EPA/630/R-98/001A
June 1998
DRAFT FINAL
REPORT ON THE SHRIMP VIRUS
PEER REVIEW AND RISK ASSESSMENT WORKSHOP
Developing A Qualitative Ecological Risk Assessment
Submitted to:
William H. van der Schalie
and
H. Kay Austin
National Center for Environmental Assessment-Washington Office
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
Prepared by:
Eastern Research Group, Inc.
110 Hartwell Avenue
Lexington, MA 02173
Work Assignment No. 1-06
Under Contract No. 68-C6-0041
-------
DISCLAIMER
This document is for review purposes only and does not constitute U.S. Environmental
Protection Agency policy. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
-------
CONTENTS
EXECUTIVE SUMMARY i
1. INTRODUCTION 1-1
1.1 ISA Report Overview 1-1
1.2 Peer Review Workshop Process 1-5
1.3 Qualitative Risk Assessment Methodology 1-6
2. QUALITATIVE RISK ASSESSMENT 2-1
2.1 The Risk Assessment Process 2-1
2.2 Qualitative Risk Assessment Results 2-2
2.3 Risk Management Relevance 2-16
3. ACTIONS FOR REDUCING UNCERTAINTY 3-1
3.1 Diagnostic Methods 3-1
3.2 Surveys of Wild Shrimp Populations 3-2
3.3 Epidemiology of Shrimp Virus Transmission 3-2
3.4 Field Epidemiological Studies 3-3
3.5 Lower Priority Risk-Relevant Research Areas 3-3
4. SUMMARY 4-1
4.1 Qualitative Risk Assessment Process 4-1
4.2 Comprehensive Risk Assessment Needs 4-2
4.3 Research Needs 4-2
4.4 Additional Areas of Concern 4-2
5. REFERENCES 5-1
APPENDIX A. Breakout Group Reports A-l
A-l. Report of the Aquaculture Breakout Group A-2
A-2. Report of the Shrimp Processing Breakout Group A-12
A-3. Report of the "Other Pathways" Breakout Group A-27
A-4. References A-41
-------
APPENDIX B. Peer Review Experts and Breakout Discussion Assignments
APPENDIX C. Premeeting Comments Prepared by Workshop Experts
APPENDIX D. Workshop Agenda
APPENDIX E. Presentation Materials on the Risk Assessment Process Developed by
the Aquatic Nuisance Species Task Force
APPENDIX F. Summary Materials Presented by Workshop and Breakout Group
Chairs
APPENDIX G. Report to the Aquatic Nuisance Species Task Force: Generic
Nonindigenous Aquatic Organisms Risk Analysis Review Process
APPENDIX H. Observers' Comments and List of Observers
-------
LIST OF FIGURES
Figure 1. Proposed shrimp virus conceptual model 1-3
Figure 2. Risk assessment model from the Report to the Aquatic Nuisance Species Task
Force 1-7
Figure A-l. Conceptual model: Virus sources and pathways for aquaculture A-3
Figure A-2. Conceptual model: Virus sources and pathways for shrimp processing A-13
Figure A-3. Flow diagram for shrimp processing A-14
Figure A-4. South Carolina commercial white shrimp landings and values A-24
-------
EXECUTIVE SUMMARY
This report highlights issues and conclusions from the shrimp virus peer review workshop
sponsored by the U.S. Environmental Protection Agency (EPA) in cooperation with the Joint
Subcommittee on Aquaculture (ISA), held January 7-8, 1998, in Arlington, Virginia. The goals
of the workshop were to:
Complete a qualitative assessment of the risks associated with shrimp viruses, following
the general risk assessment process developed by the Aquatic Nuisance Species Task
Force.
Evaluate the need for a future, more comprehensive risk assessment.
Identify critical risk-relevant research needs.
The workshop focused on the scientific and technical aspects of the likelihood that nonindigenous
viruses will become established in wild shrimp populations in the Gulf of Mexico and southeastern
Atlantic coastal regions and on the potential ecological consequences of establishment. The
workshop included 22 experts with varied backgrounds, including shrimp biology, toxicology,
virology, marine ecology, ecological risk assessment, and shrimp aquaculture and processing.
Prior to the workshop, participants received several background documents (ERG, 1997; ISA,
1997; RAM, 1996 [Appendix G]) and prepared written premeeting comments that all participants
reviewed (Appendix C). At the workshop, participants were divided into three groups, each of
which was charged with evaluating the risks associated with one of the following categories of
viral pathways:
Aquaculture
Shrimp processing
Other potential sources
The risk that shrimp viruses pose to shrimp aquaculture operations was not considered as part of
the scope of the workshop due to the limited time available; however, workshop participants
believed that the risks to shrimp in aquaculture should be given special attention as part of a
subsequent technical or management workshop.
The qualitative risk assessment was conducted using the modified Aquatic Nuisance Species Task
Force risk assessment approach (RAM, 1996; Appendix G). In developing the qualitative risk
assessment, participants considered the following:
Likelihood of viruses being present in the pathway
Ability of the viruses to survive transit in the pathway
Colonization potential of the viruses in native shrimp
Spread potential of the virus within native shrimp populations
Consequences of establishment
-------
In general, workshop participants agreed that viruses could be associated with pathways leading
to coastal environments and that they could survive in these pathways. Participants concluded
that there is potential for viruses to colonize native shrimp in localized areas, such as an estuary or
embayment, near the point of entry into the marine system. Some participants also noted that
repeated viral introductions to an area will increase the risk of colonization.
Participants had widely divergent views on the potential for viruses to spread beyond the initial
local area of colonization. This divergence largely reflects the high uncertainty associated with
this aspect of exposure. Participants considered the potential for localized colonization and
subsequent spread to be a critical aspect of evaluating the potential establishment of viruses in
native shrimp.
Workshop participants discussed the impact that virus establishment could have on local shrimp
populations (e.g., within an individual estuary). They determined that initial kill rates might be
high but that the population would be likely to recover rapidly due to reintroduction of shrimp
from other locales or compensatory increases in reproduction. Workshop participants concluded
that the risk from viral introductions to the entire population of native shrimp along the
southeastern Atlantic coast and within the Gulf of Mexico is relatively low, although there is a
high degree of uncertainty associated with this evaluation.
The ability of workshop participants to address broader ecological risks in a comprehensive
manner was limited by the time and information available. However, some participants thought
that the issue of broader ecological risks is important and merits further consideration.
Workshop participants identified areas where further research and information would improve the
assessment of risks and could help evaluate current conditions. They also identified actions for
reducing uncertainty that should be given the highest priority, including:
Improved diagnostic methods
Surveys of wild shrimp populations for presence of the four nonindigenous viruses and for
genetic composition
Experiments to reduce uncertainties surrounding virus transmission and virulence
Field epidemiological studies
Participants identified other areas where additional research is needed to improve the ability to
estimate risks to wild shrimp populations, including:
Viral persistence
Compensatory mechanisms
Monitoring of imported shrimp
Development of suitable population models
-------
Targeted surveys of nonpenaeid species to determine if they are susceptible to or carriers
of nonindigenous viruses
Workshop participants believed that, given the existing knowledge base, it is currently not feasible
to conduct a more comprehensive, quantitative assessment of the risks associated with shrimp
viruses. Participants generally agreed that, at present, qualitative evaluations could be made, but
there is a great deal of uncertainty associated with this type of process. Participants determined
that there is a need to continue efforts to gather available data on shrimp virus effects and a need
to conduct a systematic research effort that could be used to reduce the uncertainty of any
subsequent risk assessments.
Workshop participants identified the following areas of concern where additional efforts should be
focused:
Management implications of shrimp viruses
Risks of shrimp viruses to aquaculture operations
Risks of shrimp viruses to nonpenaeid species
in
-------
1. INTRODUCTION
This report highlights issues and conclusions from a shrimp virus peer review workshop
sponsored by the U.S. Environmental Protection Agency (EPA) in cooperation with the Joint
Subcommittee on Aquaculture (ISA), held January 7-8, 1998, in Arlington, Virginia. The goals
of the workshop were to:
Complete a qualitative assessment of the risks associated with shrimp viruses, following
the general risk assessment process developed by the Aquatic Nuisance Species Task
Force
Evaluate the need for a future, more comprehensive risk assessment
Identify critical risk-relevant research needs
The workshop focused on the scientific and technical aspects of the likelihood that nonindigenous
viruses will become established in wild shrimp populations in the Gulf of Mexico and southeastern
Atlantic coastal regions and on the potential consequences of such establishment.
This section provides an overview of the recently published ISA report that formed the basis for
the workshop, a description of the workshop process, and a discussion of the qualitative risk
assessment approach used at the workshop. Section 2 of the report summarizes discussions held
during the workshop on several aspects of the qualitative risk assessment process, and it contains
a risk characterization developed by the workshop chair and breakout group chairs following the
workshop's conclusion. Section 3 discusses actions for reducing uncertainty that were identified
by participants during the workshop. The reports of each breakout group are contained in
Appendix A.
