EPA 903-R-98-023
                                    CBP/TRS 216/98
                                    October 1998
  A Probabilistic Ecological Risk
  Assessment of Tributyltin in the
    Chesapeake Bay Watershed
     Chesapeake Bay Program
EPA Report Collection
Regional Center for Environmental Information
U.S. EPA Region III
Philadelphia, PA 19103
Printed on Recycled Paper for EPA by CBP

-------
                                   itr' [or Environmental
                                      19103
A Probabilistic Ecological Risk Assessment
                of Tributyltin in the
           Chesapeake Bay Watershed
                       October 1998
                    Chesapeake Bay Program
                   410 Severn Avenue, Suite 109
                   Annapolis, Maryland 21403
                      1-800-YOUR-BAY
              http://www.chesapeakebay.net/bayprogram
       Printed by the U.S. Environmental Protection Agency for the Chesapeake Bay Program

-------
                                   October 1998
                                   Final Report

A Probabilistic Ecological Risk Assessment of Tributyltin in the Chesapeake Bay Watershed
                               Lenwood W. Hall, Jr.
                                  Mark C. Scott
                                 William D. Killen

                              University of Maryland
                          Agricultural Experiment Station
                        Wye Research and Education Center
                                  P. O. Box 169
                           Queenstown, Maryland 21658

-------
                                      ABSTRACT




       The goal of this study was to conduct a  probabilistic ecological risk assessment for




tributyltin (TBT) in esruarine areas of the Chesapeake Bay watershed by using the following distinct




phases: problem formulation, analysis and risk characterization. This probabilistic ecological risk




assessment characterized risk by comparing the probability distributions of environmental saltwater




exposure concentrations with the probability distributions of species response data determined from




laboratory studies.  The overlap of these distributions was a measure of risk  to aquatic life.




Comparative  risk from TBT exposure was determined for various basins (tributaries) in  the




Chesapeake Bay watershed.




       Tributyltin saltwater exposure data from the Chesapeake Bay watershed were available from




over 3,600 water column samples from 41 stations in nine basins from 1985 through 1996. Most of




the stations were located in the Virginia waters of Chesapeake Bay, primarily the James, Elizabeth




and York Rivers. In Maryland waters of the Bay, various marina, harbor and river systems were also




sampled.  As expected, the highest environmental concentrations of tributyltin  (based on 90th




percentiles) were reported in and near marina areas. The  sources of TBT causing these high




concentrations were primarily boat hulls and painting/hull cleaning operations. Lower concentrations




of TBT were reported in open water areas ,such as the Potomac River, Choptank River and C and




D Canal, where the density of boats was minimal. Temporal data from a ten year data base (1986-




1996) from two areas in Virginia showed that TBT water column concentrations have declined since




1987 legislation prohibited the use of TBT paints on recreational boats (<25m).




      Acute saltwater and freshwater TBT toxicity data were available for 43 and 23 species,




respectively.  Acute effects for saltwater species were reported for concentrations exceeding 420

-------
ng/L; the lowest acute value for a freshwater species was 1,110 ng/L.  The acute 10th percentiles for




all saltwater and freshwater species were 320 and 103 ng/L, respectively. The order of sensitivity




from most to least sensitive for saltwater trophic groups and corresponding acute 10th percentiles




were as follows: zooplankton (5 ng/L), phytoplankton (124 ng/L), benthos (312 ng/L) and fish (1,009




ng/L).  For freshwater species, the order of sensitivity from most to least sensitive trophic groups and




corresponding acute 10th percentiles were: benthos (44 ng/L), zooplankton (400 ng/L), and fish (849




ng/L). Chronic data for both saltwater and freshwater species were limited to a few species in each




water type. Based on these limited data, the saltwater and freshwater chronic 10th percentiles were




5 and 102 ng/L, respectively. Limited mesocosm and microcosm studies in saltwater suggested that




TBT concentrations  less than  50 ng/L did not impact the structure and  function of biological




communities.




       The saltwater acute (320 ng/L) and chronic (5 ng/L) 10th percentiles were used to determine




potential ecological risk because all exposure data were from saltwater areas of the Chesapeake Bay




watershed. Highest ecological risk was reported for marina areas in Maryland waters of Chesapeake




Bay and for areas in Virginia such as the Elizabeth River, Hampton Creek and Sarah Creek. Low




ecological risk was reported for areas such as the Potomac River, Choptank River, C and D Canal




and Norfolk Harbor. Regulation of TBT on recreational watercraft in  1987 has successfully reduced




water column concentrations of this organometallic compound. However, various studies have shown




that TBT may remain in the sediment for  years and  continue to be source for  water column




exposures.

-------
                               ACKNOWLEDGMENTS









       We would like to acknowledge the U. S. Environmental Protection Agency's Chesapeake Bay




Program Office for funding this study through grant number CB993589010. The "Toxics of Concern




Workgroup" of EPA's Toxics Subcommittee is also acknowledged for their support. The following




individuals are acknowledged for providing data: Michael Unger (Virginia Institute of Marine




Science), Peter Seligman (U. S. Navy) and Aldis Valkirs (Computer Sciences Corporation).
                                         111

-------
                              TABLE OF CONTENTS


                                                                            Page
ABSTRACT	i

ACKNOWLEDGMENTS	iii

TABLE OF CONTENTS	iv

LIST OF TABLES	vi

LIST OF FIGURES	viii

1.  INTRODUCTION  	1
       1.1 Problem Formulation	3
             1.1.1 Stressor Characteristics	 3
             1.1.2 Ecosystems at Risk	4
             1.1.3 Ecological Effects	4
             1.1.4 Endpoints  	5
             1.1.5 Cumulative Stressors Potentially Impacting Aquatic Communities	6
             1.1.6 Conceptual Model	6

2. EXPOSURE CHARACTERIZATION	8
       2.1 Introduction	8
       2.2 Tributyltin Loadings in the Chesapeake Bay Watershed	8
       2.3 Chemical Properties of Tributyltin	9
       2.4 Measured Concentrations of Tributyltin in the Chesapeake Bay Watershed 	11
             2.4.1 Data Sources and Sampling Regimes	11
             2.4.2 Methods of Tributyltin Analysis	12
             2.4.3 Methods of Data Analysis	12
       2.5 Measured Concentrations by Basin	13
       2.6 Temporal Trends 	13
       2.7 Summary of Exposure Data	14

