronic values. In fact, a CCC calculated using IC25 values is similar to the CCC
calculated using statistically significant differences.
11
-------
. _ 5.00 -
4.8
4.50-
4.00-
]? 3.50-
^^
>
y 3.00-
(5
I2'50'
0 2.00-
TJ
o 1.50-
en
a
1.00-
0.50-
n on -
o
•
•
0 0
• °
O I
^ w
9
w , ,
• Feh
o hvertebrates
0 10 20 30 40 50 60 70 80 90
% Rank of GMCV Qc
> °
100
ȣ
Figure 3. Plot of low DO effect (GMCVs for growth) against percentile rank of each
value in the data set. Percentile rank was adjusted 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 calculated FCV. This value is the chronic value that
is higher than the values for 95% of the species represented. 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. Arrows refer to less than and greater than
without causing unacceptable effects on total larval survival for the entire recruitment
season. This is accomplished with a larval recruitment model8 and applying biological
and hypoxic effect parameters for each species for which sufficient data are available.
The level of impairment to cumulative seasonal larval recruitment that has been selected
as acceptable is 5%. This does not mean that a population cannot withstand a greater
percentage effect with no significant effect on recruitment. Rather, the 5% means that
this level of effect should be insignificant relative to recruitment in the absence of
hypoxic events. Many juveniles will eventually be eaten as prey or otherwise harvested
as adults. The 5% impairment is intended to minimize the effect of hypoxia on the
ultimate fate of juveniles. On the other hand, this may not be the case for certain highly
sensitive species or populations that are already highly stressed, for example an
endangered species. This may also not be the case where there are other important
Once the larvae are "recruited" into the juvenile life stage, the juvenile survival criterion
established above is applied.
12
-------
3.0 n
2.5
i
•£• 2.0
O
§ 1.5
1.0-
0.5-
0.0
D
Larvae-96 hr
A A
Larvae-24 hr
.o
• °
0 o Juveniles
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 data for the juveniles are from Table 1. The data for the larvae are listed in
Appendix D. Data points are plotted as absolute values even though some are less
thans.
natural or anthropogenic stressors that contribute to a loss of the larval life stage. In such
situations, it may be that a 5% loss in larval recruitment from DO alone is not protective
enough, and environmental risk managers may need to evaluate the province-wide 5%
protection goal in light of their site-specific factors that may contribute to a cumulative
loss in seasonal larval recruitment. States and authorized Tribes may choose a different
level of acceptable impairment, but they must justify doing so and show that the new
level of impairment still protects and maintains designated uses.
The equations that compose the model and the major assumptions used in its
application are presented and explained in detail in Appendix E. The life history
parameters in the model include larval development time, larval season, attrition rate, and
vertical distribution. The magnitude of effects on recruitment is influenced by each of the
four life history parameters. 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 that fall within the window of
hypoxic stress. The third life history variable, natural attrition rate, gages the impact of
slower growth and development 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
13
-------
development translates directly to decreased recruitment. Finally, the vertical distribution
of larvae in the water column determines the percentage of larvae that would be exposed
to reduced DO under stratified conditions.
For the purpose of the Virginian Province criterion, certain simplifying assumptions
have been made. The recruitment model assumes that the period of low DO occurs
within the larval season (hypoxic events always begin at the end of the development time
of the first larval cohort), and that hypoxic days are contiguous. The Province-wide
application of the model also assumes that a new cohort occurs every day of the spawning
season, and that each cohort is equal in size. These assumptions can be easily modified
and the model rerun using site-specific information. The model does not require that a
fresh cohort be available every day. If the model is run "longhand" as presented in
Appendix E, then its use is very flexible. Successful calculation of the recruitment
impairment only requires knowing the total number of cohorts available during a
recruitment season (i.e., it does not matter whether they were created daily, weekly,
monthly, etc.) and whether a cohort is exposed to hypoxia. If necessary, one also could
use cohorts of various initial sizes. Assuming a fixed rate of cohort introduction and size
simplifies the calculation of the total number of cohorts and the calculation of hypoxic
effects on larval survival. The model application for the Virginian Province is further
simplified by assuming that none of the life history parameters change in response to
hypoxia. These parameters are only changed when a different species is modeled,
although, as with cohort frequency and size, they can be easily changed for a site-specific
application to adjust for latitudinal changes in life history requirements.
The dose-response data used in the model are presented in Figure 5. Data are
available for nine genera and represent 24 hr exposure responses, except for the Say mud
crab (D. sayi). These species were selected based in part on the ability to spawn and test
them in the laboratory. In addition, they represent a range of sensitivities to hypoxia by
water column species. The summary response curve for D. sayi represents the more
sensitive transition from zoea to megalopa. These tests were necessarily longer (7 to 11
days) than the other tests to allow sufficient time for development to megalopa. Although
some enhanced sensitivity in these tests may be from the longer exposures to low DO,
mortality also appeared to be primarily associated with the molt to megalopa (which
occurred over a 24 hr period for a given individual). When the model was run for
Dyspanopeus, the assumption was made 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 exposure. The model run for Dyspanopeus also includes a second,
less sensitive, dose-response curve for the early life history larval stage for non-megalopa
exposures of this species. Model runs for the other eight larval genera were conducted
using only one life history stage.
Also included in Figure 5 is a final survival curve (FSC). The data points in the
FSC are calculated in the same way that the FAVs and FCVs were calculated, using the
data from the four most sensitive genera (Cancer, Morone, Homarus, and Dyspanopeus).
14
-------
100
§
1
CO
5?
90-
80-
70-
60-
50-
40-
30-
20-
10-
Rnal Survival Curve
Ro= 0.122
L = 100
k = 0.021
—*— Menidia
—3K— Ralaemonetes
—I— Scianops
-•— Libinia
—n— &ythropanopeus
-o— Cancer
—0— Morone
—A— Homarus
A Dyspanopeus
0.5 1.0 1.5 2.0 2.5 3.0 3.5
Dissolved Oxygen (mg/L)
4.0
4.5
5.0
Figure 5. Twenty-four hr dose-response curves for nine genera used in the larval recruitment model.
Dark solid line is the regression line of best fit for the FSC. See text for explanation of FSC and of
P0, L, and k. The Solver routine in Microsoft® Excel 97 was used to determine P0 and k.
The FSC will be used later for establishing DO limits for larval survival during cyclic
exposures.
The results of the model runs for each genus9 are summarized in Figure 6. The
complete data along with the biological parameters used for each genus are presented as
part of Appendix E. For the purpose of the Virginian Province, many of the values for
the biological parameters were selected to be deliberately conservative. For example, we
have selected recruitment seasons and larval development times that more likely represent
the northern portion of the Province. To support site-specific applications, Appendix F
shows several examples of how recruitment curves would be expected to change based on
changes to the model's biological parameters. Lengths of recruitment season and larval
development are particularly important especially because they are expected to change
Each genus, except for Palaemonetes, is represented by only one species. Final criteria values
calculated using the 1985 Guidelines are based on genus mean values. Therefore, all references to final
calculated values use genus rather than species.
15
-------
D)
5.0 -|
4.5-
3.5-
2.5-
1 2.0 H
1.5 ^
1.0-
0.5 -
0.0
Rnal Recruitment Curve
Po = 2.80
L = 4.64
k = 0.0222
10 15 20 25 30 35 40 45 50 55
Time (days)
60
A Cancer
p Dyspanopeus
X Eurypanopeus
A Homarus
O Libinia
m Menidia
+ Morons
<> Palaemonetes
H- Scianops
Figure 6. Plot of model outputs that protect against greater than 5% cumulative impairment of
recruitment. Input parameters for each genus are explained in Appendix E. The solid line is the
regression line of best fit for the FRC. See text for explanation of FRC and of P0, L, and k. The Solver
routine in Microsoft® Excel 97 was used to determine P0, L, and k.
significantly with latitude. Recruitment season gets longer and development time often
shortens as one moves south. This combination can significantly shift a recruitment
curve down and to the right. For this reason, it is expected that the final recruitment
curve (FRC) presented here for the Virginian Province may be overprotective for many
sites. Therefore, FRCs using site-specific biological parameters are recommended.
An FRC was calculated in the same way as the FSC, using the four most sensitive
recruitment curves out of the nine available curves. The four most sensitive curves were
for the genera Morone, Homarus, Dyspanopeus, and Eurypanopeus. The equation for the
FRC (and the FSC in Figure 5) was derived 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(t) =
P0L
P0+e
-Lkt
(L-P0)
Equation 1
16
-------
For Figure 6, P(t) is the DO concentration atstime t, P0 is the y-intercept, and L is the
upper DO limit. P0 and L were 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 Figure 5, the variables t and L represent DO concentration and
the upper limit for survival (100%), respectively. In this latter case, L is always 100%,
because this is always the upper limit tor survival.
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) exposure 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
o.o -
5n
A cj
•"i" A.r\ -
t 1'°
E 15 J
SJ, o n .
O) w-U
5?
O 9 ^ •
•a
a)
> 90-
%
.J2 1 R -
5 ^
1 n -
o ^ -
nn -
A
if
Grov
^
s
vth
rf"
S
Lar
<***
val p
,— -*
opuli
• "
ation
— «=
sun
••rf— •
Aval
Juvenile survival
5 10 15 20 25 30 35 40 45 50 55 60 65 70
Exposure Time (days)
Figure 7. Plot of the final criteria for saltwater animals continuously exposed to low DO.
The upper dashed line is the CCC for growth. The lower dotted line is the CMC for
juvenile (and adult) survival, and the curve between the two is the FRC from Figure 6
representing protective for larval survival. All of the lines are truncated at 1 day. The
cyclic portion of the criteria addresses exposure less than 24 hr.
17
-------
juvenile survival and larval growth limits depends in part on 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 daily average DO. 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 intervals 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 interval, the number of
days is recorded that the DO is between the interval's limits. For example, in interval a,
the DO is below 4.8 mg/L and above 4.3 mg/L from July 13 through 18 and again from
July 23 through 25, for a total of 7 days. This number of days is then expressed as a
fraction of the total number of days that would be allowed for the DO minimum for each
interval. For interval a, the allowed number of days is 15 (using the FRC in 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 1 (as is the
case in our example), then the DO conditions do not meet the desired protective goal for
larval survival. If the sum is less than 1, then the protective goal has been met.
5.8
5- 5.3
"B>
f 4.8
§4.3
O
o> 3.8
"o
CD q Q
O5 O.O
2.8
2.3
CCC
— — — — — — — O)
O> D)
3 3
O 10
O>
I
O>
r— I— CM
o>
<
6
co
Figure 8. A hypothetical representative DO 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 recruitment. The DO
minimum and the duration for each interval are determined for each interval.
18
-------
Table 3. Dissolved oxygen and duration data from a hypothetical persistent time series (Figure 8).
Range (mg/L) 5 Np. Days No. Days Fraction of
Within Range Allowed Allowed
Interval Below Above
a 4.8
b 4.3
c 3.8
d 3.3
4.3
3.8
3.3
2.8
7
3
1
1
15
9
4
1
TOTAL
0.35
0.30
0.20
1.00
2.05
The Below and Above columns show the range of DO covered by each interval. Number of Days Within Range refers to the
duration that the observed DO 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 DO minimum of the interval. These fractions are totaled to
evaluate whether the larval survival protective goal has been met.
The current recruitment model is a first attempt at providing a method that
incorporates duration of exposure in the derivation of DO criteria. A model that could
integrate 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).
Less Than 24 Hr Episodic and Cyclic Exposure to
Low Dissolved Oxygen
The criteria for continuous exposure to low DO do not cover exposure 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 continuous CCC. The
approaches described for treatment of nonconstant (e.g., cyclic) conditions 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 that 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 conservative
estimate of the safe DO concentration for exposures less than 24 hr would be to simply
use 2.3 mg/L. However, TTD data indicate that this would be overprotective. Data are
available for two saltwater juvenile fish (Brevoortia tyrannus and Leiostomus xanthurus),
one freshwater juvenile fish (Salvelinus fontinalis), and three larval saltwater crustaceans
(D. sayi, Palaemonetes vulgaris, and Homarus americanus), providing a total of 33 TTD
19
-------
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 these 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
sensitivity 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 endpoints (e.g., LC5 is a 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
generate 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 Equation 2 to generate a criterion for saltwater juvenile animals for exposures less than
24 hr (Figure 10).
Cyclic Growth Effects
The CCC for continuous exposure was derived based on growth effects data (mostly
from bioassays on larvae, Table 2). The simplest way to determine effects from cyclic
exposure to low DO is to compare 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., 2000). 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 between the same low DO concentration and one that was
above the continuous CCC (usually 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 3 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
20
-------
0.6 -,
0.5 -
0.4 -
o 0.3 -I
0.2 -
0.1 -
0.0
0.0
2.0 -,
1.8 -
1.6 -
1.4 -
1.2 -
1.0 -
0.8 -
0.6 -
0.4
0.2 -
0.0
0.0
y = 0.191X - 0.064
R2 = 0.835
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Dissolved Oxygen Concentration
Causing Effect Observed at 24 hr (m g /L)
: 0.392X + 0.204
R2 = 0.678
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Dissolved Oxygen Concentration
Causing Effect Observed at 24 hr (mg /L)
Figure 9. Slope (A) and intercept (B) versus low DO effect values at 24 hr from
time-to-death (I'll)) 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.
21
-------
2.5
2.0-
1z
o
1.5-
1
8
w
1.0-
0 0.5-
0.0
Tlme-to-"CMC"
y = 0.370Ln(x) +1.095
8 12 16
Exposure Time (hr)
20
24
Figure 10. Criterion for juvenile saltwater animals exposed to low DO for 24 hr or less.
The line represents the same protective limit as the CMC for juveniles for continuous
exposure. The line is a logarithmic expression with a slope and intercept calculated
from the regressions in Figure 9 at the DO concentration of 2.3 mg/L (the CMC).
90-
80
70-
o> 60 -
u>
O 50-
o 40-
o
O 30 -
20-
10 -
0
Percentage Reduction in Growth Relative to Control
y = 0.778X
R2 = 0.815
y = 0.5x
0
10
20
30
40
—T—
50
—i—
60
70
—i—
80
90
Constant Exposure
Figure 11. Plot of test results from growth experiments pairing constant low DO exposure with
exposures to various cycles of low DO 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"
relationship from a slope of 0.5 (see text for explanation). Species used and the experimental conditions
are listed in Appendix H.
22
-------
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 origin10) is 0.778, a factor of 1.56 greater.
Thus greater growth impairment occurs from cyclic exposures than expected. One
hypothesis 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).n
Figure 12 shows a dose-response for growth of larval lobster (H. americanus) over a
range of constant DO concentrations. The data are from 10 tests (see Appendix C) with
durations ranging from 4 to 29 days. The percentage growth reduction is relative to a
control response. Growth reduction effects are considered instantaneous; therefore, the
percentage reduction can be applied to any time period. Data for the lobster are
emphasized because it was the most sensitive species tested for which growth was
measured. Its use is consistent with the 1985 Guidelines (Stephan et al., 1985), which
allows a criterion to be established using data for a sensitive economically or ecologically
important species.
To evaluate a cycle for chronic growth effects, the above relationship between cyclic
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 intervals12 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 5 hours. Interval b lasts a total of 3 hours (bl before plus b2
after interval c). Similarly, interval a lasts for a total of 4l/2 hours. Each of these time
intervals is multiplied by 1.56 to adjust for the cyclic effect.
A recent publication of these data (Coiro et al., 2000) clearly demonstrates that the growth
reduction differences between constant and cyclic exposures are more or less constant across all of the DO
concentrations tested. In other words, the ratio between constant and cyclic response should remain
consistent across all concentrations. Thus the slope can be forced through zero.
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
recommend assuming constant exposure conditions for these latter situations.- •
12Any 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.
23
-------
c
us
o
•a
Q)
DC
.C
1
O
v^
100-
90-
80-
70-
60-
50-
40-
30-
20-
10-
0 -
•
•
***- A
** •
* X"X--.
*»
*-.• •
y = -23.1x + 138.1 * **v^.*
* *
"*
1.5 2.0 2.5 3.0 3.5
Dissolved Oxygen (mg/L)
4.0
4.5
Figure 12. Plot of dose-response data for growth reduction in American lobster
(Homarus americanus) exposed to various continuous low DO concentrations.
Percentage growth reduction is relative to a control. The dashed line is a linear
regression through the data points. Data are from Appendix C.
^
j.
c
o>
D)
g-
O
T3
(U
"o
01
_w
Q
6.5 •
6.0-
5.5-
5.0-
4.5-
4.0-
3.5-
3.0-
2.5-
•> n -
99
«
•
.
a1
k
•
•
b1
_
%.«^"
b2
c
_
r —
i
32
• •*••
• »
• 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)
24
Figure 13. A hypothetical representative DO 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 growth.
The range of DO, the mean DO, 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).
Interval
al-a2
bl-b2
c
DO Range
(mg/L)
4.8-4.0
4.0 - 3.5
3.2-3.5
; f
DO Mean
(mg/L)
4.40
3.75
3.35
f-
% Daily
Reduction
in Growth
36
51
61
Actual
Duration
(far)
4;5
3
5
Cyclic
Adjusted
Duration
(far)
7.0
4.7
7.8
Reduction
for Duration
11
15
18
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 Rgure 12. These in turn are normalized for the cyclic
adjusted duration.
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
constant 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 cumulative) for an estimated growth reduction for the cycle. The total percentage
reduction in our example is 44%. This reduction is greater than 25%;13 thus our
hypothetical cyclic hypoxic event does not meet the protective goal for growth.
Cyclic Larval Recruitment Effects
To evaluate cyclic exposures for their potential impact on larval recruitment to the
juvenile life stage, two pieces of information are needed: (1) a set of larval TTD curves
to estimate the expected daily mortality for a given low DO cyclic exposure and (2) 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)
from the FSC in Figure 5 and the sensitivity-dependent relationships for TTD slopes and
intercepts in Figure 9. Creation of a series of larval TTD curves followed the same
procedure 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
cycle. 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). The greatest effect datum lies between the 15%
13
'See footnote 8.
25
-------
4.5 -,
4.0-
3.5-
3.0-
0)
0)
2.0
"O
s
o 1.5
en
CO
1.0-1
0.5-
0.0
5
10
15
25
10 12 14 16 18 20 22 24
Time (hr)
Figure 14. Time-to-death (TTD) curves generated for the Final
Survival Curve "genus." Data to generate the curves were taken from
Figures 5,9A, and 9B. The numbers adjacent 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.
6.0
5.5-
5.0-
-i
I 4.5-
e
m
O)
5" 4.0-
o
•n
m
| 3.5-
Q
30 -
2.5-
2.0 -
•
1
0
a
b
c
] d
i
i
i
i
i
i
»
.
•
• • •
• •
• •
F ccc *
12:00 15:00 18:00 21:00 0:00 3:00 6:00 9:00 12:00 15:00 18:00
Figure IS. The same hypothetical DO time series as Figure 13. This time the portion of
the curve below 4.8 mg/L is divided into several arbitrary intervals to estimate effects on
mortality. The DO minimum and its duration for each interval are listed in Table 5.
26
-------
Table 5. Dissolved oxygen and duration data from the intervals selected from the hypothetical
cyclic time series in Figure 15.
Interval
DO Minimum for interval (mg/L)
Duration of Interval
(hr)
4.3
3.8
3.3
3.0
2.8
15
11.5
9
6
4
These data are plotted in Figure 16 to estimate the expected mortality occurring for recruitment modeled species during
the cycle.
4.5 i
4.0 -
3.5-
3.0-
2.5 -
2.0-
o 1.5 H
CO
co
5 1.0-
0.5 -
I
0.0
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (hr)
Figure 16. The DO minima and the durations listed in Table 5
superimposed 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 25% was
selected.
and 25% mortality curves. For the purpose of this example, we will select the 25%
mortality curve. Therefore, the hypothetical cycle of DO is expected to cause 25% 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 die only
once).