1.1 ISA REPORT OVERVIEW
Dr. Kay Austin of EPA's National Center for Environmental Assessment, and a member of the
ISA Shrimp Virus Work Group, gave an initial presentation that discussed the Work Group's
efforts to date and events leading to the workshop. She provided an overview of the purpose,
scope, and findings of the Work Group's report, entitled "An Evaluation of Potential Shrimp
Virus Impacts on Cultured Shrimp and on Wild Shrimp Populations in the Gulf of Mexico and
Southeastern U.S. Atlantic Coastal Waters" (ISA, 1997; ISA Report). Highlights of her
presentation follow.
New, highly virulent viruses have been documented in foreign shrimp aquaculture. Consumer
demand for shrimp continues to grow and, to meet this demand, the United States has greatly
increased shrimp importation from areas of the world where shrimp viruses are known to be
endemic. Recent events have prompted calls for investigation into the actual risks to U.S.
domestic resources. These events have included catastrophic viral outbreaks in shrimp
aquaculture both in the United States and abroad, recent appearances of these organisms in
1-1
-------
shrimp in commercial retail stocks, and new information on the susceptibility of shrimp and other
crustaceans to these organisms. While some of these viruses have severe and lethal effects in
crowded aquaculture conditions, they are not known to pose threats to human health.
The U.S. shrimp industry (harvesting and processing alone) is valued at $3 billion per year.
Imported shrimp account for over 80 percent of the market. In 1995, imports exceeded domestic
production by a ratio of four to one, amounting to 720 million pounds (in tails). The largest share
of these imports comes from Latin America and Asiaareas of the world where shrimp viruses
are endemic. Domestic aquaculture operations, in contrast, account for a much smaller portion of
the U.S. market, ranging from 2 million pounds in 1991 to 4 million pounds in 1994.
The ISA, which is under the auspices of the President's Office of Science and Technology Policy,
formed the interagency Shrimp Virus Work Group in March 1996 to assess the risks associated
with these emerging viral pathogens. Four federal agencies are represented in the Work Group:
the National Marine Fisheries Service (NMFS), the U.S. Environmental Protection Agency
(EPA), the U.S. Fish and Wildlife Service (USFWS), and the U.S. Animal and Plant Health
Inspection Service (APHIS). ISA charged the Work Group with developing a federal interagency
strategy to address the shrimp virus issue and to identify relevant research on viral stressors, their
potential mode of transmission, and their potential for introduction to U.S. shrimp resources.
The Work Group recognized that the shrimp virus problem presents some unique issues in risk
assessment. They determined that the problem is a complex one that moves beyond the
traditional single-chemical, single-species assessment process. The shrimp virus problem involves
potentially nonindigenous viral stressors and has great potential to significantly impact the U.S.
shrimp industry and other ecological components of coastal systems.
During its initial evaluation of the problem, the Work Group decided to base its approach on
EPA's ecological risk assessment guidelines, which were published in draft form in 1996 (U.S.
EPA, 1996). Because the Work Group determined that not enough information was available to
complete an actual risk assessment, it followed a problem formulation approach that enabled the
Work Group to summarize risk-relevant information available prior to January 1997 and to
identify data gaps and critical research needs.
During its problem formulation activities, the Work Group developed a proposed management
goal and identified potential viral sources, potential viral and other environmental stressors, and
potential ecological effects. The Work Group also reached consensus on assessment endpoints
and developed a conceptual model (Figure 1) that illustrates the linkages between human
activities, viral stressors, and assessment endpoints of concern. The Work Group's report (ISA
1997; ISA Report) was published in June 1997.
1-2
-------
VIRUS SOURCI? AMD PATHWAYS
Aquacutture
Shrimp
Processing
f
Other
Sources/Pathways
I
STRESSORS
VIRUSES
Taura Syndrome
White Spot
Yellow Head
IHHNV
Other Anthropogenic
Stressors
(e.g., harvesting,
contaminants,
habitat destruction)
Environmental and
Ecological Factors
(e.g., temperature,
salinity, predation)
EXPOSURE
Penaeid
Shrimp
Life
Cycle
f~ Pmtozoea
ggs
OCEAN
VIRAL EFFECTS
Individual
Mortality
Viral Effects
on Other
Species
Indirect
Ecological
Effects
Population
Effects
±
Assessment Endpoint:
Ecological structure and function of
coastal and near-shore marine
communities as they affect penaeid
shrimp populations
Assessment End point:
survival, growth, and
reproduction of Penaeid
shrimp
Figure 1. Proposed Shrimp Virus Conceptual Model.
1-3
-------
Significant findings of the report include:
Viral disease has been associated with severe declines in wild shrimp harvests in the Gulf
of California. Populations of the blue shrimp, Penaeus stylirostris, and other less
dominant species plummeted coincident with the observed occurrence of IHHNV disease
in wild shrimp populations in the Gulf of California. The Work Group found that this is
the best piece of epidemiological information suggesting a link between introduced viruses
and declines in wild shrimp populations. There remains considerable debate, however,
regarding the validity of this association of disease and effects.
Nonindigenous shrimp viruses have not been documented in U.S. shrimp populations; until
recently, detection efforts have been minimal. Sampling techniques may have been
inadequate, and the correct technology may not have been available to adequately detect
the viruses.
Numerous viral disease outbreaks have occurred in U.S. shrimp aquaculture since 1994,
and frozen shrimp in commerce have been found to be contaminated with these viruses.
Laboratory studies show that all life stages of shrimp are potentially at risk from at least
one of the four viruses covered by this report.
Harvesting practices in foreign aquaculture could put U.S. domestic populations at risk.
The Work Group learned that when an outbreak occurs in some foreign aquaculture
operations, the affected crop is often harvested immediately and exported to avoid severe
crop and monetary losses.
Shrimp may be contaminated from a number of possible sources. The Work Group
identified aquaculture and shrimp processing as two potentially important sources that
may affect wild shrimp populations. The Work Group also considered a number of other
possible sources, such as live and frozen bait shrimp, ballast water, and natural spread by
mechanisms such as hurricanes, floods, or animals. Research and display facilities may
also be a source of exposure to wild populations.
Species other than shrimp may be at risk from these viruses. Viral disease could result in
alterations to ecosystem structure or function, potentially affecting a wide range of
endpoints, such as predator-prey relationships, competition, and nutrient cycling. Many
other economically and ecologically important organisms that occupy coastal areas feed
on juvenile shrimp, and impacts to these organisms could be serious if the wild shrimp
populations on which they feed decline. Other organisms may be susceptible to disease
themselves or serve as carriers of these viruses.
During July 1997, ISA and EPA sponsored public meetings in Charleston, South Carolina;
Mobile, Alabama; Brownsville, Texas; and Thibodaux, Louisiana to gather stakeholder input on
the shrimp virus issue and the ISA report. Stakeholders included individuals from the wild shrimp
1-4
-------
fishery industry, the shrimp aquaculture industry, the shrimp processing industry, environmental
organizations, regulatory and resource management agencies, and the general public. The
minutes of these stakeholder meetings were published in October 1997 (ERG, 1997).
1.2 PEER REVIEW WORKSHOP PROCESS
At the beginning of the workshop, the workshop chair, Dr. Charles Menzie of Menzie-Cura &
Associates, reviewed the agenda (Appendix D), explained the workshop's format, and reviewed
the workshop's goals, which were to:
Complete a qualitative assessment of the risks associated with shrimp viruses, following
the general risk assessment process developed by the Aquatic Nuisance Species Task
Force
Evaluate the need for a future, more comprehensive risk assessment
Identify critical risk-relevant research needs
Dr. Menzie explained that the workshop report would be used to provide input to a proposed
workshop to identify potential risk management options. The proposed workshop, sponsored by
ISA and NMFS, is tentatively scheduled for July 1998.
Prior to the workshop, ERG provided each expert with the ISA report (ISA, 1997), the minutes
of the stakeholder meetings about the ISA report (ERG, 1997), and a copy of a qualitative risk
assessment process for nonindigenous organisms (RAM, 1996; Appendix G). ERG asked panel
members (Appendix B) to review the material and prepare written comments to address questions
on the following topics:
Management goals, assessment endpoints, and the conceptual model
Viral stressors and factors regulating shrimp populations
Viral pathways and sources
Stressor effects
Comprehensive risk assessment and research needs
The charge to experts and experts' premeeting comments are contained in Appendix C.
Overheads prepared by the chairs that summarize the premeeting comments are contained in
Appendix F. Peer review experts were divided into three breakout groups, each of which was
charged with evaluating the risks associated with one of three viral pathways (aquaculture, shrimp
processing, and other potential sources).