3. ECOLOGICAL EFFECTS  	15
       3.1 Mode of Toxicity	15
       3.2 Methods of Toxicity Data Analysis	15
       3.3 Effects of Tributyltin from Laboratory Toxicity Tests  	16
             3.3.1 Acute Toxicity of Tributyltin	17
             3.3.2 Chronic Toxicity of Tributyltin  	17
       3.4 Mesocosm/Microcosm Studies  	18
       3.5 Summary of Effects Data	19

                                       iv

-------
Table of Contents (continued)                                                    Page

4. RISK CHARACTERIZATION	20
      4.1 Characterizating Risks	20
      4.2 Risk Characterization of Tributyltin in the Chesapeake Bay Watershed	21
      4.3 Uncertainty in Ecological Risk Assessment  	22
            4.3.1 Uncertainty Associated with Exposure Characterization	22
            4.3.2 Uncertainty Associated with Ecological Effects Data 	23
            4.3.3 Uncertainty Associated with Risk Characterization	24

5. CONCLUSIONS AND RESEARCH NEEDS  	25

6. REFERENCES 	27

  TABLES 	39

  FIGURES	56

 APPENDICES

      Appendix A - Tributyltin risk characterization by basin and station

-------
                                  LIST OF TABLES




                                                                                 Page




Table 1. Summary of three tributyltin data sources used for this risk assessment	40




Table 2. Summary of tributyltin exposure data for all basins and stations.




       Maximum concentrations and 90th percentile values (minimum




       of four detected concentrations) are presented by station and basin	41




Table 3. Summary of TBT sample preparation procedures and analytical methods for




       the various monitoring studies	43




Table 4. Saltwater acute TBT toxicity data measured as TBT (ng/L). Concentrations marked




       with an asterik * were converted from reported compounds to TBT. The abbreviations




       used are: N=nominal, M=measured, S=static, R=renewal, FT=flowthrough,




       LC=life cycle, ELS=early life stage and NR=not reported	44




Table 5. The 10th percentile intercepts for freshwater and saltwater tributyltin toxicity




       data by test species and trophic group.  These values represent protection of




       90% of the test species	48




Table 6. Freshwater acute TBT toxicity data measured as TBT (ng/L).  Concentrations




       marked by an asterisk* were converted from reported compounds to TBT.




       The abbreviations used are: N=nominal, M=measured, S=static, R=renewal,




       FT=flowthrough, LC=life cycle, ELS=early life stage and NR=not reported	49




Table 7. Saltwater chronic TBT toxicity data measured as TBT (ng/L). Concentrations




       marked with an asterisk* were converted from reported compounds to TBT.




       The abbreviations used are: N=nominal, M=measured, S=static, R=renewal






                                         vi

-------
                                                                                Page




       FT=flowthrough, LC=life cycle, ELS=early life stage and NR=not reported	52




Table 8. Freshwater chronic TBT data measured as TBT (ng/L).Concentrations




       marked with an asterisk* were converted from reported compounds to TBT.




       The abbreviations used are: N=nominal, M=measured, S=static, R=renewal




       FT=flowthrough, LC=life cycle, ELS=early life stage and NR=not reported	53




Table 9. The percent probability of exceeding the TBT acute and chronic saltwater




       10th percentiles for all species	54
                                         Vll

-------
                                  LIST OF FIGURES




                                                                                  Page




Figure 1. Ecological risk assessment approach	57




Figure 2. Location of the 41 stations where TBT was measured from 1985 to 1996.




       See key to map where stations are described  	58




Figure 3. Temporal trend of 90th percentile concentrations of TBT for Sarah Creek




      from 1986 to 1996	60




Figure 4. Temporal trend of 90th percentile concentrations of TBT for Hampton Creek




      from 1986 to 1996.:	61




Figure 5. Distribution of TBT acute saltwater toxicity data	62




Figure 6. Distribution of TBT acute freshwater toxicity data	63




Figure 7. Distribution of TBT chronic saltwater toxicity data	64




Figure 8. Distribution of TBT chronic freshwater toxicity data	65
                                         Vlll

-------
                                      SECTION 1
                                   INTRODUCTION

       Improvement and maintenance of water quality were identified  as the most critical elements

for the restoration and protection of the Chesapeake Bay in the 1987 Chesapeake Bay Agreement

(Chesapeake Executive Council, 1988). Another goal of this Agreement was the development and

adoption of a Chesapeake Bay Basinwide Toxics Reduction Strategy in order to achieve a reduction

of toxic substances consistent with the Water Quality Act of 1987. The Chesapeake Bay Basinwide

Toxics Reduction Strategy contained various commitments in areas such as research, monitoring and

toxic substance management that were directed to overall reduction of toxic chemicals in the

Chesapeake Bay watershed (Chesapeake Bay Executive Council, 1988). One commitment specified

for the creation of a Toxics of Concern List (TOC) for the Chesapeake Bay. This TOC list was

designed to prioritize over 1000 chemicals that may be impacting aquatic life or human health in

Chesapeake Bay by using a risk based ranking system and  direct future research efforts and

management.

       The first TOC list was completed in 1990 and was revised in 1996 (U. S. EPA, 1991; U.  S.

EPA, 1996). The revised list as currently proposed is currently under review. The proposed revised

TOC list was developed using a chemical ranking system that incorporates sources, fate, exposure

and effects of chemicals on living resources and human health in Chesapeake Bay (Battelle, 1989).

The TOC list contains both a  list of primary toxics of concern as well as a secondary list (chemicals

of potential concern).  For both the 1990 and 1996 TOC lists, tributyltin (TBT) was identified as a

primary toxic of concern.  Tributyltin enters the aquatic environment primarily as an antifouling paint

additive used on boat hulls although  loading  from drydocks during painting and hull cleaning

-------
 operations can also occur. Antifouling paints containing TBT prevent the growth of fouling organisms



 such as tube worms and barnacles on boat hulls. Tributyltin is one of the most effective biocides ever



 used in antifouling  paints but unfortunately,  toxic qualities of TBT that make it effective in



 controlling fouling organisms (target species) also poses a risk to non-target species in the aquatic



 environment. Imposex (development of male characteristics in female snails) in female dog welks at



 low ng/L concentrations and shell thickening and reduced productivity of oysters at 100 to 500 ng/L



 concentrations have raised concern because these concentrations have been reported in the aquatic



 environment of the Chesapeake Bay watershed (Huggett et al., 1996).