Now all that is needed is to translate the expected 25% mortality into the number of
allowable days for this hypothetical cycle to occur. This is accomplished using the FSC
and FRC curves in Figures 5 and 6, respectively. The information in Figure 5 is for
percentage survival, but it can be converted easily into percentage mortality. Thus the
27
-------
information shows the expected cohort mortality to occur for a given DO concentration.
For the example, 25% mortality occurs at a DO concentration of 3.7 mg/L. From the
equation used to fit the data in Figure 6, the 3.7 mg/L is allowed to occur for up to 9 days
without significant impairment to seasonal recruitment. Thus the cycle that resulted in an
estimated 25% daily mortality to larvae can be repeated for up to 9 consecutive days
without exceeding a 5% reduction in seasonal larval recruitment. All of the above can be
simplified by merging the information from the FSC and FRC into one cyclic translator
figure using the DO axis that is common between Figures 5 and 6. This is shown in
Figure 17.
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
previously. 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, with three
notable exceptions. The first is a single datum (LC50 of 1.9 mg/L) for the Atlantic
Cyclic Translator
0 5 10 15 20 25 30 35 40 45 50 55 60
Daily Cohort Mortality (%)
Figure 17. A plot that combines the information from Figures 5 (Final Survival Curve)
and-6 (Final Recruitment Curve) into a single cyclic translator to convert expected daily
mortality from cyclic exposures into allowable number of days of those cycles.
28
-------
2.5 n
2.0-
I
O
Q.
o
o
1.5-
1.0-
0.5-
0.0
E. affinis
M menidia
B. tyrannus N o
\
o
Q
o
o
o
o o o
0.1
10
Time to Effect (hr)
100
1000
Figure 18. A plot of the other juvenile/adult mortality data from Appendix J (open symbols) along with
the proposed DO criteria for juvenile/adult survival (solid line).
menhaden, B. tvrannus, 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
confounding factor. In addition, the authors provided no information on control response
for either the Atlantic menhaden or the Atlantic silversides.
The third set of data above the criteria is a series of values at 0.5 hr for the copepod
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
methods, including the number of replicates and the number of animals in each replicate,
or On the response in the control. Thus, it is difficult to adequately assess the significance
of these results. However, in the absence of data to the contrary, it is worth noting that
the DO limit for juveniles and adults may not be protective of copepods. Alternatively,
one could consider that short-lived species with high reproductive outputs (such as
copepods) may be more appropriately protected in a manner similar to larval recruitment.
29
-------
In this case, all of the E. affinis LC50 values would fall below the criterion provided by
the larval recruitment (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 with which to compare these values is the
TTD curve for 50% mortality from Figure 14. There are two series of data points for
LC50 values for larval rock crab (Cancer irroratus) for exposure durations of 2 and 4
hours; each has some values above the 50% TTD curve (Vargo and Sastry, 1977). The
more sensitive values in these sets are for tests run at 25°C; thus the animals 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 to LC35, LC40 to LC65, and LC90 to LCI00. As
with the LC50 values in Figure 19B, these values are plotted along with TTD curves
(10%, 50%, and 90% mortality) from Figure 14. All of the LC5 to LC35 values are close
to 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 LC100 values are below the 90% TTD
curve. This one value is for 100% mortality of striped bass larvae (M. saxatilis) that
occurred after a 2 hr exposure to 1.90 mg/L DO. However, there are two other striped
bass tests where 100% mortality of the larvae did not occur until 24 hr of exposure to
similar low DO.
There are fewer other data on sublethal effects than on lethality effects (Appendix J).
The sublethal effects included reduced feeding, growth, locomotion, and bivalve
settlement, as well as delays in hatching and molting. However, none of these values
indicate that the CCC would not be protective against these effects.
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. Because most of the effects
observed occurred <2.3 mg/L, the 24 hr acute limit CMC would be protective. 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 reduced
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.
30
-------
^>
E
q
Q
6.0 ,
5.0
4.0 -
3.0 -
o
o
e 2.0 -
IU
1.0 -
0.0
0.01
8
0.10 1.00
Time to Effect (days)
10.00
100.00
6.0
5.0
-. 4.0
E
q
S. 3.0
3 2.0
£
1.0
0.0
F. heteroclitus
10% embryo
mortality
C. bosquianus
5% mortality
0.01
0.1
10%TTD
50% TTD
90% TTD
10
100
Time to Effect (days)
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% TTD curve from
Figure 14. Figure 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
from Figure 14.
31
-------
The following effects are reported at less than the 2.3 mg/L protective limit. In the
red morph of green crabs (C. 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 k 1 hr
exposure (Breitburg, 1994). No avoidance was observed at 3.0 mg/L. This same author
reported 100% avoidance in larval bay anchovy (Anchoa mitchilli) at 0.75 mg/L
following a 1 hr exposure. Reduced locomotor activity occurred in daggerblade grass
shrimp (P. pugio) at 1.8 mg/L (Hutcheson et al., 1985). Burrowing in the northern
quahog (M. mercenarid) 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 solidissima) 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 to 0.24 mg/L (Breitburg,
1992).
The following low DO effects on feeding are reported in a bivalve and four
polychaetes. In eastern oyster (Crassostrea virginicd) early postsettlement stage (436 |am
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 postsettlement animals (651 jam 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. virens 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 during 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
naked goby (G. bosc) larvae, avoidance of the sea nettle (Chyrsaora quinquecirrhd)
predator 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 exposure to 2.0 mg/L (Breitburg et al., 1994).
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
unknown, 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
concurrent DO exposure conditions if it is to be useful in deriving criteria. Ten studies in
32
-------
the Virginian Province have provided concurrent DO measurements. The DO
observations often are only point measurements, not continuous records, and they rarely
provide information on DO conditions prior to the observed effects. The biological
effects reported include alterations in the following: presence of fish and crustaceans,
diel vertical migration 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 Island
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 below 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 3-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. DO declines below these concentrations resulted in further exclusion of these
animals, which has implications for the secondary productivity of these waters. Reduced
species number implies reduction of community resilience, should this condition persist.
The consequences of habitat crowding on animals occurring in adjacent waters are
unknown.
Hypoxia-induced changes in the distribution offish 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
sensitivity to hypoxia; that is, 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 Oithona colcarva was
disrupted by hypoxia (Roman et al., 1993). In mid-Chesapeake Bay during the summer,
these copepods typically occurred near the bottom during the day and migrated 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 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 daytime
33
-------
avoidance of the surface waters, 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. base), 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 severe 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 absent, 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
returning during the weeks following 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 having low DO ^0.7 mg/L.
In the study with the eastern oyster (C. virginica) (Osman and Abbe, 1994),
mortality 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 2
weeks of two experiments). Mortality was proportional with depth, which corresponded
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
deployment 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,
although 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
number 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 infaunal organisms migrate toward 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 below 1 mg/L. These studies support 2 mg/L as the hypoxic
34
-------
effect threshold for the macrobenthos, which is consistent with the global literature (Diaz
and Rosenberg, 1995). ^ # „
To summarize, demersal finfish community biomass has been observed to diminish
at DO <3.7 mg/L, and species richness to diminish at <3.5. These effects become
increasingly pronounced with further DO decline. Below 2.0 mg/L, migration of the
infaunal 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 species and life stages (i.e., disruption of diel vertical migration in
copepods, reduced growth and survival of newly settled oysters, and lethality in larval
goby) as demonstrated in parallel laboratory studies (Breitburg, 1992; Roman et al., 1993)
or by other workers (Baker and Manri, 1992, 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
(Tyson and Pearson, 1991) and a review (Diaz and Rosenberg, 1995) on this subject.
Results of hypoxia effects studies were not cited for species that do not commonly occur
in coastal and estuarine waters between southern Cape Cod, MA, and Cape Hatteras, NC,
during the spring to autumn period that 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; for example, American lobsters
tested at 5°C (McLeese, 1956). Data were not used if they are probably not reliable.
Examples include indications that the test animals may have been stressed, for example,
American lobster tested at 25 °C that were not fed during an 8- to 10-week acclimation
period (McLeese, 1956); excessive control mortality (>10% for juveniles or adults and
>20% for early life stages); uncertain DO exposure concentration, 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 measurements, for example, 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
35
-------
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
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 completion of this document, a papeFappeared (Secor and Gunderson,
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 striped bass. However, a combination of low DO (ca. 3.5
mg/L) and high temperature (26°C) resulted in 100% mortality of A. oxyrhincus 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 1 and 3 ppt; therefore, it is likely that these data are
more appropriately associated with freshwater criteria, which are higher than those for
saltwater (see Implementation section).
Virginian Province Criteria
The recommended criteria for ambient DO for the protection of saltwater aquatic
life in the Virginian Province: Cape Cod to Cape Hatteras are summarized in Table 6.
These criteria are briefly described below:
(2) Protection of Juvenile and Adult Survival from Persistent Exposure
This limit is derived following the Guidelines procedures and is analogous to the
CMC, except that a protective DO concentration limit is expressed as a minimum as
opposed to a maximum, as would be the case for a toxicant. This limit represents the
floor below which DO 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. Refer to Table 1 and Figure 2 for an
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
CCC for a toxicant. This limit represents the ceiling above which DO conditions should
support both survival and growth of most aquatic species from Cape Cod to Cape
Hatteras. Refer to Table 2 and Figure 3 for an 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 number of days of persistent exposure.
36
-------
Table 6. Summary of Virginian Province saltwater dissolved oxygen criteria.
Endpoint
Persistent Exposure (24 h or greater continuous low
DO conditions) * % a
Episodic and Cyclic Exposure (less than 24 h
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 = 0.370*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 in growth must not exceed 25%
and
Gred, = -23.1 * DO,r + 138.1
where:
Gredj = growth reduction (%)
DOj = allowable concentration (mg/L)
tj = exposure interval duration (hours)
i = exposure interval •
Larval Recruitment
Effects3
(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
' ti(allowed)
and
D0i =
13.0
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)
" Model integrating survival effects to maintain minimally impaired larval populations.
37
-------
(3) Protection of Larval Recruitment Effects from Persistent Exposure
This limit is derived from a generic larval recruitment model. The limit represents
allowable DO conditions below the CCC, provided the exposure duration does not exceed
a corresponding allowable number of days that ensure adequate recruitment during the
larval recruitment season. The cumulative effects of all exposure interval durations 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 duration for
each interval of a specific DO concentration. Refer to Table 3 and Figure 6 of this
document for an 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 juveniles tested in the laboratory. It provides a degree of protection equivalent
to 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 DO conditions that
must be maintained on an hourly basis (e.g., 1-hour minimum, 2-hour minimum). 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
recruitment limit described in (6) below. Refer to Figure 10 of this document for an
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 American lobster, and comparisons of the effects of cyclic exposure
versus constant exposure on growth for a variety of species. It provides a degree of
protection equivalent to the CCC, but for exposure durations shorter than a day. The
limit represents the DO conditions that maintains a daily percent growth reduction not
greater than 25%. The cumulative effects 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 DO concentrations. 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 various growth-inhibiting conditions. The limit applies to DO
conditions that may occur as a recurring 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 recruitment season, provided the larval recruitment limit is
met according to (6). Refer to Table 4 and Figure 12 of this document for an explanation
of the derivation of this growth limit. The larval recruitment limit can be substituted in
whole for the growth limit.
38
-------
(6) Protection of Larval Recruitment Effects from Episodic or Cyclic Exposure
A •>. »
This limit is derived from the modeled relationships between daily cohort mortality
and the allowable number of days at a given maximum daily larval cohort mortality that
protects against greater than 5% cumulative impairment of recruitment over a recruitment
season. It provides a degree of protection equivalent to 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
mortality 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 cohort mortality. This limit provides additional information that should be used in
conjunction 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
includes a DO minimum of 3.0 mg/L for 6 hours results in a daily cohort mortality of
almost 25% (see Figure 16). Assuming this represents the maximum daily cohort
mortality for the cyclic pattern, the allowable number of days for the cyclic exposure is 9
(see Figure 17). 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 1-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,
respectively.
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 associated 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 (Thursby et
al., 2000). 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 primarily 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
39
-------
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
regulators 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 adverse 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. 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 toxicants. 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. irroratus and E.
affinis). The limits provided here should be sufficient under most conditions where
aquatic organisms are not otherwise unduly stressed.
Many programs that monitor coastal DO with electronic equipment cannot measure
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 assessing
field DO conditions should take this into account. Criteria for DO can be appropriately
used in a risk assessment framework. The approach outlined in this document can be
easily used to compare DO conditions among areas, and determine if the DO conditions
are adequate to support aquatic life. Environmental 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 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, five 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 characterized,
the more reliably one can decide 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 assessment of
conventional sampling procedures along the Atlantic and Gulf coasts has
suggested that hypoxia in their estuarine waters is substantially more
widespread than previously believed (Summers et al., 1997). Deciding what
data can adequately characterize hypoxia is a matter of risk management.
Cyclic conditions may require measurements every 30 min for several days,
whereas persistent hypoxia may need only several measurement a week.
40
-------
Decisions also have to be made about the number and locations of sampling
sites to properly represent argiven area..-, ,
Biological effects—Potential biological effects are most difficult to predict
when DO lies between the limits for juvenile and adult survival and larval
growth. Deciding whether concentrations between these limits are acceptable
will depend in part on several biological parameters related to the recruitment
model. How to best represent these issues is a risk-management decision. The
5% impairment level for seasonal larval recruitment was selected to be
consistent with the protection provided to juvenile and adult life stages, but a
different percentage (higher or lower) may be valid for a site-specific DO
criteria. The biological effects data represent the expected range of sensitivity
to hypoxia for the Virginian Province. In certain site-specific situations, data
on additional species more representative of the site may be desirable.
Deletion of data from the current data set, however, should be done with
caution. The fact that a species (e.g., American lobster) may not be present at a
more southern site does not mean that it does not represent sensitive species in
the community that could not be tested. In addition, the lengths of recruitment
season and larval development period may be adjusted to be consistent with
conditions expected at a site.
Spatial extent—After environmental managers have found a hypoxic area, they
must decide whether it is small enough relative to nearby unaffected areas to
allow the coastal region as a whole to meet the criteria.
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 are greater than 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 their biological communities. If they
are more like freshwater organisms, freshwater criteria should be applied. If
they are more like saltwater, then saltwater criteria apply.
Threatened and endangered species—In cases where a threatened or
endangered species occurs at a site, and sufficient data exist to suggest that it is
more sensitive at concentrations above the criteria, it is appropriate to consider
development of site-specific criteria based on this species.
41
-------
42
-------
References
Armstrong RS. 1979. Bottom oxygen and stratification in 1976 and previous years. In: Swanson
RL, Sindermann CJ (eds). Oxygen Depletion and Associated Benthic Mortalities in New York
Bight, 1976. NOAA Professional Paper 11. Washington, DC: U.S. Department of Commerce.
pp. 137-148.
Baker SM, Mann R. 1992. Effects of hypoxia and anoxia on larval settlement, juvenile growth,
and juvenile survival of the oyster Crassostrea virginica. Biol Bull 182:265-269.
Baker SM, Mann R. 1994a. Description of metamorphic phases in the oyster Crassostrea
virginica and effects of hypoxia on metamorphosis. Mar Ecol Prog Ser 104:91-99.
Baker SM, Mann R. 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 AJ, Studholme AL, Olla BL. 1987. Behavorial responses of red hake, Urophycis chuss, to
decreasing concentrations of dissolved oxygen. Environ Biol Fishes 19:261-268.
Beverton RJH, Holt SJ. 1957. On the dynamics of exploited fish populations. UK Min Agric
Fish, Fish Invest (Ser 2) 19. 533 pp.
Bittinger ML, Morrel BB. 1993. Applied Calculus. 3rd ed. Reading, MA: Addison-Wesley.
818pp.
Breitburg DL. 1990. Near-shore hypoxia in the Chesapeake Bay: Patterns and relationships
among physical factors. Estuarine Coastal Shelf Sci 30:593-609.
Breitburg DL. 1992. Episodic hypoxia in Chesapeake Bay: Interacting effects of recruitment,
behavior, and physical disturbance. Ecol Monogr 62:525-546.
Breitburg DL. 1994. Behavioral response of fish larvae to low dissolved oxygen concentrations
in a stratified water column. Mar Biol 120:615-625.
Breitburg DL, Steinberg N, DuBeau S, Cooksey C, Houde ED. 1994. Effects of low dissolved
oxygen on predation on estuarine fish larvae. Mar Ecol Prog Ser 104:235-246.
Brungs WA. 1971. Chronic effects of low dissolved oxygen concentrations on fathead minnow
(Pimephales promelas). J Fish Res Bd Canada 28:1119-1123.
Burton DT, Richardson LB, Moore CJ. 1980. Effect of oxygen reduction rate and constant low
dissolved oxygen concentrations on two estuarine fish. Trans Amer Fish Soc 109:552-557.
Carpenter JH, Cargo DG. 1957. Oxygen requirement and mortality of-the blue crab in the
Chesapeake Bay. Technical Report XIH. Chesapeake Bay Institute, The Johns Hopkins
University.
43
-------
Chesney EJ, Houde ED. 1989. Laboratory studies on the effect of hypoxic waters on the survival
of eggs and yolk-sac larvae of the bay anchovy, Anchoa mitchilli. Ch. 9. In: Houde ED, Chesney
EJ, Newberger TA, Vazquez AV, Zastrow CE, Morin LG, Harvey HR, Gooch JW (eds.).
Population Biology of Bay Anchovy in Mid-Chesapeake Bay. Maryland Sea Grant Final Report.
pp. 184-191.
Chesson, P.L. 1984. Lecture Notes in Biomathematics 54:76-89.
Coiro LL, Poucher SL, Miller DC. 2000. Hypoxic effects on growth of Palaemonetes vulgaris
larvae and other species: Using constant exposure data to estimate cyclic exposure response. J
Exp Mar Biol Ecol 247:243-255.
Das T, Stickle WB. 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 DM, Ranasinghe JA. 1992. Effects of low dissolved oxygen events on the macrobenthos
of the lower Chesapeake Bay. Estuaries 15:384-391.
D'Avanzo C, Kremer JN. 1994. Diel oxygen dynamics and anoxic events in an eutrophic estuary
of WaquoitBay, Massachusetts. Estuaries 17:131-139.
Davis RM, Bradley BP. 1990. Potential for adaptation of the estuarine copepod Eurytemora
affinis to chlorine-produced oxidant residuals, high temperature, and low oxygen. In: Jolley RL
et al. (eds). Water Chlorination: Chemistry, Environmental Impact and Health Effects, Vol. 6.
Boca Raton, FL: Lewis, pp. 453-461.
DeFur PL, Mangum CP, Reese JE. 1990. Respiratory responses of the blue crab Callinectes
sapidus to long-term hypoxia. Biol Bull 178:46-54.
De Silva CD, Tytler P. 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 RJ, Neubauer RJ, Schaffner LC, Pihl L, Baden SP. 1992. Continuous monitoring of
dissolved oxygen in an estuary experiencing periodic hypoxia and the effect of hypoxia on
macrobenthos and fish. Sci Total Environ (Suppl 1992): 1055-1068.
Diaz RJ, Rosenberg R. 1995. Marine benthic hypoxia: A review of its ecological effects and the
behavioural responses of benthic macrofauna. Oceanogr Mar Biol Ann Rev 33:245-303.
Dittel AI, Epifanio CE. 1982. Seasonal abundance and vertical distribution of crab larvae in
Delaware Bay. Estuaries 5:197-202.
Fogarty, MJ, Sissenwine MP, Cohen EB. 1991. Recruitment variability and the dynamics of
exploited marine populations. Trends Ecol Evolut 6(8):241-246.
Ginzburg LR, Person S, Akcakaya HR. 1990. Reconstructibility of density dependence and the
conservative assessment of extinction risks. Conserv Biol 4(1):63-70.
