CBP/TRS 107/94
March 1994
903R94051
Chesapeake Bay
Benthic Community
Restoration Goals
QH
96.8
.54
C5
Chesapeake Bay Program
Printed on recycled paper
-------
-rfcO S^T.
<^ *«.
USEpAReg,onffl
1650 Arch St.
Philadelphia, PA 19103
-------
Chesapeake Bay Benthic
Community Restoration Goals
March 1994
U.S. EPA Region lit
K?;7iooo] Center for Environmental
h; forrr.atior?.
1G50 Arch Street (3PM52)
Phils-dfiiphin, FA 19103
Printed by the U.S. Environmental Protection Agency for the Chesapeake Bay Program
-------
CHESAPEAKE BAY BENTfflC COMMUNITY
RESTORATION GOALS
Prepared for
U.S. Environmental Protection Agency
Chesapeake Bay Program Office
Annapolis, Maryland
and
The Maryland Governor's Council on Chesapeake Bay Research Fund
Chesapeake Bay Research and Monitoring Division
Tidewater Administration
Maryland Department of Natural Resources
Annapolis, Maryland
Prepared by
J. Ananda Ranasinghe1
Stephen B. Weisberg1
Daniel M. Dauer
Linda C. Schaffner3
Robert J. Diaz3
Jeffrey B. Frithsen1
, Inc., 9200 Rumsey Road Columbia, Maryland 21045
2Dept. of Biological Sciences Old Dominion University, Norfolk, Virginia 23529
•Virginia Institute of Marine Science, Gloucester Point, Virginia 23062
December 1993
-------
FOREWORD
This document "Chesapeake Bay Benthic Community Restoration Goals" was prepared by
Versar Inc., for Mr. Rich Batiuk of the Chesapeake Bay Program Office, United States
Environmental Protection Agency under Contract Number 68-D9-0166 and Dr. Paul Miller
of the Chesapeake Bay Research and Monitoring Division, Tidewater Administration,
Maryland Department of Natural Resources under Contract Number CB92-006-004 by the
Maryland Governor's Council on Chesapeake Bay Research Fund. The purpose of the report
is to develop restoration goals for Chesapeake Bay benthic infaunal communities.
111
-------
ACKNOWLEDGEMENTS
We gratefully acknowledge the contributions of Tim Morris and Nancy Mountford of Cove
Corporation for updating the taxonomy and helping to identify and resolve taxonomic
conflicts among the many Chesapeake Bay data sets. We also thank B. Richkus, L. Scott, F.
Holland, S. Jordan, R. Eskin, and M. Luckenbach for reading and commenting on early
drafts of the report, and R. Newport and C. DeLisle for assisting in document production.
-------
EXECUTIVE SUMMARY
Benthic macroinvertebrate assemblages have been an integral part of the Chesapeake Bay
monitoring program since its inception due to their ecological importance and their value as
biological indicators. The condition of benthic assemblages reflects an integration of
temporally variable environmental conditions and the effects of multiple types of
environmental stresses. As such, benthic assemblages provide a useful complement to more
temporally variable chemical and water quality monitoring measures.
While assessments using benthic monitoring data have been useful for characterizing changes
in environmental conditions at individual sites over time, and for relating the condition of
sites to pollution loadings and sources, the full potential of these assessments for addressing
larger management questions, such as "What is the overall condition of the Bay?" or "How
does the condition of various tributaries compare?" has not yet been realized. Regional-scale
assessments of ecological status and trends using benthic assemblages are limited by the fact
that benthic assemblages are strongly influenced by naturally varying habitat elements, such
as salinity, sediment type, and depth. Such natural variability confounds interpretation of
differences in the benthic community differences as simple responses to anthropogenic
environmental perturbations. An additional limitation is that different sampling
methodologies used in various programs often constrain the extent to which the benthic data
can be integrated for a unified assessment.
The objective of this project was to develop a practical and conceptually sound framework
for assessing benthic environmental conditions in Chesapeake Bay that would address the
general constraints and limitations just described. This was accomplished by standardizing
benthic data from several different monitoring programs to allow their integration into a
single, coherent data base. From that data base a set of measures (Chesapeake Bay Benthic
Restoration Goals) was developed to describe characteristics of benthic assemblages expected
at sites having little evidence of environmental stress or disturbance. Using these goals,
benthic data from any part of the Bay could be compared to determine whether conditions at
that site met, were above, or were below expectations defined for reference sites in similar
habitats.
The approach used to develop these restoration goals was similar to that used by Karr et al.
(1986) to develop an index of biological integrity for freshwater fish. A set of candidate
attributes believed to have properties that differentiate high and low quality assemblages were
first identified, and reference sites believed to be "minimally impacted" were designated.
Properties of the biotic assemblages at these sites were then compared to assemblage
properties at all other sites. Properties that differed significantly between these two groups
of sites were selected as metrics to be included in the restoration goals. An index was
developed to assist managers in identifying the extent to which these restoration goals were
being achieved. The Restoration Goals Index (RGI) is calculated as the average score of
metrics, after each metric is scored as 5, 3, or 1, depending on whether its value at an
vii
-------
individual site approximated, deviated slightly, or deviated strongly from its value at the best
reference sites.
The restoration goals were developed based on available data from seven benthic survey
projects: the Maryland and Virginia Chesapeake Bay Benthic Monitoring Programs, U.S.
EPA's Environmental Monitoring and Assessment Program (1990), the Maryland and
Virginia Biogenics studies, a James River study, and a study in the Wolf Trap area of the
Virginia Bay. These seven projects were selected for several reasons: each provided data
readily available on electronic media; collectively they provided sample representation in all
salinity habitats of Chesapeake Bay; and all used a 0.5 mm sieve in sample processing,
which was a critical aspect of the study, since the numbers and types of organisms collected
depend on the mesh size used to sieve the sediment.
The attributes incorporated into the restoration goals included metrics from each of the
following five categories:
• benthic biodiversity measures
• measures of assemblage abundance and biomass
• life history strategy measures
• measures of activity beneath the sediment surface
• feeding guild measures
Restoration goals were developed independently for eight habitat classes defined by salinity
and sediment type to ensure that natural differences in benthic communities related to these
habitat factors did not confound interpretation of the indices. The eight habitat classes were
determined by cluster analysis of the composite data set.
Restoration goals were developed using data from only the summer period, July 15th through
September 30th. This restriction avoided seasonal variation that would confound
interpretation of benthic community responses to environmental degradation. The summer
sampling period was common to six of the seven benthic survey projects. Using data from a
different season would have reduced the data available because the various programs differed
substantially in the extent of sampling during other seasons of the year. An index developed
for summer was desirable because benthic communities are expected to show the greatest
response to pollution stress during the summer.
Three approaches were used to validate the goals and the accompanying index. First, the
Restoration Goal Index was computed for all samples taken from each reference site to test
whether expectations of RGI values greater than three were met. This test indicated a high
degree of correct classification; classification efficiency was more than 95% in five of the
viii
-------
seven habitat classes. The lowest correct classification efficiency for reference sites was
92.3% in the high mesohaline mud habitat class. Second, RGI values were computed for all
samples taken from degraded habitats to test whether expectations of RGI values less than
three were met. This test used data that had been excluded from development of the RGI;
therefore, it was an independent validation test. A high level of classification efficiency was
observed in this test; classification efficiency was 85% or better for degraded sites in five of
the six habitat classes in which data from degraded sites were available. The one habitat
class that did not validate as well was tidal freshwater. For the third validation test, sites
that were sampled more than once during the summer of any year were identified, and the
RGI was computed for each visit. RGI values at each site were evaluated for differences in
status between visits within each year to ascertain the stability of the index. Instability of the
index would indicate an unacceptable signal-to-noise ratio in the attributes. The results
indicated that the RGI index was relatively stable. The correlation between RGI values for
the first and second visits exceeded 80% for all habitats.
The validation results indicate that these preliminary restoration goals are effective for
distinguishing between sites of high quality and those of lower quality in six of the seven
habitats for which data were available for goal development. The only habitat class for
which the restoration goals did not validate well was tidal freshwater. Although restoration
goals validated well, additional analysis and development of goals appears to be appropriate
before the goals are applied rigorously for environmental management purposes. Steps for
further goal development are recommended.
IX
-------
TABLE OF CONTENTS
Page
FOREWORD iii
ACKNOWLEDGEMENTS v
EXECUTIVE SUMMARY vii
I INTRODUCTION 1
Background 1
Benthic Indicators 1
Statement of the Problem 3
Objectives of this Report 5
H APPROACH TO SETTING GOALS 6
Approach 6
Develop Compatible Data Sets 8
Define Temporal and Spatial Strata 9
Identify Reference Sites and Degraded Sites 17
Select Restoration Goal Attributes 19
Identify Restoration Goal Values 30
Develop the Restoration Goal Index 30
m DISCUSSION AND RECOMMENDATIONS 35
IV REFERENCES 42
APPENDICES
A LIST OF TAXA
B TAXA DELETED FROM ORIGINAL DATA
C STATION LOCATION MAPS
22\epa93\1627-003\9043-r
XI
-------
I. INTRODUCTION
Background
A large number of environmental monitoring efforts have arisen from the need for
information about the condition of ecological resources and the responses of those resources
to anthropogenic activities. Although chemical measures continue to be a part of most
monitoring efforts, interest in biological measures of environmental condition is increasing.
Information about biological resources augments chemical monitoring information and can be
more useful to environmental resource managers. Most environmental regulations and
contaminant control measures are designed to protect biotic resources, since these are the
primary concern of environmental resource managers and the public. The condition of
biological resources, therefore, provides a more direct measure of the effectiveness of
environmental regulations for protecting the environment.
Another reason for interest in biological measures of environmental condition is their ability
to integrate temporally variable environmental conditions and multiple environmental
stresses. Chemical conditions, especially those describing water quality, are often variable
and difficult to characterize. Biological resources integrate this temporal variability and
facilitate environmental characterization. The condition of biological resources also
represents an integrated response to multiple environmental stresses. In addition to
responding to chemical exposure, biological resources also respond to nonchemical stresses
such as habitat loss, diversion of water flow, and additions of terrigenous sediments.
Benthic Indicators
Foremost among the biological indicators suggested for assessment of environmental
conditions in estuaries are those based upon the abundance, biomass, species composition and
richness of bottom-dwelling (benthic) invertebrates. The attributes that make benthic
assemblages reliable and sensitive indicators of ecological condition (Bilyard 1987) include
• limited mobility - Members of benthic assemblages generally have limited
mobility, and cannot avoid adverse conditions (Gray 1979); therefore, the
condition of these communities reflects local environmental conditions.
• habitat - Benthos live in sediments, where exposure to contaminants and low
dissolved oxygen concentrations generally is most severe.
• life-span - Benthic macroinvertebrates generally have life spans ranging from
months to several years; therefore, population and community level responses to
environmental stress or disturbance are reflected in a reasonable period of time
(Wass 1967).
1
-------
• response to multiple stresses - Benthic assemblages are taxonomically diverse and
their members encompass multiple feeding modes and trophic levels; thus, they
display a wide range of physiological tolerances and respond to multiple types of
stress (Pearson and Rosenberg 1978; Rhoads et al. 1978; Boesch and Rosenberg
1981).
• integrated response - Because of their diversity and life-span, benthic
assemblages integrate environmental conditions present for weeks or months
prior to a sampling event. This ability to integrate local conditions provides
information that cannot be obtained from point-in-time physical and chemical
measures of water quality.
For these reasons, benthic macroinvertebrate assemblages are considered good indicators of
environmental conditions and have been extensively used to describe local ecological status
and trends in a wide range of aquatic environments (Dauer et al. 1988, 1989; Holland et al.
1988, 1989).
The utility of benthic invertebrates as biological indicators is also derived from their
ecological and economic importance. Ecologically, benthic invertebrates are some of the
most important components of estuarine habitats and may represent the largest standing stock
of organic carbon (Frithsen 1989). They are important links between primary producers and
higher trophic levels (Virnstein 1977; Holland et al. 1980, 1989; Dauer et al. 1982; Baird
and Ulanowicz 1989; Diaz and Schaffner 1990). Components of the benthos, such as
polychaete worms and shrimp-like crustaceans, contribute significantly to the diets of
economically important bottom-feeding juvenile and adult fishes such as spot and croaker
(Chao and Musik 1977; Homer and Boynton 1978; Virnstein 1979; Homer et al. 1980).
In addition to their trophic importance, the activities of the benthos significantly affect
oxygen, nutrient, and carbon cycles and may control the coupling of benthic and pelagic
processes (Kemp and Boynton 1981; Boynton et al. 1982; Officer et al. 1984). The
burrowing and sediment reworking activities of polychaetes and bivalves influence the depth
to which oxygen penetrates the sediment-water interface, affecting the rate at which nutrients
and contaminants are lost from the sediments to the overlying water column (Rhoads and
Young 1970; Aller 1980; Blackburn and Henriksen 1983). The feeding activities of the
benthos can directly affect planktonic components of the estuarine ecosystem and the
concentrations of particles in the water column. Large filter-feeding bivalves effectively
remove plankton and other suspended material from the water column and, thus, can improve
water clarity (Cloern 1982; Officer et al. 1982; Holland et al. 1989). For example, a filter-
feeding model (Holland et al. 1989) indicated that, during each day in the summer,
suspension feeding bivalves potentially filter nearly all (84 to 100%) of the water overlying
shallow (less than seven meters deep) regions in the Potomac and Patuxent rivers.
-------
Statement of the Problem
Benthos have been included as integral components of Chesapeake Bay monitoring programs
for many years. The state of Maryland has monitored benthos regularly since July 1984,
collecting an average of 450 samples per year. This program builds upon other monitoring
data that date back to the 1970s (Ranasinghe et al. 1993). The Commonwealth of Virginia
instituted a similar program in March 1985, which collects about 240 samples per year
(Dauer et al. 1989). In addition to these programs, which are specific to the Chesapeake
Bay, national programs also collect samples in the Bay. For example, the U.S.
Environmental Protection Agency's Environmental Monitoring and Assessment Program
(EMAP) has collected approximately 60 samples from the Chesapeake Bay and its tidal
tributaries each year since 1990 (Weisberg et al. 1993).
Existing monitoring programs have thoroughly described the benthic communities at various
sites within the Chesapeake Bay system. This information has been used to characterize
environmental conditions in the Bay and to demonstrate changing conditions (Holland et al.
1989; Dauer 1991). Benthic assessments, however, have yet to reach their full potential to
provide information to environmental resource managers about triggers and endpoints for
restoration activity.
The interpretation of information produced by monitoring benthic communities is currently
limited because there is no clear definition of the characteristics expected for benthic
assemblages in nondegraded habitats. Without this definition, it is difficult to quantitatively
identify benthic assemblages that indicate degraded environmental habitats. These
expectations were recently defined for a few benthic community attributes for the Virginia
portion of Chesapeake Bay (Dauer 1993). Most assessments to date, however, have been
limited to examining differences in condition at near-field and far-field sites with similar
habitat type, or examining changes in condition over time at specific sites. These types of
assessments do not lend themselves to addressing general management questions, such as
What is the overall status of the Bay across all sites? Are certain large portions of the Bay
more degraded than others? or What remediation activities are needed to improve degraded
habitats?