        Although TBT has been identified as a toxic of concern in the Chesapeake Bay watershed,



 a quantitative probabilistic ecological risk assessment has not been conducted for this organometallic



 compound.  The objective of this study was to apply EPA's Ecological Risk Assessment paradigm



 for assessing ecological risk of TBT in the Chesapeake Bay watershed. Procedures described in the



 following documents were used for this assessment: Report of the Aquatic Risk Assessment and
                                                                                         i


 Mitigation Dialogue Group (SET AC, 1994), the EPA Framework for Ecological Risk Assessment



 (U.  S. EPA, 1992) and a recent paper entitled "An ecological risk assessment ofatrazine in North

-*,

 American surface waters" (Solomon et al., 1996). This probabilistic risk assessment characterizes



 risk by comparing probability distributions of environmental saltwater exposure concentrations with


 the  probability distributions of species response data (determined  from laboratory studies). The



 overlap of these distributions is a measure of potential risk to aquatic life in Chesapeake Bay. This



 approach has a number of advantages over a quotient method (comparing the most sensitive species



 with the highest environmental concentrations) because it allows, if not exact quantification, a least



 a strong sense for the magnitude and likelihood of potential ecosystem effects of TBT in Chesapeake

-------
Bay. An implied assumption of this approach is that protecting a large percentage of species will also




preserve ecosystem structure and function. Various studies in basic ecology (Tillman, 1996; Tillman




et al., 1996) and of ecological effects of pesticides conducted in aquatic mesocosms (Solomon et al.,




1996) support the concept that in ecological risk assessment, some effects can be allowed at the




population level provided that these do not impair ecosystem structure and function and keystone




species are not impacted. The final result of the risk characterization is expressed as the probability




that exposure concentrations of TBT (within a defined  spatial and temporal range) will exceed




concentrations deemed protective of aquatic life in the Chesapeake Bay watershed.









1.1 Problem Formulation




       This ecological risk assessment has the following distinct  phases: Problem Formulation,




Analysis and  Risk  Characterization (Figure 1).  The problem formulation phase  involves the




identification of major issues to be considered in the risk assessment. The analysis phase reviews




existing data on exposure (environmental monitoring of saltwater TBT concentrations) and ecological




effects ( primarily laboratory toxicity studies). The risk characterization phase involves estimation of




the probability of adverse effects on aquatic populations and communities in potentially impacted




areas of the Chesapeake Bay watershed.




       The problem formulation phase of this risk assessment identified the following  major issues




to be addressed: stressor characteristics, ecosystems at risk, ecological effects, endpoints,  stressors




impacting aquatic communities,  and a conceptual model  for risk assessment.




       1.1.1 Stressor Characteristics




       The chemical and physical properties of TBT are described in detail in the Exposure section

-------
of this report. In the problem formulation phase of this risk assessment, the solubility, degradation




persistence in water  and sediment, metabolism and  bioconcentration potential  of TBT  were




considered important.



       Solubility of TBT in the water column is influenced by such factors as the oxidation-reduction




potential, pH, temperature, ionic strength and concentration and composition of dissolved organic




matter (Clark et al., 1988). Maguire et al. (1983) has reported that TBT solubility ranges from 750




to 31,000  ug/L at pH of 2.6 to  8.1. Microbial degradation of TBT to the less toxic di- and




monobutyltin compounds has been reported as the most important process limiting pesistence of TBT




in the environment (U. S. Navy, 1997). Degradation half lives ranging from 4 to 19 days in seawater




have been reported while sediment half lives  on the order of months to years have been observed (U.




S. Navy, 1997, De Mora et al., 1989).  Significant TBT metabolism potential exists in fish and




crustaceans with minimal metabolic potential  in mollusks.   Mollusks also exhibit the highest




bioaccumulation factors and highest tissue burdens while fish and crustaceans generally accumulate




lower burdens of TBT.




       1.1.2 Ecosystems at Risk




       The aquatic ecosystem addressed in this risk assessment was the estuarine portion of the




Chesapeake Bay watershed.  Most of the exposure data for TBT  were reported for the Virginia




waters of Chesapeake Bay, primarily the James, Elizabeth and York Rivers. Various marina, harbor




and river systems were also sampled in the Maryland waters of Chesapeake Bay. Exposure data were




available for over 3,600 samples collected at 41 stations between 1985 to  1996.




       1.1.3 Ecological Effects




       A comprehensive review and synthesis of the TBT aquatic toxicity literature was conducted

-------
by using literature searches (AQUIRE etc.), various regulatory review documents such as the U. S.




EPA  water quality criteria report (U. S. EPA, 1997) and other TBT review documents (e. g. Hall




and Pinkney; 1985,  Champ and Seligman, 1996 and U. S. Navy 1997, among others). The ecological




effects data used in this risk assessment were derived from 43  saltwater species  tested in acute




studies and 4 species tested in chronic toxicity tests (where chronic values were reported). Only




saltwater toxicity data were used for assessing risk because exposure data were only available from




saltwater areas in the Chesapeake Bay watershed. The acute saltwater 10th percentile (protection




of 90% of the species) was 320 ng/L for all species. The 10th percentile by trophic group from most




sensitive to least sensitive was: zooplankton (5 ng/L), phytoplankton (124 ng/L), benthos (312 ng/L),




and fish (1,009 ng/L). A 10th percentile of 5 ng/L was determined from the limited saltwater chronic




data (same value reported for zooplankton tested in acute tests). Limited microcosm and mesocosm




studies showed that TBT concentrations less than 50 ug/L generally did not impact the structure and




function of biological communities.




       1.1.4 Endpoints




       The selection of appropriate endpoints is a basic element of the risk assessment process. In




ecological risk assessment, it is recognized that individual organisms are part of the food wed and are




therefore somewhat expendable as they are either consumed or being consumed. Therefore, the focus




of ecological risk assessment is the protection of population, community or ecosystem function, rather




than individuals. This acknowledges the fact that a population is less sensitive than its most sensitive




member and likewise that communities and ecoystems are less sensitive that their most sensitive




populations. A consensus of recent ecological risk assessments has lead to an important conclusion-




some effects  at the organism and population level can be allowed if these effects are restricted in

-------
space and time and keystone species are not impacted (Solomon et al., 1996; Giddings et al., 1997).




       The Framework for Ecological Risk Assessment has defined  two types of endpoints:




assessment endpoints and measurement endpoints (U. S. EPA, 1992). Assessment endpoints are the




actual environmental values that are to be protected (e.g. fish or shellfish populations).  Measurement




endpoints are the measured responses to a stressor that can be correlated with or used to protect




assessment endpoints (Suter, 1990). With each higher level of testing, measurement endpoints differ




while assessment endpoints remain the same.