44
-------
Gleason TR, Bengtson DA. 1996a. Growth, survival and size-selective predation mortality of
larval and juvenile inland silversides, Menidia beryllina (jpisces: Atherinidae). J Exp Mar Biol
Ecol 199:165-177.
Gleason TR, Bengtson DA. 1996b. Size-selective mortality of inland silversides: Evidence from
otolith microstructure. Trans Am Fish Soc 125:860-873.
Haas LW. 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.
Hammen CS. 1976. Respiratory adaptations: Invertebrates. In: Wiley M (ed.). Estuarine
Processes, Vol. 1. Uses, Stresses, and Adaptations to the Estuary. New York: Academic Press.
pp. 347-355.
Heath AG. 1995. Water Pollution and Fish Physiology. 2nd ed. Boca Raton, FL: Lewis. 359 pp.
Herreid CF H 1980. Hypoxia hi invertebrates. Comp Biochem Physiol 67A:311-320.
Hillman NS. 1964. Studies on the distribution and abundance of decapod larvae in Narragansett
Bay, Rhode Island, with consideration of morphology and mortality. MS Thesis. University of
Rhode Island. 74 pp.
Holeton GF. 1980. Oxygen as an environmental factor of fishes. In: Ali MA (ed). Environmental
Physiology of Fishes. Plenum Press, pp. 7-32. LOCATION?
Homer DH, Waller WE. 1983. Chronic effects of reduced dissolved oxygen on Daphnia magna.
Water Air Soil Pollut 20:23-28.
Howell P, Simpson D. 1994. Abundance of marine resources in relation to dissolved oxygen in
Long Island Sound. Estuaries 17:394-402.
Hughes GM. 1981. Effects of low oxygen and pollution on the respiratory systems offish. In:
Pickering AD (ed). Stress and Fish. New York: Academic Press, pp. 121-146.
Huntington KM, Miller DC. 1989. Effects of suspended sediment, hypoxia, and hyperoxia on
larval Mercenaria mercenaria (Linnaues, 1758). J Shellfish Res 8:37-42.
Hutcheson M, Miller DC, White AQ. 1985. Respiratory and behavioral responses of the grass
shrimp Palaemonetes pugio to cadmium and reduced dissolved oxygen. Mar Biol 8:59-66.
Johnson DF. 1985. The distribution of brachyuran crustacean megalopae in the waters of the
York River, lower Chesapeake Bay and adjacent shelf: Implications for recruitment. Estuarine
Coastal Shelf Sci 20:693-705.
Jones MB, Epifanio CE. 1995. Settlement of brachyuran megalopae in Delaware Bay: an analysis
of time series data. Mar Ecol Prog Ser 125:67-76.
45
-------
Jordan S, Stenger C, Olsen M, Batiuk R, Mountford K. 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 DL. 1987. Dissolved oxygen and fish behavior. Environ Biol Fishes 18:81-92.
Kuo AY, Park K, Moustafa MZ. 1991. Spatial and temporal variabilities of hypoxia in the
Rappahannock River, Virginia. Estuaries 14:113-121.
Llanso RJ. 1991. Tolerance of low dissolved oxygen and hydrogen sulfide by the polychaete
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 Shelf Sci 35:491-515.
Llanso RJ, Diaz RJ. 1994. Tolerance to low dissolved oxygen by the tubicolous polychaete
Loimia medusa. J Mar Biol Assoc UK 74:143-148.
Lutz RV, Marcus NH, Chanton JP. 1992. Effects of low oxygen concentrations on the hatching
and viability of eggs of marine calanoid copepods Mar Biol 114:241-247.
Lutz RV, Marcus NH, Chanton JP. 1994. Hatching and viability of copepod eggs at two stages of
embryological development: Anoxic/hypoxic effect. Mar Biol 119:199-204.
McLeese DW. 1956. Effects of temperature, salinity and oxygen on the survival of the American
lobster. J Fish Res Bd Canada 13:247-272.
McMahon BR. 1988. Physiological responses to oxygen depletion in intertidal animals. Amer
Zool 28:39-53.
Morrison G. 1971. Dissolved oxygen requirements for embryonic and larval development of the
hardshell clam, Mercenaria mercenaries. J Fish Res Bd Canada 28:379-381.
Myers RA, Barrowman NJ. 1996. Is fish recruitment related to spawner abundance. Fish Bull US
94:707-724.
Myers RA, Bridson J, Barrowman NJ. 1995. Summary of worldwide spawner and recruitment
data. Canadian Technical Report of Fisheries and Aquatic Sciences, No. 2024.
Osman RW, Abbe GR. 1994. Post-settlement factors affecting oyster recruitment in the
Chesapeake Bay, USA. In: Dyer KR, Orth RJ (eds). Changes in Fluxes in Estuaries. Denmark:
Olsen and Olsen. pp. 335-340.
Paul JF, Gentile JH, Scott KJ, Schimmel SC, Campbell DE, Latimer RW. 1997.
EMAP-Virginian Province Four-Year Assessment Report (1990-93). EPA 60Q/R-97/XXX. U.S.
Environmental Protection Agency, Atlantic Ecology Division, Narragansett, Rhode Island.
Pihl L, Baden SP, Diaz RJ. 1991. Effects of periodic hypoxia on distribution of demersal fish and
crustaceans. Mar Biol 108:349-360.
46
-------
Pihl L, Baden SP, Diaz RJ, Schaffner LC. 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 on Mysidopsis bahia in two modified chronic
tests. Memorandum to David J. Hansen. U.S. Environmental Protection Agency, Atlantic
Ecology Division, Narragansett, Rhode Island.
Poucher S. 1988b. Chronic effects of low dissolved oxygen on Menidia menidia. Memorandum
to David J. Hansen. U.S. Environmental Protection Agency, Atlantic Ecology Division,
Narragansett, Rhode Island.
Poucher S, Coiro L. 1997. Test Reports: Effects of low dissolved oxygen on saltwater animals.
Memorandum to D.C. Miller. U.S. Environmental Protection Agency, Atlantic Ecology Division,
Narragansett, Rhode Island. July 1997.
Poucher S, Coiro L. 1999. Data print out of ICp values for effects of dissolved oxygen on growth
of saltwater species. Memorandum to G.B. Thursby. U.S. Environmental Protection Agency,
Atlantic Ecology Division, Narragansett, Rhode Island.
Reid DG, Aldrich JC. 1989. Variations in response to environmental hypoxia of different colour
forms of the shore crab. Carcinus maenas. Comp Biochem Physiol 92A:535-539.
Reish DJ. 1966. Relationship of polychaetes to varying dissolved oxygen concentrations. Section
HI. Paper 10. Third International Conference on Water Pollution Research. Munich, Germany.
Ricker WE. 1954. Stock and recruitment. J Fish Res Bd Canada 11:559-623.
Roman MR, Gauzens AL, Rhinehart WK, White JR. 1993. Effects of low oxygen waters on
Chesapeake Ba> zooplankton. Limnol Oceanogr 38:1603-1614.
Rombough PJ. 1988a. Respiratory gas exchange, aerobic metabolism, and effects of hypoxia
during early life. In: Hoar WS, Randall DJ (eds.). Fish Physiology, Vol. XI: The Physiology of
Developing Fish. Part A. Eggs and Larvae. New York: Academic Press, pp. 59-161.
Rombough PJ. 1988b. Growth, aerobic metabolism and dissolved oxygen requirements of
embryos and alevins of steelhead Salmo gairdneri. Can J Zool 66:651-660.
Saksena VP, Joseph EB. 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 PA. 1973. Distribution and abundance of decapod crustacean larvae in the York River
estuary and adjacent lower Chesapeake Bay, Virginia, 1968-1969. Chesapeake Sci 14:235-257.
Sandifer PA. 1975. The role of pelagic larvae in recruitment to populations of adult decapod
crustaceans in the York River estuary and adjacent lower Chesapeake Bay, Virginia. Estuarine
Coastal Mar Sci 3:269-279.
Sanford LP, Sellner KR, Breitburg DL. 1990. Covariability of dissolved oxygen with physical
processes in the summertime Chesapeake Bay. J Mar Res 48:567-590.
47
-------
Savage MB. 1976. Burrowing activity in Mercenaria mercenaria (L.) and Spisula solidissima
(Dillwyn) as a function of temperature and dissolved oxygen. Mar Behav Physiol 3:221-234.
Secor DH, Gunderson TE. 1998. Effects of hypoxia and temperature on survival, growth, and
respiration of juvenile Atlantic sturgeon, Acipenser oxyrinchus. Fish Bull 96:603-613.
Setzler-Hamilton EM, Hall L, Jr. 1991. Striped bass Morone saxatilisi. In: Funderburk SL,
Mihursky JA, Jordan SJ, Riley D (eds). Habitat Requirements for Chesapeake Bay Living
Resources, 2nd ed. Chesapeake Bay Program Report, Annapolis, MD.
Shepard MP. 1955. Resistance and tolerance of young speclded trout (Salvelinus fontinalis) to
oxygen lack, with special reference to low oxygen acclimation. J Fish Res Bd Canada
12:387-446.
Simpson DG, Johnson MW, Gottschall K. 1995. A study of marine recreational fisheries in
Connecticut. Cooperative Interagency Resource Assessment. In: Final Report to U.S. Fish and
Wildlife Service, Project F54R. Study of Marine Fisheries in Connecticut. Fisheries Division,
Bureau of Natural Resources, Connecticut Department of Environmental Protection, Hartford,
CT. pp. 87-114.
Stephan CE, Mount DI, Hansen DJ, Gentile GH, Chapman GA, Brungs WA. 1985. Guidelines
for deriving numerical national water quality criteria for the protection of aquatic organisms and
their uses. NITS Publication No.: PB85-227049.
Stickle WB, Kapper MA, Liu L, Gnaiger E, Wang SY. 1989. Metabolic adaptations of several
species of crustaceans and molluscs to hypoxia: Tolerance and microcalorimetric studies. Biol
Bull 177:303-312.
Stickle WB. 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.
Strobel CJ, Buffum HW, Benyi SJ, Petrocelli EA, Reifsteck DR, Keith DJ. 1995. Statistical
Summary: EMAP-Estuaries Virginian Province - 1990-1993. U.S. Environmental Protection
Agency, National Health and Environmental Effects Research Laboratory, Atlantic Ecology
Division, Narragansett, RI. EPA/620/R-94/026.
Strobel CJ, Heltshe J. 1999. Application of indicator evaluation guidelines to dissolved oxygen
concentration as an indicator of the spatial extent of hypoxia in estuarine waters. Chapter 2 In:
Jackson L, Kurtz J, Fisher W (eds.). Evaluation Guidelines for Ecological Indicators. U.S.
Environmental Protection Agency. Office of Research and Development. In press.
Summers JK, Weisberg SB, Holland AF, Kou J, Engle VD, Breitberg DL, Diaz RJ. 1997.
Characterizing dissolved oxygen conditions in estuarine environments. Environ Monitor Assess
45:319-328.
Theede H, Ponat A, Hiroki K, Schlieper C. 1969. Studies on the resistance of marine bottom
invertebrates to oxygen-deficiency and hydrogen sulphide. Mar Biol 2:325-337.
48
-------
Thursby GB, Stern EA, Scott KJ, Heltshe J. 2000. Survey of toxicity in ambient waters of the
Hudson/Raritan Estuary, USA: Importance of Small-Scale Variations. Environ Toxicol Chem. In
press.
Tyson RV, Peason TH. 1991. Modem and Ancient Continental Shelf Anoxia. Geological Society
Special Publication No. 58.
U.S. EPA. 1985. Ambient Water Quality Criteria for Cadmium - 1984. U.S. Environmental
Protection Agency. Office of Water Regulations and Standards. Criteria and Standards Division.
Washington, DC. EPA 440/5-84-032.
U.S. EPA. 1986. Ambient Water Quality Criteria for Dissolved Oxygen. U.S. Environmental
Protection Agency. Office of Water Regulations and Standards. Criteria and Standards Division.
Washingtion, DC. EPA 440/5-86-003.
van Montfrans J, Peery CA, Orth RJ. 1990. Daily, monthly and annual settlement patterns by
Callinectes sapidus and Neopanope sayi megalopae on artificial collectors deployed in the York
River, Virginia: 1985-1988. Bull Mar Sci 46:214-229.
Vargo SL, Sastry AN. 1977. Acute temperature and low dissolved oxygen tolerances of
Brachyuran crab (Cancer irroratus) larvae. Mar Biol 40:165-171.
Vargo SL, Sastry AN. 1978. Interspecific differences in tolerance of Eurytemora affinis and
Acartia tonsa from an estuarine anoxic basin to low dissolved oxygen and hydrogen sulfide. In:
McLusky DS, Berry AJ (eds.). Physiology and Behaviour of Marine Organisms. Proceeding of
the 12th European Symposium on Marine Biology, Stirling, Scotland, September 1977. Oxford:
Pergamon Press, pp. 219-226.
Vernberg FJ. 1972. Dissolved gasses: Animals. In: Kinne O (ed.). Marine Ecology: A
Comprehensive, Integrated Treatise on Life in Oceans and Coastal Waters, Vol. I. Part 3:
Environmental Factors. London: Wiley-Interscience. pp. 1491-1526.
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 RA, Hennekey RJ. 1972. Effects of dissolved oxygen on two life stages of the
mummichog. Prog Fish Cult 34:222-225.
Wang WX, Widdows J. 1991. Physiological responses of mussel larvae Mytilus edulis to
environmental hypoxia and anoxia. Mar Ecol Prog Ser 70:223-236.
Welsh BL, Welsh RJ, DiGiacomo-Cohen ML. 1994. Quantifying hypoxia and anoxia in Long
Island Sound. In: Dyer KR, Orth RJ (eds.). Changes in Fluxes in Estuaries: Implications from
Science to Management. Fredensborg, Denmark: Olsen and Olsen. pp.131-137.
49
-------
-------
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.
Species
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 <0.55 0.72
marsh grass shrimp 0.84 1.02
summer flounder 1.10 1.10
winter flounder 1.44 1.46
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, et al., 1980
Poucher and Coiro, 1997
Burton et al., 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
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
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
A-l
-------
Appendix A. Continued
Species
Common name
24hrLC50 96hrLC50 Reference
Morone saxatilis
Palaemonetes pugio
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
Palaemonetes vulgaris
striped bass 1.96 1.96
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 <0.48 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
Poucher and Coiro, 1997
Appendix A: page A-2
-------
I
I
T)
i
2
_
g
I
H
ll
03 ^
^
P
n
O
V
S
Life Stage
Common name
.2
03
f-
ON
0\
„
a
a
1
Poucher
NO
"5
S
~
§
cs
01
en
*
s
x<
Jo
juvenile, <24
mysid shrimp
Americamysis bahia
t—
ON
ON
„
a
a
•§
C3
Poucher
>n
eN
"
CM
cs
CO
*
g
P*H
Ui
juvenile, <24
mysid shrimp
Americamysis bahia
t--
ON
ON
„
g
a
1
Poucher
ON
O
V
CM
CM
CM
en
en
-*•
s
tL
juvenile
T5
O
1
Ampelisca abdita
t*-
„.
S
a
1
Poucher
CM
en
2
ON
O
ON
O
en
"*
S
B-
J
I
•8
four spine stickleba
Apeltes quadracus
r--
o\
o\
„
a
a
1
Poucher
5j
ON
CM
"
ON
en
CM
*
Jg
E
juvenile
Atlantic menhaden
Brevoortia tyrannus
o
OO
ON
*C3
-------
I
0.
•
I
D,
§
g
1
II
to ^3
U B
o 3
J J
d,
£
S
£*
1 S
w N~''
S-
11
Is
t
o
"S
0
es
C/3
3
name
P
1
s
•g
cu
g.
ON
s"
5
1
o
1
vo
I**
JO
""
_
^
C1
fn
i
cn
^.
S
L,
J
S
3
-a
o
g
•g.
Q
i
E
1
o
s
americanus
Pleuronectes
g.
ON
6"
6
1
1
s
vo
0
cn
"
o
o\
o
en
CN
^
5
M
g
_3
•o
o
J
"&
S
1
1
o
C
J>
1
americanus
Pleuronectes
-
g.
ON
O
3
C3
J
3
CO
0
85
o
o
ON
s
1
^
jg
M
1
"^
C
1
g
t/3
g
.C
|
J
Prionoliis cat
CO
o\
*C3
0
u
00
CO
CO
ON
1
55
_
O
o
O
0
d
en
^.
S
o
1
— '
t
u
E
•c
S
HW harrisii
Rilhropanope
g.
ON
O
6
i
«
(D
|
CO
CM
CO
o
o
ON
""
0
d
en
CM
g
J2
1
•~>
t_
•o
c
0
c:
£
1
aquosiis
Scoptlialmus
g.
o\
o
a
•o
c
o
1
s
(^
d
cn
o
„.
c*J
CM
CM
en
^
g
fi>
1
'"•
el
1
o
H
!l
Spisula solidi
g.
ON
O
6
l
CJ
1
ON
"
^-
^,
CM
cn
o
t ,
5
O
1
^3
fi<
3
O
CO
|
Stenotomus c
S;
ON
O
'S
r>
cn
„.
D
1
•->
f
1
Taiitoga oniti
cs
I
o
es
CO
f
I
53
-------
u
•8
i
8
IS
C§
^
5
o
£
0
JS
"I
.a
''S
H
>,
e
1
•8
TJ
"3
e
hH
1
«
Q
g
w>
>>
X
O
1
>•
"o
1
o
^^
5
en
"es
S
'S
"^
-*-»
jrt
|
TS
CO
o
.•§"
•^«
cu
CO
*u
*S
J3
!H
v ?
.*
•a
a
Oi
S
1
i
fe 55 ^%
o y g
o w ~
.a « 3
o - *a
o •*" —>
'S
o
1
w "S
1 S
"gi
§1
ilai
'e O g
rP 2 "*— '
W
e
.2 -o?
2|
s S
o
Q
Pi ^
•S S
£> 'S
"5 S
« £1
a s
03 g
H
S
•a
W
C*^
1
c
HH
CB
G
a
J
i
4J
W
rt
oo
oo
ON
S
o
D
£
^
z
o
S
tt
1
o o
o —
^ 0,
« o
q in o Tf
vd co co oi
so
s)
"*j*
s*
« 2
•g
^
^
>>w 00
01
in
oi
CO
o
CO
13
JJ ~°
| 00
.S, V
1"
w
^
'w
•S
1
•2
1
c
«
^
oo
CO
ON
jf
3
£
^
z
VO
CO
vo
co'
O 0
in o
01 01
r- vo
~*. ^
co oi
w E2
S •*'
oo
01
VO
01
cs
CO
6
CO
•o
_5J ~3
1 »
•H,^
I"
w
3
'w
.2
1
.^
mericamys
•*.