The development of expectations for benthic assemblages in relatively nondegraded estuarine
areas is an important step toward using measures of benthic condition in estuarine
assessments. These expectations
• provide a means to assess measures of benthic abundance, biomass, and species
richness quantitatively;
• establish criteria with which to determine the extent of degraded habitats in
Chesapeake Bay and identify those bottom habitats most in need of water quality
or habitat restoration;
-------
• provide a well defined endpoint for restoration activities; and
• permit intermediate determinations of progress (or the lack thereof) toward
attaining goals.
Defining expectations for benthic communities in nondegraded habitats in Chesapeake Bay is
a fundamental prerequisite to providing state and federal agencies with the information
needed to plan and evaluate the effectiveness of restoration activities.
Historically, efforts to identify characteristics of benthic assemblages associated with
environmentally nondegraded and degraded areas have been limited. In part, this was due to
the strong influence of habitat on the composition of benthic communities. The complexities
of identifying characteristics for each type of habitat (where habitat is defined by salinity,
sediment type, and depth) were a stumbling block, and most benthic assessments were
limited to comparisons of benthic assemblages within similar habitats. Comparisons across
habitats were infrequently attempted. Benthic assemblages are used more frequently in site-
specific assessments, where natural variability is minimized, and in studies of trends, where
variability is minimized by periodically returning to the same site or area (Pearson 1982;
Rosenberg and Moller 1979). One limitation is the paucity of data with which to develop
expectations for nondegraded environmental conditions. Few living resource monitoring
programs collect adequate data, over large enough array of habitat types, for sufficiently long
periods of time for investigators to feel confident about defining "expectations" for living
resources.
Recently, there have been several efforts to develop living resource community measures
with broader application. For example, Karr et al. (1986) developed the Index of Biotic
Integrity (IBI) to characterize the condition of freshwater fish communities in streams.
Hilsenhoff (1982) developed an index that reflects the condition of benthic macroinvertebrate
assemblages in streams. Word (1978) developed the Infaunal Trophic Index (ITI) based upon
the species composition of benthic communities in Southern California coastal areas. Dauer
(1993) developed expected values for several benthic infaunal community measures for the
Virginia portion of Chesapeake Bay. More recently, a benthic index was developed for the
U.S. EPA's Environmental Monitoring and Assessment Program (EMAP) using benthic
infaunal measurements from estuaries between Cape Cod and the mouth of the Chesapeake
Bay (Weisberg et al. 1993).
Although some of these approaches are promising and resolve the interpretational needs of
their specific studies, none are directly applicable to all the benthic data collected for
Chesapeake Bay monitoring programs. For instance, the method employed by Dauer (1993)
was based on data from the Virginia Benthic Monitoring Program, which samples only a
subset of benthic habitat types present in Chesapeake Bay. Applying the EMAP benthic
index in its present form requires using standard EMAP sample collection and sample
processing protocols; however, several different types of sampling gear and processing
protocols were used in the projects for which Chesapeake Bay benthic macroinfaunal data are
-------
available. More importantly, the EMAP index was derived to interpret benthic data for the
range of estuarine habitats present in the entire mid-Atlantic region. A greater degree of
sensitivity for describing conditions within Chesapeake Bay is probably possible if a similar
index is developed using only data from Chesapeake Bay habitats.
Objectives of this Report
This report describes an effort to define the expectations for benthic communities in
nondegraded bottom habitats of the Chesapeake Bay. The specific objectives were to
• use existing data to establish expectations (restoration goals) for benthic
communities in nondegraded bottom habitats of the Chesapeake Bay;
• develop an index that measures goal attainment; and
• identify areas in which future research may be helpful for refining, and
decreasing uncertainty in, the identified restoration goals.
This report describes development of the Chesapeake Bay Benthic Community Restoration
Goals, and a Restoration Goal Index (RGI) that can be used to assess whether goals are being
met. The approach used to develop the goals and the index, and the results of analytical
efforts, are detailed in Chapter II. A discussion of results and recommendations for the
future are presented in Chapter HI.
-------
II. APPROACH TO SETTING GOALS
Approach
The approach used for developing Chesapeake Bay Benthic Restoration Goals was similar to
that used by Karr et al. (1986) to develop an index of biological integrity (IBI) for freshwater
fish and includes three complementary activities: selecting community attributes that differ
in degraded and undegraded areas (goal attributes); identifying values that differentiate
between degraded and undegraded areas for each selected attribute (setting goals); and
combining attribute results so that the condition of communities in areas of unknown status
can be determined (developing an index). Attributes are selected by identifying a set of
reference sites that are believed to be "minimally impacted." Properties of the biotic
assemblages at these sites are then compared with the same properties at all other sites, and
properties that differ significantly between these two groups of sites are selected. Goals for
each of these attributes, or metrics, are established based on the response at reference sites.
The index is the average score of metrics after each metric is scored as 5, 3, or 1, depending
on whether its value at an individual site approximates, deviates slightly from, or deviates
strongly from its value at the best reference sites.
The general IBI approach was adopted here because it results in a quantitative statement of
condition that can be used to compare assemblage status over space or time. In addition,
the use of multiple attributes provides; a more reliable indicator of condition than indices
based on single attributes. Incorporation of multiple attributes also provides flexibility in
index application, since indices can be developed even if particular data sets are missing
information about selected attributes. Flexibility is particularly valuable in working with
benthos, since not all types of data are collected by all benthic sampling programs in
Chesapeake Bay.
Implementing this approach involved seven steps (Figure 1):
• Modify the data sets to be used for goal development to ensure their
compatibility. This step included activities such as ensuring common taxonomic
nomenclature and level of identification.
• Define the temporal and spatial strata for which goals will be developed.
Benthic macroinvertebrate assemblage composition and abundance vary with
season and habitat. These effects must either be corrected for or eliminated by
stratification.
• Identify reference and degraded sites for each of these strata. The reference sites
are the basis for selecting goal attributes and setting goals; degraded sites are
used to validate the goals.
-------
Develop Compatible
Data Sets
Define Temporal and
Spatial Strata
identify Reference
Sites and Degraded Sites
Select Restoration
Goal Attributes
Identify
Restoration Goal Values
Develop Restoration
Goals Index
Validate Restoration
Goals and
Restoration Goal Index
Figure 1. Steps in the development of Chesapeake Bay benthic community restoration goals
7
-------
• Select benthic assemblage attributes to be included in the restoration goals
(restoration goal attributes).
• Determine goal values for each of the restoration goal attributes based on their
levels at reference sites.
• Develop a scheme for combining restoration goal attributes into a Restoration
Goal Index.
• Validate the restoration goals and Restoration Goal Index based on values for
restoration goal attributes at degraded sites.
Each of these steps is described in detail below.
Develop Compatible Data Sets
Data from seven benthic survey projects were used for goal development: the Maryland
(Ranasinghe et al. 1992) and Virginia (Dauer et al. 1989) Chesapeake Bay Benthic
Monitoring Programs, the U.S. EPA's Environmental Monitoring and Assessment Program
(EMAP; Weisberg et al. 1992), the Maryland (Reinharz and O'Connell 1981) and Virginia
(Nilsen et al. 1982) biogenics studies, a James River study (Diaz 1989), and a study in the
Wolf Trap area of the Virginia Bay (Schaffner, unpublished data). Sampling information for
these projects is provided in Table 1.
These seven projects were selected for several reasons: each provided data readily available
on electronic media; collectively they provided sample representation in all salinity habitats
of Chesapeake Bay; and all used a 0.5-mm sieve in sample processing, which is important
because the numbers and types of organisms collected depend on the mesh size used to sieve
the sediment. In addition, the selected projects were restricted to those in which the
organisms were identified to the lowest possible taxonomic level.
In addition to limiting the data sets to those with common properties, several data
transformation procedures were needed to ensure complete data compatibility. This was
necessary because sample collection and processing methods varied to some extent among the
projects, and corrections were needed to ensure that differences in the data reflected true
differences among the sampled benthic communities rather than differences in sampling
methodology. Standardization procedures included applying uniform naming conventions
across projects, eliminating organisms that were not sampled quantitatively, and standardizing
biomass measures.
Taxonomic differences among projects were eliminated by cross-correlating the species lists
of the seven projects, identifying differences in nomenclature, and consulting the taxonomic
expert for each of the projects to resolve discrepancies. These experts were available for six
8
-------
o
Is
«
•8
1
a
£
ca-a
•
ct
*-^
O
d
a
a
1
Modif
V
1
1
1
§
S
5
1
1
1
S
i
i
O — f»
a
00
CO
ts
2 -
81
II
l
CO
ll
I
i-
aryl
1
Bi
M
jo .S;
.2 >
Wolf Trap
-------
of the seven projects. Development of taxonomic consistency was aided by the fact that
many of these experts had previously exchanged samples and had noted and resolved
differences in nomenclature. The taxonomic list, after standardization, is provided in
Appendix A.
Four groups of organisms that are not truly indicative of bottom habitat conditions and which
were found in the species lists of some projects were eliminated from the data. These four
groups included algae, vertebrates (larval fish), pelagic invertebrates, and epifauna (Table 2,
Appendix B). They are not sampled reliably or accurately by the sampling methods used for
benthos. The first two were eliminated because they are also not invertebrate. Organisms
considered predominantly epifaunal were excluded for several reasons: the sampling
methods did not sample epifauna quantitatively; the exposure of epifauna to pollution insults,
particularly chemical contaminants in sediments, is different from the exposure experienced
by infauna; and the presence of epifauna is most often associated with the occurrence of shell
or structures such as T>ryozoan colonies, irrespective of habitat condition.
Biomass measures were available for the three largest of the programs, but not in the same
format. Biomass measurements were made for all taxa in the Virginia monitoring program,
for 22 species in the Maryland program (Ranasinghe et al. 1992) and for 51 taxa and feeding
groups in EMAP (Weisberg et al. 1993). In addition, biomass was measured as dry weight
for EMAP and as ash-free dry weight for the Maryland and Virginia monitoring programs.
To normalize for this difference, EMAP data were converted to ash-free dry weights using
conversion factors developed from data used to calculate length-weight regressions for the
Maryland Chesapeake Bay Benthic Monitoring Program. The conversion factors and regres-
sion results are presented in Table 3. Finally, abundance and biomass measurements were
standardized to values per square meter to account for the differences in surface area sampled
with the various types of collection gear used in the different projects.
Define Temporal and Spatial Strata
Temporal Stratification
The abundances and diversity of benthic organisms in Chesapeake Bay vary seasonally; most
organisms exhibit a large recruitment pulse in the spring, a smaller recruitment pulse in the
fall, and reductions in the summer and winter (Holland et al. 1977; Mountford et al. 1977;
Holland et al. 1980, Holland 1985; Ranasinghe et al. 1992). Cumulatively, these species-
specific changes in abundance and diversity over time result in relatively consistent seasonal
variation in community abundance and diversity (Figure 2). This seasonal variation could
confound the interpretation of benthic response to environmental degradation if not accounted
for in developing benthic restoration goals. This is particularly true given the markedly
unequal distribution of sampling effort among seasons in the different projects from which
the benthic data were drawn.
10
-------
Table 2. Groups of organisms eliminated from the data. Additional information
is provided in Appendix B.
GROUP
Algae
Porifera
Hydrozoa and Scyphozoa
Platyhelminthes
Nematoda
Hirudinea
Organisms identified only as "Mollusca"
Cephalopoda
Merostomata (Limulus)
Cladocera
Ostracoda
Copepoda
Branchiura
Cirripedia
Mysidacea
Isopoda : Cymothoidae
Isopoda : Sphaeromidae
Amphipoda: Hyperiidae, Caprellidae
All Zoeae & megalopae
Tardigrada
Bryozoa
Entoprocta
Asteroidea, Echinoidea
Chaetognatha
Vertebrata
REASON FOR ELIMINATION
Not invertebrates, Epifaunal
Epifaunal
Epifaunal
Epifaunal or Parasitic
Meiofaunal
Parasitic
Too little information
Pelagic
Megabenthic, Epifaunal
Meiofaunal, Planktonic
Meiofaunal
Meiofaunal, many Planktonic
Pelagic, Parasitic
Epifaunal
Epifaunal
Ectoparasitic on fish
Epifaunal
Pelagic
Pelagic, Meiofaunal
Meiofaunal
Epifaunal
Epifaunal
Epifaunal
Pelagic
Megafauna, not invertebrate
11
-------
Table 3. Results of regression analyses used to derive factors for the conversion of
EMAP dry weights to ash-free dry weights. The independent variable was
the dry weight, and the dependent variable the ash-free dry weight of each
individual organism. Intercepts were forced through the origin. The
conversion factors were applied to EMAP dry weights by taxonomic group
as indicated.
Taxonomic Group with dry
weights to be converted
Nemerteans
Polychaetes and Oligochaetes
Crustaceans
Bivalves, Gastropods, Insect
Larvae, Phoronids,
Hemichordates
Taxon in
Regression
Nemerteans
Polychaetes
Arnphipods
and Isopods
All taxa
Conversion
Factor
0.83464
0.62087
0.76622
0.73374
N
57
244
143
444
r2
0.978
0.948
0.990
0.950
To resolve this problem, restoration goals were developed for a single season. Summer was
chosen for this purpose because it was a sampling period common to six of the seven
projects for which data were available. EMAP, which purposely collected the specific kinds
of reference data needed for this project, sampled only during the summer. Considerably
less data would have been available to develop restoration goals for other times of the year.
Summer was also selected as the most appropriate period for developing restoration goals
because it is the period when assemblage attributes are expected to show the greatest
response to pollution stress. Episodic low dissolved oxygen events, identified as a major
factor affecting the occurrence of depressed living resources in Chesapeake Bay (Jordan et al.
1992), are most frequent in the summer. In addition, the adverse effects of contaminants are
greatest during the summer because of low dilution flows and high temperatures.
Summer was defined as extending from July 15th through September 30th based on mean
abundance and diversity data collected in the Maryland Chesapeake Bay Benthic Monitoring
Program (Figure 2) and analyses of the frequency of sampling visits. This definition
maximized the number of sampling visits available within a window when species abundances
12
-------
(Z*«uj/#) eouepunqy
o
§
o
8
W)
E
13
-------
were relatively stable. In addition, visit-to-visit and year-to-year salinity differences at each
station were smallest during this period. The total numbers of samples for each project, and
for the summer, are provided in Table 4. Sampling location maps for each project are
presented in Appendix C.