       The assessment endpoints for this risk assessment are the long term viability of aquatic




communities in the Chesapeake Bay (fish, invertebrates etc.). Specifically, the protection of at least




90% of the species 90% of the time (10th percentile from species susceptibility distributions) from




acute TBT expsoures is the defined assessment endpoint. Measurement endpoints include all acute




TBT toxicity data (survival, growth and reproduction) generated from saltwater laboratory toxicity




studies.




       1.1.5 Cumulative Stressors Potentially Impacting Aquatic Communities




       When assessing the potential impact of TBT on aquatic communities in the Chesapeake Bay




watershed, it is important to remember that both biotic (food quality and quantity, disease) and abiotic




factors (water quality, other contaminants,  physical habitat alteration)  influence the status of




biological communities. As discussed above, individuals within the various biological communities




are more sensitive to contaminant stress than the community as a whole. Therefore, individual losses




due  to a stressor such as TBT may or may not impact the viability  (persistence, abundance,




distribution) of the population depending on all the factors influencing the population.




       1.1.6 Conceptual Model

-------
       Problem formulation is completed with the development of a conceptual model where a




preliminary analysis of the ecosystem at risk, stressor characteristics and ecological effects are used




to define the possible exposure and effects scenarios. The goal is to develop a working hypothesis




to determine how the stressor might affect exposed ecosystems. The conceptual model is based on




information about the ecosystem at risk and the relationship between the measurement and assessment




endpoints. Professional judgement is used in the selection of risk hypotheses.  The conceptual model




describes the approach that will be used for the analysis phase and the types of data and analytical




tools that will be needed. Specific data gaps and areas of uncertainty will be described later in this




report.




       The hypothesis considered in this risk assessment was:




•      TBT  may cause permanent reductions at the population and community level   for fish,




       benthos, zooplankton  or phytoplankton  in the Chesapeake Bay watershed and these




       reductions may adversely impact community structure and function.

-------
                                     SECTION 2
                          EXPOSURE CHARACTERIZATION

2.1 Introduction

       The potential for exposure of aquatic organisms to TBT is an important component of a

probabilistic ecological risk assessment. Exposure data are used in conjunction with effects data (see

next section) to conduct a risk characterization. The exposure analysis for TBT  considers use rates,

sources, loadings, chemical properties and spatial/temporal scale of measured concentrations (data

sources, sampling regimes, analytical methods and data analysis).



2.2 Tributyltin Loading in the Chesapeake Bay Watershed

       The major source of TBT to  Chesapeake Bay is from the use of antifouling paint on

watercraft hulls. Loading of TBT into the aquatic  environment from either industrial or sewage

treatment plant effluents was reported to be minimal (Huggett et al.,  1996). Based on this

information, h is not surprising that the highest concentrations of TBT have been measured in areas

with the greatest number of watercraft. Highest concentrations were generally found in marinas that

had a high density of boats painted with TBT and a low flushing rate.  Two types of watercraft are

generally considered when determining loading of TBT- recreational and  commercial. Estimates

conducted for the State of Virginia reported that ~ 70% of the TBT entering Virginia waters came

from recreational watercraft while ~ 27% was from large commercial vessels such as freighters and

tankers (Huggett et al., 1996). The  remaining 3%  came from  miscellaneous sources such as the

military.

       The logical focus for controlling TBT input to the Chesaepeake Bay's aquatic environment

-------
was to restrict use on recreational watercraft. With this consideration in mind, both Virginia and




Maryland passed legislation to restrict use of TBT on vessels <25m in length in 1987. Longer vessels




and aluminum craft hulls were exempted but could only use paints that release TBT at a rate of <5




ug cm"2 d"1. A number of other states followed the actions of Virginia and Maryland and federal




legislation was established in 1988.




2.3 Chemical Properties of Tributyltin




       Tributyltin compounds used in antifouling paints consist of a tin atom covalently bonded to




three butyl  moieties and an associated anion such as chloride  or oxide or copolymers such as




methacrylate/methyl methacrylate.  Copolymer paints allow manufacturers to formulate paints that




have better controlled leach rates in seawater (e. g. tributyltin methacrylate). Due to the hydrophobic




nature of the TBT antifouling coating, seawater interacts with the copolymer at the surface  which




initiates a hydrolysis reaction that cleaves TBT from the copolymer backbone and releases it into the




water.




       The most important process limiting the persistence of TBT in the aquatic environment is




microbial degradation to the less toxic dibutyltin (DBT) and monobutyltin (MET) compounds.




Degradation half-lives in various seawater experiments were reported to vary from 4 to 19 days (U.




S. Navy, 1997). Lee et al. (1989) also reported that phytoplankton were active in degrading TBT to




the less toxic DBT and MBT; both 2 and 6 day half-lives were reported. Photolysis and chemical




degradation were reported to be insignificant in the degradation  of TBT in seawater (U. S. Navy,




1997). Degradation of TBT in sediments is much slower than in the water. Various studies have




reported half-life values in sediments to range from months to years in anaerobic sediments (Stang




and  Seligman, 1986; De Mora  et al. 1989; Dawson et al.  1993). These data suggest that that

-------
sediments may remain a source for TBT after limiting water column concentrations through




regulation.




       Due to the hydrophobic nature of TBT it partitions rapidly to particulate material in the water




column and bottom sediments. Partitioning rates are dependent on TBT concentration, suspended




sediment, load, pH,  salinity and organic carbon (U. S. Navy, 1997). Langston and Pope (1995) have




reported a partitioning coefficent of 25,000 L/kg at a water column concentration of 10 ng/L (a




realistic environmental concentration). Various investigators have reported  that partitioning




coefficients can vary between 340 to 39,000 (Valkirs et al. 1986,1987; Harris et al. 1996).




       Laughlin (1996) has reported that TBT is accumulated by nearly all taxa that have been




evaluated. Mollusks were reported to have the highest bioaccumulation factors and highest tissue




burdens. Fish and crustaceans generally accumulate lower burdens of TBT because they possess the




active cytochrome  P-450 enzyme system that oxides TBT to less toxic components (Lee, 1996).




Bioaccumulation factors (BCFs) range from about 200 in some fish tissues to 100,000 in American




oysters and mussels (U. S. Navy, 1997).  Salazar et al. (1987) reported that bioconcentration factors




showed an inverse relationship to concentrations in the field, with lower exposures leading to higher




bioconcentration factors in bivalves.