*<*
^ o
XO O
t- —
o o
t-~ 01
O4 *n
c- o
oi --
is
o
g1
CO
g
CO
Tf
r-
oo
CN
00
1
o
*s>
^
r-
5
oi
CO
<>
r-
o
§
CO
o
in
QJ
Si
S
"«
J3
^
S
•g
8
o
1
-------
» "S
"E S
n R
CO §
EH
g
w
(A
<§
.S
'S
C3
C
g
p
c
0
•3
u
a
CO
—
jj
This report;
Poucher & Coi
1999
in
t-
en
oo
vft
!n
O O g
VO C7» •-*
1— 00 OO
in en en
en ^
CO
•"*
s •
1
i
s
„§
d4 a3
M 3 a\
5
1
*o
o o o o
Sill
8a\ a\ •^t
o\ — • en
»n en en cN
— t^
fC vo
p^
o
c;
c\
cs
I
CO
1 2
J3 en
"i
o
•a
i
>-»
CO
•-*
g
5"
i
2 . 2
Ha e fc.g
o o o o
§•1 §• 2! §•
J^S J .2§S .2
-COON 4S pCOON J3
fr— ' Qt — * E— ' E-H &H •— ' t~*
co <; t^. ^
-nO O *-* t"- OO
\O (3 °O VO OO O\
O O O J co co co co O O O" ^ O O" O" O S
CNCN-^-O ^MOOO cnenvoo r-ootNo\o
en vo f^ — * enoooNON en>ovo^^ vo vo t^- o^ — *
OOvoo oo — * ON en OO\ONVO enTj-Tj-enO
•^•"^j-enen vocor-oo ^•c^'^fen es»n<— ivo^r
•^J-eneN'— • ^- en cs ^^ M-encs'—* en eN eN '-• <— «
vo rf vo oo en r~ M-
vo* vi vo vj c-^
-------
u
1
2
£
fsi
1 « J
.a « <•?
g a d
2 « 1
6 -
^
1
0
11 1
Is
ii
l"ai
Hi O g
Q z ~
e
o -5
a &
11
a
|j;%-
JBf **
£> «
ii "
la s
CO §
H
Cfl
s2
V*
1
o
g
g
g
s
I
j
o
OH
a
CO
is
<
z
0
en
col
S i
O dof
oo i— i in
t** Ox *~H
en cs c-i
•* - 0
t— vo m
,
o
S
en
1
es
"a m
t 2
S cs
0 O ®J
^" G^ Si O
in" vo" oo O
•^ O en F";
in •*' en w
r-;
^
.
o\
§
cn
60
ui
|1
w
J
g
1
i
s
1
S
^s
g
1
1- !_
tfS r§
&L W
g|2 ||l
in eo
in -*
2 g
en r-;
CO CO
in CN
•^ oo
odd o o o g
111 Ills
£^3 SJn^S
— q ox g JQ
*"•«•* S in
O Ox
S Si
en m
a a,
In tS
11 11
«3 S
J J
a s
S S
•g '5
E E
§ 3
1 1
£ S
o 3
g g
1 1
•S •§
S3 EC
2
*!
a3
III
„
en
o
°
CO
—
o o d^
\o o\ -^ o
<— ' ^ OO ~-<
rf cs in —<
fri c4 — < —*
' —
t^
o
04
o\
§
en
»n
& T*-
If
CL, to
a
1
"g
E
g
1
1
5
&o
g
g
1
g
.J
S*i
£3
» §
Si
__
en
en
O
1
en
2
vo m
r^ en
$ —
es
^
S
en
1
.2
|1
•Ejin
o
CO
O
g
O
'C
g
<
g
|
1
13
C
2
1
a
fc.6
si
Ox w O Ox
Ox *^" 3 Ox
OX pC O OX
— H O, — <
q
en
o
in
O O
en ox
•* in
in e-<
en cs
vq
ox
rq
oo
en
t
CO
JS
§"— VD
=> S
.Sj in
c3
£
G
1
<
g
-1
K
^j
C
2
1
g"
'o
.. U
IS
gig
H eu — <
Ox
"*
5
vq
"
CO
vo
en
O O O O
cn vn i— « en
rj- »n vo oo
i— • — * en *— •
,— < fs) (s) yQ
^j- cn eN — <
s-o
r-^ in
r^*
•— ^
§
en
o
DJ3
•J5
*e5
jg
'I
U
TD
i
c
o
-2
"§
-2
3
S
1
"C
o
S
<
2
m
c^-
en
O O
<
•3
I
-------
t
u
o
60
s.
u
X
1
1
o
t,
1
e.
BS
•SS C3 ^
III
o ~ -
'^i
pi
^
1
ts
g
|1
Bo5
§1
~eo 3
° w "a
1S<§
o
c
15 &
Is
Q
^e
c* ^*
a £
'£
u
C3
Tn
1
!S
"S
H<
o
§
e
c
g
3
1
Cl
a
**
o
I
P
1
t^
CO CO
^ t*~
c^ ^^
o do
*S *51 >O
e^ e^ e*
-M VO tT)
CN »n r^
|c*r ^j^ ^
•^f en °o
en oi —
P
VO
00
in
s
en
"o
0
JS
i
a.
E
1
C/l
1
.c
1
g^
C
.§>
s:
a..
s
V
=:
Q
i
<2
s
*o
..0
11
to ^ ON
S <2 2
en
eM
CO
^
vo
0000"
"^ C*4 ^" VO
t"* CO T*1* ^
r^- CN vo cj
en en CN o»
.o
C^ ^1
vo "^
t~
in
CJ
en
2
VO
V
&
E
•c
G4
U
1
-c
S3
e3
_to
C
J>
S»
S
u
3
s:
1
eg
•c
1
H
Z
en
•*'
CO
ts
CO
o dg
^ ^ o
CO ON O
CN
2
VD
"v
a,
1
1
1
JZ
2
a
•8
^
>
S
SJ
5
i
1
^
a
2
.2
iS
z
1
CO
(i..
O\
o"
5v
3
* K
~- CO
vo en
t-
o
^?
o\
M
en
8
O
I
t«
•3 m
i 2
o.
.E
CO
ob
.C
1
•8
.§>
>
s
u
3
s:
i
1
2
..3
,6
av S
ON ™
^ H CM -*
s
S
O O
en
o\
O
en
t-
o" f
C-; •*
o
— O O
co in in
vo en cs
I
§
I
on
"I
ea
o.
2
oo
1
o.
I
-------
v~>
U
&
1
I
.2
,5,22
£
U
_><
•o
c
CL.
Cl,
SS
•a 8
CO -S
I
O U O O
D
H
o'lsS
eN O O
11
I a
O
O O 0
O O O
'en o ^ »n «^ co
t-^ iri -sj- en en oi
Tf
01
§1]
c-J
a
o
£2
o
cs
en
1
O
&
11
•g
>, e
II
•8
"8
§
I
•8
•
i
5
I
5
•f
s
"?
-------
VO
6
I
a.
u
o.
5B« H
> M ^
i* 2*
ts'i
15
^ia
5-8
Q. 0
• •« Q
^3 a
0
a
.s
a
o o o"
t^ TJ- o
CM PI 10
J2.
•§
I
a
I
J
K .2
11
II UlJ
8S
I
i«
o 2
O c
w •= g
•o =• e
i^i
S
-52 -1 o
!!!
z_ai g z
-------
1
o\
•a
a
«
rt
§
g
O
1
"5
5
^
o
H-l
_o
.a
1
t*
*rf
a
£
"a
CO
OS
hJ
°O
£>
"?
2
"w
S
CO
"S
0
•
d
X
T3
S
c.
e.
I
1
O
1
CO
i
,§
§
g
su
H
'So
1
2?
|
"«
03
ei
S
£
*J
0>
s
jgr
g
I
6
•|
Be
00
- |88888E8
O" O" O" 'O* O" 0 "^ o"
t-tJ-tfctl-ll-iI-ioSt-l
UOOOlUUc^}
•S •§ •§ -o •§ •§ g^
01 o »n
ON tn \o
V V
01 o tn
ON to \o
— < o) — "
V V
o)2{CS£g8S
V V V
cs— 'dol^^clcsol
0
IT)
)M*4 i i i i i r i i
j^cncocScncncscsm
^•5
^
s
01 TJ- -^- »n ^j g- en
~t£<2<raDra3a>n' iiiJ^
W W W W M M 00 O^^CO
£E£fc£££-°V'S>'5'e
«ea«c3eae3rt^9iSie5
jj
^3 JS XI X) X) .O ^
222222*^^ >»°
ooooooSj^^1
•O-O-OT3-a-O-o.02 S)"
lllilll'|l-g|
CO CO CO ' CO CO CO IS £"« C<
S g ^,
a'&o'o'a'^-i'o 1 £.§
IliillMII !
£3 Sa £5 EJ^ <^ 55^ S^ ^ -ft "Q !5
dClQQQQtt][i,OO>U
s g. g. g. & &
o g- o g g- o
111 1 1 1
1 1 1 1 1 1
S g 2 ^ en 01
r4 M cs c4 en cs
V
O TO O CO O\ O\
0 « g g « 0
01 en tn en
— — < ol ol ol en
Ml M M M M M
tS w to to M to
11 1 §11
« t3 « « e3 «
Ui Ur t-r Ul I-i (-(
(U H> to O (L> 5>
"w "w ^« "K "« to
O O O O O O
C C C C C C
1 1 ! 1 1 1
<<<<<<
H H S S 2 S
.1 •§ .1 •§ .§ .1
111111
=: K K s: =: g
o o ^ o o o
!U »U Ilj I^j »U >U
CD
g?
d-
Q
§
&
-------
2
&
o.
Q
cu
O
1
K
|
CS
1
*g
CO
1
I
1
£
-a
5
CO
0
•S
CO
Common Name
c.
_.
ON
ON
O
a
c
a
•1
3
£
5
04"
04
S
I
"S
American lobster
to
1
a
1
ON
O
a
i
1
3
<2
en
cs"
S
04
04
m
a
en
to
I
American lobster
t^
|
i
Homarus ai
ON
O
a
c
o
1
3
£
CS
O\
CS
en
cs
en
1
SO
1
American lobster
to
1
ju
Homarus ai
t~-
ON
o
a
c
CJ
u
3
<2
OC
en
C
CN
OC
""
O-
CM
U
ex.
P:
E
™
American lobster
GO
1
S
Homarus ai
r*-
o\
o\
o
a
c
cs
e>
J
£
o
04
in
C
CS
oc
ri
c
r*
-
V
u.
~
*H
rt
longnosc spider crab
.c
Labinia did
t--
ON
O
a
C3
O
3
S
jr
-j
i-»
-
ti.
0
c
1
longnose spider crab
•2
Labinia did
2
o
a
CS
U
3
£
C
V
C-J
™
d.
~
E
"S
longnosc spider crab
.2
Labinia did
O\
o
a
c
cs
1
3
£
0
o
o
O
M
V
§
V
c
OJ
~*
~
0
1
s.
CJ
c
long fin squid
:=
Loligo peal
2
B
a
c
cs
1
3
en
en
V
en
en
M
V
in
V
WJ
cs
T^-
cs
rn
cs
0
cS
6
3
inland sitverside
2
^
Menidia be
o\
g
a
cs
£
3
£
5
04
en
6
en
ig.
•0
O4
1
inland silverside
2
1-
Menidia be
2
2
a
Ti
C3
U
U
1
S
01
oo
04
en
T3
0
•C
"5
inland silverside
|
!>
Menidia be
ON
O
a
"E
C3
1
3
<£
o
04
ON
""
en
en
T3
1
f
inland silverside
S
:g.
Menidia be
ON
OO
ON
g
1
0
.5
K
r--
C)
V
f>»
o
V
o
o
V
O4
04
1
04
en
o
en
J3
O
I
Burry's octopus
b
Octopus bit
r-
ON
ON
2"
a
1
S3
o
1
ON
en
V
ol
T-H
S
•n
tj>
01
en
o
en
i
2
*o
^
V
o.
£
daggerblade grass shi
5.
a
Palaemonel
r~
ON
2
"S
CJ
"O
C
o
1
V)
<— 4
V
o
V
o
en
o
en
t
00
*e3
marsh grass shrimp
~e
Q
^
3
3
Palaemonel
t-.
ON
ON
2"
O
CJ
1
1
S
£
S
CS
o
cs
en
CS
1
•w
cs
marsh grass shrimp
'§
C3
8
Palaemonel
2
o"
a
c
cs
•§
£
ON
r-
V
cs
2
en
o>
cs
.2
•a
0
vo
marsh grass shrimp
'C
i?
g
Palaemonel
2
0
a
"O
c
cs
•§
3
£
£
>n
tj.
04
en
o
B
0
.c:
marsh grass shrimp
•^
-2f
a
Palaemonel
ON
2
a
"g
cs
1
1
s
V
04
^.
04
en
O
en
o
a
1
•o
0
ii
vo
marsh grass shrimp
•§
^>
a
Palaemonel
ON
2
a
1
1
<2
8?
V
tn
cs
cs
en
1
o
3
marsh grass shrimp
'C
.IP
a
Palaemonel
ON
g-
a
cs
u
3
s;
V
0
en
o\
cs
en
o
CJ
1
-o
o
J=
marsh grass shrimp
"c
"S
S
Palaemonel
r-~
ON
2
a.
1
J3
CJ
3
£
vo
CS
o
cs
en
cS
o>
1
•o
o
_=
1
CO
CO
bb
1
g
'C
"a1
S
Palaemonel
P-.
ON
2
a
T3
Cj
3
£
ON
V
vo
C-4
0)
O
S
•a
o
V
marsh grass shrimp
•r;
•S?
s
^
a
Palaemonel
-------
H
t
£ £
j> P o o
MH PH CU CU
VO ON ON
V) eN -^ OJ
c^ in co co
rJ — cs
cs CN eN
en en en en ( ,
— JL, en o _!, en
en en en
VI
%
ON
— -H CO VO
w
73"
S S a
e
^
S 2 5J
•S « t3
2 S 6
2 2 S
1
t^- i> r^ r*- r^
OS OS OS Os O\
8 g g g
OS OS
.§ .1 .§ .§ .§ M .§
aaaaaaaaaa
g g
g g
a a a
Po
O CO CD fl>
- 3 •§ •§ "§ •§
££££111111
u o u
.8 £ £
000
III
a
•s
g g
•s -3
u u
•O T3
I
<—' en
U3
en en —< >
o o qj o
"o
ff) _H ^H —I
W) W) Bb 60 60
cs ro a _« eg cs
73 73
.0
1
^
0
2
J3
2
o
^
o
8
-n
-^
o
g
"§
^
o
8
1
-^
O
g
|
1
I
5T
•s
1
'i
I
a
"5
-D
•o
1
CO
f
O
i
CO
1
•§ 1
o o
11
>1 >,
£ £
•o
Po
r>
04'
,
^f
oi
OS O O4
O 00 0)
CO O4 O4
04 04 04
i i i
O OS O
CO CO CO
O O OS
CO CO O4
SO
r-
0
V
f-
o
V
f- O O
— ' O4 T
04 04 O
V
04 04 04
O OS O
04 — 04
04 04 04
CO CO CO
CO CO CO
oo in oo
OS -- f;
— oi oi
oo in oo
os :2 ?~
^^ oi oi
inc~-r*-ocoeocooooos — COCOOOO4
^TOs
-------
Q
I
c.
Q
o
I
o
c*
CJ
1
O
>
*S
CO
i
O
Q
P
Pi.
w
H
-a
i
"S
i
CO
&0
S
,2
C3
Common N
S
"o
g.
co
-,'
& S; S
ON ON ON
222
656
•§ "g 1
ij jl j=
III
— ; en CS
ON O
•7* _H CS
OO CN ON
CS <"">
en cs
en en en
c? ra ca
w « to
lit?
Q « 3
U. - I.
Ill
e e n
cs ca e3
^o to ^o
111
< < <
a g g
G tj
£ T3
O *o
bO _. cs
w "• "S
g « ^
3 ij £
J3
2 o
•§ 1 '1
*S« u ^;
« > .«
Hi
== ^
:! ^ '«
^ ^ .1
.5 T2 ^2
311
ON ON
ON ON
2 2
6 6
•§ •§
•g -g
s s
en
CS
en
cs'
°5 t
_,
CS Q^
CS
vo xn
1/1 4
^^ „«
£ £
•= 5
o o
striped bass
striped bass
5 :=
'o 'o
2 2
3 3
1 1
ON ON
ON ON
2 2
6 6
1 1
1 I
3 3
CO TT
ON
*— f 0^
00 -
*n r~-
4 4
•« •«
« «
striped bass
striped bass
5 :5
'§ "a
S v
5 g
1 1
g S S
222
o o o
U U U
•o -o -o
o a5 a
JS J= J3
3 =J 3
000
CU CU &r
„
^4
V
co in
»n cc
1-4 -4
V
OO OO ON
"1 ™ ^
V
ft in
CS CS ^Q
ct S M
en en en
o o ci
en en en
0 0 0
« C3 «
t z z
£ S S
TJ -O T3
O O O
J= J= JZ
V V V
a.
^3
w a. n,
g S S
& -s •%
•§ ss n
5 2 2
J3 OJ) 00
U J= -C
111
•§ e E
•i i i
§. § 1
t^ t-3 to
•fii •£.! -fii
sj Si Si
333
5J S3 5i
CJ « C5
ON ON
ON ON
2 2
o o
0 0
•0 T3
S £
O O
0 0
vo in
vo -
CS CS
CS CS
CO CO
O ON
CO CS
n
ct ^
en en
ON ~
cs en
m
"o 1
C? "°
'S "3
"a ^
« ^v
1" £
•C 'C
x: j=
bD 00
£2 i2
i i
.g .g
S, S,
1 g
S §
^ ^
1 |
1 1
(•X, Q^
ON ON
ON ON
2 2
6 5
1 1
a S
1 £
O O
3 3
£ £
VO f-
ON oq
O
co ^
ON CS
CO _ ,
O co
CO
to
C3
& CO
-S CL)
*o ^
*£ «
5 "I
"v S
1 t
•c -n
CA U3
marsh grass
marsh grass
•s -e
g, a
§ §
«n <*>
•Si ii
u v
1 1
1 1
o c
ON ON
ON ON
2 2
6 6
"§ "E
1 1
O O
3 3
£ £
CS OO
"> °^
»-* o
vo ON
in oo
cs »-•
cs cs
CO CO
0 J.
CO CO
tO
ra *«
» |
T3 T!
> w
1 g.
a. cu
CO CO
CO CO
00 00
X! -S
a a
i i
1 1
§ §
a s
Si Si
3 3
a a
s §
0 .JS
Cx, CL,
>"
5
O
I
"o
«
S
o
c
(3
»
3
O
•g
d
52.
!
•f
>•
s
ies Mean Acu
S.
CO
£
•a
%
W
C
rt
£
0
1
£
o
p
-------
Appendix E. Explanation of Larval Recruitment Model and
How It Is Used
I. Introduction
The recruitment model is a discrete time, density-independent model consisting of several
equations that allow the cumulative impact of low DO to be expressed as a proportion of
the potential annual recruitment of a species. The model is run by inputting the necessary
bioassay and biological information selecting DO durations to model, and then iteratively
assessing various DO concentrations until the desired percentage recruitment impairment
is obtained.1 The resulting pairs of duration and concentration become the recruitment
curve. The applications of the model in this document were for the purpose of deriving a
Final Recruitment Curve. Alternately, one can assess the expected impairment for a given
site by inputting the DO interval that represents the minimum for that site, and inputting
the number of days that the site is experiencing DO concentrations less than the CCC.
The model can be set up to handle any number and types of life history stages. We have
chosen to model larval recruitment to the juvenile stage. For one application
(Dyspanopeus sayi) we set up a two life-stage model. All other species were modeled
using a single life stage. A variety of assumptions were made in the application of the
model to each of the species used. Some of the assumptions result in likely
overprotection, and some in underprotection. An implicit assumption that the model
makes is that the various over- and underprotection issues more or less cancel each other
out.
Without a clear functional relation between stock size and recruitment (which typically
does not exist for marine species), the only prudent course of action is to assume that
recruitment is density independent (Ginzburg, 1990; Fogarty et al., 1991). Few would
argue that density-dependent processes do not exist for marine organisms. However,
these relations are typically extremely difficult to characterize. Myers et al. (1995) fit four
standard theoretical recruitment functions to stock recruitment data for a considerable
number of marine fish species. In a few cases, there is a reasonable fit to the data, but in
many other cases there would appear to be virtually no relation between stock size and
recruitment. Using a density independent model will be conservative or pverprotective if
the population does have compensatory capacity (Fogarty et al., 1991), whereas a
density-dependent model could be vastly under protective if the chosen functional
relation is not accurate. Although fisheries resource managers make use of density
dependence and "compensatory reserve" in calculating maximum sustainable yields, the
collapse of many of the major groundfish stocks in the northwest Atlantic has likely
resulted in a reevaluation of the assumptions and application of the concept of maximum
sustainable yield.
'in Microsoft Excel® this is most easily accomplished using the Goal Seek command.