Three experts in ecological risk assessment were selected as breakout group leaders: Dr. Wayne
Munns (EPA Office of Research and Development), who facilitated discussions on aquaculture;
Dr. John Gentile (University of Miami), who facilitated discussions on shrimp processing; and Dr.
Anne Fairbrother (Ecological Planning and Toxicology, Inc.), who facilitated discussions on
1-5
-------
"other potential sources." (See Appendix B for breakout group assignments.) After the
workshop, Dr. Menzie prepared the qualitative risk assessment (Section 2), Dr. Munns prepared
the report of the Aquaculture Breakout Group (Appendix A-l), Dr. Gentile prepared the report of
the Shrimp Processing Breakout Group (Appendix A-2), and Dr. Fairbrother prepared the report
of the "Other Pathways" Breakout Group (Appendix A-3). Workshop participants had a chance
to review and comment on the report prior to preparation of the final document.
1.3 QUALITATIVE RISK ASSESSMENT METHODOLOGY
Mr. Richard Orr, of the U.S. Department of Agriculture, Animal, and Plant Health Inspection
Services (USDA-APHIS), provided participants with an overview of the qualitative risk
assessment methodology to be used at the workshop. The process is based on the Aquatic
Nuisance Species Task Force (ANSTF) risk assessment approach (RAM, 1996, Appendix G),
which provides a qualitative assessment of the probability and consequences of establishment of a
nonindigenous species in a new environment. Mr. Orr noted that the methodology may be used
as a subjective evaluation, or it may be quantified to the extent possible depending on the needs of
the analysis. He reviewed an assessment on black carp to illustrate the application of this process
to a nonindigenous species. Both documents were provided to workshop experts as background
information prior to the workshop.
Mr. Orr explained that the risk assessment model is divided into two major components: the
"probability of establishment" and the "consequences of establishment" (see Figure 2, which
contains the risk assessment model from the Report to the Aquatic Nuisance Species Task Force).
These components of the model are further divided into basic elements that serve to focus
scientific, technical, and other relevant information for the assessment. Mr. Orr discussed how the
following elements could be used to estimate the probability of establishment of viral pathogens in
wild shrimp populations:
Probability of the nonindigenous organism being on, with, or in the pathway
Probability of the organism surviving in transit
Probability of the organism successfully colonizing and maintaining a population where
introduced
Probability of the organism spreading beyond the colonized area
The following elements are used in the ANSTF approach to evaluate the consequence of
establishment of a nonindigenous species (see page 22, Appendix G):
Economic impact
Environmental impact
Impact from social and/or political influences
For the purposes of the Shrimp Virus Peer Review Workshop, only environmental impacts were
evaluated. Economic and perceived impacts of establishment may be considered at the proposed
workshop on risk management options, scheduled for July 1998.
1-6
-------
Mr. Orr stressed that it is critical for the qualitative risk assessment to capture and communicate
the uncertainty that surrounds the available information about shrimp viruses.
1-7
-------
era
c
re
s
fa
fa
O
O-
-
O
Risk Assessment Model
Risk =
Standard Risk Formula
Probability of
Establishment
Consequence of
Establishment
oo
O
3-
Elements of the Model
I.
O
O
VI
H
fa
VI
Risk =
Organism
with
Palhway
Entry
Potential
Colonization
Potential
Spread
Potential
Risk Management
SS
Economic
Impact
Potential
Non-SS
Environmental
Impact
Potential
Impact
(Social &
For model simplification the various elements are depicted as being independent of one another
- The order of the elements in the model does not necessarily reflect the order of calculation
-------
2. QUALITATIVE RISK ASSESSMENT
2.1 THE RISK ASSESSMENT PROCESS
Workshop participants began the risk assessment process by reviewing the management goal and
assessment endpoints presented in the ISA report (ISA, 1997). Participants evaluated the risks
associated with aquaculture, shrimp processing, or other potential sources. In the breakout
groups, participants considered the ecological risks associated with each identified viral pathway.
The evaluation of each pathway was conducted independently. It is important to note that
participants did not attempt to rank the relative risk of the three identified sources.
Each breakout group evaluated both the potential for establishment of the viruses via the
identified pathways and the potential ecological consequences of establishment. The breakout
groups considered the four following elements of the potential for establishment of viruses via the
identified pathways:
Association of nonindigenous viruses with the pathway
Entry of nonindigenous viruses into coastal waters via the pathway (including survival)
Colonization/infection of shrimp at the local level
Spread of nonindigenous viruses to the shrimp populations at large
To determine the probability of establishment of nonindigenous viruses, the breakout groups rated
each of these elements low, medium, or high. The consequences of establishment were similarly
rated. During their deliberations, the breakout groups were asked to identify the level of
uncertainty (ranging from very uncertain to very certain) associated with each of the elements
described previously.
Using the method set forth in the ANSTF report (Appendix G), workshop participants estimated
the overall risk by compiling the risks associated with the individual elements of the process (i.e.,
[1] the four elements of the probability of establishment and [2] the consequence of
establishment). The probability of establishment is determined by the lowest ranking of any of the
four elements. For example, if elements under the probability of establishment had rankings of
high, high, low, and medium, the overall probability of establishment would be considered low,
because establishment is determined by the lowest likelihood of any one element. Rankings for
the probability of establishment and the consequence of establishment may be combined into an
overall level of risk as follows:
1-9
-------
If the Overall
Probability of
Establishment Is:
High
Medium
Low
High
Medium
Low
High
Medium
Low
And the
Consequence of
Establishment Is:
High
High
High
Medium
Medium
Medium
Low
Low
Low
Then the Overall
Risk Ranking Is:
High
High
Medium
High
Medium
Medium
Medium
Medium
Low
These rankings, which are based on judgment, should not be considered separately from the
discussion and rationale provided by the workshop participants. As noted in the ANSTF report
(RAM, 1996), "the strength of the Review Process is not in the element-rating but in the detailed
biological and other relevant information statements that motivate them."
After evaluating the probability of establishment for their respective pathways and the
consequences of establishment at the local and regional (e.g., Gulf of Mexico) population levels,
the three breakout groups presented their findings in a plenary session. Following the conclusion
of the expert workshop, the breakout group chairs and the workshop chair met to develop a risk
characterization for the assessment.
2.2 QUALITATIVE RISK ASSESSMENT RESULTS
This section summarizes discussions held during the workshop on several aspects of the risk
assessment process:
Management goals and assessment endpoints that frame the assessment (Section 2.2.1)
The probability of establishment of shrimp viruses (Section 2.2.2)
The consequences of establishment (Section 2.2.3)
A characterization of the risks resulting from a combination of the probability and
consequences of establishment (Section 2.2.4)
The reports of the three breakout groups are contained in Appendix A.
2-2
-------
2.2.1 Management Goals and Assessment Endpoints
Workshop participants were asked to evaluate the completeness and adequacy of both the
management goal and the assessment endpoints identified in the ISA report (ISA, 1997). In the
ecological risk assessment process, the management goal is intended to reflect the management
context of the assessment, while the assessment endpoints are explicit expressions of the
environmental values to be protected, which serve as the focal points for an assessment.
The management goal identified in the ISA report is to:
Prevent the establishment of new disease-causing viruses in wild populations of shrimp in
the Gulf of Mexico and southeastern U.S. coastal waters while minimizing possible
impacts on shrimp importation, processing, and aquaculture operations.
A number of participants thought that the management goal should be broadened to include risks
to aquaculture operations. Participants concurred that these risks are important but, because of
the limited time available for workshop discussions, they agreed that risks to aquaculture
operations would not be considered during the workshop. Participants recommended instead that
risks to shrimp in aquaculture operations and management of those risks be the subject of a
separate workshop.
The ISA report identifies two assessment endpoints:
Survival, growth, and reproduction of wild penaeid shrimp populations in the Gulf of
Mexico and southeastern U.S. Atlantic coastal waters
Ecological structure and function of coastal and near-shore marine communities as they
affect wild penaeid shrimp populations
Workshop participants elected to focus their efforts on the first assessment endpoint (direct
effects to wild shrimp populations) for the following reasons:
Risks to wild shrimp populations are of primary concern
Information on secondary effects is even more limited than information on direct effects
on shrimp
There was limited time available at the workshop for evaluating all possible direct and
indirect effects.
Participants recognized the potential for direct effects on organisms other than penaeid shrimp
and the potential for indirect effects; however, these effects were not discussed in detail during the
workshop. They are, however, a potential concern for resource managers.
2-3
-------
2.2.2 Probability of Establishment
This section summarizes breakout group discussions concerning the elements of the probability of
establishment (association with pathway, entry potential, colonization [infection] potential, and
spread potential).