       Although the potential for sediment-bound TBT to cause risk to sediment dwelling aquatic




biota exists, the focus of this risk assessment was an evaluation of risk to aquatic biota from




exposures to surface water concentrations. Probabilistic risk assessment techniques for assessing risk




of aquatic species to  sediment exposures is still developmental and contains a higher degree of




uncertainty than water column  exposures.  By using surface water concentrations in this risk




assessment, the results can be more closely related to regulatory issues such as the proposed U. S.





                                           10

-------
Environmental Protection Agency's chronic marine water quality criteria of 10 ng/L TBT ( U. S.




EPA, 1997).








2.4 Measured Concentrations of Tributyltin in the Chesapeake Bay Watershed




       2.4.1 Data Sources and Sampling Regimes




       Tributyltin exposure data (seawater measurements) were available from three primary data




sources from 1985 to 1996 for over 3,600 samples at 41 stations (Figure 2, Tables 1 and 2).




Approximately 92% of the measurements were from Virginia waters of Chesapeake Bay. The




remaining samples were measured from marinas, harbors and river systems in Maryland waters of the




Chesapeake Bay. The data sources are briefly described below:




Hall et alData (Hall et al. 1987, 1988, 1989r and 1992)




       These data were collected from 1985 to 1989. During the 1985-86 effort, samples were




collected monthly from July through June at eight stations located in four small and large marinas,




a large harbor, two major river systems and a heavily used shipping channel (Hall et al., 1987). For




the other three studies, samples were collected bi-weekly from June-September of 1986, 1988 and




1989 at six stations in or near marinas in Back Creek (Annapolis, MD) and one location in the Severn




River near the confluence of Back Creek and the Severn River (Hall et al., 1988, 1989, and 1992).




Navy Data (Valkirs et al.r 1995)




       Samples were collected during the summer of 1986 and/or quarterly from 1988 to 1992 at




12 stations in the Elizabeth River, 10 stations in Norfolk Harbor and one station in Hampton River




(see Figure 2 and Table 2).




VIMS Data (Unger, personal communication)






                                          11

-------
       These data were collected monthly at five stations in Hampton River and four stations in




Sarah Creek (York River basin) from 1986 through 1996 (See Figure 2 and Table 2).




       2.4.2 Methods of Tributyltin Analysis




       A summary of preparation procedures and analytical methods for the various TBT monitoring




studies is presented in Table 3. Detection limits were generally less than 5 ng/L for all studies (except




the Hall et al. 1987 - Hall et al 1985 data). In all cases, the water samples (grab samples) collected




for analysis were unfiltered and iced for preservation. The analytical method used  for all studies




except  the Navy monitoring was Gas Chromatograph - Flame Photometric Detector (GC-FPD)




(Unger et al., 1986). In all cases, TBT was speciated from the degradation products MET and DBT.




       2.4.3 Methods of Data Analysis




       Approaches for handling values below the detection limits include assigning these values as




zero, one-half the detection limit or the detection limit (MacBean and Rovers, 1984; Giddings et al.,




1997). For this risk assessment, TBT values below the detection limit were assumed to be log-




normally distributed.  The distribution of exposure data was calculated based on the measured values




and the concentrations of the non-detects were assumed to be distributed along a lower extension of




this distribution. For example, if 80 out of 100 were reported as non-detects, the 20 measured values




were assigned ranks from 81 to 100 and the frequency distribution was calculated from these 20




values.   For the very few cases where more  than one value was available at the same time and




station, the highest value was used in the frequency distribution.




       For data sets arranged by basin or station with four or more values above the detection limit,




log-normal distributions of exposure concentration were determined as follows. The observations in




each data set were ranked by concentration and for each observation the percentile ranking was






                                           12

-------
calculated as n/(N+l) where n is the rank sum of the observation and N is the total number of




observations including the non-detects. Percentile rankings were converted to probabilities and a




linear regression was performed using the logarithm of concentration as the independent variable and




normalized rank percentile as the dependent variable. Although non-detected observations were not




included in the regression analysis, they were included in the calculation of the observation ranks.




The 90th percentile concentrations (exceedence of a  given value only 10% of the time) were




calculated for sampling stations (or basins) based on the calculated log-normal concentration




distributions.









2.5 Measured Concentrations by Basin




       A summary of maximum concentrations and 90th percentiles of individual stations and pooled




stations by basin or drainage is presented in Table 2. Maximum TBT concentrations by basin ranged




from below detection limit in the Choptank River and C and D Canal to 1801 ng/L in the Back Creek




marina area in Maryland.  The 90th percentile values by basin for locations with at least 4 detected




concentrations, ranged from 4.1 ng/L in Norfolk Harbor to 387 ng/L in Pier 1 Marina in Maryland.




As expected, the highest 90th percentiles (138 to 387 ng/L) were reported in marina areas with high




densities of boats using TBT paints; much lower values were reported in open water areas such as




the Potomac and Choptank Rivers.









2.6 Temporal Trends




       Tributyltin saltwater monitoring data were available  from 1986 to 1996 in Sarah Creek and




Hampton Creek in Virginia to assess temporal trends in TBT (M. A Unger, personal communication).






                                            13

-------
These data showed a clear trend of decreasing concentrations in the water column after the 1987




legislation (effective in 1988) restricted the use of TBT on recreational watercraft in Chesapeake Bay




watershed (Figure 3 and 4). For Sarah Creek, 90th percentile values dropped from approximately




40 ng/L in 1987 to approximately 9 ng/L in 1996 (Figure 3). The reduction of 90th percentiles values




in Hampton Creek was from 160 ng/L in 1987 to approximately 15 ng/L in 1996 (Figure 4).