E-l
-------
Theoretical density-dependent recruitment functions have been available for nearly 50
years following the seminal works of Ricker (1954) and Beverton and Holt (1957).
However, density-dependent functions are not readily available for marine fish species.
Myers et al. (1995) fit up to four standard theoretical stock recruitment functions,
including the functions described in Ricker (1954) and Beverton and Holt (1957), to
stock recruitment data from hundreds of fish stocks. While this work represents a
valuable contribution to fisheries science, the stock recruitment data were often highly
variable and Myers' et al. (1995) did not suggest that these functions accurately reflected
recruitment in these fish stocks. Additionally, Myers' and Barrowman's (1996) findings
indicated that recruitment increased with increasing stock size, which is counter to the
assumptions of "compensatory capacity." Ginzburg et al. (1990) proposed that "when
available data sets are insufficient for reconstructing reliable measurements of density
dependence, conservative estimates of extinction probabilities can be made from models
that simply omit density dependence."
Recruitment in many marine species is characterized by occasional strong recruitment
events (dominant year classes), with more frequent years of poor recruitment.
Recruitment success is log-normally distributed—with predominantly low recruitment
(Fogarty et al., 1991). The factors which lead to good years or bad years are not well
understood, at least not so that one can predict in any given year whether recruitment will
be good or bad. Often the occasional good recruitment events are important for
maintaining a population through several years of poor recruitment. This is often referred
to as Chesson's "storage effect" (Chesson 1984). What is not understood in the least is
how additional anthropogenic stress might influence the likelihood of that one good event
occurring. For instance, the role of low DO in reducing the likelihood of a strong
recruitment event cannot presently be measured, but could potentially be more important
than an incremental reduction in survival probability would suggest.
II. Model Equations
Recruitment Under Nonhypoxic Conditions
Under nonhypoxic conditions, the number of recruits from each cohort is expressed by
the following equation:
Equation El
Where:
D =
number of individuals of a cohort surviving to the next life stage (juveniles in our
application).
initial size of cohort.
attrition rate expressed as a percentage.
larval development time in days.
E-2
-------
The total number of recruits for npnhypoxic conditions is then determined by multiplying
NR by the total number of cohorts.2 % ,
Recruitment Under Hypoxic Conditions
To account for effects of hypoxia on total recruitment during a spawning season, the total
number of cohorts have to be segregated into those that are not exposed to hypoxia and
those that are at least partially exposed during their developmental period. For the former,
Equation El is applied. For the latter, Equation El is modified to account for DO effects
on the initial cohort size. These modifications are performed using several intermediate
calculations, but the overall equation is:
Equation E2
Where:
Sv = the proportion of a cohort that is unexposed to hypoxia (e.g., the percentage of the
cohort in the upper portion of the water column, and assumed to not be exposed)
surviving to the next life stage.
SH = the proportion of a cohort that is exposed to the hypoxic event and survives to the
next life stage.
Equation E3
Where variables are the same as described above and:
p - the proportion of a cohort that is exposed to hypoxia (e.g., "l-p" = the percentage of
the cohort in the upper portion of the water column, and assumed to not be
exposed).
a = the attrition rate expressed as a percentage ("natural" mortality due to predation,
etc.).
Note that this equation is Equation El (the recruitment model for nonexposed cohorts)
multiplied by the proportion of the population that does not experience hypoxic
conditions.
SH=(.N0p(ERSURV)](l-af
Where variables are the same as described above and:
Equation E4
ERSURV = the exposed proportion of a cohort surviving at a given DO concentration using
laboratory exposure-response data.
Alternately, if various initial cohort sizes are used Equation El would be run for eacn N0 and the resulting
list ofNR values summed.
E-3
-------
Note that for model applications with more than one life stage dose-response (i.e.,
Dyspanopeus—two life-stage model) separate SH values are calculated for each life stage.
The model can account for indirect effects of hypoxia such as delayed development,
where an increase in the time spent as a larva means that the natural attrition rate is
applied for a longer period of time. When such is the case, D in Equation E4 is replaced
by D' (Equation E5).
Equation E5
Where variables are the same as described above and:
D '= New development period (days) due to exposure to low DO.
E = Duration of the hypoxic exposure in days.
ERjevet - Exposure response for change in development period for a given DO
concentration, expressed as a percentage. Note that for the purpose of this
application of the model ERdevel was set at 100%.
Equations similar to Equation E5 can be added for other biological attributes. For
example, if data become available to justify increased predation due to larvae avoiding
hypoxia and thus becoming more concentrated in other areas, then the attrition rate can be
made a function of DO by incorporating a value for "ERattrition."
The total number of recruits under hypoxic conditions is determined by summing NR, for
all cohorts. The percent recruitment impairment due to hypoxic conditions is calculated
as follows:
Impairment = •
-*100
Equation E6
in. Model Assumptions
The application of any model requires 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. For the purpose of our application the recruitment season is
divided into 24 hr time periods. This means that a new cohort of larvae (those released
within a 24 hr period) are available each day of the recruitment period. The model does
not require any knowledge of how often cohorts are produced. All that is required is a
knowledge of how many cohorts are produced in a recruitment season and which of these
are exposed to hypoxia. Assuming a fixed rate of cohort introduction simplifies the
calculation of total number of cohorts, as well as the number and degree of cohorts
exposed to hypoxia.
E-4
-------
Figure E-l demonstrates pictorially the number of cohorts associated with a hypothetical
species that has a 30-day recruitment season and a 10-day larval development period.
This means that a total of 21 cohorts are possible during the modeled recruitment season.
The model assumes that the hypoxic event begins at the end of the first cohort's
development period. This assumption maximizes the number of cohorts that are exposed
during a hypoxic event. In this example, the hypoxic event lasts 6 days. Fifteen of the 21
cohorts are exposed to hypoxia anywhere from 1 to 6 days. The remaining six cohorts are
not exposed.
The above example demonstrates that for any hypoxic event most (and sometimes all) of
the exposed cohorts experience hypoxia for only a portion of their larval development
period. Unlike juveniles (Figure 1, main text), the sensitivity of larvae to hypoxia
increases with increasing length of exposure (Figure E-2). The concentration of DO that
causes a given percentage mortality increases by 14% (the slope of the line in E-2 is 1.14)
when the exposure increases from 24 to 96 hr. Therefore, the model accounts for
exposures longer than 24 hr by increasing the percentage mortality by a factor that is a
function of the duration of the hypoxic event. This function was developed in three steps
which are shown in Figures E-3 to E-5.
Figure E-3 is the same set of curves show in Figure 14 (main text), but the time axis has
been extended to cover several weeks. For each time increment greater that 24 hr, we
plotted the relationship between the 24 hr value and the value for that time increment
(Figure E-4). For example, for a 14-day exposure, we plotted the 24 hr values for each of
the percentage mortalities with their corresponding values at 14 days. This resulted in one
of the lines in Figure E-4. The slopes of the lines in Figure E-4 increase as the number of
days of exposure increases. This relationship is plotted in Figure E-5 (top curve). The
lower curve in Figure E-5 represents an adjustment to the calculated curve that forces the
curve through the slope from Figure E-2 (solid square). This was done by changing the
intercept of the log regression until the curve passed through this point. The other three
data points (plus signs) are from three test series from which there are data for 24 hr and
greater than 94 hr responses. Although these latter points are based on a very limited
number of tests, they do serve to show that the adjusted curve in Figure E-5 is probably
reasonable.
Each hypoxic event results in individual cohorts with different exposure durations. In the
example in Figure E-l, two cohorts are exposed for 1 day, two for 2 days, two for 3 days,
and so on. The model makes a simplifying assumption with respect to applying the
adjustment for mortality based on duration of exposure. To avoid having to calculate
separate adjustments for each individual cohort (depending on its individual length of
exposure), the model assumes an "average" duration of exposure. This average is equal to
one-half of the exposure period for exposures that are less than the development period,
and equal to one-half of the development period for exposures equal to or greater than the
development period. Figure E-6 shows that the effect on the resulting recruitment curve is
negligible when comparing curves generated using this "average" exposure duration with
curves generated long-hand using a separate adjustment to survival for each duration
from 1 day through the total number of exposure days.
E-5
-------
IV. Model Input Parameters
The input parameters for each application of the model consist of three main parts. The
first is the DO dose-response data for the organisms being modeled. The second is
various biological parameters for that species. The final set of input data is associated
with the hypoxic event itself. These parameters, as well as all of the calculation fields, are
shown for each species in the output tables at the end of this appendix.
Two Life-Stage Model
The model can be applied to any number of life history stages. This application for the
mud crab Dyspanopeus sayi incorporates two life stages. The first is for zoeal larvae and
the second is for the transition from zoea to megalopa. The model assumes that once a
zoeal larva has made the development transition to megalopa, then there is no further low
DO effect (the model only applies the late larval to megalopa dose-response curve for one
24 hr time period).
DO Response
The 24 hr exposure response data for life stages one and two for D. sayi are listed under
ERSURV. The PO and k values are variables in the logistic function (Equation 1, main text).
ERdeve, is the percentage used to calculate D' in Equation E5, and accounts for any
increase in larval development period due to exposure to low DO. There are insufficient
quantitative data to determine this value; therefore, in the current application we have
assumed no effect (i.e., the value has been set at 100%).
Population Parameters
Five population parameters have to be input for each species modeled. These are:
R = the length of recruitment season in days. For the purpose of our application,
recruitment season is the sum of the spawning season and the length of the larval
development period. The value 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.
This takes into information in 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).
D = duration of larval development in days. The development time of 21 days was
estimated from field data (Hillman, 1964), as well as from laboratory observations
made during EPA's DO testing with D. sayi.
E-6
-------
N0 = initial cohort size. This was arbitrarily set at 100 for each cohort. Any value can be
used. The absolute value of this parameter does not matter unless one chooses to
model using unequal cohort sizes.
a = rate of natural attrition in percent. This is the loss per day due to predation and other
natural causes. This parameter influences the percentage impairment only if there is
delayed development. Since we are assuming no effect of DO on development rate,
the attrition rate has been arbitrarily set at 5% (it could just as well be set at zero).
p = the percentage of a cohort exposed to a hypoxic event. For D. sayi, the model
assumes that only 75% of the available mud crabs are exposed to low DO 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 in the more sensitive later stages. A general assumption
regarding vertical (and horizontal) distribution is that zoea do not successfully avoid
hypoxia, although one could account for avoidance by making p a function of DO
concentration.
Hypoxic event
A hypoxic event consist of three input values, the duration of the event in days (E), the
DO value for that event (mg/L), and the "average" duration to use for cohorts exposed for
part of their development period. To determine recruitment curves for a given species, we
preselected values for E that covered the entire potential exposure days. We then
manually entered DO values for each duration until the desired percentage impairment
was reached. These two columns are the paired x,y values for plotting the recruitment
curve. The duration for partial exposures is either E/2 for exposures less than the
development period (D) or D/2 for exposures equal to or greater than D (duration is set at
ldayforE=l).
One Life-Stage Model
One life-stage models were used for each of the other either species. Input parameters are
the same as for the two life-stage model except there is only one set of P0 and k values for
V. Model Calculations
Once the input parameters are set, the spreadsheet automatically calculates several
intermediate values for exposure to hypoxia. The model then compares the total
recruitment with exposure to that without exposure to hypoxia to calculate the percentage
impairment.
E-7
-------
Two Life-Stage Model
The two life-stage model was only used for the Say mud crab D. sayi. All other species
used the one life-stage version. The differences are slight and are listed below in the
paragraph under One Life-Stage Model.
Survival Attributes
This section contains three columns. The first is the slope for adjusting survival rate
based on duration of exposure. This is calculated using the Duration for Partial Exposure
from the section for the Hypoxic Event in the equation for the adjusted curve in Figure
E-5. The second and third columns are the percentage survivals for life stages one and
two, respectively, for the given DO concentration. These survivals are calculated using
Equation 1 from the main text with the 24 hr PO and k values from above (recall that L in
Equation 1 is set at 100% survival). These 24 hr values are adjusted for the duration of
exposure by dividing the k value by the calculated slope in the first column of this
section.
Cohort Information
This section calculates the total number of possible cohorts and how they are distributed
into exposed and unexposed categories based on the recruitment season (R), larval
development period (D), and duration of hypoxia exposure (E). It also incorporates the
assumptions that a new cohort is present each day and the hypoxic event begins at the end
of the first cohort's development period. The total number of cohorts that are possible is
equal to R-D+1. This can be see graphically in Figure E-l, where R = 30 days and D = 10
days. The maximum exposure days for a given species are equal to the total number of
cohorts minus one (the first cohort is assumed to be never exposed). The actual number of
cohorts exposed is equal to D+E-1 until the maximum1 number of cohorts that can be
exposed is reached (total possible minus one). The number of cohorts that are exposed
during the transition to the second life stage is equal to the number of exposure days. The
number of partial exposures that do not include a transition to life stage number two is the
difference between the last two numbers (# exposed minus # exposed during transition).
Survival Distribution
This field calculates SH and Su from Equations E4 and E3, respectively. Separate SH
values are calculated for cohorts exposed for a portion of their development period
(partial exposed cohort) and those that are exposed through the transition to the next life
stage (LS2 exposed cohort).
Cohort Specific Hypoxia Survival
This is just the sum of each of the SH values with S0 to yield the NR, values for each life
stage.
E-8
-------
Hypoxia Survival Table '-
There are three columns in this section. The first is the number of individuals surviving
from all of the cohorts that were not exposed to the hypoxic event. This is essentially
Equation El times the number of unexposed cohorts (total number minus number
exposed). The second column is the NR. for partial exposure times the number of cohorts
partially exposed (exposed for less than the duration of larval development). The final
column is the N R. for life-stage two exposures time the number of cohort exposures that
included the transition to the second life stage.
Seasonal Recruitment
This section calculates the total number of individuals that can be expected at the end of
the recruitment season with (sum of the three values in the previous section) and without
exposure to hypoxia (Equation El times the total number of cohorts). The difference
between these two values is expressed as a percentage of the "without hypoxia"
total—this is the % impairment. For the current application of the model, this value was
set at 5%.
One Life-Stage Model
The one life-stage model calculations are similar to those for the two life-stage model
except where life-stage two was used above, the number and effects of a full exposure (E
equal to or greater than D) are used in the one life-stage model. For example, under
Survival Attributes, the last column is Larval Survival Adjusted for Full Exposure rather
than life-stage two % survival. Likewise, under Cohort Information the last two columns
are # Partial Exposures and # Full Exposures (referring to whether a cohort will be
exposed during part or all of its larval development period).
VI. Model Output Tables
Tables E-l to E-9 used to create the recruitment curves shown in Figure 6 (main text) are
appended at the end of this appendix. The DO dose responses are based on the data
presented in Figure 5 of the main text (with the exception of life-stage one for
Dyspanopeus sayi—which is not plotted in Figure 5). The initial cohort size and the
attrition rate for each model run were arbitrarily set at 100% and 5%, respectively. The
population parameters R, D, and p were selected based on the species being modeled.
Where a variety of information was available for a given species (e.g., different values for
different latitudes), we selected the more conservative of the values (i.e., those that were
representative of the more northern areas of the Virginian Province. Appendix F shows
the relative effects of various changes to these latter three population parameters on a
species' recruitment curve. It is important to carefully consider what the appropriate
values for the population parameters should be on a site-specific basis.
The population parameters for the Say mud crab (Dyspanopeus sayi) were selected based
on the literature as described above. The parameters for the flat mud crab (Eurypanopeus
E-9
-------
depressus) were assumed to be the same. We found no species-specific information for
the spider crab (Libinia dubia), so we chose to use the same R and D values as the mud
crabs, and assumed they would be equally distributed above and below the pycnocline (p
= 50%). The parameter values for the fish (Menidia beryllina, Morone saxatilis, and
Scianops ocellatus) were selected in consultation with Dr. David A. Bengtson of the
Department of Fisheries, Animal and Veterinary Science, University of Rhode Island,
Kingston, RI. The fish larvae were assumed to be equally distributed above and below the
pycnocline. The values for the American lobster (Homarus americanus) and the Atlantic
rock crab (Cancer irroratus) were selected in consultation with Dr. J. Stanley Cobb,
Department of Biological Sciences, URI, Kingston, RI. Larvae of these two crustaceans
generally spend most of their development time in the upper areas of the water column,
and as such may only rarely experience hypoxia. However, since they are in the bottom
waters at hatch (eggs carried on the abdomen of the mother), we selected p = 20% as a
reasonably conservative value. Finally, the R and D values for the grass shrimp,
Palaemonetes spp., were chosen based on field and laboratory observations by EPA
personnel at Narragansett, RI. We assumed that their larvae were evenly distributed
throughout the water column (p - 50%).
E-10
-------
o
_Q
•c
O
o
O
10
15
Time (days)
20
25
30
Figure E-l. Representation of a recruitment season for a hypothetical species with a season lasting 30 days
and a larval development period of 10 days. The hypoxic event begins at the end of the first cohort and
continues for 6 days. Each horizontal bar represents a single cohort. There are 21 cohorts. Fifteen are
exposed for a portion of their development period and six are unexposed.
E-ll
-------
Larval Data
5.0 n
4.5 -
4.0 -
Ib 3.5 -
E,
3 3'° •
i! 2.5 -
$
q 2.0 -
Q
1.5 -
1.0 -
0.5 -
y = 1.14x
R2 = 0.87
0.5
1.0
1.5
i
2.0
—i—
2.5
—i—
3.0
3.5
—i
4.0
D.O. 24 hr LCx (mg/L)
Figure E-2. A comparison between larval response of various species at 24 hr and 96 hr exposure durations
(n = 64). Data represent % mortality ranging from 5% to 95% for eight different species. Two species are
fish (Menidia beryllina and Morone saxatilis); the other six are crustaceans (Dyspanopeus sayi,
Eurypanopeus depressus, Homarus americanus, Libinia dubia, Cancer irroratus, and Palaemonetes
vulgaris).
E-12
-------
TTD-based on FSC
1s>
E,
d
d
7.5 -
6.5 -
5.5 -
4.5 -
3.5-
2.5-
•\ R -
D
O
j5 A
[§g X
JP
JP^ X
AJJX^
If "
j£o
p
o
o
A
X
X
-
o
n
o
A
X
Bl
O
+
o
A
X
X
o
+
D
O
A
X
X
—
O
+
D
A
X
X
.
—
o
+
n
A
X
X
.
—
o
"*"
n
A
X
X
-
_
0
+
D 5% Mortality
0 10%
A 15%
X 25%
X 50%
-75%
-85%
0 90%
+ 95%
Mortality
Mortality
Mortality
Mortality
Mortality
Mortality
Mortality
Mortality
10 20 30 40 50
Time (days)
60
Figure E-3. Time-to-death curves based on the Final Survival Curve (Figure 5) and the data in Figure 9A
and B. These are the same data as in Figure 14 of the main text, the time axis has just been extended.
Recall that Figure 14 only goes up to 24 hr. Figure E3 extrapolates this to several weeks. Because it is an
extrapolation one should not read too much into the absolute values of the DO in the figure. Note that this
figure is giving hypothetical values of DO for a given response by a single cohort. The recruitment model
gives results that are intended to represent the response of all cohorts in a given recruitment season. Note
also that this long term data is "corrected" (see Figure E5) before it is used in the model.
E-13
-------
7.5
7.0 -
6.5 -
6.0 -
5s 5.5 -
E, 5.0 -
O 4.5 H
o> 4.0 -
O 3.5 -
3.0 -
2.5 -
2.0 -
1.5
—gg— 63 day
—«—56 day
A 49 day
—4—42 day
—•—35 day
—t-—28 day
—0—21 day
^—14 day
——7 day
% 6 day
-X 5 day
—£s—4 day
O—3 day
—D— 2 day
1.5 2.0 2.5 3.0 3.5 4.0
24 hr LCx (mg/L)
4.5
5.0
Figure E-4. Data from Figure E-3 replotted for each time interval versus the corresponding value from
24 hr.