Table 4. Number of samples and number of summer samples for each of the seven
benthic survey projects contributing data
Data Source
Chesapeake Bay Benthic
Monitoring Program - Maryland
Chesapeake Bay Benthic
Monitoring Program - Virginia
EMAP-Near Coastal
Biogenics - Maryland
Biogenics - Virginia
James River
Wolf Trap
TOTAL
Number of Samples
3505
487
111
50
53
117
24
4347
Number of Summer Samples
950
116
96
0
21
28
4
1215
Spatial Stratification
Even under optimal conditions of water and sediment quality, the composition of benthic
communities differs substantially according to bottom salinity and sediment type (Boesch
1973, 1977a; Dauer et al. 1984, 1987; Holland et al. 1989). Benthic communities may also
be affected by water depth, because of the nature and magnitude of the hydrodynamic forces
influencing bottom habitats. Habitat-specific restoration goals were developed to ensure that
natural differences in benthic communities related to such habitat factors did not confound
interpretation of the indices being developed.
Habitat strata expected to be relatively homogeneous were delineated using a three-step
process. First, cluster analysis was used to identify affinities between sites based on species
abundance. Second, each of the sites was assigned to a habitat class defined by salinity,
substrate, and depth; these classes were drawn onto the cluster output from step 1. Third,
14
-------
the initial habitat classes were reduced by eliminating or merging those that did not relate
well to the abundance and biomass cluster patterns. Each of these steps is explained in detail
below.
Cluster analyses were performed separately on taxon abundances and taxon biomass. Single
linkage cluster analysis of stations and strata using Iog10-transformed Oog10 (mean + 1))
overall mean station abundances per m2 were applied to all benthic macroinfaunal taxa
encountered at each station over all seven projects. The Ecological Analysis Package
(Ecoanalysis, Inc. 1988) was used for cluster analysis with the Bray-Curtis similarity
coefficient and flexible sorting (/8=-0.25) (Boesch 1977b). The analysis was repeated on
Iog10-transformed overall mean taxa biomass per m2 for all stations for which these data
were available (Table 1).
Initially, ninety possible habitat classes were derived as a factorial combination of six salinity
classes, five substrate classes, and three depth classes. The six salinity classes were defined
according to a modified Venice System (Symposium on the classification of brackish waters,
1958): tidal fresh (0.0-0.5 ppt), oligohaline (0.5-5.0 ppt), low mesohaline (5-12 ppt), high
mesohaline (12-18 ppt), low polyhaline (18-25 ppt) and high polyhaline (> 25 ppt). Stations
were allocated to salinity habitats based on the mean salinity over all summer visits. Each
salinity habitat was subdivided into five sediment classes on the basis of silt-clay sized
particle content by weight as follows: sand (0-5% silt-clay content), muddy sand (5-25% silt-
clay content), mixed sand and mud (25-75% silt-clay content), sandy mud (75-90% silt-clay
content), and mud (90-100% silt-clay content). The salinity-sediment habitats were
subdivided into depth classes for the tentative habitat classification scheme. Tributary
stations were assigned to shallow (0-5 m water depth) and deep (>5 m water depth) habitats;
mainstem Chesapeake Bay stations were assigned to shallow (0-2 m water depth), medium
(2-10 m water depth), and deep (> 10 m water depth) habitats.
A comparison of the initial habitat classes to the clustering pattern defined by the benthic
assemblages confirmed the dominant influence of salinity on benthic communities in
Chesapeake Bay (Dauer et al. 1984, 1987; Holland et al. 1989). A relationship between
sediment characteristics and the abundance and biomass clustering was discernible within the
three high salinity groups, but only at the broad levels of mud (40% or more silt-clay
content) and sand. The effect of depth was not apparent in any of the groupings. Based on
these results, the ninety initial habitat classes were reduced to eight habitat classes for which
restoration goals would be defined (Table 5). The number of samples available for analysis
in each habitat class are presented in Table 6.
15
-------
Table 5. Habitat classification scheme and definitions
A. CLASSIFICATION SCHEME
HABITAT CLASS
Tidal Freshwater
Oligohaline
Low Mesohaline Sand
Low Mesohaline Mud
High Mesohaline Sand
High Mesohaline Mud
Polyhaline Sand
Polyhaline Mud
B. DEFINITIONS
(i) Bottom Salinity
Description
Tidal Fresh
Oligohaline
Low Mesohaline
High Mesohaline
Polyhaline
Salinity (ppt)
0.0 - 0.5
0.5 - 5.0
5.0 - 12.0
12.0- 18.0
> 18.0
(ii) Sediment Grain Size Composition
Description
Sand
Mud
Silt & Clay Sized Particle Content by
Weight (%)
0-40
> 40
16
-------
Table 6. Number of summer samples available for each habitat class
Habitat Class
Tidal Freshwater
Oligohaline
Low Mesohaline Sand
Low Mesohaline Mud
High Mesohaline Sand
High Mesohaline Mud
Polyhaline Sand
Polyhaline Mud
Number of Summer Samples
95
116
65
286
237
199
54
163
Identify Reference Sites and Degraded Sites
Reference sites that represent unimpacted or least impacted conditions were identified to
evaluate the validity of benthic community attributes as indicators of habitat status. It was
also desirable to identify sites that were clearly degraded, so that data from those sites could
be excluded from restoration goal development and used later for validation purposes.
Ideally, a number of degraded and reference sites would be identified for each habitat class.
Condition of a site was defined on the basis of bottom water dissolved oxygen concentrations
and sediment contamination. These properties were chosen since they represent two of the
most important sources of pollutant exposure to benthic invertebrates. In addition, they are
readily quantifiable with available data for many of the sites at which benthic data were
collected.
The criteria used to define reference and degraded sites are listed in Table 7. Contaminated
habitats were identified using threshold contaminant concentrations (ER-M values) above
which biological effects are frequently observed (see Long and Morgan 1990 for details and
lists of contaminant ER-M values). For many EMAP sites, results of a 10-day acute
sediment bioassay using the amphipod Ampelisca abdita provided supplementary toxicity
information. For dissolved oxygen, the criteria differed slightly depending on the nature of
dissolved oxygen data available, as explained in Table 7. Affected EMAP sampling stations
17
-------
Table 7. Criteria for designating degraded and reference sites. The criteria vary
between projects because of the different types of information available.
The ER-M value for a chemical is the concentration above which biological
effects are observed frequently (Long and Morgan 1990).
A. DEGRADED SITES
Sites meeting any of the following criteria were considered degraded.
Chesapeake Bay Monitoring Program
EMAP
Dissolved Oxygen
Availability of data for at least five summer site
visits, and bottom dissolved oxygen
concentrations less than 2 ppm measured on
80% or more of these visits.
Bottom dissolved oxygen concentrations below
0.3 ppm recorded on any occasion, or 10% of
the continuous observations less than 1 ppm, or
20% of the observations less than 2 ppm, or
bottom dissolved oxygen observations of less
than 2 ppm observed for 24 consecutive hours.
Chemical Contaminants
The measurement of any chemical contaminant
concentration exceeding the ER-M vaJue on any
visit.
The measurement of any chemical contaminant
concentration exceeding the ER-M value on
any visit, and test survival significantly
different from and less than 75% of control
survival for the A. abdita bioassay.
B. NONDEGRADED REFERENCE SITES
Sites were required to meet all of the following criteria to be considered a reference site.
Chesapeake Bay Monitoring Program
EMAP
Dissolved Oxygen
Availability of data for at least five summer
visits to the site with no bottom dissolved
oxygen concentration measurements less than 2
ppm, and more than 80% of the measurements
greater than 5 ppm.
Summer bottom dissolved oxygen
measurements never less than 1 ppm, 90% of
the observations greater than 3 ppm, and 75%
of the observations greater than 4 ppm.
Chemical Contaminants
Sediment contaminant data available with no
measured concentration exceeding the ER-M
value.
Sediment contaminant data available with no
measured concentration exceeding the ER-M
value, and bioassay test A. abdita survival
greater than 75% of, and not significantly
different from, control survival.
18
-------
were identified based on criteria applied to continuously recorded dissolved oxygen data that
were available for many of the EMAP benthic sampling sites. In contrast, low dissolved
oxygen concentration habitats among Chesapeake Bay Benthic Monitoring Program sites
were identified using point-in-time dissolved oxygen measurements made at the time of
sample collection, supplemented by point-in-time measurements from the Water Column
Physical-Chemical Characterization Components of the Maryland and Virginia Chesapeake
Bay Water Quality Monitoring Programs.
Application of these criteria resulted in designing 99 samples as being from relatively
undisturbed regional reference sites and 135 samples as being from sites with known habitat
degradation (Table 8). Because biomass and depth distribution data were not collected with
all samples at all sites, the number of samples for which community data other than
abundance were available was fewer than for abundance (Table 8; types of measures are
discussed further below). For three of the eight habitat classes, there were more than ten
samples in each of the reference and degraded categories. For each of the tidal freshwater
and oligohaline classes there were only two samples available from degraded sites; for
polyhaline sand there were none. The absence of samples from degraded sites within any
habitat class only limits validation efforts; for each habitat class with only one or two
samples from degraded sites, there were more than 10 samples considered to have been taken
from reference sites; therefore, restoration goals could be developed. Only for the low
mesohaline sand habitat class was goal development not possible. For this habitat class, no
samples could be classified as being from reference sites, probably reflecting the relatively
small spatial extent of this habitat in the Bay.
The number of samples classified was far fewer than the total number of samples in the data
base. In part, this was due to the conservative nature of the criteria and the presence of a
large "grey zone" in the criteria applied to define exposure to low dissolved oxygen. Several
sites could not be classified because exposure information, generally contaminant data, were
unavailable. The conservative nature of the classification approach dictated that a site be
clearly acceptable with respect to both contaminant and dissolved oxygen exposure.
Select Restoration Goal Attributes
Restoration goal attributes were selected by first identifying a set of candidate attributes that
are believed to be properties of benthic assemblages at sites of high environmental quality
19
-------
$
(A
a £•
00
^
A
s
I
K
J
d
.2
•B
J
•c
.a
f
Measures
•CL,
pq
.2
j
j
"8
•o
w>
£
1
•S
2
8
a
Degraded |
i
HABITAT CLASS
0
CN
CS
1 '
M
WO
1 Tidal Freshwater
o
^— <
-
a
i
cd
!
0
o
o
o
o
o
I Low MesohaUne Sand
o
Os
O\
OS
«
Low Mesohaline Mud
o
o
»— 1
00
-
-
Ji
High MesohaUne Sand
Tt
CO
CO
1— 1
CO
CO
1 High MesohaUne Mud
o
o\
o
2
o
Os
»— «
•w
a
5
i
I
£
o
CO
T~^
CO
e
CO
1 Polyhaline Mud
^
©
Tt
oo
CO
•
1 TOTAL
20
-------
based on the literature and experience of the investigators, and then comparing the value of
these candidate attributes at reference sites with their values at all other sites. Attributes that
differed between categories of sites were incorporated into the restoration goals.
Identification of Candidate Attributes
A list of 24 candidate attributes potentially indicative of benthic habitat status was developed
from among five categories of benthic macroinfaunal community attributes based on the
literature, and the knowledge and experience of the authors (Table 9). Attributes in each
category were indicative of different aspects of benthic community structure or function.
The five attribute categories were
• benthic biodiversity measures
• assemblage abundance and biomass measures
• life history strategy measures
• measures of activity beneath the sediment surface
• feeding guild measures
Brief descriptions of the attribute categories, the candidate attributes in each category, and
the rationale for including them, are provided below.
L Benthic biodiversity measures.
The number of different kinds of benthic organisms supported by the habitat at a particular
location is often considered indicative of relative habitat "health." Two benthic
macroinfaunal community measures that reflect biodiversity were considered for use as
restoration goal attributes:
• the mean number of taxa collected per sample
• the Shannon-Wiener diversity index (Shannon and Weaver, 1949)
21
-------
Table 9. Candidate attributes investigated in each of the categories.
Benthic biodiversity measures
The mean number of taxa per sample
The Shannon-Wiener diversity index
Assemblage abundance and biomass measures
Total benthic infaunal community abundance per m2
Total benthic infaunal community biomass per m2
Life history strategy measures
The percentage of abundance contributed by equilibrium taxa
The percentage of biomass contributed by equilibrium taxa
The percentage of abundance; contributed by opportunistic taxa
The percentage of biomass contributed by opportunistic taxa
Measures of activity beneath the sediment surface
The percentage of benthic biodiversity deeper than 5 cm below the sediment
surface
The percentage of benthic biodiversity deeper than 10 cm below the sediment
surface
The percentage of benthic abundance deeper than 5 cm below the sediment surface
The percentage of benthic abundance deeper than 10 cm below the sediment
surface
The percentage of benthic biomass deeper than 5 cm below the sediment surface
The percentage of benthic biomass deeper than 10 cm below the sediment surface
Feeding guild measures
The percentage of benthic abundance contributed by carnivores and omnivores
The percentage of benthic biomass contributed by carnivores and omnivores
The percentage of benthic abundance contributed by suspension feeders
The percentage of benthic biomass contributed by suspension feeders
The percentage of benthic abundance contributed by deep deposit feeders
The percentage of benthic biomass contributed by deep deposit feeders
The percentage of benthic abundance contributed by interface feeders
The percentage of benthic biomass contributed by interface feeders
The percentage of benthic abundance contributed by suspension and deep deposit
feeders
The percentage of benthic biomass contributed by suspension and deep deposit
feeders
22
-------
Other available measures of diversity, such as Sanders' rarefaction index, which measures
the number of taxa for a fixed number of individuals (Sanders 1968), were not used because
the low numbers of organisms encountered in anoxia-affected environments would preclude
consistent application of these measures to data from all samples and sites. Diversity
indices, such as Shannon-Wiener, have recently declined in popularity (Green 1979) because
they are redundant and more difficult to interpret than the number of taxa encountered. Both
measures were treated as candidates, however, because the number of taxa encountered is
related in a complex manner to the area sampled (Connor and McCoy 1979) and seven
different types of gear differing in sampling area were used to collect samples for the benthic
survey projects that contributed data to this study. We were concerned that use of the mean
number of taxa per sample might be influenced more by the type of sampling gear than by
habitat status. Ewing et al. (1988) showed that when different types of gear are employed at
the same location, the Shannon-Wiener diversity index values are less different between
sampling devices than are the numbers of taxa encountered, even when standardized by
sampling area. To account for all possibilities, both attributes were included as candidates
for restoration goal development. Both measures would be expected to have higher values at
reference sites, in most cases.
2. Assemblage abundance and biomass measures.
Overall community abundance and overall biomass are measures of the total biological activ-
ity at a location. They also indicate the occurrence and availability of food for organisms at
higher trophic levels; thus, they provide information about the potential contribution of the
benthic community to energy flow in an ecosystem and on the relative size of the benthic
community compartment within the ecosystem. Two candidate attributes were identified in
this category:
• the total number of organisms present (standardized to numbers per m2 surface
area)
• the total community biomass (standardized to ash-free dry weight per m2 surface
area)
In most cases, both measures would be expected to have higher values at reference sites
although more exceptions are likely for abundance than for biomass. In addition, eutrophic
23
-------
sites (which should be considered degraded) may support high abundance and biomass
(Pearson and Rosenberg 1978).
3. Life history strategy measures.
Relatively short-lived, tolerant taxa with relatively high reproductive and recruitment
potential (opportunistic taxa) often dominate disturbed or stressed habitats (Boesch 1973,
1977a; Pearson and Rosenberg 1978; Rhoads et al. 1978; Dauer 1991, 1993; Dauer-et al.