Workshop participants recognized that differences among the four viruses could result in
variations in the risk rankings associated with the elements comprising the probability of
establishment. For example, if one virus were to survive longer than another virus in the marine
environment, it could affect the entry potential ranking. However, the breakout groups decided
to consider the viruses as a single group but to identify any unique differences that might alter risk
rankings. A summary of the characteristics of the four viruses is contained in Table 2-1.
2.2.2.1 Association with the Pathway
Breakout groups concluded with moderate to high certainty that there is a high likelihood that
viruses are present in the aquaculture pathway, shrimp processing pathway, and some of the other
potential pathways.
Aquaculture. The occurrence of nonindigenous viruses in U.S. aquaculture operations is well
documented. As summarized in the ISA report, TSV has been identified in disease outbreaks in
Hawaii, Texas, and South Carolina (Lightner, 1996a, 1996b). IHHNV was first identified in
Hawaii (Lightner et al., 1983a, 1983b) and was subsequently observed in farms in South Carolina,
Texas, and Florida (Fulks & Main, 1992). WSSV and YHV also have been documented at a
shrimp farm in Texas (Lightner, 1996a, 1996b). WSSV and YHV are considered to be of Asian
origin; TSV and IHHNV are thought to have originated in Latin America. Workshop participants
noted that the origins of these viruses are not always traceable to their ultimate sources, but it was
suggested that their introduction to the United States may have resulted from the importation of
infected shrimp from other regions of the world (e.g., Latin America and Asia).
Shrimp Processing. Shrimp viruses can be brought into the United States with imported shrimp
that are subsequently processed or used for other purposes (e.g., feed, bait shrimp, and retail
sale). Of the shrimp processed in the United States, 80 percent of the total crop is foreign and 20
percent is domestic in origin. Pathogenic viruses have been identified in imported shrimp sold in
this country.
Other Pathways. Other "primary" pathways described in the ISA report and considered by
workshop participants include ballast water, bait shrimp, animal vectors, and shrimp feed. There
appears to be no data on the occurrence of shrimp viruses in ballast water (or any of its
components). Nonetheless, it is known that many organisms are discharged routinely with ballast
water (including species of mysid shrimp, some of which have colonized bays and estuaries with
devastating effects). There is therefore a high probability that ballast water could
2-4
-------
Table 2-1. Virus persistence, virulence, and infectivity
Persistence
(1 = most, 4 = least)
Virulence to Gulf Species
(1 = most, 4 = least)
Penaeus setiferus
Larvae
Post-larvae
Juvenile
Adult
Penaeus duorarum
Larvae
Post-larvae
Juvenile
Adult
Penaeus aztecus
Larvae
Post-larvae
Juvenile
Adult
IHHNV
3.5
1
TSV
3.5
2
YHV
1.5
3
wssv
1.5
4
Relative Infectivity
+
ND
+
ND
+
ND
++
+
+
+
ND
+
+
ND
ND
++
ND
ND
++
ND
ND
++
ND
ND
++
++
ND
ND
++
+
ND
ND
++
+
INFECTIVITY
ND = No data
+ = Infectious
++ = Mortality
= Tried but negative
2-5
-------
contain shrimp viruses, whether free living, attached to participate matter, or in dead or infected
shrimp.
Anglers use shrimp as bait when fishing in estuaries for fish that eat shrimp. They purchase bait
from bait shops or they use shrimp sold in grocery stores for human consumption. Bait shrimp
generally are smaller than those sold for human consumption and are often considered
substandard. They may originate from aquaculture facilities that have harvested their shrimp prior
to full growout because of a viral outbreak. Some participants thought that Latin American and
Asian producers may freeze these small shrimp and ship them to the United States for sale as bait,
while the larger, uninfected shrimp will be sold at premium prices for human consumption.
Therefore, there is a high probability that some imported bait shrimp may contain viruses.
Both live and frozen shrimp may be sold as bait. However, only native species of aquaculture
shrimp may be harvested and sold as live bait. Some states (e.g., South Carolina) allow the use of
nonnative farm shrimp as frozen bait. Native shrimp used in aquaculture are known to sometimes
carry indigenous viruses (such as BP, another baculovirus) but, to date, there is no evidence that
they carry nonindigenous viruses. Furthermore, any of these shrimp that are harvested early due to
perceived disease problems are likely to be sold as frozen bait rather than as fresh bait. Therefore,
there is a low probability that live shrimp used for bait will carry nonindigenous viruses.
Shrimp feed is made from soy protein, fish protein (including anchovies and menhaden), shrimp
heads, and other types of shrimp and crustaceans (e.g., Anemia). Because the heads and other
body parts of infected shrimp can carry a high concentration of viruses, workshop participants
believed that there is a high probability that the shrimp parts used as an ingredient in shrimp feed
may be contaminated with the viruses. Although viruses may be associated with this pathway,
workshop participants concluded that they are likely to be destroyed during processing of the
shrimp feed (see Section 2.2.2.2).
Animal vectors such as gulls and freshwater and marine invertebrates were considered as another
possible source for viral entry. For example, gulls and other scavengers, such as raccoons, are
often seen feeding on dead shrimp and other organic matter associated with aquaculture facilities
that have undergone viral outbreaks. Workshop participants believed there was a high probability
of viral association with this pathway.
Workshop participants considered a number of other pathways to have a low probability for viral
association; therefore, these pathwaysnatural spread of the viruses, research and display
facilities, human sewage, fishing vessels, hobby and ornamental displays, live seafood distribution,
other crustacean aquaculture, and incidental introductionswere not discussed in detail at the
workshop.
2-6
-------
2.2.2.2 Entry Potential
Entry potential includes the probability of viruses surviving in transit and the probability of their
transport to coastal waters. Each breakout group recognized that the entry potential of
nonindigenous viruses depends on the pathway of arrival. For example, the survival and entry
characteristics of viruses found in shrimp processing effluents may be quite different from those
found in ship ballast waters. In addition, the breakout groups recognized that entry potential
depends on location. For example, viruses associated with shrimp that are raised, processed, or
disposed of in locations far inland are less likely to reach coastal waters than are viruses that are
associated with shrimp that are raised, processed, or disposed of along the coast. Workshop
participants evaluated subpathways within each of the major pathway categories (aquaculture,
shrimp processing, and other source pathways) and described entry potentials for viruses as
ranging from low to high. Participants found the level of certainty associated with these
evaluations to be quite variable.
Aquaculture. The Aquaculture Breakout Group considered the six subpathways from aquaculture
to wild shrimp stocks identified in the conceptual model contained in the ISA report. Many
breakout group members believed that the escapement subpathway (including both accidental and
intentional releases, as well as "escape" via transport of shrimp tissue by the predatory activities
of other animals) was the most likely route of release of viruses to the environment and that
viruses were likely to survive when transported via this pathway. (As discussed in the following,
however, some breakout group members believed that the sediment and effluent pathways, which
the group tabled because of a lack of crucial data, may also be important.)
The Aquaculture Breakout Group noted that the entry potential via escapement (and other
pathways) is likely to be related to the conditions in the pond (i.e., the presence and degree of
infection by the viruses), the life stage of the shrimp (e.g., postlarvae may be more likely than
adult shrimp to escape by passing through engineering controls), and the design of pond control
systems. They concluded with relatively high certainty that the probability of surviving in transit
would be high if conditions are favorable but assigned a low probability of survival if they are not.
The Aquaculture Breakout Group had considerable discussion about the ability of viruses to
survive in pond effluents and sediments. There is suggestive evidence about this potential
pathway. TSV has been documented in water, but not specifically in effluent waters. A
workshop observer (R. Laramore) communicated results of an experiment that suggest that caged
shrimp exposed in infected ponds developed disease. (Shrimp developed disease when exposed
within 1 to 2 days to experimentally inoculated water, but they did not develop disease when
exposed on days 3 to 5 following inoculation [R. Laramore]). In 1995, HSF, Ltd., and the
Arroyo Aquaculture Association conducted several trials in which cages were floated within a
shrimp growout pond that had experienced a TSV epidemic and with pond water in tanks. The
cages were suspended above the pond bottom and stocked with juvenile P. vannamei.
2-7
-------
No TSV was detected in shrimp exposed for 30 days under these conditions (F. Jaenike, personal
communication). These results suggest that TSV may be transmitted during the acute but not the
chronic stages of the disease. Other data suggest that IHHNV can survive in water in an infective
state for at least 24 days (Glover et al., 1995). One participant noted that viruses can spread
quickly from pond to pond on aquaculture facilities, but it is not known how this transmission
occurs. Based on this information, the aquaculture breakout group estimated that there is a
medium potential that effluents released from infected farm ponds are a viable pathway for
exposure to native populations; however, the breakout group was very uncertain about this
estimate.