E-14
-------
I 0.9
0.7-
0.5
24 hr vs Other LCx, Based on Rgure 9 & Larval FSC
y=0.19Ln(x) + 1.00 -«
Adjusted for Actual 24hr vs. 96 hr Ratio
y = 0.19Ln(x)+0.90
10
20
30 40
Time (days)
50
60
70
Figure E-5. The linear slope for each line in Figure E-4 was plotted against its corresponding time value
(solid circles). A logarithmic regression was run through these points (dashed line). This curve was forced
through the slope for 24 vs 96 hr from Figure E-2 (1.14) by changing the intercept of the logarithmic regression.
The other data points (plus signs) represent slopes for much smaller data sets for 24 hr vs 7 day (n=8), vs 10
day (n=l) and vs 15 day (n=l).
E-15
-------
Recruitment Curves: Incorporating Mortality
Adjustments for Length of Exposure
5.0 n
4.5 -
4.0 -
3.5 •
i3-0-
£2.5 -
P2.0-
1.5 -
1.0 -
0.5 -
0.0 -
C
-annnnnnnnnn
• • B ™ High Hypoxia Sensitivity
.goeeooooooooooo
o O 8 * " Low Hypoxia Sensitivity
) 5 10 15 20 25
Time (days)
Figure E-6. A comparison between recruitment curves for two hypothetical species with R = 30 days and D =
10 days. One species has a high sensitivity to hypoxia and the other a low sensititivity. The solid symbols are based
on "average" durations of exposure for cohorts that are exposed for only a portion of their development period
(E/2 is E < D and D/2 if E > D). The open symbols represent recruitment curves where each duration (e.g., 1 day,
2 days) was accounted for separately.
E-16
-------
OJ
"o
S3
2
o
o
o
as
UJ
_0)
.a
Attributes
Survival
o
1
1"
OJ
£
2
to
c
_g
as
rs
a.
S.
Response
O
Q
>- £
l€i
2 O o.
ro § uf
£§03 »
5 8 -o = "5 =
1 1 s a i g
W ^ 3 ^~ fit
k.
•2 _ £ ^
Q
"o "x "H ^
§*ll
ii^
8.1s o
«ii|
oaia
R D
Length of Duration of
Recruitment Larval '
Season Development f
(days) (days)
| ^ TJJ I.
*•" ra g o
I '
cc
UJ
^ ^- ^»
SSi
O O 10
1- T- T-
CM CM CM
§§§
xp >^ %^
o o o
§o o
o o
s s s
o o o
o o o
III
o o o
0 O O
O 0 O
1 1 1 i 1 1 1 11 1 1 1 ! 1 1 1 i 1 S 1 1
O CO tD O) W 1O ^ O CV1 LO ^ OJ f* CO If) (p CO O t~ CO ^h
O O T~ T" T"T"OJOJWWWO1OJCOC*3COCQCOCOC3CO
oinomotnomomomoLnotnoinOLOo
CMCMeOcO^>^tf)lOO)OO«**«,»»»
§00000000000000000000
oooooooooooooooooooo
COCOCOCOCOCOCOPJCOCOCOCOCOCOCOCO CO CO CO CO CO
iissisiiiiiiiiilsssss
OOOOOOOOOOQOOOOOOOOOO
ooooooooooooooooooooo
ooooooooooooooooooooo
ooooooooooooooooooooo
g
§
CO
10
CM
CO
*
c£
o
o
o
CO
o
g
•^•
o
0
o
o
w
r*-
CO
o
CO
CO
*
#
o
o
o
.
s?
o
o
0
o
o
0
^£
5^
ra
03
m
CO
*
s^
0
o
o
.
^
o
o
o
o
o
0
g
to
s
o
Tj*
m
*
0
cS
o
o
o
.
^
o
o
o
1
o
o
0
g
-------
CM
•5
CM
>
HI
a
_L
.a
'o
0)
a
CO
f
°
(5
stributlon
3
j>
>
3
Cohort Information
T:
O>O>O)O)O>O>OO)OJOlO)O>O>O)O)O)OJ2O)O>criO)OJOJC)a>C>O>O!
u>iqioiqu>tqu^u;viiniou)tou)tntAwiAiAU)u>iniou}Su)§@Su)
(C3(D«>(DCD
6^8^888^86988 CRRI88 IB 88 ^^-888
in id in io in id id id id id id id id in id in id id id id id id id id id id id id id id
sssasssssssssaasssssssasssssss
SSSISSBSSSSSS-aESSSSElSKKS-o*-*--
csiMoic4cxioic\ic^cxioicsicvicsicue4oiCKicxicsioic\icacxic>JMCJcOcsic>icsi
oooooooooooooooooooooooooooooo
gggggggggggggggggggggggggggggg
gggggggggggggggggggggg'g'ggggggg
COCOCOCOCOCWC3COCOCOCOCOCOCOCOCOCOCOCOCOC3COCOCTCOCOCOCOCTCO
-------
.">. «> " "J"3. U> U> "0 W 10 10
_ 0
= •§ = -3-
UJ S | 8 I
CO
CO.
1.1 s a_
»
h-h-r^-r^r--r^i^t^t^r*-r^-r*-r^-r^-h-t^r^r*-
800000000000000000000000000000000
OOC3OOOOOOOOOOOOOOOOOOOOOOOOOOOOO
Dl!II
oc >-5
a
s
OC
O
a
-
s -s
6?^ 1
="8 o
« "
oooooooooooooooooooooooooooo
oooooooooo
OOO
OOOCSOOOCDCSOOOOOOOO
EC
LU
ooo'oo'oocDodoo'ooo'ooooocDo'oo'oo'o'ooo'ooo
OCSC3OOOOOOCSOOOOOOOOOOC3OOOOOC3OOC3OOO
OOOOOOOOOOOOOOOOOCSOOOOOOOC3OC3OOOOO
OOOOOOOCDOOOOOOOOOC3OOeDOOOO. ^OOOOOOO
OOOOOOOOOOOOOOOOOOOOOOOOO O' C3 O O O C3 O O
-------
CM
"5
u
s«
pppppppppppppppppppppppopooppoooo
'*''*
5 W cQcocococOcococococQcocococQcococococococococOcQcocQcocococo
Q*X u_iou3iqioioiou3U3iou>tou3i^inioioioujU5ioiou)ioioioioioiflio_
•» O COCDCOCOCOCOCOCDCOCOCOCQCOCOCOCDCOCOCOCOCOCDCDCOC0CDCDCOCDCD
"•PL COCDCDCDCDCDCOCDCOCDCDCDCOCDCDCDCDCDCDCDCOCDCDCDCOCOCOCOCDCDCDCDCD
MmtOtOU3tOtOU}U3lOtOU3lOtOtOtOM^WtOtOlOLOlOtOlOtOLOiOtOtOtOtO
_ .— — u3^Tfi«iou><'*Tfu>u>Loioioioioioiou>iou}u3u>u>toioioioioiniou>
_ Xx CMCMCyCMCMOJCvlCJCMCMCMWWCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM
rn ^~ 2 cocomajajiDcococaenmcDcdcdcdcQcdcd-Q — — —"—*—* -1 "
*" S fr comcococacoococococacocococDcocococo
^ X 22222S22H^!5!*^^^^^^!555!^^5!^^!?!!?!!?!!?!
»_ &
_____ _____ _ _ _ rfU5O>
* I |f "
g -x-
^ f-ns,. o> o ^ to a o> o 01
^f O & f?8 CMCM^;CDf^<3COCOO>C)>O>O>O>O>OJOjb>O5O>O>O>O>O>O>O>O)O>O>O>O>OOO
— CJ .S* Q. ^CMTOCOCOCOCOCOCOCTCOCOCOCOCOWCOCTcicOCOCOCTCOCTCOOCOCOP?-^^^^1
I
C c/j cdcocdcda5a5cdcdo5coa5a5cdcdacococncdmcdcdc6cdcDcdcococdcococdcd
^2
01 "a
_. ,.
CM CSJ CM CM CM
CvlCMCMCMCMCVICMCMCMCMCM
O O O O O O O
CM CM CM (M CM CM CM
O O -r- CM
O
fa '= " o g
5 O (D ~ 3t
OT Q ^ •« CO
n_ ° "SI — c -J
°- O. (8 y CB
*di^-s^2
! tJs o
i S-1 g
! &Is
O ^- 0)
ll s
§•§ g-
2 o fl
Isl-
: x Q
i-D
OOOOOO
(O CD CO CO --CD
CD CO
-------
CVJ
'o
Attribut
Survi
• 13 3
: (D -=; w
M"-1
° 0) °0 0
5 = -0 c "5 =
•g g g £ .S -g g
- «
° .5 1 2 « S
-
WCMCMCM
to us in
xic
I
LU
0)
_
Q. •» «-
c-oooooooooooooooooooooooooooooooo
i-oooooooooooooooooooooooooooooooo
o !
°fUt
5 > ~"
o «
_
j: £ c «
.
i
0>
a:
O
a
SlPt
eooc
ooooooooooooooooooooooooooooooooo
OOOC3OC3C3OOOOOOOCSC3C>C3OOOOOOOOOOOOC5OC>
ocsoocJoooooooocJooooooooooooooooooo
-------
o
CM
qq-qqqqqqqqpqqqqqqqqqqq-qqqqoqqqqq
CDtDCOCO
OOCOCOCOCOCO
It
g S
is!
t
§
5
S
.2
S
a.
H
3
w
.
cococococococococococococoaDroffiraaDcdcocdcdcDQOcdcocdcocdcd
iiricMc^c^c^e^c^csJev'e^c'c'^c^c^c^
cocdro
ffl
OOOOOOOOOOOOOOOOOOOO-r-WCT^J-tOtOt—
||
III
-------
.a
CO
*-1 _ ;
12 -| ji o '•! i
f«P£i
i §
CM <0 CM U> f»-
.g
s.
1-s I-
05 oj o" o> d o --" ^ «
•*^ ff CM CM W W
o--CM«'Win
-------
B
o
c
oooooooooooooooooooooooooooooooooooc
CToo«c>"»-ffW'»«*^^j'rr«*rr-*Tt^«>(oiotoiAtoiotoiou)uiinu)ujiri
CJWWCJWWWCVJCJCVIOJ^C^CJCJWCJOJwSSjOJwSjOjSioJWWOJOJCjSwoJ
eGOOOOOGOOGv-v^v-<»^v-v^^vv^v?v^v^v^v^oioiNOJC>JO)OIoioJOJOJ
a
£•
II
I!
UJ
JD
o
Cohort M
-------
CM
"5
£
w
ooocoS.i-cjiot:3o>ocyro^5i'»^^^v*«-<^'«tf-^-
•* C — •=• Jo Irl^^r-rZ-r^T^T^T^-r-T^v^
o> If § S *
g-cof-ao-
co <
.O ID
Jig|
j*§~
1*5,
m 'I I -I
8>I £
oopooooooooooopoopooooooooo_ooop
CL ^ UJ ~
c
_0
-------
CM
75
CM
o
S
it
cjoooooooooooooooooooooooooooooooo
coeocococococococococacococococOGOcococOcocococoeocococooococDco
eocdcococococdcocdcdcocdcdcdcdcoQofficdcooocQGoaSGOcocdcocpcD
-------
CD
C?
CL
'S2
I
w
T5
TO
0)
-Q
§
LU
_0>
-Q
Attributes
Survival
1
LU
o
g
Q.
2
a)
S.
S.
|
"5
£
Response
O
Q
1- £
llSi
(0 § "- g-
gS 3 m
0 CD ° "5 0
§ S -o c •« ^
'i§ °«-|l s
1 1 1 s « |
CO < ^
PP
"o 'x "c ^J*
0 §.§ 1>
0 .g-UJ £.
"° M £ „
Lul'lll
l^t"
atif-ii
°" "o
§ o?
co ~ 'S p
j-ll S
= 5 to
"5 0)
Q -° § g- =?
»^ "c
° c c -~-
rr-' — .— * w >»
C H flj ^
"J cc
rr* ? "§ S
"~* D) g) g)
CO
o:
S.
•*& ^
c^ fis
OJ O
o o
o o
gg
So
If)
o o
o o
** "fr
o o
CO CD
§ §
o o
o o
o o
o o
ifllillllflliilllliiill^ii
eetcgggggg'gggisegtisteftgfiseis
gggg.ngaSBSBSSBBBBESSSEBE
inotnouiOLooiooioooooooooooooooo
gggggggggggggigggggggggggg
mLou)Lnu?ioinioix>tOLOU)LOu>LOioLOLniou>iou?mtnmtn
OOOOOOOOOOOOOOOOOOOOOOC3OOO
oooooooooooooooooooooooooo
OOOC3OOOOOOOOOOOOOOOOOOOOOO
oooooooooooooooooooooooooo
oooooooooooooooooooooooooo
oooooooooooooooooooooooooo
oooooooooooooooooooooooooo
ooooooooooooooooo p_ oooooooo
-------
CM
"S
CM
CD
c»
g
1
o
CC
o)ooo'1~''~CSJCVlcvlcvlcv'cv)CXICV'cv'cv'cvicv'cvlcv)
COinCDO>O3COCOcqit-;C\{TOinCOtx-COO>p'I-C\IC\)CO
'**OCf)CDCOOC\lr3'lOCOCOOJOC\l'^'^'lOCOCO Is"- CD CO O5 O5 O O T~ T—
CCCOCOCOCOCOCQrocQCOCQCOCOCQCQrocOCOCQCQCDCOCQCOCOCQCO
3 CO
CO ^
nJ o
r
lii
at -g 5 o
CD
-a
•55
.g
t
-Q
1
to
III
£ g
fi-
co
o S
I
S
•ffl
'5
III
o-
ti
o
^
tl s
§ •§ g-
2 O £
x 3 w
I s &
S O.Q
UJ
QJ -2
l!
^•inajt^cocnoi-ojcoTi-incDcotoin^-cowi-oooot^-coin
^— *--^--«— i— •«— cMC\]ojc\JOjcycycvic\Jc\lc\iojc\Jcvjc\iT-T-t— i— T-
^ lOCD^corooT-cxjco^fincot^-oococococofflcocococococococo
— T-T-^T--i-OJOJWOJC\]WWOJC\JWCVICJCSJOJCOCVJWOJOJOJOJOJ
^KSS^^S^K^^^^^^^^^S^^^S^^S^^
3>OJO)O>O)a>O>OJO>O>O>O>O>O)O>O>c35O>OJO)OJO>
CVJCMCMQJCMCMCMOJCMCMCMCMCMCUCMCMCNlCXIOlCvTCMCM
-------
CM
"5
tes
Survival
:5
« -
o
is
CO
*!"!
Q. tq
O O) O) 05 O) O) O O) O) O) O) O5 O) O) O) O) O) O) O) O> O)
'
I
LU
o
'x
•S _ £
C CO 3
o ••=» 2
o
Q
C *55 3
=
OOU>OU>OU>OU>OU>O*U)OIOOU>OU>OU>
CMCMWCOCOCOOOCOOOOOCOCOCOCOCOCOCOCOCO^-*
01
I
I
CD
2
CO
Q-
I
3
Q_
£
sP vp sP
tfx 5^ O"*
ooo
«J> vP N.P vP vp vP vP vP xp xP *»P «>^ vP
O^ O^ O^- 0s" O** 0s* 0s" 0s* O^ 0s* &^ O** O"*
ooooooooooooo
tointotomu>tflLf)tr>totoLnLOLotnLou?Lntoioio
xp -x? xP xP xP xP G>O>OG>OO>O>OO)G)O>O>O>0>OO>CnO>CnO>
•"^••^••^••^••q-^tTf'^t>a-'a-TfTr-M-'<3-rf'»Tr'
-------
Cricrio'oOOO'!—
ioio.io_u>u?io.u}U}U>io. to. tqto. tnio. toio.toto. toto
cy cvi cvl r^- to CM co ^" CD co co ^~ to co o co to t°*~ o o CM
c\j c\i CM c\i co' ^t ^f to' to' co' to' h^ r^ i^ po* co' co" co" co oi cri
toiotoioiotoioioioioioioioioioiotototointo
CO CD CO CO. CO CD CO CO. CO CO CO CO CO CO CD CD CO CO CO CD CO
o' o o" o" o' o" o" o' o o' o o" o" d d o" o' o' o o" o"
ooooooooooooooooooooo
CMCMCMCMCMCMCMCMCMCJCMCMCMCMCMCMCMWCNicMCM
CJCVlCMCJC\lWCMC\icMC\icMCvicMCMCMC\icMCMCMCMCNi
CMWWCMCMCNCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM
•s
I
w
I
a.
rn
*
T>
8.
s
i
si
CD
CD £
8
,£ S
a: a
CO CD
C =
•= CD :
3 CO O
U- O .E
t:
-
a. o. o
i ,>j O
co UJ
I
t:
o
t
CO
•
o
3 01
S 5-
o
£
I!
IS
-------
_ .£.
1 is
1 a **5£* as
•* *
si is
5 =
q q to_ q io q to q to q to q q q q q q q q q q q q q q q q q q o o o o o o o
0 o § Is, CM CT CT 5? K> 5 b> 5 o e3 6 o o — -.. -^ --.--.--.--.--.
^!^g T^T^T^T^T^r^r^^T^T^T^WWT^T^OiCM'tNCaC^OiCMOlCMCMOlOlOlCMCMCUCMCMCMCMM
UJ -2 _-.__.-
= J»1 =
O. S g. 8 O..S g g g | g g | | g g g g g g g g g g- g g g g g- g o" | g' g" o" o o g- g g g o
"S <2
.2 o 5"
*"- " ^* " uj Iri Iri to" to" io" wj 5 io" tri tri 5 5 tri io" 5 5 to" tri 5 io' 5 io" to" to" trf to" 10" to" to" to' to" 10" to to
ggggssssilgg§°°ggogs||g§gig|i|ggsssg
°- -<3
5 <
CC _
^JS !
a 111 1 1 I g i I 1 1 1 I 1 I 1 1 1 1 1 1 g S § 1 i i I 1 1 1 ° a
UJ I
—. o'o'ooooooooooooooo'oo'oooooo'ooocaoo'oocioc
a
si
rr
g g g g g g g g g g g g g g 8 g S S g g S g S S S § S 8 S S g. S S 8 S
dddo'dddciddcDddddddcJdddddddddddoooooo
•8
-------
® °. °. ooooooooooooooooooopooooooooo
u>iotAioiou)tAir>iAu>totou>ir>u>tr>u>toiou>iA*u>to(ou? in" to' u> u> iri 10' tri u
llilllllllslllll-rfll-flllll-^^^^^^?
I^J5"rj!j:^1o5"^"ioco^j'U>T-^ocMvcM^^a-^'roT-o^'^^'V'«j-^-^-i^-^-vJC\ICMCM£IC\1C\]C\1CMC\1C\)OJCMCMC\
M0JOr-C\ie9^lOU!)O)cSc>C)C>C>T-WT^*^CMC\iCMrfCQe9<"J:^:
W 73
s!
IB
I
CO
111
B
•a
I
a
u 3 e>
III
gggsssssssssssssgssssssgsgssssssssss
-------
CM
*O
Attribu
Survi
rvival of
D_
% Survival
j.p. ". Adjusted fo
*fdfo; 'FU||
u^t'°n|of Exposure
Partial F
..
Exposu
re
S
,
for
ival
tmen
to
tial
sure
,
lo
S
di
D
P
ci CM o> CM
-t-f co co en
CO CO CO 00
COCOGOCO
COOOCOCOCO
aocM^rcqoqocM
COOOCOOOCOO>OCT>O5O3O5O5
I^-t-CO
-------
o
CM
CD
t
I«
fE O.