1992). Large, relatively long-lived (equilibrium) taxa often dominate community biomass in
undisturbed or unstressed habitats (Warwick 1986; Dauer 1993). Four candidate attributes
were identified based on this life history perspective:
• percentage of abundance contributed by equilibrium taxa
• percentage of biomass contributed by equilibrium taxa
• percentage of abundance contributed by opportunistic taxa
• percent of biomass contributed by opportunistic taxa
High percentages of opportunistic taxa would be expected in degraded habitats and low
percentages in undegraded habitats. The converse would be expected for equilibrium
species. It was expected that the opportunistic measures would be more consistently
applicable because equilibrium species tend to be larger and rarer, and even when present,
they are more likely to be missed in sampling efforts.
Taxa with adequate available life history information were classified as opportunistic or
equilibrium species if their life history characteristics (Boesch 1973; Grassle and Grassle
1974; McCall 1977; Rhoads et al. 1973; Gray 1979; Rhoads and Boyer 1982; Warwick
1986; Dauer 1991, 1993) warranted it. No assignments were made for taxa with inadequate
life history information or with intermediate life history characteristics. Sites with no
organisms were considered to have 100% opportunistic, and 0% equilibrium membership.
The list of opportunistic taxa identified is presented in Table 10, and the list of equilibrium
taxa in Table 11.
24
-------
Table 10. List of opportunistic taxa
Annelida : Polychaeta
Asabellides oculata
Capitella spp.
Gtycinde solitaria
Heteromastus filiformis
Hypereteone heteropoda
Leitoscoloplos foliosus
Leitoscoloplos fragilis
Leitoscoloplos robustus
Leitoscoloplos spp.
Mediomastus ambiseta
Neanthes succinea
Paraprionospio pinnata
Polydora cornuta
Spiophanes bombyx
Streblospio benedicti
Mollusca : Bivalvia
Corbicula fluminea
Gemma gemma
Mulinia lateralis
Nucula proximo
Arthropoda : Amphipoda
Ampelisca abdita
Ampelisca spp.
Ampelisca vadorwn
Ampelisca verrilli
Ampeliscidae
Byblis serrata
Corophiwn lacustre
Leptocheirus plwnulosus
Arthropoda : Insecta
Chironomus spp.
Cladotanytarsus spp.
Coelotanypus spp.
Glyptotendipes spp.
Polypedilwn tripodura
Procladius sublettei
Tanypus spp.
Annelida : Oligochaeta
Aulodrilus limnobius
Aulodrilus paucichaeta
Aulodrilus pigueti
Aulodrilus pluriseta
Bothrioneurum vejdovskyanum
Branchiura sowerbyi
Haber cf. speciosus
Ilyodrilus spp.
Ilyodrilus templetoni
Isochaetides curvosetosus
hochaetides freyi
Limnodrilus cervix
Limnodrilus claparedeanus
Limnodrilus hoffineisteri
Limnodrilus profundicola
Limnodrilus spp.
Limnodrilus udekemainus
Oligochaeta
Potamothrix spp.
Potamothrix vejdovskyi
Quistadrilus multisetosus
Tubificid imm. with cap. chaetae
Tubificid imm. w/o cap. chaetae
Tubificidae
Tubificoides benedeni
Tubificoides brownae
Turbificoides diazi
Tubificoides gabriellae
Tubificoides heterochaetus
Tubificoides maureri
Tubificoides spp.
Tubificoides wasselli
25
-------
Table 11. List of equilibrium taxa
Coelenterata : Anthozoa Mollusca : Bivalvia
Ceriantheopsis americanus Anadara ovalis
Anadara transversa
Annelida : Polychaeta Cyrtopleura costata
Asychis elongata Dosinia discus
Chaetopterus variopedatus Ensis directus
Cfymenella torquata Macoma balthica
Diopatra cuprea Mercenaria mercenaria
Gtycera americana Mya arenaria
Macrocfymene zonalis Rangia cuneata
Spisula solidissima
Arthropoda Tagelus divisus
Alpheus heterochaelis Tagelus plebeius
Biffarius biformis
Callianassa setimanus
Echiurida
Squilla empusa
Thalassema spp.
Echinodermata : Ophiuroidea
Microphiopholis atra
4. Measures of activity beneath the sediment surface
The large, equilibrium species that often dominate community biomass in undisturbed or
unstressed habitats usually live deep within the bottom sediments (Schaffner 1990). In
contrast, the smaller, opportunistic, species dominant in disturbed and stressed habitats
usually live within one or two centimeters of the sediment-water interface (Warwick 1986;
Schaffner et al. 1987; Dauer 1991, 1993). It was considered possible, therefore, that the
depth distribution of organisms representing benthic macroinfaunal communities would
provide information related to the "health" of bottom habitats. Six measures of the depth of
the benthic community within the sediment were considered as candidate attributes:
• the percentage of total species present deeper than 5 cm below the sediment-
water interface
• the percentage of total species present deeper than 10 cm below the sediment-
water interface
26
-------
• the percentage of benthic abundance deeper than 5 cm below the sediment-water
interface
• the percentage of benthic abundance deeper than 10 cm below the sediment-water
interface
• the percentage of benthic biomass deeper than 5 cm below the sediment-water
interface
• the percentage of benthic biomass deeper than 10 cm below the sediment-water
interface
For all these candidate attributes, higher values would be expected in undegraded habitats
than in degraded habitats.
5. Feeding guild measures
Mature stable benthic assemblages typically contain a diverse set of feeding guilds, including
carnivores, deposit feeders, and suspension feeders. In contrast, disturbed communities are
often dominated by a single feeding group, such as surficial interface feeders. Word's
(1978) Infaunal Trophic Index is based on the idea that large, deep-dwelling taxa considered
equilibrium species are usually suspension feeders or deep deposit feeders. Based on these
ideas, ten feeding guild measures were considered as candidate attributes:
• the percentage of benthic abundance contributed by carnivores and omnivores
• the percentage of benthic biomass contributed by carnivores and omnivores
• the percentage of benthic abundance contributed by deep deposit feeders
• the percentage of benthic biomass contributed by deep deposit feeders
• the percentage of benthic abundance contributed by interface feeders
• the percentage of benthic biomass contributed by interface feeders
27
-------
• the percentage of benthic abundance contributed by suspension feeders
• the percentage of benthic biomass contributed by suspension feeders
• the percentage of benthic abundance contributed by suspension and deep deposit
feeders
• the percentage of benthic biomass contributed by suspension and deep deposit
feeders
To create the candidate measures, it was necessary to categorize species into specific trophic
groups. Species were classified as being carnivores/omnivores, deep deposit feeders,
interface feeders or suspension feeders based upon literature descriptions of feeding behavior
(Jorgensen 1966; Bousfield 1975; Fauchald and Jumars 1979; Dauer et al. 1981) as well as
the collective judgement of the authors.
Evaluation of Candidate Attributes
The evaluation of candidate attributes for suitability as indicators of habitat condition was
performed by comparing values for candidate attributes between reference sites and all other
sites within each habitat category. To be incorporated as a restoration goal, an attribute
needed to respond in a manner that was consistent with the hypothesis for its consideration as
a candidate. For instance, for species diversity to be incorporated, values at reference sites
should be higher on average than at all other sites. This comparison was performed in two
ways. First, a t-test was conducted to examine for differences in mean condition. Second,
the range of values over all reference sites was compared with the range over all other sites.
Figure 3 shows a comparison between values for an attribute that met these criteria and
values for one that did not.
Investigator judgement was used to supplement the analytical tests in the selection of
attributes, primarily due to the limited amount of reference site data available for a number
of the habitat classes. For instance, some parameters such as total biomass were consistently
higher at reference sites in all habitat classes, but not significantly so in some others. In this
case, the significance of the test was considered less important than the consistency of the
response. Investigator judgement was also permitted in an effort to include attributes from
28
-------
1.2-
1.1-
1.0-
H0.0-
i 0.8-
O
O 0.7
O 0.6-
3 0.5-
§ 0.4
cc
"-0.3
0.2-
0.1-
o.o-
O CD O ©O
2 3
Shannon-Weaver Diversity Index (Log2)
i
4
i
5
O
O
2
oc
u.
OO OOOO
o o
10
r-t—f—r-r-, I T r-r-r T | i I . , . . -r T T ^ -r—r
20 30 40
Carnivore/Omnlvore Abundance (X.)
50
60
Figure 3. Examples of range comparisons between reference (open circles) and other
sites (filled circles). Top: Example for an attribute that was selected. The
range for reference samples is much smaller than the range for "other"
samples in a manner consistent with ecological theory. Bottom: Example for
an attribute that was not selected. There is hardly any difference between
reference and "other" sample ranges.
29
-------
among all five attribute categories, and to avoid over-representation of attributes from a
single category.
Ten of the 24 candidates were selected as restoration goal attributes for at least one habitat
class. The selected attributes for each habitat class are summarized in Table 12. Restoration
goal attributes could not be selected for the low mesohaline sand habitat due to a lack of
reference sites in this habitat class.
Identify Restoration Goal Values
Restoration goals for each selected attribute within each habitat class was identified as the
fifth percentile value for the reference sites within that class (Table 12). The values at
reference sites were used for establishing goals because they represent the response at the
least disturbed sites in Chesapeake .Bay. Consideration was given to establishing higher
values, such as the median response at present day reference sites, as restoration goals given
the possibility that all sites in the Bay are impacted relative to historic reference conditions.
The lower value was selected, however, because of our inability to define historic reference
condition and the concern that historic conditions may no longer be achievable. The fifth
percentile value for reference sites was selected to allow for the possibility that we
misclassified a small fraction of the reference sites because of incorrect information or
because the variables used for classification did not include all of the potential stresses to
benthic macroinvertebrates in the Bay.
Develop the Restoration Goal Index
To provide managers with a quantitative tool to assess the extent to which restoration goals
were being achieve, an index that combines these attributes into a single value was
developed. Development of this Restoration Goals Index (RGI) was patterned after the
approach used to develop the Index of Biotic Integrity (IBI) by Karr et al. 1986, in which
each metric is scored as 5, 3, or 1, depending on whether its value at a site approximates,
deviates slightly from, or deviates greatly from conditions at the best reference sites. For the
RGI, the threshold for differentiating between a value of 1 and 3 was defined by the
restoration goal for the attribute. The threshold between a value of 3 and 5 was defined as
the median value among the reference sites for that habitat class. These threshold values are
defined in Table 12.
30
-------
a
.83
s
1
.s
c
I
1
I
s
o
i
&
-§
q
o*
g
g.2
I
0>
I
S
nce*
% Equilibriu
Abunda
% of
> 5cm belo
diment Surf
Ab
%
31
-------
The restoration goal attribute values are combined into a Restoration Goal Index (RGI) by
computing the mean attribute score across all attributes for which values are available. This
differs from the approach used in the IBI, in which values are summed. Averaging permits
greater flexibility to compute the RGI even in instances where one or more attributes are not
measured. For example, if biomass was not measured, as is often the case for many benthic
survey projects in Chesapeake Bay, the RGI could be computed even if the goal attributes for
the habitat class included biomass measures.
Based on this approach, an RGI value of 3 represents the minimum restoration goal. Values
less than three indicate unacceptable benthic community status, whereas values of three or
more indicate habitats that meet or exceed the restoration goals. The range of the RGI score
extends from 1 to 5.
Validating the Restoration Goals Index
Three approaches to index validation were employed. First, the RGI was computed for all
samples taken from each reference site to test whether expectations of values of three or
more were met. This test included data used for development of the index and, therefore,
was not independent. The test was necessary to ensure that selecting the fifth percentile of
the attribute values at reference sites, rather than the lowest value, as the first threshold did
not cause the RGI to be excessively conservative.
The results of this test indicated a high degree of correct classification (Table 13);
classification efficiency was greater than 95% in five of the seven habitat classes. The
lowest classification efficiency for reference sites was 92.3% in the high mesohaline mud
habitat class.
Second, RGI values were computed for all samples taken from degraded habitats to test
whether expectations of RGI values less than three were met. This test used data excluded
during development of the RGI and, therefore, was an independent validation test.
A high level of classification efficiency was observed in this second test; classification
efficiency was 95% or better for degraded sites in four of the six habitat classes for which
data from degraded sites were available (Table 14). The two habitat classes that did not
validate as well were tidal freshwater and low mesohaline mud. For tidal freshwater, only
two data points, both from the Anacostia River were available for validation. Although it is
32
-------
tempting to conclude that the failure to validate in tidal freshwater was a result of having so
few validation data points, very few attributes showed much difference between reference
and nonreference sites in the attribute selection stage, suggesting that factors of importance in
tidal freshwater are different than in the other habitats. The limited data available for this
habitat category, however, constrains our ability to investigate this issue further.
Table 13. Classification rates for the RGI at reference sites
Habitat Class
Tidal Freshwater
Oligohaline
Low Mesohaline
Mud
High Mesohaline
Sand
High Mesohaline
Mud
Polyhaline Sand
Polyhaline Mud
Number of Samples
15
22
6
11
13
19
13
% Correctly Classified
93.3
95.5
100.0
100.0
92.3
100.0
100.0
Table 14. Classification rates for the RGI at degraded sites
Habitat Class
Tidal Freshwater
Oligohaline
Low Mesohaline
Mud
High Mesohaline
Sand
High Mesohaline
Mud
Polyhaline Sand
Polyhaline Mud
Number of Samples
2
2
29
1
34
0
67
% Correctly Classified
0
100.0
86.2
100.0
97.1
-
95.5
33
-------
For the third validation test, sites sampled more than once during the summer for any year
were identified, and the RGI was computed for each visit. RGI values at each site were
evaluated for differences in status designation among visits within each year to ascertain the
stability of the index. Instability of the index would indicate an unacceptable signal-to-noise
ratio in the attributes. The results indicated that the RGI index was relatively stable. For all
seven habitats, the correlation between the first and second visits to these sites exceeded 0.8,
and for six of them it exceeded 0.9 (Table 15). In addition, for all but the high mesohaline
mud habitat (78.2%), more than 80% of the sites that were sampled twice in the same
summer classified the same with respect to goal attainment (Table 15). Most of the sites that
classified differently between the two visits were sites where not all the restoration goal
attributes were quantified during sampling because biomass or depth distribution data were
not available. For all these sites, RGI values were similar for both visits but were close to,
and on either side of, the critical value of 3.
Table 15. Classification consistency for sites visited more than once in the same
summer.