Shrimp Processing. The Shrimp Processing Breakout Group identified two subpathways for
which there is a moderate to high potential for viruses to enter coastal areas: untreated effluents
from shrimp processing facilities and solid wastes from disposal facilities near coastal areas that
receive waste from shrimp processing facilities. The breakout group concluded with high
certainty that there is a low potential for viable shrimp viruses to survive in effluents that are
treated and disinfected at municipal facilities and, therefore, there is a low potential for entry of
viable shrimp viruses to coastal areas from this pathway.
The Shrimp Processing Breakout Group estimated that approximately 50 percent of shrimp
processing liquid effluent is untreated and that virus-contaminated discharges may therefore be
released regularly into the environment. The breakout group was very certain that the probability
of the organism surviving in transit, and therefore entering the environment through this pathway,
is high.
Because of the uncertainties associated with the amounts of material reaching landfills, the types
of vectors, and the threshold amount of virus required to infect the wild and aquaculture
populations, the Shrimp Processing Breakout Group found it difficult to assess the probability of
establishment of shrimp viruses from solid waste disposal facilities. Most breakout group
members generally agreed that the shells, and particularly the heads, of foreign farmed and wild
shrimp are highly likely to contain viruses. Considering these factors, breakout group participants
concluded that these viruses are likely to persist for some time in landfill settings. Land crabs and
seagulls are thought to be primary vectors for moving viruses from the landfills to estuarine
waters. When these animals consume virus-contaminated materials, some of the viruses (TSV
and IHHNV) can pass through their digestive systems in an infective state. It was noted that TSV
remains infective following gut passage in gulls, waterboatman, and other insects. It is not
known whether the concentrations and frequency of virus introduction from these vectors is
sufficient to infect wild and aquaculture shrimp populations. The breakout group was reasonably
certain that there is a moderate probability of entry potential from coastal landfills to estuaries.
Other Pathways. The "Other Pathways" Breakout Group found that the entry potential of viruses
in ballast water and bait shrimp is high. The group determined that while it is not likely that the
freezing process used for bait shrimp will significantly reduce the virulence and infectivity of the
virus, the effects of freezing may be virus specific. For shrimp feed, participants concluded that
2-8
-------
the probability of survival in transit depends on whether or not the feed meal is heat treated to
temperatures sufficient to inactivate all viruses. It is thought that some of the viruses (e.g., TSV)
may survive and maintain infectivity, even when heated to temperatures greater than 100 °C.
While most of the fish meal produced in the United States is subjected to temperatures that appear
to be sufficient to kill the viruses, breakout group members were unable to provide published data
that would confirm this supposition. Moreover, several participants believed that other countries
do not always heat-treat their meals, which would increase the potential for viable viruses to be
present in the feed. The "Other Pathways" Breakout Group concluded that the transit survival
probability is low for heat-treated feed and high for untreated feed. In contrast, the Shrimp
Processing Breakout Group was very certain that feed was processed at temperatures sufficient to
inactivate the viruses. Additional research will be necessary to resolve this issue.
2.2.2.3 Colonization Potential
Workshop participants agreed that the potential for viruses to colonize coastal areas is one of the
most critical aspects of evaluating the potential for establishment. Workshop participants
concluded that there is a high potential for viruses to be associated with many of the pathways
identified in this report and a low to high potential that these pathways could lead to introduction
of viruses. The breakout groups were certain about association of viruses with these pathways
and their entry potential through the pathway; however, they had a high degree of uncertainty
about colonization potential. Breakout groups believed that, for most subpathways, there is either
a low or medium likelihood that, once introduced, viruses would be able to colonize native shrimp
at a local level (i.e., within specific estuaries or embayments). In support of their conclusions, the
breakout groups identified the following factors:
Colonization potential is likely to be related to the magnitude of the source and the
frequency of introductions. Therefore, large, frequent sources may have a greater
likelihood of colonization than small, intermittent sources.
Colonization potential is likely to be related to the medium in which the viruses are
introduced. For example, viruses introduced within live or dead shrimp are thought to
have a greater likelihood of colonization than are viruses introduced via water.
There is no clear evidence to suggest that colonization has occurred in wild shrimp
populations, despite a history of outbreaks in aquaculture operations, the presence of
shrimp processing operations, discharges of ballast water, and the use of bait shrimp.
(Recent evidence suggests, however, that WSSV-like viruses found in wild shrimp
populations in South Carolina coastal waters may not differ from Asian isolates of the
virus [Lo et al., in press]).
2-9
-------
2.2.2.4 Spread Potential
The breakout groups viewed the potential spread of viruses beyond the initial locus of
colonization as an area of uncertainty. During plenary discussion of the reports from the
individual breakout groups, workshop participants generally believed that there is a medium
probability that viruses could spread beyond the initial locations of colonization.
Breakout groups identified a number of factors significant to evaluating the potential spread of
introduced viruses, such as the degree of interaction that would occur among individual wild
shrimp and the spatial scale over which shrimp might "mix." Stocks of P. setiferus in the
southeast Atlantic are thought to be fairly genetically homogeneous, as are the northern and
southern populations in the Gulf of Mexico. Workshop participants believed that this suggests
the potential for substantial interaction over broad geographic regions, which would promote the
spread of viral infection. However, genetic homogeneity may not be the case for other penaeid
species. The potential for spread also depends in large part on the time course of the disease, as
well as the density of shrimp in wild populations. Breakout group members determined that low
shrimp densities are likely to hinder disease spread, whereas high densities are likely to promote
transmission. Spread potential is also host dependent and virus specific. It was noted that TSV
and IHHNV have low spread potential, and the spread potential of YHV and WSSV is currently
unknown. A WSSV-like virus has been found in a variety of crustaceans in southeastern Atlantic
waters, but it is unknown at this time if it is the same as the Asian strain of WSSV. (Recent
evidence suggests, however, that WSSV-like viruses found in wild shrimp populations in South
Carolina coastal waters may not differ from Asian isolates of the virus [Lo et al., in press]). As
noted in the ISA report, the presence of other stressors (e.g., low dissolved oxygen and extreme
salinity) is also likely to influence the potential for spread of the disease. When WSSV is detected
in wild stocks in Asia, it is known to be distributed over wide geographic areas, which suggests
that viral disease can spread from an original locus of colonization.
2.2.3 Consequences of Establishment
In continuing to assess the risks to wild populations of shrimp viruses, the breakout groups
evaluated the potential ecological effects associated with the establishment of pathogenic shrimp
viruses. The breakout groups approached this step of the qualitative risk assessment process by
considering the available information on the direct effects of viruses on shrimp. Breakout groups
also examined possibly analogous situations based on experience with other diseases and
invertebrates. Breakout groups discussed possible effects on ecological structure and function
but, due to the limited time available, gave primary attention to direct effects on wild shrimp
populations. In the absence of documented information or firsthand knowledge, experts relied
primarily on professional judgment to evaluate the consequences of establishment. The breakout
groups concluded that there is a high degree of uncertainty in assessing the consequences of
establishment.
2-10
-------
2.2.3.1 Direct Consequences to Shrimp Populations
In considering the possible consequences of shrimp viruses to shrimp populations at the local level
and at the scale of the entire populations or stock, breakout groups evaluated three types of
effects:
Mortality of the infected animal
Reduction in reproductive rates
Alteration of the genetic structure of the population
Mortality Effects. Breakout group experts concluded that the direct consequences of the
establishment range from low to medium and that effects on the mortality of shrimp are more
likely to occur at the local level than at the scale of the entire population or stock. The breakout
groups determined that the probability of colonization at a local level is greater than the
probability that viruses would spread beyond the local level to a regional population. It is thought
that WSSV and YHV are more likely than IHHNV or TSV to cause acute mortality but that
IHHNV and TSV are more likely to become endemic.
Reproductive Effects. Breakout group experts focused primarily on factors that would affect
reproductive output or recruitment. Experts were aware of no information describing adverse
viral effects on the reproductive potential of infected individuals (indicating a potentially
important data gap). One expert noted that reproductive output of infected P. vannamei brood
stock appears to be unaffected by viral infection (F. Jaenike, personal communication). However,
in contrast to the previous statement, individual growth impairment in offspring of P. vannamei
infected with IHHNV has been documented (Fulks & Main, 1992). Assuming that fecundity of
female Penaeus is an increasing function of size (a phenomenon common in other invertebrate
species), workshop participants considered that stunted growth of offspring could result in
reduced reproductive output of the second generation. Individual growth impacts could therefore
cause population-level effects, although an analysis of any changes in reproduction on shrimp
population dynamics would be required to support this conclusion. Workshop participants noted
that epidemiologic models show that, in "r-selected" species, effects on reproduction can have
greater effects on population size than mortality effects. (Penaeid shrimp can be characterized as
"r-selected" organisms because they display an annual life history pattern with high reproductive
output and high mortality during early life stages.)