OOOOOC3OOOOOOOOOOOOOOOOOOOOOO
in 10 to in to 10" in in to" to to in to irj 10 irj 10 10 to 10 u> ID* m" 10 tri tri 10 u
roCTCTWrtOTCQCQCOCOCOrooOCOCOCOCOCOCTCOCOCOCOCOCOCOCT
CO CO CO CO tO COCOCQCOCOCOcOCOCOCOCOCOCOOOCOCOQOCOCOCOCOfflCt
cooocofficococococococdaococdcocoGdcdco'cdcdcdcdcdodcdco*
S|g^|||||iIiiIiii|iIIIii. ilis
g_ g g g g g g g g g g g g g g g g_ g g g g g g g g g g
CO-B
f'
•
§
I
I
UJ
0}
I
= |
"fci
,0
CO
CO UJ
3 "Se
HI
0.
I
ll
I<3
OOOOOOOOOOOOOOOOOOOOT-CMCO^lOCON-CO
-------
Appendix F. Sensitivity Analysis of Larval Recruitment Model
*
Several figures are presented below to demonstrate the relative effect of changing the
population parameters for the larval recruitment model. Examples are given for each
parameter except attrition rate, which is not relevant unless delayed development can be
documented. However, the model treats delayed development as a lengthening of the
development period (D), the effect of which is shown below. The first five figures show
effects on an individual species, the Say mud crab Dyspanopeus sayi, which used a two-
stage life history version of the model. The sixth figure shows effects on the inland
silverside Menidia beryllina, which uses a one-stage version of the model. The final
figure shows the effect of species selection on the Final Recruitment Curve.
Changing the larval development period (D) or the recruitment season (R) alone does not
have as large an effect on the recruitment curve for D. sayi as changing both (Figures F-l
to F-3). (The ranges for D and R were chosen to represent those that might easily occur
within the Virginian Province; however, local experts should be consulted when
attempting to adjust these values for site-specific recruitment curves.) As one moves
south along the east coast it is likely that both R and D will change. Recruitment seasons
are likely to lengthen, while development periods might be expected to shorten (both are
temperature dependent). These two parameters should be relatively easy to determine for
a site-specific application of the models in this document.
Figure F-4 shows the effect of changing only the percentage of each daily cohort exposed
to low DO (e.g., what percentage is below the pycnocline). This also is a site- and
species-specific issue. Its effect on the recruitment curve is similar to that of changing
both R and D. The parameter is important to assess because as the probability of a
population being in the upper water column increases the effects of hypoxia are reduced.
Figure F-5 shows the effect of increasing the acceptable percentage impairment on a
recruitment curve for the Say mud crab D. sayLTbis can further compound any effects on
a recruitment curve resulting from changes to other biological parameters. Clearly,
careful consideration must be given to what parameters will best represent populations
within a given area. If managers can justify the potential for a greater percentage
impairment to a population from exposure to hypoxia, then the DO recruitment criterion
becomes less restrictive.
Figure F-6 shows how the recruitment curve for larval inland silversides (Menidia
beryllina) could change as one moves from the Northeast through the mid-Atlantic and
down to Florida. Note that all of the curves are below the 2.3 mg/L CMC. However, the
graph does serve to show the magnitude of shifts that can occur for individual species
when changing only the recruitment season. If similar parameter information were
available for all of the species, then a generalized FRC could be calculated for each
region.
Finally, Figure F-7 shows the potential effect of eliminating some species on the Final
Recruitment Curve. This example is provided in order to demonstrate modifications that
could be made to the Virginian Province FRC to adjust for site-specific species
occurrence issues. In the example shown, striped bass (Morone saxatilis) is eliminated
F-l
-------
because it may not be exposed to hypoxia since recruitment of this species usually occurs
in the early spring, before hypoxia occurs, and often in tidal freshwater areas (Setzler-
Hamilton and Hall, 1991). American lobster (Homarus americanus) is eliminated in this
example because it does not occur in the more southern portions of the Virginian
Province. However, the data presented in this document are not just representing the
individual species for which we have such information. The data on the effects of low DO
are intended to represent the range of sensitivity expected to occur in the communities of
saltwater organisms within the Virginian Province. There are a large number of species in
the environment which we cannot test in the laboratory. Thus, care should be taken before
eliminating any data from the data set, and sufficient information must be provided to
justify the change. A species that might be eliminated may represent the sensitivity of a
species that is present in the community of concern, but has not been or cannot be tested.
F-2
-------
Changing D Only
~ 5.0,
_j
a 4-°-
§> 3.0.
O 2.0 -I
T3
5
o 1.0 J
CO
CO
Q 0.0
0 5 10 15 20 25 30 35 40 45 50
Time (days)
Figure F-l. Effect of changing larval development period on recruitment curve of the Say mud crab
Dyspanopeus sayi.
Changing R Only
o>
0)
I
T3
CD
"5
CO
CO
Q
5.0 n
4.0-
3.0-
2.0-
1.0-
0.0
0 5 10 15 20 25 30 35 40 45 50
Time (days)
Figure F-2. Effect of changing larval recruitment season on recruitment curve of the Say mud crab
Dyspanopeus sayi.
F-3
-------
Changing R & D
^
~D>
C
CD
D)
>>
X
0
"D
S
o
CO
to
Q
5.0 n
4.0-
3.0-
2.0-
1.0-
0.0
o«£SSS g 2 2 i
-CP^^J^^AA^"^
o°cfAffSi
§^
o
g
o R=66, D=28
n R=80, D=21
A R=94, D=14
0 5 10 15 20 25 30 35 40 45 50
Time (days)
Figure F-3. Effect of changing both larval recruitment season and larval development period on
recruitment curve of the Say mud crab Dyspanopeus sayi.
5% Impairment
CD
CU
!
T3
0)
"I
CO
CO
b
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0 5 10 15 20 25 30 35 40 45 50
Time (days)
Figure F-4. Effect of changing percentage of a daily cohort exposed to low DO on the recruitment
curve of Say mud crab Dyspanopeus sayi. Recruitment season was 66 days. Larval development
period was 21 days.
F-4
-------
Changing % Impairment
2"
a
c
0)
D)
X*
o
T3
CD
"5
CO
CO
Q
5.0 n
4.5 -
4.0 -
3.5-
3.0-
2.5 -
2.0 -
1.5 -
1.0 -
0.5 -
-**^ 1 I I I I
n AAJD°°
n^o°#
£p Aff^
^°
D5%
A 10%
o 15%
0.0
0 5 10 15 20 25 30 35 40 45 50
Time (days)
Figure F-5: Effect of changing the acceptable percentage impairment for seasonal recruitment curve for
Dyspanopeus sayi.
Menidia beryllina: 5% Recruitment Impariment
2.5-,
2.0 .
CD
0)
CO
TJ
(D
"o
to
CO
1.5.
1.0-
0.5-
0.0
10
20 30
Time (days)
40
50
Figure F-6. Effect of changing latitude (and therefore recruitment season) on the recruitment curve for inland
silverside Menidia beryllina. The recruitment season for the Northeast was 42 days, for North
Carolina/Virginia was 180 days, and for Florida, 300 days. All other parameters for each of the curves were
the same.
F-5
-------
I
•o
>
o
(D
-------
Appendix G. Time-to-Death Curves Used to Generate the Regressions in
Figures 9A and 9B
G-l
-------
G-2
-------
Dyspanopeus say/
2.5-,
2.0.
i.
s
o>
O
"§ 1.0
o
(A
in
0.5.
0.0
LT10! y = 0.511 Ln(x) + 0.892
r2=0.98
: y = 0.493Ln(x) + 0.672
r2=0.99
T50: 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.
G-3
-------
Palaemonetes vulgaris
2.5
2.0
f 1.5
o
01
I
"5
U)
J2
Q
1.0
0.5.
0.0
LT10: y = 0.449Ln(x) + 0.856 A
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
Time (hr)
20
25
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 from this study. Solid lines are
logarithmic regressions of the four data sets. Regressions were calculated using Microsoft® Excel 5.0.
G-4
-------
Homarus americanus
4.0
3.5.
_ 3.0.
I
£ 2.5
c
o
O)
12-°
•o
§1
1.5.
g
5 1.0
0.5.
0.0
LT10: y = 0.487Ln(x) +1.827
r2=0.71
LT50: y = 0.363Ln(x)-H.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.
G-5
-------
O
1.6 =
1.4.
1.2.
0.8.
0.6.
0.4.
0.2.
0.0
Brevoortia tyrannus
LT5: y = 0.156Ln(x) + 0.804
R2 = 0.91
LT50: y = 0.084Ln(x) + 0.618
= 0.96
LT95: y =*D.043Ln(x) + 0.469
R2 = 0.90
20
40 60
Time (hr)
80
100
Leiostomusxanthurus
LT5: y = 0.058Ln(x) + 0.542
R2 = 0.96
LTBO: y =?T).050Ln(x) + 0.478
R2 = 0.97
LC95: y = 0.043Ln(x) + 0.421
R2 = 0.95
40 60
Time (hr)
80
100
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.
G-6
-------
I
c
0)
O)
T3
I
s
(A
1.6-
1.4.
1.2.
1.0.
0.8.
0.6.
0.4.
0.2.
0.0
Salvelinus fontinalis - small fingerlings
y = 0.204Ln(x) + 0.97
0
y = 0.21 OLn(x) +0.844
R2 = 0.91
1 A
y= 0.150Ln(x) +0.884
R2 = 0.97
' = 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
0.235Ln(x) + 1.177
0.95
= 0.207Ln(x) + 1.013
R2 = 0.87
Figure G-5. Time-to-death curves for LTSOs of small fmgerlings (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.
G-7
-------
Salvelinusfontinalis- large fingerlings
2.0 .
1.8.
f 1-4"
8 1.2.
| 0.8 4
g 0.6.
"o
"> n*
= 0.345Ln(x) +0.773
= 0.990 O
= 0.260Ln(x) +• 0.709
R2 = 0.945
y=0.200Ln(x) +0.645
= 0.984
= 0.180Ln(x) + 0.606
R2 = 0.963
0.165Ln(x) + 0.616
1^ = 0.913
0.2.
0.0 .
0 5 10 15 20
Time (hr)
25 30
Salvelinusfontinalis - large fingerlings
O]
03
I
O
$
Q
y= 0.180Ln(x) +0.720
B #= 0.949
y=0.168Ln(x) + 0.725
= 0.856
y = 0.161 Ln(x) +0.660
= 0.989
200
Figure G-6. Time-to-death curves for LT50s of large fingerlings of the freshwater brook trout Salvelinus
fontinalis. Data are for fish acclimated to different concentrations of low dissolved oxygen (2.5 to 10.7
mg/L) and then exposed to different concentrations of low D.O. (A), and for fish acclimated to 7.1 mg/L
in the dark and then given different light pre-treatments (B). Data are from Shepard, 1955. Solid lines are
logarithmic regressions of the four data sets. Regressions were calculated using Microsoft® Excel 5.0.
G-8
-------
Appendix H. Growth Data for Constant Versus Cyclic Exposure to Low Dissolved Oxygen (Coiro et aL, 2000)
Species
Dyspanopeus sayi
Dyspanopeus 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
Paralichtkys dentatus
Paralichthys dentatus
Paralichthys dentatus
Paralichthys dentatus
Paralichthys dentatus
Life Stage
larval
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 low/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
DO
Minimum
(mg/L)
4.5
3.6
2.6
1.5
4.2
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
51
69
21
56
75
0
13
55
3
13
64
35
45
18
31
33
47
Appendix H: page H-l
-------
-------
\-\ •aSed :\ xxpuaddy
jajsqoi UHouauiy jo uosuBduioo "i xrpuaddy
-------
:i xipusddy
-------
"I~1SFRLu? P3;lsH 3-re suopipuoo
3Apisuss jsoui aqj si (sui{ pips ppq—snuvouzwv sruvuiojf) jajsqoj UBOUSUIV
jo ssiosds IEJ3A3S joj (pjjuoo oj aApBpi juauiiredmi aSBjiraojQd) qjA\oj§ jo joy -[-I
O'S 0> OB
O'l-
0
•01.
•03
•03
0
•» 3
KB?
ros g
hot I
08
06
001-
OH
-------
"O *T3
a~ cr dr
2 g. I"
=5- §• §
S'
a3
eft r^
a a
Si "i ~
Co Co C^
£•£•5
a a a
a* & £>5~
5 If §.
ft ft ^1"
3 5 »'
1 §
I
a
a. S;
§- §
»• f
a a a a
3552
a"
1
§
3233325!
<•>" o" o" D" D' D'
a a"
a a a' a a a
a a a a a a
a a a a a a
g£3333333o>>>
3 3 IV* Mp?C?PCJO^E3E3E3
H^r-ti-ti-ti-t^tr-t frrt C3 P3 9
3ScncnwwwcnwH3a>OO
3 3-3-3-3*CT3-Crg 2. D. S
E? Q rm rtn rrft no no no rto co ^ ^
- 03 03
S 3
oc
ilt§(§t§c§w'<§(§8s?»»>
S. B. S. S. S. 3 3 3 3 3 3 3 3
g1 g5 Lo S S
S S in la in
S
03
03 03
CO CO CO co>O o'o'o'o"
o>. al cr cr — ~
os 03 03 a.Q.o.Q.a.o.o.c.c.S
*&
S 8
&§-
cooo
'
g.S.f|-g-B-3-g-g-&
""§'§" 3" 3" 3' 3" 3
"O *O 13 13 13 13 *O
w w
CD Q G> CD CD O O
3 3
3 3
O O
3 CO
1 -1
Ct C5
0\ "Co
O O
CU Q.
-O T3
3 3
W W W CO
i CO 03 03 03 05 TA
TO TO TO TO TO S.
O O CD O O 03
_._._! ^^ _ TO
i i i i <"
4x 4i. OJ to ,..
to to
-J oo oo
A
en TO TO TO TO TO
. ^ to (jj tO U3 U)
to oo
O [O
TO OTOW3 = 3
' O
i T3
§
*. * . .
M_ S
i 1
•i *
03
cr
vo
A
oo
10 H-
p vo
to o\
-tv y» vo
b b 43k 4^*
10 JO fo 0\ ,U O O O
o\ a\ to oo V to
oo
VO vo
O\ ON
to
L.J OJ CO CO
° c; gg o
tt\ \J\ (-/» CJS
to 43.
o o
to
en
•a
n
S.
O
o
5
z
S3
1
03
ET
n
i
•a
^-, co
1 =
^ 3.
SB *o 2.
s J? E
§ z
& z
K o
89 *
J?
g
5"
o
o
I
I'
OQ
3
el-
s'
-------
•o
I
O O
5t- ft*
§ §
£ £
S S
S S
S; S;
a a
<*> ^
O CP
£ i
II
OO -^
^ 1
r 40% mortality 0
r 5% mortality 1
P P.
1 S
o re
l-l t-t
ja o
3 3
R£
OQ OQ
0 0
Ln L/I
OOOOOOOOlClCltaCaCataCataCaCaCaCa
S;K;S:S:S;s;s:s:^'^*^ "* "* "* ^ ^ °* "* °* ">
a* a' a" a" a* a* S" S" 5" S" ^ Q Q ^ ^ o o ^ G o
ftc^ftftftc^tftf^f^e^vjwOvjOQQvsOO
cinrsonocs'Sori-'-'s-'.-t-i^-i!^-'^
SSSSScSStScSSSsaaSaa'aS'S'S'
Vit^c*jVitot^c*iO5C*jc*j ^3- »"J (31 . <^f t^- t~j- t^f ;^f (3- >>>>>>>>>
&3
goooooooooaSSSSHSSSS
BB»»»B»»BOOOOOOOOOO
Cy^O^
O
N>
ONOOOOOv^^^O^—OOOOOOOOOOOOOOOOOOO
3-=P3-Er=r=r5"5'g'g"g"sr=r3"S-3^3-3'B-3-
4x(JlN-ootO4^-^-LatO»-*C-o ui
33>^S>33h.jO33
0030003^00
3.aoaaao»aa
B.E.gg.g.e.g-5e.g.
>5<5 5?<5'5'<§' 5?'-i*n^Qil^'*r^r*r^'^lrt''^1'H'
OOOOOOOOOOs ^o^vo^o^ovo^ovovo
OQ OQ CTO TO OQ OQ OQ OQ (TQ Op fj
»»»».»3.
to
OJ
I
a
a-
£a
£T
o
1
to
3
o
S.
3
OO
£
5
1
«
•0
OO
o
Co
ft
i
a
g
<^i
1
c menhaden
S
3
0
S.
3
OO
?
§
Ul
1
1
CO
o
to
ft
§
a
g
&9
sr
o
1
to
3
0
S.
3
00
OO
0>
5
o
I
CO
o
CD
ttl
§
a
cS
I**
3
ic menhaden
to
i
b
£.
3
OO
?
5
I
?
CO
O
Co
n.
1
a
i
§
&
s
o
1
3
to
3
id
S.
3
OO
r
§
Ui
O
1
S
VO
oa
O
ta
»
§
a
a'
g
&3
3
ic menhaden
to
3
o
g
OQ
OO
F
o
o
•J
p
CO
o
CS3 &3
c5 nj
11
a a
s i
§ g
3 3
ic menhaden
c menhaden
(0 to
3 3
3 3
S o
1 I
OO OO
F FT
ss
tA vo
Ui
.— o
g£
? ?
11
Co Co
5 1%
1 1
a a"
"§. "3
i g
g g
3 3
ic menhaden
ic menhaden
to to
B 3
3 3
0 0
3 3
ere era
to to
OO OO
sr ?
o o
Ln LA
f"l >!— •
-J O
0 0
S 1=
1 1
o o
CJ JU
vO vO
§ §
en
S
S
o
3
z
ffi
S
»"
ffl
u
2.
5
§'
H
O
3
§
1
3 9
M
*^5
rf
n
|
lile/adult
-------
&
•|
i
8
1
o.
1
o
o
Ul
sr
G
o
§
Vargo and
GO
•3
5
OO
5>fi!giglgl&glglg3gl&
'S'S^^^'S'S'S'S'S'S
2323322^^2°
!|!|!|!|!|!|!|l|!|!|s|
00000000000
O.O.O.O.O.O.O.O.O.O.O.
g3MMMMC3E3G3G3C3GJ
0.0.0.0.0.0.0.0.0.0.0.
eeeececCCCC
3 S.3 £.3 EL3S.3 E. 3
O C9 O Q O O
^"-wwoo^^^wo
ppppppppppp
QQGQQQGQGQQ
ooooooooooo
ooooooooooo
ggggggggggg
0.0.0.0.0.0.0.0.0.0.0.
•3<3<3-3-5-3'3<3'5>3<3*
vo vo vo vo vo vo vo vo vo vo vo
CO OO OO OO OO OO 00 OO OO OO OO
gl
'S
2
1
53"
CO
o
Cu
s.
T*
a
5
P
G
o
2
Vargo and
8
^
VO
OO
a1
•§
1
I
O
o.
1
r*
3
0
u,
p
G
o
a
Vargo and
8
,3
VO
OO
gl g]
•5 9.
1 a
II
CD CD
O O
O. Cu
II
r* •-••
£4
«- a
P to
n H
0 0
8 *
Davis and
Vargo and
8 3
•3 5s
vo *-
So g
a
g
<8
OT
o'
p
3
sr
s!
S I'
£ ™
o
0
Ul
Baker and
|
i
>--
vo
vo
to
p
i
^
1
g
a
o.
tr
3
•a
o
c
(M
1
^*
8
to
tp
to
OO
o
G
o
p
Poucher ai
5.
O
o_
o
VO
5
s? ?
3. f
^a •• »
& ^
I|
cm cr
* o
S 3
& =r
II
to
° S
S- B-
u» ^
o 3
1
5?
**
A o
S «
J2 °°
Carpenter
Theede et
r |
s Q
S (n
_O
VO
O 05
a a
1 S
1 I
t-3 Co
•§ -1
1 1
a* o*
c c
CD O
O O
3 3
cr cr
II
O ON
to to
S" 5"
CA O
e?3 ®
1 3
a. o
1 1
>- o
S JS
Carpenter
Carpenter
II
p p
ftft ftrt
p p
VO VO
t-n t^i
O O
S" S"
11
f 1
cr cr
CD O
O O
i-i-
II
to M
3- 3"
^^ ff«
0 0
3 3
o o
B. £2.
•§'•§'
^ o
o o\
Carpenter
Carpenter
1 1.
n o
.5 ,5
P p
VO VO
-O -J
en
a
S.