Habitat Class
Tidal Freshwater
Oligohaline
Low Mesohaline Mud
High Mesohaline Sand
High Mesohaline Mud
Polyhaline Sand
Polyhaline Mud
Number of Site- Year
Combinations with
Multiple Summer
Samples
17
32
82
77
55
1
36
Percentage
of Sites with
Unchanged
Annual
Status
100.0
93.8
84.2
81.8
78.2
100.0
88.9
Correlation
between
RGI values
(r2)
0.98
0.97
0.93
0.95
0.90
1.00
0.83
34
-------
III. DISCUSSION AND RECOMMENDATIONS
The objective of this project was to develop a practical and conceptually sound framework
for assessing benthic environmental conditions in Chesapeake Bay. The results suggest that
the primary objective of this effort has been met in at least a preliminary manner:
• A set of restoration goals have been identified that can be applied to data from
all of the major benthic sampling programs in Chesapeake Bay, despite their
differences in sampling techniques. Application of the restoration goals to
current and future monitoring efforts has been enhanced by establishing a set of
common taxonomic conventions that can be applied across these major
programs.
• The calibration and validation results indicate that the restoration goals are
effective in distinguishing between sites of high quality and those of lesser
quality in six of the seven habitats for which data were available for goal
development. For these six habitat classes, correct classification rates for
reference samples exceeded 90%; for samples from degraded sites the rates
exceeded 85%.
The only habitat class for which the restoration goals did not perform well was tidal
freshwater. Although reference sites were identified correctly 93% of the time, the degraded
sites did not validate well. There are two possible explanations for the failure of the goals to
validate in tidal freshwater. The first is that only two degraded-site data points were
available for validation. Both of these were from the Anacostia River, which may not
represent typical degraded sites in this habitat class. The validity of this explanation could
be assessed if more data from degraded sites within this habitat class were available.
The second possibility is that the tidal freshwater fauna are inherently different from fauna in
the other estuarine habitat classes and the attributes incorporated into the goals are not
applicable to tidal freshwater benthic communities. Low biomass, shallow-dwelling,
opportunistic oligochaetes and chironomids overwhelmingly dominate the tidal freshwater
fauna. Estuarine depth distribution and life history parameter degradation effect paradigms
that were developed based on work conducted primarily in higher salinity environments may
be ineffective for tidal freshwater habitats. The paradigms that are presently used in nontidal
freshwater environments (Plafkin et al. 1989) differ substantially from those used in
35
-------
estuaries. This explanation is consistent with the fact that only two attributes were found to
discriminate reference sites from all other sites during goal development. It is also consistent
with the findings of EMAP, where development of a benthic indicator was least successful in
tidal freshwater environments (Weisberg et al. 1993), and with the high variability observed
for these environments in the graphical models of expected values developed for the Virginia
portion of Chesapeake Bay (Dauer 1993).
Confidence in the habitat-specific goals was not equally high among the six habitat classes
that exhibited acceptable levels of validation. Validation was highest for the high mesohaline
mud and the polyhaline mud habitat classes. For these classes, all five attribute categories
and as many as nine individual attributes were included in the restoration goals; moreover,
for each of these habitat classes there were more than ten degraded sites that could be used
for validation. The high level of validation, together with the fact that data for calibration
and validation in these two habitat classes were acquired from several different projects,
indicates that these particular goals can readily accommodate differences in sampling
program methods.
Classification results in the oligohaline, high mesohaline sand, and polyhaline sand habitat
classes were encouraging; classification efficiency was greater than 95% samples taken from
both degraded and reference sites. For each of these habitat classes, however, fewer than
three samples from degraded sites (Table 7) were available for validation. The paucity of
sites for validation raises questions concerning the degree to which the available sites
represent the full range of degraded conditions in these habitat classes within the Chesapeake
Bay. For the high mesohaline sand and polyhaline sand habitat classes, the classification is
probably robust because goal attributes from all of the attribute categories were included;
however, goal attributes from only three of the five attribute categories were incorporated for
the oligohaline habitat, which might limit its general applicability.
Acceptable classification results were obtained for the low mesohaline mud habitat class;
however, data available for selection of goal attributes consisted of only six samples from a
single river system. Although the attributes and attribute threshold values for this class were
generally consistent with those in a number of the other habitat classes, it is not clear to what
degree the habitat-specific benthic community characteristics from that single system may or
may not represent those from the same habitat class in all other segments of the Bay.
Resolution of this issue would require broader sample coverage of this habitat class
throughout the Bay.
36
-------
Most of the potential shortcomings of the restoration goals developed here arise because goal
development was carried out using existing data, rather than new data collected specifically
for the purpose of restoration goal development. Although the data limitations do not
preclude application of these goals, they do suggest that additional refinement is both
necessary and appropriate if the goals are to be applied broadly and regularly in Chesapeake
Bay environmental management programs. A number of necessary refinement activities are
suggested:
1. Obtain additional data from reference and degraded sites to enhance calibration and
validation of the restoration goals
The habitat-specific data limitations described above allow relatively specific data needs to be
defined. The additional data would contribute to improved confidence in the goals presented
in this report:
• Additional reference site data should be obtained for the low mesohaline mud
habitat. Only six reference samples were available for analysis and they were
all obtained from the same tributary.
• Additional degraded site data are required from the oligohaline, high
mesohaline sand, and polyhaline sand habitats. No more than two samples
were available for validation in these habitats in this project.
• Data from both reference and degraded sites are needed for the low mesohaline
sand and tidal freshwater habitats. No data of either type were available to
even begin developing goals for the low mesohaline sand habitat. For the tidal
freshwater habitat, goal development was largely unsuccessful, and it is
necessary to determine whether this was a function of the calibration/validation
data sets, or whether it is a property of the different types of biota inhabiting
the habitat.
Beyond these specific needs, there is a general need for more reference site information for
validation of the goals for all habitat classes. At present, the validation has been primarily
limited to examination of degraded sites. Although the Restoration Goal Index was
examined for each of the available reference sites, this activity was somewhat circular given
that these same samples were used to calibrate the goals.
37
-------
In developing the additional reference site information for validation, it is important that such
data be gathered from a number of different subsystems within the Bay. Development of the
Restoration Goals has been based on the assumption that one can identify expected properties
of an assemblage for a particular habitat type, and that the expectation will be the same in all
locations in the Bay where that habitat type is found. This assumes that undisturbed sites in
mesohaline mud of a tributary like the Chester River have the same expected biomass levels
as a mesohaline mud reference site in all other Bay tributaries. Although this assumption is
reasonable within the context of establishing broad multi-attribute goals, it certainly needs to
be confirmed if the goals are to be applied as management tools. The goals developed so far
are particularly vulnerable to this assumption, since most of the reference data tended to be
from locations clustered in a small number of tributaries.
The additional data needs identified above can be obtained in several ways. The most
obvious is to collect benthic invertebrate and exposure information at sampling locations
selected specifically to meet those defined needs. In this project, however, we used data
from only the largest, most spatially diverse programs being conducted in the Bay.
Numerous site-specific impact assessment efforts have sampled near-field and far-field
environments. Although data from many of these studies may be difficult to obtain, they
may contain the kind of supporting information needed to classify sites as degraded or
reference and to offer a more cost-effective means of meeting the identified data needs.
Future restoration goal development efforts may also benefit from the fact that EMAP
collects the type of exposure information needed to classify some of its sites as reference or
degraded, and it has already collected samples at approximately 120 sites for which data
were not available at the time this work was performed.
2. Obtain samples from reference sites outside of the Chesapeake Bay system
Reference sites for this project were all developed from data within Chesapeake Bay. Thus,
the definition of restoration goals is relative to present conditions and may be conservative
relative to historic conditions, as all sites in Chesapeake Bay are arguably now in some way
anthropogenically influenced. At a minimum, the presently proposed goals provide a
benchmark against which to assess future conditions. We suggest, however, that to better
understand whether the present restoration goals are conservative, it will be necessary to
examine data from less anthropogenically affected reference sites nearby, but outside of
Chesapeake Bay. Such data might be found for reference areas along the coastal bays of
38
-------
Maryland and Virginia, in areas such as Assateague. Alternatively, it may be possible to use
historical data for Chesapeake Bay for such comparisons.
3. Conduct specific sampling to quantify eutrophication effects on benthos
At high levels of organic input, benthos respond with a reduction in biomass and abundance
as eutrophication leads to hypoxia; at lower or intermediate levels, however, the response is
likely to be increased benthic community abundance and biomass, with the observed maxima
at some distance, in space and time, from the source (Pearson and Rosenberg 1978, Dauer
and Conner 1980; Ferraro et al. 1992). In developing the restoration goals, total biomass
and abundance were treated as a monotonic response with higher levels considered indicative
of healthier environments. We would prefer to have developed a bimodal scoring system for
these attributes, but available information is inadequate to identify specific threshold values
above which total abundance or total biomass can be considered "enriched" for any habitat
class. A special sampling program instituted at varying distances from the outfalls of
wastewater treatment or food processing plants might be an appropriate way to better
quantify the effects of increased organic inputs on total benthic abundance and total benthic
biomass.
4. Add early warning indicators as candidate attributes
The restoration goals as they are presently developed are structural measures of an
assemblage that change only after mortality and/or species replacement has occurred.
Benthic assemblages may experience functional alterations (e.g. changes in nutrient cycling
or production of biomass) that have important implications for ecosystem function before
structural responses become apparent. Further development of the restoration goals should
incorporate measures that reflect alterations that occur at lower levels of perturbation. These
measures might include physiological measures, which might have to be determined through
laboratory studies, or may include the presence of sentinel species or assemblages that are
the first members of the communities to show a response to stress. In particular, studies are
needed that identify sublethal response by individuals and communities to low dissolved
water column oxygen, enrichment conditions, and pollutant stress.
39
-------
5. Evaluate alternative weighting schemes for the attributes
At present, each of the attributes in the RGI is weighted equally. It is not clear, however,
that each of the attributes is equally sensitive or equally reliable as an indicator of condition.
Further analyses could be conducted to evaluate the merits of alternative weighting strategies.
6. Define significant deviation from the restoration goal
During this project, we concentrated on identifying attributes, and their average values, that
allowed a site to be considered to represent reference conditions. It is inappropriate,
however, to conclude that all samples for which the RGI was less than 3 represent degraded
sites. Within any sampling program, natural sampling variability confounds interpretation
based on any individual sample. The second step in the validation effort, in which index
values for multiple samples taken from a single sampling location within a single summer
were evaluated, documented that this variability was not excessive; however, no statistical
analyses were conducted to establish expected levels of variability and expected ranges of
goal values.
These issues will need to be addressed if the RGI is to be applied to generate maps of
condition of the Bay. For instance, an area might be classified on the basis of its average
RGI score; alternatively, it may be classified on the basis of percentage of samples (e.g., >
75%) exceeding an RGI of 3. Possibly an RGI greater than 3 could be required in order to
ensure that only those areas that are meeting goals are classified as such. Choosing among
these alternatives is in part philosophical and relates to how protective biological criteria
should be (Dauer 1993), but further examination of the variability of the response would
provide useful input to that debate.
7. Identify a minimum number of restoration goals attributes
Calculation of the RGI is based on average attribute score, rather than on the sum of the
values. Use of the average was incorporated to maintain flexibility, since not all attributes of
the RGI are measured by all benthic sampling programs in the Chesapeake Bay and there
was a desire to make the index applicable to as many programs as possible. We do not wish
to imply, however, that data for a single attribute is sufficient to establish that restoration
40
-------
goals are being met. Multiple attribute indices have been shown to have less variabilty and
greater responsiveness to a wide array of system perturbations than do single attributes (Karr
et al. 1986). Further work is suggested to identify the minimum number of attributes that
are required for each habitat class in order for the RGI to be applied.
8. Extend the development of restoration goals to other seasons
The strategy adopted for temporal stratification, restricting restoration goal development to
the summer, was selected largely for practicality, based on the nature of the data available
for goal development. Summer was also believed to be the most appropriate period for
developing goals, given the need for anticipated maximum community response to degraded
habitat conditions. Limiting the restoration goals to the summer, however, excludes much of
the benthic data available for Chesapeake Bay from the goal development process. The
development of more robust goals with higher levels of precision might be possible if all data
were employed for the development process. Such analysis would require identification of
appropriate procedures for averaging or detrending seasonal changes in community attributes.
41
-------
IV. REFERENCES
Aller, R. C. 1980. Diagenetic processes near the sediment-water interface of Long Island
Sound. I. Decomposition and Nutrient Element Geochemistry. Adv. Geophys.
22:237-250.
Baird, D. and R. E. Ulanowicz. 1989. The seasonal dynamics of the Chesapeake Bay
Ecosystem. Ecol Monogr. 59(4):329-364.
Bilyard, G.R. 1987. The value of benthic infauna in marine pollution monitoring studies.
Mar. Pollut. Bull 18:581-585.
Blackburn, T. H. and K. Henriksen. 1983. Nitrogen cycling in different types of sediments
from Danish waters. Limnol. Oceanogr. 28(3):477-493.
Boesch, D. F. 1973. Classification and community structure of macrobenthos in the
Hampton Roads area, Virginia. Mar. Biol 21:226-244.
Boesch, D. F. 1977a. A new look at the zonation of benthos along the estuarine gradient.
pp 245-266 In: Coull, B. C., ed. Ecology of marine benthos. Columbia, SC:
University of South Carolina Press.
Boesch, D. F. 1977b. Application of numerical classification in ecological investigations of
water pollution. Prepared by the Virginia Institute of Marine Science for the
Corvallis Environmental Research Laboratory, Office of Research and Development,
U.S. Environmental Protection Agency. Corvallis OR. EPA-600/3-77-033.
Boesch, D.F. and R. Rosenberg. 1981. Response to stress in marine benthic communities.
pp. 179-200 In: G.W. Barret and R. Rosenberg, eds. Stress effects on natural
ecosystems. New York: John Wiley & Sons.
Boynton, W. R., W. M. Kemp, C. G. Osbome, E. Spalding, C. W. Keefe and K. N.
Wood. 1982. Estuarine community dynamics in relation to power plant operations.
Prepared for the Maryland Dept. of Natural Resources, Power Plant Siting Program,
by the University of Maryland, Chesapeake Biological Laboratory, Solomons, MD.
42
-------
Chao, L.N., and J.A. Musik. 1977. Life history, feeding habits and functional morphology
of juvenile sciaenid fishes in the York River estuary, Virginia. Fish. Bull. 75:657-
702.
Cloern, J. E. 1982. Does the benthos control phytoplankton biomass in South San
Francisco Bay? Mar. Ecol Prog. Ser. 9:191-202.
Connor, E. F. and E. D. McCoy. 1979. The statistics and biology of the species-area
relationship. Am. Nat. 113:791-833.
Dauer, D.M. 1991. Long-term trends in the benthos of the lower Chesapeake Bay. In:
J.A. Mihursky and A. Chamney, eds. New perspectives in the Chesapeake System:
a research and management partnership, 527-536. Solomons, MD: Chesapeake
Research Consortium.
Dauer, D. M. 1993. Biological criteria, environmental health and estuarine macrobenthic
community structure. Mar. Pollut. Bull. 26(5):249-257.
Dauer, D.M., and W.G. Conner. 1980. Effects of moderate sewage on benthic polychaete
populations. Est. Mar. Sci. 10:335-346.
Dauer, D.M., C.A. Maybory, and R.M. Ewing. 1981. Feeding behavior and general
ecology of several spionid polychaetes from Chesapeake Bay. J. Exp. Mar. Biol.