2.7 Summary of Exposure Data




       Highest environmental concentrations of TBT (based on 90th pereentiles) in the Chesapeake




Bay watershed were reported in and near marina areas. Sources of TBT responsible for these high




exposures were boat hulls and/or painting and hull cleaning operations.  As expected, lower




concentrations of TBT were reported in open water areas such as rivers where  the density of boats




was minimal.  Temporal trends analysed from a 10 year data base (1986 to 1996) in two areas in




Virginia  showed that TBT water column concentrations have  declined after 1987 legislation




prohibited the use of TBT paints on recreational watercraft (<25m). Due to the long half-life of TBT




concentrations in  sediment and  the equilibrium shift occurring with lower  water  column



concentration,  sediments will continue to be a source for TBT.
                                          14

-------
                                       SECTION 3
                                ECOLOGICAL EFFECTS

3.1 Modes of Toxicity

       Both aquatic plants and animals have enzyme systems that can metabolize TBT to less toxic

derivatives. Plants such as eel grass, diatoms and dinoflagellates have been shown to metabolize TBT

(Lee, 1996). Diatoms produce a series of hydroxylated derivatives and it is likely that the algal

dioxygenase  system is involved. For animal species, crustaceans, annelids and fish have enzyme

systems that rapidly metabolize TBT. The hydroxylation of TBT that occurs in aquatic vertebrates

and invertebrates is controlled by the microsomal cytochrome P-450 systems present in the hepatic,

intestinal and kidney tissues of these organisms. These hydroxylated derivatives are conjugated to

sulfate or carbohydrate by phase two enzyme systems, which facilitates the elimination of TBT (Lee,

1996). Mollusks have low cytochrome P-450 content and mixed function oxygenase activity and

therefore exhibit TBT accumulation with slow depuration rates. Various TBT effects in mollusks

include shell thickening, reduced growth rates and imposex (Champ and Seligman, 1996).



3.2 Methods of Toxicity Data Analysis

       The primary toxicity benchmark used for this risk assessment was the 10th percentile of

species sensitivity (protection of 90% of the species) from acute and chronic exposures. The implied

assumption when using this benchmark is that protecting a large percentage of the species assemblage

will preserve ecosystem structure and function. This level of species protection is not universally

accepted, especially if the unprotected 10% are keystone species and have commercial or recreational

significance. However, protection of 90% of the species 90% of the time (10th percentile) has been
                                           15

-------
recommended by the Society of Environmental Toxicology and Chemistry (SET AC,  1994) and others




(Solomon et al., 1996; Giddings et al. 1997). Recent mesocosm studies have reported that this level




of protection is conservative (Solomon et al.,1996; Giddings, 1992).




       Tributyltin toxicity data were analyzed as a distribution on the assumption that the data




represented the universe of species. An approximation was made since it is not possible to test all




species in the universe.  This approximation assumes that the number of species tested (N) is one less




than the number in the universe.  To obtain graphical distributions for smaller data sets that are




symmetrical (normal distributions) percentages were calculated from the formula (100 x n/(N + 1))




where n is the rank number of the datum point and N is the total number of data points in the data




set  (Parkhurst et al., 1994).  This formula compensates for the size of the data sets as small




(uncertain) data sets will give a flatter distribution with more chance of overlap than larger (more




certain) data sets. In cases where there were multiple data points for a given species, the lowest value




was used in the regression analysis of the distribution. When data were available for multiple life




stages of a species the lowest values were generally reported for early life stages. Using the lowest




value therefore provides a conservative approach for protecting the most sensitive life stage of a




species.  Data were plotted using Sigma Plot (Jandel Corporation, 1992) .








3.3 Effects of Tributyltin from Laboratory Toxicity Tests




       Acute and chronic TBT toxicity data used in this risk assessment were obtained  from the




AQUIRE database through 1995, U. S. EPA water quality criteria documents (U. S. EPA, 1997),




literature review documents (Hall and Pinkney, 1985; Hall and Bushong, 1996; Champ and Seligman,




1996) and manual  searches of grey literature from academia, industry and government sources.






                                           16

-------
Only data that met the various criteria established by the U. S. EPA for use in water quality criteria




development were used in the analysis (acceptable control survival, complete description of test




methods etc.). Tributyltin acute and chronic toxicity data by water type (saltwater and freshwater )




are  discussed below. However,  only the  saltwater  acute and chronic data were used for risk




characterization since all exposure data were from Chesapeake Bay saltwater environments.




       3.3.1 Acute Toxicity of Tributyltin




       Acute saltwater TBT toxicity data were available for 43 species, including five algal species,




five zooplanlcton species, 24 benthic species and nine fish species (Table 4, Figure 5) The range of




acute toxicity values was 420 ng/L for the mysid, Acanthomysis sculpta to 330,000 ng/L for an algal




species. The acute 10th percentile for all saltwater species was 320 ng/L (Table 5).  A breakdown of




10th percentiles by trophic group from most to least sensitive was as follows: zooplanlcton (5 ng/L),




phytoplankton (124 ng/L), benthos (312 ng/L) and fish (1,009 ng/L) (Table 5).




       Acute freshwater toxicity data were available for 23 species (Table 6, Figure 6). The data base




included three zooplanlcton species, five fish species and 11 benthic species. The range of acute




toxicity  for freshwater  species was 1,110  ng/L for  the hydra, Hydra  littoralis to greater than




114,000,000 ng/L for the adult clam, Elliptic complanata. The high value for the clam likely resulted




from shell closure during the acute exposure. The acute 10th percentile of all freshwater species was




103 ng/L (Table 5). The 10th percentiles by trophic  group from most to least sensitive were as




follows:  benthos (44 ng/L), zooplanlcton (400 ng/L) and fish (849 ng/L).




       3.3.2 Chronic Toxicity of Tributyltin




       Chronic saltwater TBT toxicity data, where chronic values were reported, were limited to four




invertebrate species (Table 7, Figure 7). These chronic values ranged from  14 to 131 ng/L. The 10th





                                            17

-------
percentile for these four saltwater species was 5 ng/L (Table 5).




       The freshwater chronic TBT toxicity data, where chronic values or no observed effect level




(NOEL) were reported, were also limited to only three species (Table 8, Figure 8). These chronic




values ranged from 137 to 253 ng/L for the cladoceran and fish species. The chronic 10th percentile




for the freshwater species was 102 ng/L (Table 5).









3.4 Mesocosm/Microcosm Studies




       Saltwater mesocosm and microcosm studies with TBT were limited (Henderson,  1985, 1986,




1988). A three month mesocosm study by Henderson (1985,1986) with TBT concentrations ranging




from 500 to 1,800 ng/L showed the following: (1) annelid worms, crustaceans and fish were




insensitive to the exposures; (2) larval stages of various animals species such as corals, anemones,




echinoderms and mollusks were most sensitive. In later three month microcosm studies with TBT




concentrations ranging from 40 to 2,500 ng/L, Henderson (1988)  reported the following: (1)




significant declines in pre-established fouling communities occurred  at 500 ng/L and higher, (2)




significant reductions in number of species and species diversity of larval forms and reductions in the




condition index of the American oyster occurred at 100 ng/L and (3) oyster condition index, species




diversity and mortality did not occur at concentrations of 40 ng/L. The conclusion from these studies




is that TBT concentrations in the environment should be less than 50 ng/L to avoid effects on aquatic




communities.