$
n
ommon I
^
1
i
en
1
en
£L
H
i
O
1
I
3 ?
Reference
I
CI>
g.
-------
I
CD
-
"g
£
5
3.
S
2
nj
»
1
1
O
g.
£*
0
o
£
e
o
o
to
H
8
o.
O
O
EL
\o
*o
R*
I"
S
•3
K
&a
S
1
•8
o
OO
oo
3
-J
to
oo
j£
's
<-n
o
0
o
CO
1
s
r-
\o
R-
pi"
5"
a:
c
=0
S
S
^
&
•8
o
oo
oo
g
3
--1
to
oo
3"
1
0
o
CO
I
o
~
0
R-
a
5
•5
S
&Q
S
£
-:
s;
•S
o
oo
CO
3
^
to
oo
cr
§
o
00
CO
|
£
S.
0
R1
%
3
2
s:
ts
g
1
O
oo
oo
-J
to
oo
-J
to
tr
Ol
o
CO
1
S-
£L
oo
^
sr
£
^
2
s
s
2
£)
o
oo
oo
^
to
CO
-^
to
cr
o
o
oo
cd
1
CD
£3.
oo
^
sr
%
s
5!
S
z
S'
1
en
O
OO
00
-J
to
oo
-o
to
sr
Lrt
O
^
CO
1
CD
£.
CO
o
5T
a
3
5
5
g
1
o
oo
00
3
3
-j
£
t?
1
o
u»
CO
1
t»
E.
VD
OO
^
ST ST
1 S
5 5^
% %
s g
3- S-
11
en en
"8 1
oo oo
CO 00
3 3
3 3
-J -O
to to
oo oo
3- 3-
o o
Ui O
O Ul
0 0
sa a;
CO CO
§ i
3 3
O CD
E. S.
00 »
O ^
?T?rsrsr^Ri Oi
-J-J-J-J-J-J-J-J^J^J-J^JO-jggg^g^^gWtOtO
to to to
ooooooooooooooooooooooooooooooooiutoo^^ului
^"|: ^f^l1^^^^^^^^^^^^^^^^^^^
BiBBiiiiBBiiBiiiBSiBiiiii
as^^g^a^fe^^fe^ss^s^^ggs^g-
cowrowwwcowd3roci3WG3co<:<:<:<:<<:<:<:<:-i<:
§§§§§§§§§§§§§§^""00000000
aS.^s.^s.aa.E.B.B.SSSgg^^^^^^^0^^
o o o o o o o o o o o o o o t^* &r yr
v^ i^ ^ \O \Q VD VO VO VO ^O '"O
^_t_tl_COOOOOOOOOOOOOCO
-O -0 -J
to to to
$
Q
2
^
-t
&
*£*.
^
1
0>
o
o.
g.
G
'
3
-------
Tautoga onitis
Tautoga onitis
I I
0 0
TO TO
e* c'
i I
O CD
LO LO
LO LO
to to
to -^
•£"• to
u>
?s
^5 °
if
p p
bo Ln
•£>• OO
c c
g. g.
5? £?
1 1
9 9
3 3
VD VD
**J *J
Tautoga onitis
I
o
OQ
I
CD
V
to
to
4^
*.
ET
8
1
p
Ln
oo
g
o
1
9
3
IS
"O.
Tautoga onitis
i
TO
C*
i
0
CO
o
1
LA
to
oo
3.
5'
R a
0 ~J
ll
II
CO
K
o
f
o
n
E.
S
Sn
Nereis diversicoi
o
polychaete worm
1
d
^.
0
§
ET
o
to
1—*
i5
n
o
o>
-
S
Mytilus edulis
blue mussel
1
0
^«
o
1
u<
o
s
CD
0
CD
CD
U
O\
My« arenaria
o
cp_
0,
1
^
ui
^^
O
?
O
A
bi
S
CD
CD
CD
£.
U
c\
Mulinia lateralis
coot clam
<;
CD*
to
-^4
"
to
O
to
"
s
CL
P
g1
?
a
Duration 1
1
i o
^P
S
JB
g
8
I
"0
i
r
-------
I
CO
g.
0<3
O
ft ft <%
8 I
I
5'
o o o cs
..
2 3 s | |-
llllllllllllll;;llllllllllllll
333
O O O
33333333333333333333
oooooooooooooooooooo
II
{B EP
1 1
«_ 00^ CO^ CO^ C0_
era era era era era ., ., ., ,.
TOTOOcBTOTOOTOTOO
cowtotototototototo
eg
Iff
TOOTOC3C3t3£3C3£3C3tf _K *-•
H-»— H- o'o'o'o'o'o'o'o's-r'
5T
tOtOtO4^4^4^.-t^tO
=-^5-?5-?SfEf
n
en
ooo
_. __ _, _, _. _.CnCnC>iC>iOtUiCftt^LnLftLnu»u»u»tnuii-»»wi\jiv=nwiwis::.l i_
OOOOOOOOOOOOOOOOOOOOOOOOOOOOO 3
oo
L/iLft
>-• p p to jo h- p to
vO^.\O4^Uti— -f* U\
OQ era era era OQ OQ era era era
cucucucucucucucu
COCOCOCOCOCOCOCXl
44<5<|f4<|^
OOOS
sC:l
LO
u
5f a
o.
era
to to
In Ln
BBBBBB g g i
(T> O TO
D. CU CU
•8 S
en en
o o o
pop
'•— ^- to
en
fe" &
•o 3.
fD
-------
S
I
I
1
f
«<"
^
3
o
^J
o
e
I
°
f
?
st
1
8
:»•
£3
S
S
g
%
•§
0
a.
cS
_
Q.
«
hi
™
H.
c?
cu
W
a
o
BT
O
f?
ET
0
>-
|
fa
£3
VO
VO
to
§>
SI
•g
o.
%
HH
V?
S.
o.
M
a
o
S.
£
P
13
1
o
5
VO
to
p
^
3
^'
s
2
g
I
i-
1
o
S
S3.
o
i-
LO
ep
w
to
S
_
o.
i
a
o
to
in
oo
1
?f*
|
§'
VO
VO
g> £>
rj n
ft ft
i 2
g g
3 3
0 0
i-i-
§C£
Jf
OQ era
EJ CP
I" "
* 3
"" TO
o a
i-1
9 ^
UJ U
to to
VO O
NJ KJ
& &
Co M
II
to ui
w o
VO W
II
o o
i" ff
1 1
P 9
o o
§" 3*
vo vo
£> §>
^ K
ni (^
S 2
g g
3 3
i-i-
CO CO
If
^- Ul
< 3
g g
to
VO LO
U> 0
to
7> to
VO
Sf
o o
to m
S3 g:
c s^
o OQ
sr o
« P
3 O<
CL Co
P S
1?
H- vo
vo -J
vo ^J
§>
^
T
|
g
3
S-
i-
CO
1
01
3
g
LO
0
g
ET
a
OJ
to
1
o
o.
CO
-1
vo
£>
rj
n>
I
g
I
a1
CO
1
Ul
<
8
o
Cn
Ef
i
o
vo
Ui
1
O
1
4
VO
P
P£
^
2
2
g
3
g-
0"
1
Ul
g
g
o
o
ET
i
OO
to
1
O
B
g,
CO
g
•5
-a
p p
^ r*
« »
3 2
2 2
g g
3 3
0 0
1 1
S S
<% <%
Ul Ul
S" S*
g g
LO LO
O O
8 S
to to
ET ET
a a
O 0
!° .-
1 1
O O
g. g.
g1 i?
•1 -3
3 =}
??????????
P^f^pj^jrc^RKKS
««««wrerereftre
iiiiiiiiii
gg§Sgg§ggg
3333333333
oooooooooo
IIIIIIIIII
S'S'STpTB'S'^^jsrjB
tA(ji*.ju^.^*.4^4i.^>.
gss§§se||s
gggg^gggrog
oooooooooo
-5tog5;5,og.-5
^^^1^^
oooooooooo
O«UP—OOIO>— OO
sssssgftSogy:
IIIIIIIIII
oooooooooo
O.Q.O.O.P.Q.O.C.Q.Q.
coc/3Cococ«oat>3Cocnoa
^^^^^^^^^^
VO vo vo VO VO vo vo VO vo vo
p p p p £
p ~ pr pr p?
re re n re ft
=• 3- ^- ^- ~
1 1 1 1 1
g g g =j K
33333
O 0 O O O
1 1 1 1 1
It It I
Ul LO LO U LO
g | | S 8
g g ro g o
00000
JO g - g JO
Hlffg?
p O O O O
*• 10 0 f to
lll<<
o o o o o
Q. o. o. a. a.
oa w co en en
«3 5 4 5 *?
vo vo vo vo vo
5?
n
S
f '
O
3
9
S3
3
a
i
s.
rq
CO
II
H
3
ri
3
1
5*
3
f
•S1 O
!§
0
p
S
f
-
09
O)
-------
CB
&
><
<—I
vj V?
2. 1.
II
< "£
(b ft
Oq (>q
a a
1 1
3* 3-
8 a
M en
ff CD"
p P
5. 0.
1. 3.
3' 3'
3 3
0 0
Z Z
|£
^2 co
S- 5"
g £.
0 0.
»• £•
g
Q
U) W
7" 9
Ui UJ
to i—
to to
9 9
to to
1>J
to
to
^
0.
3
to
^
en
O
0
en
LO
Poucher
1
p'
3
VO
VO
n
|
1
s.
S'
o"
a
g
c»
o
S"
I
Ci*
'
to
to
t-n
S
3"
O. O.
s -1"
s «
§ 0
0 CU
3- «
£.
cT
CD
•3
5*
—
In
CD
i
0.
1
3
3
lo
P
Q
1
o
%
I
5"
OQ
H"
o"
a
%
S"
o
ST
&3
CD
to
to
Ul
Ov
3-
Ul
to
% reducti
o
3
JS'
tn
n
S
o>
3
h-
LA
1
»
3
0.
?
3
3
VO
vo
to
0
g
C£
%
I
S"
o"
n
&
3T
o
f?
*i
|
CD
to
to
Ef
LA
LO
% reducti
o
3
J2*
2.
S
re
3
—
m
CO
1
"1
Ci
3
o.
I
vo
vo
to
o
1
^
s
3
>
1
o'
3*
1
(TO
•<
Ui
ac larvae
.
.
to
M
o
to
OQ
DeSilva
O
3
5
f
S5
•^i
LO
Ci
g
1
1
>a
K
1
£2.
n
1'
o.
cr
CD"
1
I
3"
0
to
o
to
to
B4
VO
o
% mortal:
*<
—
lo
Saksena
1
0
f
5
to
OOC^OOOOOOOOC^OClOOOOOOOO
1 1 l-l iiiiiiiiiiigsssess
^^^:^a,a«:^a,a,a,a,a.a^CL.a.a,a;.a,a.a,a,a,
Cl CJ *tf]C; b^. ^ G Cin'CjJCl O ^ O Q ^ Ci C5 CJ Q C5 Cs O Ci Ci
K: s: K'- K x K: K sc ;s s K as; s K K K s: K K sr s
11111 I'llllllllllllllll
MVitntntnwvivtt/>v>wviV)t/>uiu)vitntnviu)m
CDCDCDraCBCDCDCDraCDCDCDCDCDQCDCDOOCDCDCD
0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0-0.0.
£SSHIS!SSISSSSSS2!Ea"crcr cr cr cr cr
CDCDCDCDCDCDCDCDCBCDrtOCDCDCDCDCDCDCDCDCDCD
3333333333333333333333
3333333333333333333333
333333333BBB333333333B
§i§§iiiiiiiig§iii§iig§
asa|rsaal"aaaal"l"sl"|1|"aal"a
IIIIIIIIIIIIIIIIIIIIII
tototototototOtototototototoiotototototototo
o o o o o o o o o o o o o o o o o o o o o o
totototototoiotototototototototototototototo
OiOOOOOOOOOOOOOOOOOOOOO
tOtOtOtOtOtOtOtOtOtOtOtOtOtOtOtOtOtOtOtbtbtO
tototoNJtotototototototototototototototototo
^OJtO>— *OvoO°-JCVL/l^LOlO^^OV^Go^J^t^'^'tji
OvO\LriLftUtC^tCAJ^^.LOtOtOtOtOtOtOH^)~H^H^UtUi
3333333332333333S33311
OOOOOOOOOOOOOOOOOOOOJ5.J5,
v<*-<*tOtOlOlOtOtOtOtOtOtOtON3tOtOtOtOtOtOlOtO
ooggoooooooooooooooooo
CO t/3 t/1 CO CO C/2 ^ CO CO W Kl W CO CO CO CO CO CO CO W M CO
pa pa Ei cj pa M ja cs o^ ca pa opwwsftswcjppfts
ww»M*cn'*ww*w*w*«*^?*^?i^'"
§CDfOOOO>C8C5OOOCSC8OCl)CDOltT>f6OOCD
333333333333333333333
Q.O.O.CUQuQ.CuCuCi.CUCuQ.O.Cl.n.O,Q.D.Q*C^Q*Cl*
oooooooooooooooooooooo
•i-s-i'i'Slll'i'i'S'S'i'i'S'i-i-i-l'Sj^
=• ?• ==• =• =• ?• p- »• p- =r =r B- sr s- =r B- 3- 3- 5- 3- 5- S-
SSSSSSSSSSSSSSSSSCSvSvSvSCS
tototototototototototototototototofoStototo
o
1"
a.
§•
K
5
w
1
cr
3
3
3
f»
3*
I
to
O
8
S
?
LA
O
ro
o
•^j
Saksena
Q.
g
f
.^j
to
f
a.
1"
sr
£
^
2.
1
cr
3
3
3
O
a
I
to
o
g
B
ff
LA
I
to
0
-o
Saksena
!
g
f
«^j
to
a
n
2.
Crt
n
|
O
I
§
ffi
EJ
GO
IQ
CM
f f
~5
a
3
a
I
f
M
S
f
•"^ M
1 b
|0
Referen
8
-------
C5 C5 Q O C2 O O
§•§•§•§•§•§•§-
o 5££
11111111111
ft c 5
&• a- f^
S §
mil
>§ >§
O3«C/3C/aCO&QGQC/aCO
- — J» S3 » ja » » J»
O. O. Q.
OT OQ OO I
§•§•§•
v; v- «< '
.
a^a . .
3. 5. 3- 5.
vs <-< v- v; •< v< ><
33333-3
333
0-0.0.0.0.0.0.0.0-
o o o
S 3 3 w »
a* cr o* a* a*
i-S-e-
CD n a a a a w
fDCDfDfDCOfBQOCDCD
O.O.O.O.O.O.O.O.O.CL
JO » O o o «
» 5 = a 3 £
CD CB
— M to to — i— to
tototoioio'ototototo
g 8 S
to to ^- ^- to _
OJ LA VO VO O 3
to to to to to
o\ o\ ^— o >—*
£l£lvOCO-J-J-J-J
""'JQ.Q.&-O-a'0-0-0-
o o o o o o
3 3
tO tO tO LO 4^
lo bo In o 4^-
-P^ LO Ul *— •—'
VO VO VO VO VO VO VO VO VO VO VO VO VO VO
-J-J-J-O-J^J-^J^J—J—J^O-J^J-*1
totototototototototototototo
to 10 to to
§
CL
-------
S'
a.
g:
1
a
S:
f
a
5- >
TO TO
3. 3,
o o
g g
> > >
s. §. s.
§ s s
> > >
333
§TO TO
n' g- S-
> > >
333
CD TO TO
a. a. a.
iE
3 3
-pTT-rpr
S.S.2.S.S.S.a.a.§.2.g.S.
.£.S.ol
"S
a.
S
g
§• §•
t/5 CO
TO TO
O* O* O* O* _
*<'<;*<*<«<:
Q S
TO CT TO TO TO TO
0 to 3.
a
OJ tO OJ -t*. OJ
a g. a jr a
S Q1 S O O
&. w ^ u\ jx
S- 3. S- 3- 3-
g w g. g. g
i I i i i
>-. (— 4^. W tO OJ
w w ^ 1-1 >-i "^
g- o o 3- g- g- 5
w M to (v, ^ ^ j^
a B. o. s1 s- s- s-
OT Si £S. 53. 52. JS. w
no 05 WQ OQ eg pg 09
>— >— 3333333=3 3 3 33
"
SO
P
§ g
» u » w M » »
Q g CD CD « O TO
OJ OJ OJ tO OJtO tO OJ tO OJtO
t/1 OJ OJ OJ OJ ^3 OJ OJ O OJ OJ OJ C5 OJ OJ
N- N- 10 >- (On- _* N3 O ^- N-
--.-
o oo s c; oooo
M^M,3««S5S^^^^ -to-tototo
(-nOOt^j-^t^^.^
H-H- >— M^-tOtOtOtOtO^r^
OOOOtsjVOOCXOOOOOO O
to>—'•— totoJ-iotototototLifj
*-j >J ^J IO tO IO
r^ -i —i :-i (—» r-i
10 to 10 to to
00009
I
t
Ci
ff
ra
n
g"
§
o
to
1 g- 5 g- g- g-
"S -• "S •< "io
MtOtOM — — —
t^DOOOODSIOC"1
»HCTCTICT>eT>CDOOCB("}
°ffffII£f°
D-O.O-0-O-O.^ O.
111111s-!
oJtovoij t-lnovoi- *.*.'— i>.t.loboo
VO — OWOVVOOV — LOOtOWO-iOvtOOjJo
s.
s.
S1 S1 c? c?
Is e s
>TJ ho TI TJ
g g g g
"§-£?•"
T3
CD
g.
•8
g
3
SO SO
O
srg- f rrr f n
3 L _
O. O. Q.
n o
333
o. o. o. o. o.
....,,, B g i1 S1 £>
3 3 3 3 3 3 3 c
VOVOVOVDVOVOVOVOVOVOVOVDVOVO
-J-*4*J--J'*J-^J'-J-J-J-*J'^l-^-J-*J
S. 3.
S O
-------
5 a a Q o £: «. ~.
s 2 is 2 ^ 5* K* ^
£r s s ?• S S S S
gs — 2:
222
S- ?5 n[ 5
|3 J? j? c? 3
g c^ t^ to O
*5 S S *S **"
§* (§•<§• *S
£§.§.*
V? ^- Q- o1
rrl 1
U ES J5 - °s ^s
I: It I: s s
%' %' s* 5* 5*
S.
.....
S S »' »' »' »' a' "a "a
§§52555
5 5
-< "t
a" a"
co 2 SI S
2. 2. 2.
2L. £.
O- O- CL Q. Q.
CO C^> 3 §• *•
™ ST
1
1
1
I f
cr <
o. 5 o _ „
rr" 5? £3 5" ca
3 3
i
T:
o
•p
LO LO CO LO
O O O O
l*> Oi t>J tii
to — — to
K- — o O O O
S. 2.
to to
— f •?•
o d, ,U
In
o
ri
IA -j ~>
g- g- g-
>S «< <5
g-
oo
o.
Ol
0.
1
4X ISJ to
CO Q. 4^
5-1 ?
o
i
I
1
p to to p
p
b\
to
c\
P r~ r" *" r*
b\ i* ^-a Ln vo
tn o o
u g.
g o
s> s.
g
I
I
9
? S
e c:
2.
£r s-
o o
5 g
99
VO VO
vo vo
-4 -4
TO
I
I
=S ^
I
TO CTQ cr
o. cu 2. 2,
«
C3= 8.
..
g.g.
3 3 3
§ §' §' §' 3 3 3 3' §' 3 §' 3
f
0>
3
O.
"g
'S
------- |