Ecol. 54(l):21-38.
Dauer, D.M., R.M. Ewing, G.M. Tourtellotte, W.T. Harlan, J.W. Sourbeer, and H.R.
Barker, Jr., 1982. Predation pressure, resource limitation and the structure of
benthic infaunal communities. Int. Revue Gen. Hydrobiol. 67:477-489.
Dauer, D. M, T. L. Stokes, Jr., H. R. Barker, Jr., R. M. Ewing and J. W. Sourbeer.
1984. Macrobenthic communities of the lower Chesapeake Bay. IV. Bay-wide
transects and the inner continental shelf. Int. Revue ges. Hydrobiol. 69:1-22.
Dauer, D. M., R. M. Ewing and A. J. Rodi, Jr. 1987. Macrobenthic distribution within
the sediment along an estuarine salinity gradient. Int. Revue ges. Hydrobiol. 72:529-
538.
43
-------
Dauer, D.M., R.M. Ewing, and J.A. Ranasinghe. 1988. Macrobenthic communities of the
lower Chesapeake Bay. March 1985 - June 1987. Prepared for the Virginia Water
Control Board by the Old Dominion Research Foundation, Norfolk, VA.
Dauer, D. M., R. M. Ewing, J. A. Ranasinghe and A. J. Rodi. 1989. Macrobenthic
communities of the lower Chesapeake Bay. March 1985 - June 1988. Report
prepared for the Virginia Water Control Board by the Old Dominion University
Research Foundation, Norfolk, VA.
Dauer, D. M., A. J. Rodi, Jr., and J. A. Ranasinghe. 1992. Effects of low dissolved
oxygen events on the macrobenthos of the lower Chesapeake Bay. Estuaries 15:384-
391.
Diaz, R. J. 1989. Pollution and tidal benthic communities of the James River Estuary,
Virginia. Hydrobiologia 180:195-211.
Diaz, R. J. and L. C. Schaffner. 1990. The functional role of estuarine benthos. Chapter
2. In: M. Haire and E. C. Chrome, eds. Perspectives on the Chesapeake Bay, 25-
56. CBP/TRS 41/90. Gloucester Point, VA: Chesapeake Research Consortium.
Ecoanalysis, Inc. 1988. EAP User's Guide. Version 5 Edition. Ojai, California.
Ewing, R. M., J. A. Ranasinghe and D. M. Dauer. 1988. Comparison of five benthic
sampling devices. Report prepared for the Virginia Water Control Board by the Old
Dominion University Research Foundation. Norfolk, VA.
Fauchald, K. and P. A. Jumars. 1979. The diet of worms. A study of polychaete feeding
guilds. Oceanogr. Mar. Biol. Ann. Rev. 17:193-284.
Ferraro, S.P., R.C. Swartz, F.A. Cole, and D.W. Schults. 1991. Temporal changes in the
benthos along a pollution gradient: discriminating the effects of natural phenomena
from sewage-industrial wastewater effects. Est. Coast, and Shelf Sci. 33:383-407.
Frithsen, J. 1989. The benthic communities within Narragansett Bay. An assessment for
the Narragansett Bay Project by the Marine Ecosystems Research Laboratory,
Graduate School of Oceanography, University of Rhode Island, Narragansett, RI.
44
-------
Grassle, J. F. and J. P. Grassle. 1974. Opportunistic life histories and genetic systems in
marine benthic polychaetes. J. Mar. Res. 32(2):253-284.
Gray, J. S. 1979. Pollution-induced changes in populations. Phil. Trans. R. Soc. Lond. B.
286:545-561.
Green, R. H. 1979. Sampling design and statistical methods for environmental biologists.
John Wiley and Sons. New York
Hilsenhoff, W.L. 1982. Using a biotic index to evaluate water quality in streams.
Technical Bulletin No. 132. Department of Natural Resources, Madison, WI.
Holland, A.F. 1985. Long-term variation of macrobenthos in a mesohaline region of
Chesapeake Bay. Estuaries 8:93-113.
Holland, A.F., N.K. Mountford, and J.A. Mihursky. 1977. Temporal variation in upper
bay mesohaline benthic communities. I. The 9-m habitat. Chesapeake Sci. 18:370-
378.
Holland, A. F., N. K. Mountford, M. H. Heigel, K. R. Kaumeyer and J. A. Mihursky.
1980. The influence of predation on infaunal abundance in upper Chesapeake Bay.
Mar. BioL 57:221-235.
Holland, A.F., A.T. Shaughnessy, L.C. Scott, V.A. Dickens, J.A. Ranasinghe, and J.K.
Summers. 1988. Progress report: Long-term benthic monitoring and assessment
program for the Maryland portion of Chesapeake Bay (July 1986-October 1987).
PPRP-LTB/EST-88-1. Prepared for Maryland Power Plant Research Program and
Maryland Department of the Environment by Versar, Inc., Columbia, MD.
Holland, A. F., A. T. Shaughnessy, L. C. Scott, V. A. Dickens, J. Gerritsen and J. A.
Ranasinghe. 1989. Long-term benthic monitoring and assessment program for the
Maryland Portion of Chesapeake Bay: Interpretive Report. Prepared for the
Maryland Dept. of Natural Resources by Versar, Inc. Columbia, MD. CBRM-
LTB/EST-2.
45
-------
Homer, M. and W.R. Boynton. 1978. Stomach analysis of fish collected in the Calvert
Cliffs region, Chesapeake Bay-1977. Final Report to Maryland Power Plant Siting
Program, Univ. of Maryland, Chesapeake Biol. Lab. Ref. No. UMCEES 78-154-
CBL.
Homer, M., P.W. Jones, R. Bradford, Jr., and J.A. Mihursky. 1979. Fish community
studies in the Patuxent Estuary, 1978-1979: Final Report. Prepared for Maryland
Department of Natural Resources by the University of Maryland, Chesapeake
Biological Laboratory, Solomons, Maryland.
Hunsaker, C.T. and D.E. Carpenter, eds. 1990. Environmental Monitoring and Assessment
Program: ecological indicators. EPA/600/3-90/060 U.S. Environmental Protection
Agency, Office of Research and Development, Washington, DC.
Jordan, S., M. Olson, C. Stenger, K. Mountford and R. Batiuk. 1992. Chesapeake Bay
dissolved oxygen goal for restoration of living resources habitats. Final Draft,
Revised February 1992. Chesapeake Bay Program. Annapolis, MD.
Jorgensen, C.B. 1966. Biology of Suspension Feeding. Oxford: Pergamon.
Karr, J. R., K. D. Fausch, P. L. Angermeier, P. R. Yant and I. J. Schlosser. 1986.
Assessing biological integrity in running waters: A method and its rationale. Special
Publication 5, Illinois Natural History Survey. Champaign, Illinois.
Kemp, W. M. and W. R. Boynton. 1981. External and internal factors regulating metabolic
rates in an estuarine benthic community. Oecologia 51:19-27.
Long, R. R. and L. G. Morgan. 1990. The potential for biological effects of sediment-
sorbed contaminants tested in the National Status and Trends Program. NOAA
Technical Memorandum NOS OMA 52. U.S. Dept. of Commerce, National Oceanic
and Atmospheric Administration, Seattle, Washington.
McCall, P. L. 1977. Community patterns and adaptive strategies of the infaunal benthos of
Long Island Sound. J. Mar. Res. 35:221-266.
McKenzie, D.H., D.E. Hyatt, and V,J. McDonald. 1992. Ecological Indicators. Vols. I
and n. London and New York: Elsevier Applied Science.
46
-------
Mountford, N.K., A.F. Holland, and J.A. Mihursky. 1977. Identification and description
of macrobenthic communities in the Calvert Cliffs region of the Chesapeake Bay.
Chesapeake Sci. 14:360-369.
Nilsen, K. J., R. J. Diaz, L. C. Schaffner, D. F. Boesch, R. Bertelsen and M. Kravitz.
1982. The biogenic structure of lower Chesapeake Bay sediments. Report prepared
for the Chesapeake Bay Program, U.S. Environmental Protection Agency by the
Virginia Institute of Marine Science. Gloucester Point, Virginia.
Officer, C. B., T. J. Smayda and R. Mann. 1982. Benthic filter feeding: A natural
eutrophication control. Mar. Ecol. Prog. Ser. 9:203-210.
Officer, C. B., R. B. Biggs, J. L. Taft, L. E. Cronin, M. A. Tyler and W. R. Boynton.
1984. Chesapeake Bay anoxia: origin, development and significance. Science
223:22-27.
Pearson, T. H. and R. Rosenberg. 1978. Macrobenthic succession in relation to organic
enrichment and pollution of the marine environment. Oceanogr. Mar. Biol. Ann.
Rev. 16:229-311.
Pearson, T.H. 1982. The Loch Eil Project: Assessment and synthesis with a discussion of
certain biological questions arising from a study of the organic pollution of sediments.
J. Exp. Mar. Biol. Ecol. 57:93-124.
Plafkin, J.L., M.T. Barbour, K.D. Porter, S.K. Gross, and R.M. Hughes. 1989. Rapid
bioassessment protocols for use in streams and rivers: Benthic macroinvertebrates
and fish. EPA/444/4-89/001. U.S. Environmental Protection Agency, Assessment
and Watershed Protection Division, Washington, DC.
Ranasinghe, J. A., L. C. Scott and R. Newport. 1992. Long-term benthic monitoring and
assessment program for the Maryland portion of Chesapeake Bay. July 1984 -
December 1991. Report prepared for the Maryland Dept. of the Environment and the
Maryland Dept. of Natural Resources by Versar, Inc. Columbia, MD.
47
-------
Reinharz, E., and A. O'Connell. 1981. Animal-sediment relationships of the Upper
Chesapeake Bay. Report prepared for the Chesapeake Bay Program, U.S.
Environmental Protection Agency by the Maryland Geological Survey. Baltimore,
MD.
Rhoads, D. C. and D. K. Young. 1970. The influence of deposit feeding organisms on
sediment stability and community trophic structure. J. Mar. Res. 28:150-177.
Rhoads, D.C., P.L. McCall, and J.Y. Yingst. 1978. Disturbance and production on the
estuarine sea floor. Amer. Sclent. 66:577-586.
Rhoads, D. C. and L. F. Boyer. 1982. The effects of marine benthos on physical
properties of sediments: a successional perspective. In: McCall, P. L., and M. J. S.
Tevesz, eds. Animal-sediment relations, 3-52. New York: Plenum Press.
Rosenberg, R. and P. Moller. 1979. Salinity stratified benthic macrofaunal communities
and long-term monitoring along the west coast of Sweden. /. Exp. Mar. Biol. Ecol.
37:157-203.
Sanders, H. L. 1968. Marine benthic diversity: A comparative study. Am. Nat. 102:243-
282.
Schaffner, L.C. 1990. Small-scale organism distributions and patterns of species diversity:
evidence for positive interactions in an estuarine benthic community. Mar. Ecol.
Prog. Ser. 61:107-117.
Schaffner, L.C., RJ. Diaz, C.R. Olson, and I.L. Larsen. 1987. Fauna! characteristics and
sediment accumulation processes in th James River Estuary, Virginia. Est. Coast
Shelf Sci. 25:211-226.
Shannon, C. E. and W. Weaver. 1949. The Mathematical Theory of Communication.
Urbana: University of Illinois Press.
Symposium on the classification of brackish waters. 1958. The Venice System for the
classification of marine waters according to salinity. Oikos 9:311-312.
48
-------
Virnstein, R. W. 1977. The importance of predation of crabs and fishes on benthic infauna
in Chesapeake Bay. Ecology 58:1199-1217.
Virnstein, R. W. 1979. Predation on estuarine infauna: response patterns of component
species. Estuaries 2:69-86.
Warwick, R. M. 1986. A new method for detecting pollution effects on marine
macrobenthic communities. Mar. Biol. 92: 557-562.
Wass, M. L. 1967. Indicators of pollution. In: Olsen, T. A. and F. J. Burgess, eds.
Pollution and Marine Ecology, 271-283. New York: John Wiley and Sons.
Weisberg, S. B., A. F. Holland, K. J. Scott, H. T. Wilson, D. G. Heimbuch, S. C.
Schimmel, J. B. Frithsen, J. F. Paul, J. K. Summers, R. M. Valente, J. Gerritsen
and R. W. Latimer. 1993. EMAP-Estuaries, Virginian Province 1990:
Demonstration Project Report. EPA/600/R-92/100. U.S. Environmental Protection
Agency, Washington, DC.
Word, J. Q. 1978. The infaunal trophic index. In: Bascom, W., ed. Coastal Water
Research Project, Annual Report for the Year 1978, 19-39. El Segundo, California:
Southern California Coastal Water Research Project.
49
-------
APPENDIX A
LIST OF TAXA
-------
CD
CO
'c
CM CM
CM
CM
DO <>
o> .. 5 ..
CO
«*>
CM
r- r- «- 00
O>
cc 3 n o>
^o CO «-
CM «- O>
CM CO
00
O)
CM
CO
3 > CO CO
E CD CD CO
_ .C O- o
£ «-• CO Q
O ojS co
o'-f O ®
i!<*
CB "^ i
(O
'c
Q.
o
IB •»
M **
CO CO
" c E < £
J J2 2 -o .>
E "° A- r« OT
CO C Q- O CD
<- co _.£ c
CO C3> j; c
® o
^j M
w w co c 2
j Z CO LU 00
£
o
'c
o
"c
£ £
o o
£ £ £
o o o
£ £
o o
'c
E
S
£
o
o
o.
'c 'E
EE
55
"c
E
0
I
5
CO CO CO
cj u o
M
(0
_CJ
*w
CO m
1 = 0
<° CO 3
111
S -2 co
2 € I I o
e c co -Q co
O < U UJ I
c c
<
O
Q} CD CD Q) CD CD w,
Q U O U O O O
(Q CO CO CO fl3 >
ft t t t 1 t c
.
oo
>>
'
oo
eO>
.
oo
>CO
'
uo
°:ii
_ CO CO
« £ ,C
2 Q. Q.
•5OO
^ CO CO
e o» o»
Ii
+j CO
M £
CO Q.
E E
e
e
CO CO
£ £
Q. Q.
E E
_
-------
CM
CM
«- CM CM «•- CM
CM
CM «- CM CM
«- »- CO
in CM •* O
«- U)
o> «- t- co
«- T- ^- to
CM CO
CM
CO
r> co co
CM
S
t- CO
co
CM f". to CSi «- »-
«- O> CM
in ID
CM
o
O)
I
o o o o
'c 'c 'c 'c
£ •£ E E E E
33999 9
O. &. <1> fl) fl) A>
Q) 4) h» w. ^
"c '£
E E
99
£ £
o o
1 I
a re
O O
CO CO
o o
o> o>
Q Q
o
o
"c
p
c c
£
o
'c
o
CD Q>
d)
m co
o o
O O O O O O O
> .> > ^ > > >
'c c 'c c 'c "c 'c
E £ E E E E E
O
3
CO
^ a
2 S"
O> -
CO
1
TO
Sf
S€
O- O TO
O U k_
TO ~*
ETO
OD
!^ 2
m 3 S £ o
I ill His
C co co
S '£ "£
II
co co
'£ £
0] 0>
c c
o o
OOOOOOOOO
000000
II
LU 111
-------
CM
CM t-
CM
CM CM CM CM
o »-
co
oo
CM
CM
«- r- CD CO
co
CM to
CM
t- «- •- CM CM
00
CO CO «-
t
o>
O «- o
CM «-
O
m
0.