       Two microcosm studies with durations of 24 to 55 days were conducted in freshwater (




Delupis and Miniero, 1989; Miniero and Delupis, 1991). In both studies effects were immediate at




80,000 ng/L dose as Daphnia magna disappeared, ostracods increased and algal species increased






                                           18

-------
immediately and then gradually dissappeared. The lowest concentration (4,700 ng/L) to even suggest

an effect in these studies caused a temporary reduction in metobolism (oxyen consumption). This

suggested effect concentration of 4,700 ng/L is much higher that concentrations typically measured

in the environment.




3.5 Summary of Effects Data
               *
        Acute effects with saltwater species were generally reported at concentrations greater than

or equal to 420 ng/L. The 10th percentile for all species derived from the acute TBT saltwater

toxicity database was 320 ng/L. The order of sensitivity from most to least sensitive trophic group

and corresponding 10th percentiles were as follows: zooplankton (5 ng/L), phytoplankton (124 ng/L),

benthps (312 ng/L) and fish (1,009 ng/L). For freshwater acute TBT studies, effects were reported

at concentrations at or above 1,110 ng/L.  The acute freshwater 10th percentile for all species was

103 ng/L. The 10th percentiles by trophic group from most to least sensitive were as follows: benthos

(44 ng/L), zooplankton (400 ng/L) and fish (849 ng/L).

       Chronic data from both saltwater and freshwater studies were limited to a few species. The

saltwater and freshwater chronic 10th percentiles for all species were 5 and 102 ng/L, respectively.

Limited mesocosm/microcosm studies in saltwater demonstrated that TBT concentrations less than

50 ng/L generally did not impact the structure and of biological communities. Freshwater microcosm

studies suggested that concentrations as high as 4,700 ng/L only caused temporary reductions in

biological community metabolism.
                                           19

-------
                                      SECTION 4
                              RISK CHARACTERIZATION

4.1 Characterizating Risks

       The report of the Aquatic Risk Assessment Dialogue Group (SETAC, 1994) recommends

using tiers when assessing the risk of pesticides in the aquatic environment. The first tier is a

simple and commonly used  risk quotient. Risk quotients are simple ratios of exposure and effects

concentrations where the susceptibility of the most sensitive species is compared with the highest

environmental exposures. If the exposure concentration equals or exceeds the effects concentration

an ecological risk is suspected. The quotient method is a valuable first tier assessment that allows a

determination of a worst case effects and exposure scenario for a particular contaminant. However,

some of the major limitations of the quotient method for ecological risk assessment are that it fails

to consider variability of exposures among individuals in a population, ranges of sensitivity among

species in the aquatic ecosystem and the ecological function of these individual species. The quotient

method also assumes that that there is a 100% probability of co-occurrence of the stressor and the

most sensitive  organism and that the most sensitive organism is a keystone organism in the

environment. The probabilistic approach addresses these various concerns as it expresses the results

of an exposure or effects characterization  as a distribution of values rather than a single point

estimate.  Quantitative expressions of risks to aquatic communities are therefore determined by using

all relevant single species toxicity data in conjunction with exposure distributions.  A detailed

presentation of the principles used in a probabilistic ecological risk assessment are presented by

Solomon et al. (1996) and Hall et al. (in press).

       The following section will summarize the results of the risk characterization phase of this
                                           20

-------
probabilistic ecological risk assessment of TBT in the Chesapeake Bay watershed. The toxicity




benchmarks used for the risk characterization will be both the saltwater acute and chronic  10th




percentiles for all species since all exposure data were reported in saltwater environments. Both the




acute 10th percentile for all species (320 ng/L) and the chronic 10th percentile for all species (5 ng/L)




are similar to the proposed U. S. EPA acute (370 ng/L) and chronic (10 ng/L) TBT criteria (U. S.




EPA, 1997). It is also important to note that the chronic 10th percentile of 5 ng/L also equals the




acute 10th  percentile  of the most sensitive  trophic group (zooplankton) resulting from acute




laboratory toxcity tests.  Therefore, the most sensitive trophic group (zooplankton) will be protected




in this risk characterization.









4.2 Risk Characterization of Tributyltin in the Chesapeake Bay Watershed




       Potential ecological risk from TBT exposure  was characterized by using both acute and




chronic saltwater effects data (10th percentiles for all species) since all exposure data were collected




in saltwater  areas of the Chesapeake Bay. Using the acute 10th percentile as a toxicity benchmark




generally results in low risk for all areas sampled (Table 9). The greatest risk (12% exceedence) at




Back Creek marina in the Severn River would still be considered in the low risk range.  The use of




the chronic benchmark of 5 ng/L results in some significant ecological  risk  in all marinas and




Baltimore harbor (> 97% exceedence), Hampton River (73% exceedence), Sarah Creek (52%




exceedence)  and the Elizabeth River (33% exceedence). These data suggest that TBT may be posing




a risk to aquatic biota in specific areas of Chesapeake Bay associated with boating activity. The low




ecological risk reported in the Potomac River, Choptank River, C and D Canal and Norfolk Harbor




using the chronic 10th percentile indicates that potential ecologial risk is not present throughout the






                                           21

-------
entire Chesapeake Bay.









4.3. Uncertainty in Ecological Risk Assessment




       Uncertainty plays a particularly important role in ecological risk assessment as it impacts




problem formulation, analysis of exposure and effects data and risk characterization. Uncertainty in




ecological risk assessment has three basic sources: (1) lack of knowledge in areas that should be




known; (2) systematic errors resulting from human or analytical error and (3) non-systematic errors




resulting from the random nature of the ecosystem ( e.g. Chesapeake Bay watershed). The following




sections will address specific uncertainty from the above three sources as associated with exposure




data, effects data and risk characterization.




       4.3.1 Uncertainty Associated with Exposure Characterization




       The tributyltin exposure data used for this risk assessment were obtained from three different




data sources from 1985 to 1996 as described in Section 2.  The spatial scale of these  data (41




stations in 10 basins, primarily in Virginia) was somewhat limited considering that there are at least




50 major rivers that discharge into the Chesapeake Bay and numerous marinas where exposures are




unknown. Extensive exposure data from basins in Maryland  waters of Chesapeake Bay were




particularly limited on a temporal scale. Uncertainty also existed because the exposure data used in




this risk assessment were not collected from random sampling designs compatible with unbiased




statistical analysis of frequency distributions. In fact, these data were biased toward high values




because many of the stations were located in areas where concentrations were expected to be high




(near marinas and harbors) and samples were collected during  the high use season (summer) for




antifouling coatings such as TBT.