Effects on Genetic Structure and Fitness. Breakout group participants discussed the potential
effects of virus colonization on the genetic structure and fitness of wild shrimp populations. One
breakout group thought that rapid reductions in population abundance resulting from viral disease
could have unknown but potentially important effects on genetic structure by limiting genetic
variability (the "founder effect"). One participant cited evidence from Thailand indicating that
shrimp populations in the south of Thailand are much less genetically diverse than those from the
northern part of the country. It has been hypothesized that this is due to the release of shrimp
from aquaculture into the wild. One breakout group discussed whether genetic resistance to
viruses differs among populations. Further knowledge of genetic variability among Gulf Coast
2-11
-------
shrimp is necessary to make accurate predictions about which area has the highest potential for an
epizootic.
Other Information. Other information or lines of evidence that affected the experts' professional
judgments about the potential consequences of establishment are in the following:
Penaeid shrimp can be characterized as "r-selected" organisms because they display an
annual life history pattern with high reproductive output and high mortality during early
life stages. Thus, penaeid shrimp populations that suffer population reductions in one year
can exhibit rapid recovery, and this may reduce the long-term consequences of short-term
impacts. In reviewing available information, the breakout group concluded that mass
mortalities of adult shrimp may have relatively short-term impacts on standing shrimp
stocks. For example, some natural stressors on shrimp (e.g., cold temperatures or
freshwater flooding) are known to cause short-term reductions in populations at the local
level. Because of high fecundity and migratory behavior, P. setiferus is capable of
rebounding from a very low population size in one year to a large number in the next, if
environmental conditions are favorable. This has been observed off the South Carolina
coast several times in the past 50 years (Linder & Anderson, 1956; McKenzie, 1981). In
another case, an increase in reproductive output of the Honduran population of P.
vannamei was reported during a 1994 TSV outbreak. This provides anecdotal support for
the concept that demographic compensatory responses may occur in disease-depleted
populations, although it was noted that the population changes could have been caused by
other factors (Laramore, 1997).
Along with anecdotal information about the possible long-term effects of viral infections in
Latin American and Asian shrimp populations, observations by some workshop
participants suggest that direct mortality effects would be relatively transitory. Also, it
was suggested that initial outbreaks could lead to enhanced resistance to future viral
infection, based on the observation that resistance to HTHNV appears to have increased in
all populations tested since the identification of this virus in Hawaiian stocks.
It should be noted, however, that some workshop participants were concerned that the
ability of viral pathogens to persist at low levels in a population could result in long-term
adverse population effects. For example, purported virus-induced declines in the
population abundances of P. stylirostris in the Gulf of California began in 1987 and lasted
6 to 7 years, with stocks now reported to have returned to preoutbreak levels. (The role
of IHHNV as the cause of the initial population decline has been the subject of much
debate, however.)
Based on observations from aquaculture situations, it appears that local colonization of
shrimp viruses could result in local mortalities of shrimp. For example, TSV and others
viruses are known to cause mass mortality on shrimp farms. Experiments with these
viruses have documented mortality rates of up to 100 percent. In South Carolina, survival
on commercial farms affected by TSV dropped from 63 percent in 1995 (the year prior to
2-12
-------
the TSV outbreak) to 19 percent in 1996 (the year of the TSV outbreak) (P. Sandifer,
personal communication).
Lines of evidence from other crustacean species indicate an association between an
introduced biological agent and subsequent environmental impacts. For example, a
crayfish species introduced from California to Europe may likely have served as a carrier
to spread the freshwater crayfish plague throughout Scandinavia (Unestam & Weiss,
1970). Unlike short-term natural stressors (e.g., changes in temperature or salinity), an
introduced disease organism (biological stressor) is likely to persist in the population.
No empirical data exist to indicate that historical releases of shrimp virus to the Gulf of
Mexico or to southeast Atlantic coastal waters have resulted in population-level impacts.
However, no well-designed studies have been conducted to examine the epidemiological
conditions within these waters.
2.2.3.2 Effects on Ecological Structure and Function
Workshop participants observed that the introduction of shrimp viruses could affect ecological
conditions apart from any direct effects on shrimp; however, experts made only a limited attempt
to characterize these consequences, primarily because of a lack of information and because these
indirect effects were not a focus of this workshop. Despite these limitations, some of the
discussion related to this topic may be helpful to risk managers.
The Aquaculture Breakout Group discussed instances in which other invertebrate species have
experienced severe disease consequences. Participants viewed these examples as relevant to the
effects of pathogenic viruses on shrimp:
The near decimation of oysters (Crassostrea virginicd) by the protozoan pathogens
Haplosporidium nelsoni and Perkinsus marinus, called MSX and dermo disease
respectively (Haskin & Andrews, 1988; Andrews, 1996; Burreson & Ragone-Calvo,
1996), has resulted in significant changes in the oyster reef habitat throughout Chesapeake
Bay and dramatically reduced the rate at which bay water was filtered by feeding bivalves
(Kennedy, 1996).
Insect/virus associations in which high abundances of the host species promote rapid
outbreaks of viral disease, followed by dramatic declines in the host, near-disappearance of
the virus, and reestablishment of the host (S. Thiem, personal communication).
The introduction into Scandinavia of North American crayfish that were carriers of the
freshwater crayfish plague Aphanomyces astaci (Unestam & Weiss, 1970).
Workshop participants felt that these and similar examples could serve as models for extrapolating
potential consequences of viral establishment for shrimp populations. These examples may also
2-13
-------
serve as models for how ecological systems might be affected by viral outbreaks in shrimp. Either
application would require careful analysis to identify similarities and differences relative to the
shrimp virus situation.
The "Other Pathways" Breakout Group discussed the potential for viruses to affect estuarine
ecology by infecting other species of shrimp, such as grass shrimp. Grass shrimp (Paleomonetes
sp.) are an important part of the estuarine food web. Many species offish (and penaeid shrimp)
rely on this species as an important prey item. Data from Thailand suggest that grass shrimp may
be carriers of one or more of these viruses, but data on infectivity rates and effects are lacking.
On the other hand, it was noted that observations in South Carolina confirmed the presence of
large populations of apparently healthy Paleomonetes in tidal areas near TSV-infected shrimp
farms.
2.2.4 Risk Characterization
Using the Aquatic Nuisance Species approach (RAM, 1996; Appendix G), workshop participants
characterized the risk of viral introductions to wild penaeid shrimp populations by combining the
probability of establishment of the virus with that of the presumed ecological consequences (see
Section 2.1). Workshop participants assessed risks to local populations, which they generally
defined as the population within a single estuary, and they also considered the long-term effects
on the entire population of native shrimp in the Gulf of Mexico and southeastern Atlantic coastal
waters.
2.2.4.1 Risk to Local Populations
Workshop participants concluded that the probability of establishment of shrimp viruses in a local
estuary ranges from low to medium. The probability of establishment depends primarily on the
colonization potential of the particular viruses. However, the probability of establishment could
become much greater if virus is introduced repeatedly to the estuary over a long period.
Workshop participants generally believed that the impact of such an establishment on the local
shrimp population would be moderate. They noted that initial kill rates might be high, but the
population would likely recover rapidly due to reintroduction of shrimp from other locations.
Therefore, workshop participants characterized the overall long-term risk of virus introductions to
the shrimp populations in a local estuary as moderate. (The possibility of longer-term effects is
suggested in discussed in Section 2.2.3.1).
Workshop participants also considered the risks posed by shrimp viruses to other components of
the estuarine ecosystem to be moderate, although uncertainty surrounding this risk estimate is
very high. Of particular concern to participants was coinfection of important food-web species,
such as grass shrimp and crayfish. Because both penaeid shrimp and grass shrimp are important
food sources for many other estuarine organisms, participants noted that the loss of this food base
could have significant effects on other species. Following an initial viral kill of shrimp, fish or
2-14
-------
wildlife populations that depend on shrimp and other crustaceans as prey sources may take longer
to recover than shrimp populations.
Participants raised concerns about the lack of information on the transmissibility of disease from
one estuary to another through migration of diseased or infected shrimp. Participants thought
that survivors of a local epizootic could move out to sea to reproduce, possibly infect other
shrimp and offspring, and then move into adjacent or nearby estuaries. Such an event would
expand what appears to be a localized risk into large-scale risk; however, each breakout group
that evaluated the potential for spread by natural processes rated the probability of this occurrence
as low. Therefore, the risk of a local infection having large-scale consequences is characterized as
moderate.
2.2.4.2 Large-Scale Risk
Workshop participants characterized the risk from viral introductions to the entire population of
native shrimp along the southeastern Atlantic coast and within the Gulf of Mexico using the same
analysis of the establishment pathways combined with that of the potential consequences of
establishment on a large geographic scale. Workshop participants concluded that the
consequences of virus introduction to the population as a whole would be relatively insignificant,
and they characterized the risk as low.