<
CM 10 ^- CM »- r» co «- co co CM * CM «- «- u> CM »- in co cMin
•- »- CM «- CO«- CM CM «- «-
«- co co «- O oo O r* CD CM in co oo
«- «- CM «- «-
co o «- in •- «- o COCM
CO
f.
CO
»- O)
co
CO CM 00 O
in to CD *-
O) «-
CM
in
o
t
o
Q.
Q.
O
u
I
'c
Q.
O
o o
'^
in
'
o o
*>>>>>>_
cccccccc
EEiiEEEE-^
o
h_ t» w w
O O O O
305500598
0) O O 4) Q) 09 09 Q) "•
c c c
E E E *
o o
&
& « S £ S2 £
O O O O Q
C "C C
k. i. h.
CD CO CD
U O U
o o o o o
.> .2 .> 5 .> .2 > .> »- .^ .^ .- .^ «
£EEE££E£Si£EEE&
CDCOCOCOCDCDCDCDCD^IDCDCOCDv
UUUOUUUUQ^OOUUQ
O
X
I
«
I
•3
i
to
93
ID
«
CO
*-•
ID ID
C _ ^
9- te J? » -s S 2
S 8 9 >>•>>•>•>•>•
-------
(N
r- CM «-
CM CM CM CM
CM
CM
CM
IO «-
O
CM
mm «- O> OO ID «- »- O CM
CO CO «- CM
in i- •* r-
«- CO in CM
«- m *t r- u> m *t 10
CM «-
<- CD *t CM
o
o
o
co «-
O
IO CO «- «D
IO »-
o
v»
(O
co
co oo co
en
to co in
CM CM
o
Q.
Q.
o
Q.
O
o
O)
o
'c
O
£
o
CO
O
o
s I
&€
S f
(A £
o> «
III
c 'c 'c
E E E
953
Q) flJ > >
c c c
E E E
999
£ £ £
o o o
.> .> .a
E E E
(0 tO (O
u o o
o o o o
> > > >
'c 'c 'c 'c
E E E E
9999
£ £ £ £
o o o o
.> > .£ .>
E E E E _
o o o u u
£ £
6 o
> >
'c 'c
E E
99
£ £
o o
> >
E E
CO
U
a> cu
o o
.> .>
c c
E E
O O
£ £
o o
> >
E E
CO CO
U U
£ £
o o
> >
'c 'c
^*rf*-CC*^4-»-^^
'to 'to 'to fZ n to to 'to 'to
oooyOoooo
P-a-a-^jlj Q-tittQ-
V W >^
III
(U O> 0) 2. Z.
Q Q Q P ~
> >
Q. Q. Q. *? - *p
SO) V fc fc
(U
-------
CM CM CM
CM
CM CM
CM
O «-
CO «-
U) Tf
n n
«- «- CO CM
*- 00 «-
CM
CM
«- CM «- CM Xf »-
CM r-
*- 10
CM
O CM
CM
T- eo rx o
co «- «-
CO «- CM CM
«- CO i- CM
CM
CM
CM CD
CO CO
co co «-
CM
co «-
co oo «—
CD CO
o
co
co
10 o
to «-
eo
co
to
o
I
t
o
Q.
Q.
O
0) > > > >
- '- c c c
£ £
o o
o
X
•8
§
o
I
a
i
J5
v
c "55
O 0)
•o £
2 2
2
2
C C dJ
M 2 O. j= »
» ~ .2 o 3 ci £
S "z. o- * 2. °- «°
5 s S 8 £ "
-2 I § -2 1 1 -
c 2 -c o> SS 2 v
o o. Q. 3 .= .E S
2 2 2 S ~ " -
£ £ £ £
2
2
| 1
22'« S
2 « J= .E c
= s o x .; o>
2 I III I
S ^ 2 .° ° .2 .2
•o = 'QL '5. '5. '5. Q.
'n 0> M (A U) W CO
S = o o o o o
t. '5 C C C C C
±•200000
" w !? j- !? ^ >-
Q. O. Q. Q. Q. O. I
-------
CM
CM
CM CM
CM CM
CM «- CM
§
ID
CO
CO
«- co«-«-io^-»-»- rx «- «— CM o
^ CM «-
CM
CM
O CM
CM
T- ^- ^- CO T-
CO «- T- .-
i- CM «- CM
CM i-
cn
co
eo »-
in CM «- T-
co •<*•
to
co •*
co co co
CM CM
CM
co co
CM
CO
CM
co
CM
r- co to ca ro
«- «- CM
cn
en
«- o>
CM
u
1
c
t
a
Q.
O
10
'c
o
CL
Q.
O
s
a
o
I
£
o
'c
O
^
o
"c
£ £ £ £
O O O P
I I
CO
o
Q.
CD CD CD
Q O O
" CO CO
Q. t t
_ CD CD CD
Io£l
c
O
CD Q>
O O
> .>
C C
E E
oooonrjopoooooooouoooooooo
COCD>>*-IU>>>>CgcDCOCOCOCOCOCOCO>CD>>COCO
t: t •§•§ d & •§-§•£-§ t t r t t: t r t -c •£ t •§•§ t: t
coojtc:o)0>btc:to>CDO)O)O)O)aja)a3Ca)t:ta)OJ
•*^+^COCQCDCDCBCOCaCO<«-'*-'+J+J4-'4-'+J43<*-'m+^COm'£5<*-'
££ S
Q Q §
a. a. -F
t
a> o> CD
O O U
O
•8
g
o
S
f
o
'anchiata
a
S
Jdeurythoe pa
ellidae
CO -O
CO CO
Q. CO
CO c
*-• c
«» 3
ellides octocin
libregma inflat
JD co
CO U
CO CO
T5
.£
**T
CO
3 0.
o a.
o co
co co
'o. '5.
05 0)
0> CD
o o
CO CO
2
§8
2
CO
CO
+••
CO
3
U
§:*
OL Q>
CO <->
CD CO
.0 J3
E E
co co
CD GO
CO CO
*w
f\
aerosyllis taylt
i filicornis
.c .°
Q. 0.
CO CO
i pettiboneae
i setosa
o o
Q. a
CO CO
1
co
+•>
CO
o
o
ispp.
ichaetopterus
midae
o o o
a. a. a.
CO CO CO
X
I Q.
0 Q.
.O CO
co co
0) 0)
c c
CO CD
.C £
a. a.
o o
a. a.
CO CO
0)
CD
5™ co
0
a> co
I!
0 «
ca •£
CO CO
s
T)
CD
C
CD
JO
O
'a.
co
o
JO
£
4-«
CO
CO
|
s§
T3 CO
>• >•
CO CO
ibellidae
s
CO
6
S CO
. CD
< E
o- S
CO
&l
CD jj
H H
2
CO
.S2 c
E «
Si
2 •"-•
< -2
£
& |
(0 (i)
.a t5
> ^
g «
-------
CM
co oo
in
CO «- CM
CM
(O
CM
CM
«- O> «- O> «-
«- CO T-
IO
co
in
t- CM «-
co to «-
*- CM t •-
«-inincor-o>o>»-coco«-«*cM «- «o «-
to 'co 'co "55 '55 "co 'co en 'co co 'co '25 '55 'co 'co
ooooooooooooooooo
0,0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.
cocococococococococococococacococo
oopoopppppooopppo
o.o.o.o.o.o.
Q. a. a. Q.
O)
I
o
___..............o.o.o.a.o.........____
0>0>V(DCUQ>(U(UQ>CUfl)CP(l>Q>1]?Cl>Q>Q)Q3(l>(UQ}Q>(l>4}(l)fl)Q)Q)Q)Q'Q)Q)Q9
QQQQQQQQQQQQQ
QQQDQQ
|
re
£
a
co
3
!o
S Eg
5 -2 E
ue
ris
»S.^S B
•K 5? •£ S «? i
CO
o
8
O
X
i
3 E g s -g
a. E w -o =
3
O)
™ =*
S S S S § 1 11 -S1
•= -c ~ -c o S S SS =6 ^
•o "c o co co
*fOOOO£C'«=°43OOO
B3333o'222S«®
CD m ffi m O O
rix
aredeanus
§1
o o
c c
_J _J
meisteri
undicola
**- *»-
o S
£ Q.
inodrilus
inodrilus
t t
_i _j
Q.
0.
CO
inodrilus
t
_j
cemainus
(U
•o
inodrilus
c
"J3
4)
re re
•o *^
~ re
3 £
Is
C CO
3 <0
J Z
•a
'c
cu
CO
re
Z
CO
'c
O
U
CO
re
Z
CO
"5
•5
re
o.
CO
re
Z
CO
co
re
Z
CO
!o
re
'^.
re
>
CO
re
Z
re
4->
o>
5
U
o
O)
6
o>
2
45
CO
o
X
Q.
CO
re
oS
u
o
CD
6
re
c
'4^
c
0>
Q.
k.
0>
co -5
hidonais
•anais frii
o- re
O a.
-------
CN CM
i
ffl
<
CM
CD co
<0
oo
1
O
Q.
Q.
O
CD co
to «-
in
co
in *- *-
to co
.9 .9 .9 .9 .9 .9 .9 .9 .9 .9
towtowtowtototoo)
"c "c "c 'c 'c 'c 'c "c "c 'c
oooooooooo
0.0.0.0.0.0.0.0.0.0.
0.0.0.0.0.0.00.0.0.
OOOOOOOOOO
o o o o o o o
co "(o 'to "w '55 *cn
o o o o o o
o. a. o. a. o. a.
v>v>tnu>u>v>u>u>v>mu>u> ----.-..
ooooooooooooooooooooo
o.Q.a.Q.o.ooo.a.oa.o.o.Q.o.o.a.o.Q.a.o._
QQQQOQQQOQOQQQQQQQQQQOQ
0.0.0.0.0.00.0.0.0.0.0.0.0.0.0.000.00.0.0.0.0.0.
aoaaaaaaaaaaaoaaaaaaaaaaoa
c c c c c c c
E E E E E E E
9999999
6 o o 6 o o o
1111111
CQ CO tO CO CO CD ^S
u u o o o o o
5
o
a
I
O
X
3
6
I
»
v>
1
O
(O
1
(O
g>
u
to
Q..
W Q)
to jg
11
*= *3
(O W
(A
2 «
« CD
is
tO
^
— "O
I?
I s
i
£
V)
co ca
C C
co _
"c 'c
o o
._ v> in
.3 c c
*~ CD Q)
™ c. f.
(U tt tt
o. £ £
bl (/)(/>
II
o «
if
5 -
.2 3 5
~ i= • •
.2 S£ E
? 3 »- ,-
a>
cu !2
= 8
.9> 2
o E E „
CO "~" ™ to
~ "O "O *O
"5 *^ "^
fill
£???
ai
A
u>
0)
•o
jo a
cp a>
8 8
u> in
O) CU
."2 7s
'o '6
o o
t » !t *t **"
!o S S J3 15
3 3
S3 J3
33
a>
a !S
& 5 =
U) U> J3
o) a) o
•O 13 c
8— ._ c
O CO
f ^ ca
III
I
S
o
S
s
«o «
.2 2
3 .2
O »-:
co a.
o w
OT
"5
= O
5 •£ Q.
0 co a.
CO (0 n Q) "rt ^^
c c o- S -° c c
'§ 'i § 1 S 8 8
2 2 IB S-g ^g-
O O O C o 3 3
< < < < CQ ffl GO
-------
01 CM
CM CM
CM
CM
.- CM
«- u> *-
CM «- O> «-
,-,-,-.- m
co
<*
«- co
CM
r- •* i- CN i- «D CO
re
O
'C =
>-
O
v
O
5T* p |A U. n -* *- %w WJ *^ — ~ t
a&««o&fc«oo&a-| 2£i g
— «5fl)Wfl3«j-5c «,-S o OT<5^
l3|8||JBl|.8l685ls
t •-• fc:=^^t.t-SS^C O >. >. C
uiuxxx^x^^^^zzofifs:
2 •=
Q) Q> M
8O ® "7^
^ 2| 8
D 3 .£ »
<° £ 3
£ O 5
"o re re
2 S
'c 'c
a.
a.
in
Ss
-------
CM
CM «- CM CM CM «- CM
CM CM
«- eo CM
CM
CO O f» «-
•t-COU)
«- CM t-
O>IO o>
oo
oo
10 in en
|
'5 "5
iff ,ff
u u
III i 1
* 11 1 *
I 3 "5 3 §•
s* ^ cr o" Jr*
O LU tU \U O
^o
ll
— O
"5 o.
ET O.
So
s.o
c c c c c c c
P C3 O O _P _P .P
CCCOOCCCCCCCCC
OOOQ.Q.OOOOOOOOO
' ' " ' ' ' " ' '
ppppp_ _ _ __ __________
"co 'co '5 'in 'co « '<» '5 'w '5 '«o '5 "w "B S S S 8 S '5 "B 'co 'co 'co » CB "co co 'co co co "co co co "co
CCCCCQCCCCCCCCgUOUUccCCCQQCCCCCCCCC
COCOCOCOCOCDMCOCOCOWCOCOCOCUCUCDCUCUCOCOCOCOCOCDCUCOCOOOCOCOCOCOCOCO
33333«33333333*i*i*i*;*;33333« to -^ §" '£ •- E S.
0)TOiurau«o2-Sti"cOQ.-5Ep
•— CCCC.>-0~cVm
<<<< 1?
%*
sS
0) T3
OD CO
.
Lyonsia hyalina
Macoma balthica
Macoma mitchell
Macoma spp.
Macoma tenta
CO
CO
C
m
Mercenaria merci
Mulinia lateralis
Musculium
|
CO
w
cu
•».
Musculium transx
Mya arenaria
Nucula proxima
CO
i
0)
Nuculana messar
CO
Pandora gouldian
2
CD
CD
C
Pandora spp.
Parvilucina multil
-------
v- CM
«- 04
i
CD «-
O
m
Q
O
GO
CM
CM
a.
<
CM
CM
CM «- i- r- oo CM
CM r- CO
I
O
oo «-
CM «- CO CM CM
CM
co co o>
O
5
en
ID
«- t- in co «-
co CM co
co CM «- «- o>
* co co
co
O
•2
JD
"5
•c -c •=
.0 JO JO
'5 '5 '5
iff S LLJ
|
CT
111
cu
w
O
2
(I
w
I
CCCCnCCCCC
OOOOQ.OOOOO
1 s a a S 11 a a a §
& & & & &
c c- S H
t- L_ Q Q
.2 .2 o- Q.
CO O) 4) O>
CO
_
co
3 S
hw ^ ^ ^ JH c/>C/}(O(/>££c/3C/)OO
'c
I
2
O
to
O
O O O O O
(0 (O TO CD CD
•c t t r t
0)
i
§
O
1
S
S
X
CD
O.
CO
Q. -n •-
i !•§
C CD
< I
-------
CM
CM
CM
i
UJ
i- oo
If)
CM
o> in r^ «-
CM
o
CD
O
2
o>
•* CM
O
CM
CO
CM GO «- «- OO
CM
If) CM OO T- «-
O) (O
CM
CM ID «- O) «- CM
00 O5
CD T- m CM
O r-
CM CM t- rs CM *- oo «- «- CM
CM
o> to co oo
f». «- CM OJ
O. Oi O. O. O.
O O O O O
o
+3
(O
"
o
a.
a.
O
o>
I
CUCDCUCVCUCUCUCDCl}
060000006
111111111
EEEEEEEEE
999999999
CUCDQ)
OOOOOOOOO
>>>>>>>>>
E I E E E E E E E
C0(Q(OCOCO(Ofl)COCD
ooooooooo
o o o o o
'c c 'c "c 'c
E E E E E
99999
CO CB CD CD
re
re
Ire
il
re*
O Q)
_*
^
re
J3
S
2 i 3 S 3 S
re
+3 o> *d5
E C in
< < <
™ c=
+-* • E-
o t: co -» O Q. *^
_ A O. (O >D
re re re i..
222'££'£ oc
.t .± .± re re 1S
gi
co *:
._
3 =5
o. S
CD CO (O E
re => 3 n>
5 5 -o >. ~ -o
Illlll
E E 3 -5. S. t
re ro
X —i _J 3
-------
CM
CO
CO
CM «- co oo CM m
CM
co
«- CM
CM
CM
CO CM
re
w/
0.3
si s
in
i, r*H
if g8 S 1 3
o ^ « E I I I
J2 JS J2 "5 3 S "
» « « g 8 9
te £; H C C O
% % X £ £ 3 .
_i .j j 5 5 5 i
8
e
re re
o> re
- I
to to
o o
1 -2
Q. 0)
0 ™
II
iS
•8-=
It
< <
-E
re ..
utomate everm
ffarius biformis
< CO
in
i |
" 5
^ o
allianassa setinr
jceramus praeli
U ui
re
c
re
w
c
i
«
5
« re
1^
s E
£ ^
C "o
< u
-------
M
IU
<
•- 00
?
g
CO
o
o
CM
CN
CM
«- co •* «- co
M
I
3
o
O)
03 03 03 03
O O O O
'c 'c 'c 'c
E E E E
9999
03 03 03 03
03 03 03 03
O O O O
c -c c "c
E E E E
9999
0> 03 «3 03
O O O O
1 I 1 I
CO CO CO CO
u u u u
o o o o
1111
CO CO CO CO
o o u u
o o
co io
O U
43 (D Q} >>>>.>>
'c "c 'c 'c 'c c "c
EEEEEEE
0505500
Q> CD CD 0} CD CD 0)
OOOOOOO
c "c c c c
W k- t- t- k-
-
c "c
-
CO CO CO
u u u
. _
CO CO CO CO
o u o o
"c
E
Q
09
O
>
c
k.
CO
o
0) 03 03 03 03
OOOOO
'c '£ 'c 'c 'c
£ E E E E
Q $ Q 5 Q
CD CD CD CD CD
OOOOO
£ £ E £ 1
CO CO CO CO CO
u o u u u
O
X
Q.
< U UJ
£ o.
.. Q-
(A
9- °-
o ro
11
9
|
O
O 03 O.
m* &
•S » .2
* E £ m
Q, « Q. 03
2 S 2 -g
o.
ex
M
CO •
= 9-
o. o.
at
-------
§
ffi
1
co
g
00
o
9
m
co
«- CM
co
«- co
,- «- o> o>
« «
it § I
'c 'c "c 'c
E E E E
O O O O
5 § 5 §
E E E E
(0 CO (0 (0
o o o o
§ § § 5
E™ "c c *E
h_ i. t-
(0 (0 (0 (Q
o o o o
c c c c
E E E E E
99955
£ £ £ £ £
o o o o o
.> > > > >
E E E E E
ooooo
o o
•II
o o
> .>
E E
(0 TO
O O
3 o o o o o o
o O
> .2
E E
"(3
o o
> >
§ §
ooo
E E
(O
O
E E
ID (O
O O
E E E
(Q (Q (Q
000
o o
.> >
E E
re re
O O
o o
>. >
E E
(O (D
O O
o o
II
o o
I
o
O
X
0)
-C 0)
d£- a
a £!£ «
-2^-2
re S2 5
=T — p Q- Q-
S o § o o
£ £ S -o -o
n « ^ >- >-
< o o i i
1 "
o m
a a>
3
U
'a.
v>
o
U
c
U)
1
a>
O
Q.
a.
M
Q.
S
4-»
0)
U
i- ^J
re
03
*^
a.
^
£
-------
i
(O
o> « o
o o o o o
> > > > >
c 'c c c c
E E E E E
o o o o o
> > > > >
re re
U (J
CO CO
u u
<3(3
11
CO CO CO CO CO CO CO CO CO
oouoooooo
re m
U U
o
§
o
I
» 3 3
WC g to
•§ 'S cL
3 ro a.
re o.
CO TO TO {Q
'>• '> '>• 2 ._ _ _
E | E E E E E
S 8 £ g g ' -
•8-8-8 I ! S s
S
i
15
C o>
re T3
•£ W <»
V9
U)
*iL
•5 •_ re Q.
1£ !c !c lc
O U U U
0£C^JSg2fc
llfllllE*
*" c *? H frererere
t •=•=•=•=
o£ £ £ £
-------
i- at m
o>
CM *t »- «-
CM
•«*• co «-«-
If) «-
«- CM
«- CM
»- ID
CM
O
Q
O
I
CO
If)
CM
CM
«- 00
U
I
I
O
Q.
_
I
o.
O
u
•c
M
Q.
O
|
iff
o o o o
> > > >
"c c 'c 'c.
E E E E
5999
CD 0> Q> 0>
O O O O
> > > >
c "c c 'c
E E E E
9999
o>
o o o o
> > > >
E £ '£ £
to to CD CD
u u u u
o o o o
1 1 E I
CD CD CD CO
O O O (J
Q Q
a. a
a> a>
a> a>
O Q
u>
O
a.
a>
a
0)
O
spp
ervo
a
a.
*Sf
pa
ni
co c •= c
3 a
CO <0 CO S
C a> E
<5 CL a. a>
- '-5 '•& *"
en co <5
i^B
irri?- c c^
«D CO CO £.
H H I- H »- H
o.
Q.
CO
CO
C Q.
I &
Xenochi
Zavrelie
u
I
e-Z -*
O O
«I
Q. CO
a.
Q.
(O
R)
.1
•O (0
—
-------
CM
CM
§
CO
UJ
o
00
O
2
OQ
OL
<
o>
oo
O CM
CM in i- •-
O •* co
O
O
CM
O
CM
O
CM
s
I
o
c
g
°
Q.
(0
O O
Q. Q.
d> V
O Q
Q. Q.
O) O)
0) 0)
O Q
'« '« 'to *u>
o o o o
Q. Q. Q. Q.
0> 0>
O O
Q. Q. Q. tt
o) « o> a>
O) 03 0)
000
Q. Q. Q.
£0>
000
s 1
u u>
.2 >
£ «
I I
3 O
S 2 1
111
o •§ »
TO I §
-------
APPENDIX B
TAXA DELETED FROM ORIGINAL DATA
-------
Taxa not meeting benthic macroinfaunal criteria.
Porifera
Microciona prolifera
Porifera
Cnidaria
Cnidaria
Cnidaria : Hydrozoa
Bougainvillia rugosa
Cordylophora lacustris
Garveia cerulea
Hydra spp.
Hydractinia echinata
Hydrozoa
Moerisia lyonsi
Obelia spp.
Sertularia argentea
Cnidaria: Scyphozoa
Chrysaora polyps
Scyphozoan polyps
Cnidaria : Artthozoa
Diadumene leucolena
Platyhelminthes : Turbellaria
Dugesia tigrina
Euplana gracilis
Hydrolimax grisea
Stylochus ellipticus
Turbellaria
Platyhelminthes : Trematoda
Trematoda
Nematoda
Nematoda
Annelida : Polychaeta
Harmothoe extenuata
Harmothoe spp.
Hydroides
Hydroides protulicola
Lepidonotus sublevis
Lepidonotus variabilis
Polychaeta
Polydora websteri
Protodrilus spp.
Sabellaria vulgaris
Annelida : Polychaeta (Continued)
Serpulidae
Syllides spp.
Syllides verrilli
Annelida : Oligochaeta
Naididae
Nais pseudobtusa
Annelida : Hirudinea
Alboglossiphonia heteroclita
Batrachobdella spp.
Batracobdella phalera
Glossiphoniidae
Helobdella elongata
Helobdella fusca
Helobdella spp.
Helobdella stagnalis
Helobdella triserialis
Hirudinea
Illinobdella moorei
Piscicola spp.
Piscicolidae
Rhynchobdellida
Mollusca
Mollusca
Mollusca : Gastropoda
Amnicola limosa
Anachis avara
Anachis lafresnayi
Anachis obesa
Astyris lunata
Balcis intermedia
Boonea impressa
Cincinnatia winkleyi
Columbella spp.
Cratena kaoruae
Cratena pilata
Cratena spp.
Crepidula fornicata
Crepidula plana
Crepidula spp.
Cylichnella bidentata
Doridella obscura
Epitonium multistriatum
Epitonium rupicola
Epitonium spp.
-------
Molluscs : Gastropods (Continued)
Eupleura caudata
Fargoa bushiana
Gastropoda
Hydrobiidae
llyanassa obsoleta
Nassarius spp.
Nassarius trivittatus
Nassarius vibex
Nudibranchia
Odostomia spp.
Physella spp.
Pyramidellidae
Skeneopsis planorbis
Turbonilla spp.
Urosalpinx cinerea
Vitrinellidae
Molluscs : Bivslvia
Amygdalum papyrium
Anomia simplex
Anomia spp.
Crassostrea virginica
Geukensia demissa
Ischadium recurvum
Mysella planulata
Mysella spp.
Mytilidae
Mytilopsis leucophaeata
Mytilus edulis
Mollusca : Cephalopoda
Lolliguncula brevis
Arthropods : Merostomata
Limulus polyphemus
Arthropods : Hydracarina
Hydracarina
Arthropods : Pycnogonida
Pycnogonida
Arthropods : Cephalocarida
Cephalocarida
Hutchinsoniella macracantha
Arthropods : Cladocera
Alona affinis
Cladocera
Arthropods : Ostracods
Ostracoda
SarsieHa spp.
Arthropods : Copepoda
Calanidae
Calanoida
Caligoida
Arthropoda : Branchiura
Argulus spp.
Arthropoda : Cirripedia
Balanus
Balanus amphitrite niveus
Balanus balanoides
Balanus eburneus
Balanus improvisus
Cirripedia
Arthropoda : Mysidacea
Heteromysis formosa
Mysidae
Mysidopsis
Mysidopsis almyra
Mysidopsis bigelowi
Neomysis americana
Arthropoda : Isopoda
Aegathoa medialis
Cassidinidea ovalis
Cymothoidae
Edotea triloba
Erichsonella attenuate
Erichsonella filiformis
Erichsonella spp.
Paracereis caudata
Arthropoda : Amphipoda
Ampithoe longimana
Ampithoe valida
Ampithoidae
Batea catharinensis
Caprella andreae
Caprella equilibra
Caprella penantis
Caprella spp.
Caprellidae
Cerapus tubularis
Corophium acherusicum
Corophium acutum
Corophium insidiosum
-------
Arthropods : Amphipoda (Continued)
Corophium lacustre
Corophium simile
Corophium spp.
Corophium tuberculatum
Corophium volutator
Cymadusa compta
Dulichiella appendiculata
Elasmopus laevis
Ericthonius brasiliensis
Gitanopsis spp.
Melita nitida
Melita spp.
Microprotopus raneyi
Paracaprella tenuis
Parametopella cypris
Parapleustes aestuarius
Parathemisto compressa
Photis dentata
Photis pollex
Photis pugnator
Photis spp.
Pleustidae
Pleusymtes glaber
Pleusymtes spp.
Stenothoe minuta
Stenothoe spp.
Arthropods : Decapoda
Brachyuran megalopa
Brachyuran zoea
Callinectes sapidus
Caridean zoea
Crangon septemspinosa
Decapoda
Eurypanopeus depressus
Hexapanopeus angustifrons
Libinia dubia
Libinia spp.
Majidae
Neopanope say!
Ogyrides alphaerostris
Paguridae
Pagurus longicarpus
Pagurus spp.
Palaemonetes pugio
Palaemonetes sp. zoea
Palaemonetes spp.
Palaemonetes vulgaris
Panopeus herbstii
Processa vicina
Rhithropanopeus harrisii
Arthropoda : Decapoda (Continued)
Rhithropanopeus harrisii zoea
Xanthidae
Arthropoda : Trichoptera
Oecetis spp.
Tardigrade
Haplomacrobiotus spp.
Bryozoa
Aeverrillia armata
Alcyonidium spp.
Anguinella palmata
Cristatella spp.
Electra crustulenta
Membranipora membranacea
Membranipora spp.
Entoprocta
Urnatella gracilis
EcNnodermata
Echinodermata
Echinodermata : Echinoidea
Echinoidea
Chordata : Ascidiacea
Ascidiacea
Botryllus schlosseri
Cnemidocarpa mollis
Molgula lutulenta
Molgula manhattensis
Chordata : Cephalochordata
Branchiostoma caribaeum
Chordata : Vertebrata
Alosa mediocris
Teleostei
Miscellanea
Algae
-------
APPENDIX C
STATION LOCATIONS
-------
082
08
186
143
Station locations for the Environmental Monitoring and Assessment Program-Near Coastal
1990 (Goals Station Prefix EMAP)
-------
JV8
«••
t
\
CA
^3
§
o
•2
§
*5
i
-------
Station locations for the Virginian Chesapeake Bay Benthic Monitoring Program (Goals
Station Prefix MONV)
-------
Station locations for the Virginia Biogenics Project (Goals Station Prefix BIOV)
-------
-JO1
TT-OO
•ks
Station locations for the Maryland Biogenics Project (Goal* Station Prefix BIOM)
-------
KEY
Tidal Fresh / Ottgohalne
Low Mesohaline
Shallow high Mesohaine
Deep High Mesohaine
" A
Stratum locations for the Maryland Chesapeake Bay Benthic Monitoring Program 1989-
1991 (Stations MONM-101 to MONM-131)
-------
SUSOUEHANN*
CHESAPEAKE BAY
0 5 10 NAUTiCAu MILES
Calvert
'CUffs ."••
77
Rxed station locations for the Maryland Chesapeake Bay Benthic Monitoring Program 1984-
1989 (Stations MONM-001 to MONM-080)
------- |