Some participants expressed concern that the genetic structure of the population might be altered
and, if viral resistance were linked with appropriate genes, overall fitness of the shrimp could be
lowered. One participant noted that alterations to the genetic structure of the population could
make the shrimp more susceptible to future infections and to simultaneous environmental
stressors, such as weather changes or reduced estuarine salinity, thereby potentially increasing the
risk potential. Furthermore, some participants stressed that uncertainty about the long-term
ecological consequences of viral introduction will remain high until the effects of virus infection
on reproduction can be determined.
2.2.4.3 Summary
Overall conclusions by workshop participants concerning the risks posed by nonindigenous
shrimp viruses may be summarized as follows:
Based on information currently available, most workshop participants believed that the
risk to native shrimp from introduction of nonindigenous viruses is low to moderate,
although uncertainty is high.
Most participants agreed that local effects should be given a higher risk ranking than
large-scale effects because local effects are more likely to occur.
2-15
-------
Participants suggested that the large amount of uncertainty associated with this risk
characterization could be reduced through appropriate laboratory and field studies. The
lack of evidence of conclusive viral impacts on worldwide shrimp populations does not
derive from published systematic studies but rather is anecdotal. Furthermore, by analogy,
other marine invertebrates have experienced severe local impacts from exposure to
pathogens (as has been noted in oyster populations in Chesapeake Bay). Also, viruses that
have become established in terrestrial insect populations can cause cyclic epizootics and
population crashes. Therefore, participants concluded that there is an urgent need to
continue efforts to gather available data on shrimp virus effects and to conduct a
systematic research effort that could be used to reduce the uncertainty of any subsequent
risk assessments.
2.3 RISK MANAGEMENT RELEVANCE
Although this report does not recommend risk management actions, it contains information that
may help risk managers with their decisions by:
Providing insight into the pathways by which shrimp viruses could potentially enter and
become established in the marine environment
Identifying potential consequences to wild shrimp populations at local and stock levels
Suggesting specific actions and studies that can reduce the uncertainties associated with
evaluating the potential risks of shrimp viruses on wild shrimp populations
The ability to make quantitative estimates of the risks of viruses to wild populations of penaeid
shrimp is constrained by the amount and type of information that is currently available. The
majority of workshop participants believed that it is unlikely that the information required to make
quantitative estimates of risk will be available within the foreseeable future. At present,
qualitative evaluations can be made.
The ability of workshop participants to address broader ecological risks in a comprehensive
manner was limited by available information, but participants thought that this is an important
issue that merits further consideration. Furthermore, while the risks that shrimp viruses pose to
shrimp aquaculture operations was not part of the scope of the workshop, workshop participants
agreed that these risks should be given special attention as part of another technical or
management workshop.
2-16
-------
3. ACTIONS FOR REDUCING UNCERTAINTY
The qualitative risk assessment conducted during the workshop revealed several critical sources
of uncertainty. Further improvement in the ability to estimate risks to wild populations of shrimp
will require reducing uncertainty in these key areas.
Workshop participants discussed the relative importance of actions for reducing uncertainty.
Some participants stressed that, to reduce uncertainty, risk management actions need to occur in
parallel with research, monitoring, and other actions.
Most workshop participants generally believed that particular emphasis should be given to the
following actions for reducing uncertainty:
Improved diagnostic methods
Surveys of wild shrimp populations for the presence of nonindigenous viruses and for
genetic composition
Experiments to reduce uncertainties surrounding virus transmission and virulence
Field epidemiological studies
3.1 DIAGNOSTIC METHODS
Workshop participants determined that improvements to existing diagnostic methods and
development of new diagnostic tools are a very high priority. Several participants noted that
without adequate diagnostic methods, other risk assessment elements cannot be well studied or
adequately evaluated. Other participants noted that many valuable diagnostic tools currently
exist. Several key needs were identified during the workshop:
There is a significant need to develop new diagnostic procedures. Some molecular probe
applications and bioassay tests are available, although several workshop participants noted
that the sensitivity of existing bioassay tests needs to be improved. One participant also
cited the need to develop cell culture tests for Crustacea, noting that new technologies are
available to assist in developing cell cultures, but money and lack of equipment have been
major obstacles.
Tests for infectivity are needed to establish the threshold number of viruses that would be
required for colonization potential. At least two tests should be employed, such as a PCR
and ELISA or a PCR and a bioassay.
Current diagnostic applications are focused on detecting viruses in the animal itself.
Although some preliminary efforts have been made to detect viruses in environmental
media (e.g., to identify the presence of WSSV using water concentration techniques and
PCR), techniques to detect viruses in effluent streams, sediment, and other environmental
media need to be improved.
3-1
-------
There appears to be considerable variability among laboratories in the procedures for
using available diagnostic tools. Procedures for using diagnostic tools should be
standardized so that both the credibility and limitations of diagnostic tools can be
established.
3.2 SURVEYS OF WILD SHRIMP POPULATIONS
Participants identified the need to survey native shrimp populations to develop baseline
information on viruses in wild stocks. It was noted that some monitoring activity has been
conducted in the coastal waters of South Carolina and Texas. Participants generally believed that
it was important to proceed with field surveys despite the current limitations of diagnostic
methods. Participants suggested that because of these limitations, current survey efforts should
include the archiving of samples to be evaluated pending development of improved diagnostics.
Workshop participants noted that monitoring surveys should include genetic characterization of
wild populations. To date, only limited studies have been conducted. (In one study that is
underway, molecular techniques are being used to determine the degree of genetic variability
between populations of P. setiferus in the Gulf of Mexico and the U.S. southeastern Atlantic
coastal region.) Participants suggested that surveys should be focused both in areas that may
have experienced the release of nonindigenous viruses and areas where it is unlikely that prior
release has occurred.
3.3 EPIDEMIOLOGY OF SHRIMP VIRUS TRANSMISSION
Workshop participants identified a need for well-designed experiments to improve understanding
of the pathogenicity of viruses in native shrimp. In particular, studies are needed on virulence,
distribution in various shrimp tissues, and rates of transmission, susceptibility, and recovery.
Some suggested that laboratory experiments would be hindered by inadequacies in current
techniques to identify pathogens and by the absence of diagnostic methods specific to identifying
viruses in various environmental media. Given existing techniques for quantifying the amount of
virus present, participants noted that currently it is most feasible to conduct qualitative
transmission studies in which the amount of virus is estimated on a relative basis.
In other discussions, participants identified the need to understand not only mortality effects but
also the consequences of infection on shrimp reproduction and growth. It is recognized that there
are significant differences in viral pathogenesis among the four different viruses and the relative
ability of the viruses to affect mortality, growth, and reproduction.
Participants also identified the need to develop a better understanding of the transmission of
viruses from one species to another (i.e, between penaeid species and between penaeid and
nonpenaeid species).
3-2
-------
One participant stated that the most important reason to improve understanding of the
epidemiology of shrimp viruses is to help identify mitigation measures (e.g., for aquaculture as a
pathway).
3.4 FIELD EPIDEMIOLOGICAL STUDIES
In addition to laboratory-based experiments, most participants believed that a parallel effort
involving field epidemiology could yield information helpful for understanding the prevalence and
potential effects of viruses in wild shrimp populations. Field epidemiological studies may not
provide the same level of understanding of detailed mechanisms as would laboratory experiments.
Participants suggested that field epidemiological studies could make use of existing information
from Latin America and Southeast Asia. Information would be sought on:
The extent to which native shrimp populations in these areas may have been exposed to
viruses
The presence of viruses within these populations
The observed effects (or lack thereof) of viruses on shrimp abundance and recruitment
Possible ecological effects
Others suggested that the known locations of shrimp virus prevalence around the world should be
documented and mapped so that potential sources can be identified.
3.5 LOWER PRIORITY RISK-RELEVANT RESEARCH AREAS
Workshop participants identified other areas, in addition to the four priority areas listed
previously, where additional research is needed to improve the ability to estimate risks to wild
shrimp populations.
3.5.1 Viral Persistence
Some participants noted the need to develop better techniques and to conduct experiments to
evaluate the persistence of viruses in effluent streams, sediment, and other environmental media.
It was noted that experiments should couple viral persistence with viral infectivity. For example,
participants noted that IHHNV can be detected in sediments for 24 days; however, the duration of
infectivity is unknown.
3-3
-------
3.5.2 Compensatory Mechanisms
Participants felt that it is important to develop a better understanding of the compensatory
mechanisms of native shrimp species in response to viral disease outbreaks. Research is needed
to:
Understand genetics and disease resistance (i.e., the need to improve understanding of the
relationship between population genetics and the identification of disease-resistant
phenotypes and how particular phenotypes develop resistance to a particular virus).
Determine whether shrimp populations compensate for increased mortality with increased
reproduction.
Compile information on the shrimp immune-like response to viral infection. It was noted
that coupling our understanding of target-organ sensitivity with information about
resistance will improve the ability to predict which shrimp are likely to become carriers.
3.5.3 Monitoring of Imported Shrimp
Participants identified the need to monitor virus levels in imported shrimp using tests such as PCR
and bioassay. Some suggested that, in terms of risk reduction, monitoring imported shrimp
should be a higher priority than monitoring wild shrimp populations because of the high volume of
imported shrimp.
3.5.4 Development of Suitable Population Models
Suitable population models are needed to evaluate the consequences of various virus-induced
mortality or reproductive impairment scenarios. Because of the commercial importance of
shrimp, workshop participants believed that it is highly likely that population models exist for
these species. Additionally, a large body of catch statistics could be subject to time series analysis
in concert with known periods of virus outbreaks or other environmental stressors, such as storm
events. These types of data may be available for foreign fisheries as well. By using population
models, constants for infection and transmission rates, and transport and fate, a modeling
framework could be created to examine specific hypotheses. Sensitivity analyses could then be
performed to determine which parameters are most important and contribute the most
uncertainty. Research could then be directed to reduce uncertainty.
3-4
-------
3.5.5 Other Risk-Related Research Needs
Other risk-related research needs identified by workshop participants include:
Procedures for disinfection and eradication of large-scale outbreaks in aquaculture settings
Genetic and biochemical characterizations of the viruses
Research to improve understanding of factors that exacerbate expression of viral disease
under conditions of high densities and high nutrients found in aquaculture settings
Targeted surveys of nonpenaeid species (e.g., grass shrimp, crayfish, and micro-crustacea)
to determine if they are susceptible to, or carriers of, nonindigenous viruses
3-5
-------
4. SUMMARY
This section provides a brief summary of the results of the workshop. Topics include the
qualitative risk assessment process; the need for a future, more comprehensive risk assessment;
risk-relevant research needs; and areas of additional concern.
4.1. QUALITATIVE RISK ASSESSMENT PROCESS
Workshop participants conducted a qualitative assessment of risks by considering the:
Likelihood of viruses being present in the pathway
Ability of the viruses to survive transit in the pathway
Colonization potential of the viruses (in native shrimp)
Spread potential of the virus within native shrimp populations
Consequences of establishment
In general, workshop participants believed that viruses could be in pathways leading to coastal
environments and that they could survive in these pathways. Participants concluded that there is
some potential for viruses to colonize native shrimp in a localized area, such as an estuary or an
embayment, near the point of entry into the marine system. Participants had widely divergent
views on the potential for viruses to spread beyond the initial local area of colonization, and this
divergence reflected the large uncertainty associated with this aspect of exposure. Participants
considered the potential for localized colonization and subsequent spread to be a critical aspect of
evaluating the potential establishment of viruses in native shrimp.
Participants considered the consequences of virus establishment at a local level (e.g., within an
individual estuary) as well as within the offshore stocks. Workshop participants discussed the
impact of such an establishment on the local shrimp population. Initial kill rates might be high,
but the population would be likely to recover rapidly due to reintroduction of shrimp from other
locales. The risk from viral introductions to the entire population of native shrimp along the
southeastern Atlantic coast and within the Gulf of Mexico was thought to be relatively
insignificant, and workshop participants characterized this risk as low. Concern was expressed
that certain effects (e.g., effects on genetic structure of shrimp and on the ecological system) may
be difficult to assess.
4.2. COMPREHENSIVE RISK ASSESSMENT NEEDS
Most workshop participants concluded that, given the current knowledge base, it is infeasible to
conduct a more comprehensive, quantitative estimate of risk. Most participants believed that, at
present, qualitative evaluations can be made, but these are accompanied by large uncertainties.
Participants agreed that there is a need to continue efforts to gather available data on shrimp virus
4-1
-------
effects and to conduct a systematic research effort that could be used to reduce the uncertainty in
any subsequent risk assessments.
4.3. RESEARCH NEEDS
Workshop participants identified a number of areas in which further research and information
would improve the assessment of risks and the evaluation of current conditions, with particular
emphasis on the following areas:
The improvement of existing and the development of new diagnostic methods for
viruses in shrimp and environmental media. These methods are essential for all
research studies and monitoring programs and for determining if viruses are present in
imported shrimp, cultures used for aquaculture, and other possible pathways.
Surveys of wild shrimp populations. Baseline information on the presence of viruses in
native shrimp populations would provide insight into the extent to which populations
already carry viruses. Baseline information would also be useful for supporting
epidemiological studies. Baseline studies could proceed even though there are limitations
with current diagnostic methods. Well-designed studies would be enhanced by including
an examination of the genetic structure of the populations.
Epidemiology of shrimp virus transmission. Workshop participants identified a need
for well-designed experiments to improve understanding of the pathogenicity of viruses in
native shrimp.
Field epidemiological studies. In addition to laboratory-based experiments, participants
believed that a parallel effort involving field epidemiology could yield information helpful
for understanding the prevalence and potential effects of viruses in wild shrimp
populations.
4.4. ADDITIONAL AREAS OF CONCERN
Workshop participants identified the following areas of concern, in which additional efforts should
be focused:
Management implications of shrimp viruses. It was recommended that a risk
management workshop be held, focusing on impacts to natural resources and on possible
impacts on shrimp importation, processing, and aquaculture operations.
Risks of shrimp viruses to aquaculture operations. Workshop participants also
recommended that a separate workshop be held on this topic.
4-2
-------
Risks of shrimp viruses to nonpenaeid species. Because this workshop was limited to
evaluating the direct effects of viruses on wild shrimp populations, participants
recommended that additional effort be directed toward evaluating nonpenaeid shrimp
species (e.g., grass shrimp) and other species that could be impacted by the viruses (e.g.,
crabs, amphipods, and copepods).
4-3
-------
5. REFERENCES
Andrews, J.D. 1996. History ofPerkinsus marinus, a pathogen of oysters in Chesapeake Bay
1950-1984. J. Shellfish Res., 75:13-16.
Aquatic Nuisance Species Task Force, Risk Assessment and Management Committee. 1996.
Generic nonindigenous aquatic organisms risk analysis review process. Draft Final Report.
Burreson, E.M., & L.M. Ragone-Calvo. 1996. Epizootiology ofPerkinsus marinus disease of
oysters in Chesapeake Bay, with emphasis on data since 1985. J. Shellfish Res., 75:17-34.
Eastern Research Group. 1997. Minutes of the stakeholder meetings on the report of the ISA
Shrimp Virus Work Group. Lexington, MA.
Glover, K.L., L.M. Nunan, & D.V. Lightner. 1995. Measurement using polymerase chain
reaction (PCR) of the survival of Infectious Hypodermal and Hematopoietic Necrosis Virus
(IHHNV) subjected to shrimp culture techniques. World Aquaculture Society. Baton Rouge, LA.
Book of Abstracts, Aquaculture,'95, 127.
Haskin, H.H., & J.D. Andrews. 1988. Uncertainties and speculations about the life cycle of the
eastern oyster pathogen Haplosporidium nelsoni (MSX). Amer. Fish. Soc. Spec. Publ, 75:5-22.
Joint Subcommittee on Aquaculture. 1997. An Evaluation of potential shrimp virus impacts
on cultured shrimp and on wild shrimp populations in the Gulf of Mexico and southeastern U.S.
Atlantic coastal waters. Washington, DC.
Kennedy, V.S. 1996. The ecological role of the eastern oyster, Crassostrea, with remarks on
disease. J. Shellfish Res., 75:177-183.
Laramore, C.R. 1997. Shrimp culture in Honduras following the Taura syndrome virus. IV
Central American Symposium on Aquaculture. Tegucigalpa, Honduras.
Linder, M.J., & W.W. Anderson. 1956. Growth, migration, spawning, and size distribution of
shrimp, Penaeus setiferus. Fisheries Bulletin., 106:555-645.
Lo et al. In Press. Specific genomic DNA fragment analysis of different geographical clinical
samples of shrimp white spot syndrome associated virus. Diseases of Aquatic Organisms.
McKenzie, M.D., ed. 1981. Profile of the penaeid shrimp fishery in the south Atlantic. South
Atlantic Fishery Management Council.
5-1
-------
Unestam, T., & D.W. Weiss. 1970. The host-parasite relationship between freshwater crayfish
and the crayfish disease fungus Aphanomyces astaci: Responses to infection by a susceptible and
resistant species. J. Gen. Microbiol., 60:77-90
U.S. EPA. 1996. Proposed guidelines for ecological risk assessment. Fed. Reg.,
6J-.47552-47631.
5-2
------- |