                                           22

-------
       Analytical techniques differed among three the laboratories - a GC-FPD method was used




by Hall  et al and VIMS while the Navy used an AAS technique. Detection limits among the




laboratories were less than 5 ng/L for all measurements except the Hall et al 1985 data (< 20 ng/L).




Sample preparation techniques were consistent among all the laboratories;  therefore, this area of




uncertainty was somewhat reduced.




       4.3.2 Uncertainty Associated with Ecological Effects Data




       Due to the relatively small number of species that can be routinely  cultured and tested in




laboratory toxicity studies, there is uncertainty when extrapolating these toxicity data to responses




of natural taxa found in the Chesapeake Bay watershed. In the case of TBT in the Chesapeake Bay




watershed, saltwater acute and chronic toxicity data were available for 43 and 4 species,  respectively,




for use in the calculation of the 10th percentile. Although the  acute data provide a reasonable




representation of species in the Bay environment,  the chronic data were limited and  should be




expanded to reduce uncertainty in this risk assessment.




       Variability in the results of toxicity tests for a given species tested in different experiments or




by different authors is a potential source of random and systematic errors. In this assessment, the




most conservative (lowest) effect value was used when multiple data points were available for a given




species. The range of toxicity data among trophic groups differed for the acute saltwater toxicity data




as the 10th percentiles ranged from 5 ng/L for zooplankton to 1,005  ng/L for fish. Using the




distribution of susceptibility accounts for this range of data points.  Distributions will be flatter,  with




greater chance of overlap with exposure distributions, when the range is large.




       Acute  and chronic saltwater TBT  toxicity data were used in the risk characterization as




previously discussed.  The use of acute (and chronic) data for predicting ecosystem effects is often






                                            23

-------
questioned and assumed to be an area of significant uncertainty. However, Slooffet al. (1986) in




their review of single species and ecosystem toxicity for various chemical compounds, have reported




that there is no solid evidence that predictions of ecosystem level effects from acute tests are




unreliable. The result of Slooffet al. (1986) coupled with the use of a distribution of acute toxicity




data reduces some of the uncertainty associated with using acute data.




       4.3.3 Uncertainty Associated with Risk Characterization




       Many  of the uncertainties associated with the variability in the exposure and  effects




characterizations discussed above are incorporated in the probabilistic approach used in this risk




assessment  (SETAC,  1994).  Quantitative estimation of risks are analyzed as a distribution of




exposure and effects data.




       Ecological uncertainty includes   the effects of confounding  stressors such as other




contaminants and the ecological redundancy of the functions of affected species. In the Chesapeake




Bay watershed, numerous contaminants other than TBT may be present simultaneously in the same




aquatic habitats near marinas and harbors; therefore,  "joint toxicity" may occur. The concurrent




presence of various contaminants along with TBT makes it difficult to determine the risk of TBT in




isolation.




       Ecological redundancy is known to occur in aquatic systems. Field studies have shown that




resistant taxa tend to replace more sensitive species under stressful environmental conditions




(Solomon et al., 1996; Giddings et al., 1992)  The resistant species may replace the sensitive species




if it is functionally equivalent in the aquatic ecosystem and the impact on overall ecosystem function




is reduced by these species shifts. For this risk assessment, information on the ecological interactions




among species would help to reduce this area of uncertainty.





                                            24

-------
                                      SECTION 5
                       CONCLUSIONS AND RESEARCH NEEDS
       Potential ecological risk from TBT exposure was reported for various areas in Chesapeake

Bay that are in close proximity to boating or shipping activity. Highest risk was reported for marinas

in Maryland  but significant potential ecological risk was also reported for areas in Virginia such as

the Elizabeth River, Hampton Creek, and Sarah Creek. Low ecological risk was reported for areas

such as the Potomac River, Choptank River, C and D Canal and Norfolk Harbor. Temporal exposure

data has demonstrated that TBT concentrations in the water column have declined since 1987

regulation of this compound on recreational watercraft in Chesapeake Bay. Although these declining

water column concentrations are beneficial in reducing risks to aquatic biota, various studies have

showed that the TBT in the sediment may last for years and continue to remain a source for TBT

exposure in the water column.  The use of TBT by commercial watercraft (>25m) and associated

painting and hull cleaning activities also continues to contribute to TBT loading in Chesapeake Bay.

       The  following research is recommended  to  supplement existing data for  assessing the

ecological risks of TBT in the Chesapeake Bay watershed:



(1) Post-regulation exposure assessments for TBT in water and sediment for Sarah Creek and

Hampton Creek in Virginia should be continued as a long term monitoring effort since a 10 year data

base has already been established in these areas. These data will allow an analysis of TBT trends.

Similar type post-regulation monitoring activities should also be conducted in the Severn River/Back

Creek areas of Maryland where exposure data have also been collected.
                                          25

-------
(2) Chronic TBT toxicity experiments are recommended for various keystone Chesapeake Bay




species (sensitive bivalve species)  to determine the lowest observed effect concentrations (LOEC)




for these valuable species. In the current chronic saltwater toxicity data base, chronic values were




only available for four species.








(3) Biological communities  such as fish and benthos (community metric approaches) should be




evaluated in areas where potential ecological risk of TBT is reported to be the greatest (near marinas




and harbors). Imposex in gastropods should also be assessed in these areas. These data would provide




a validation step for this ecological risk assessment.
                                           26

-------
                                      SECTION 6
                                    REFERENCES

Alabaster, J.S. 1969. Survival offish in 164 herbicides, insecticides, fimgicides, wetting agents and

       miscellaneous substances.  Int Pest Control 11:29-35.

Battelle.  1989.   Implementation of a  chemical ranking system.   Draft report to  the  U.S.

       Environmental Protection Agency Criteria and Standards Division, Washington, DC.

Brix, K.V., P.P. Sweeney and R.D. Cardwell.  1994. Procedures for Conducting Acute Toxicity

       Tests  Using the Echinoderm Sperm Cell Test  to  Determine the Acute  Toxicity  of

       Bis(Tributyltin) Oxide. Elf Atochem North America, Inc. Philadelphia, Pennsylvania.

Brooke, L.T., DJ. Call, S.H. Poirer,  T.P. Markee, C.A. Lindberg, DJ. McCauley and P.O.

       Simonsoa 1986a. Acute Toxicity and Chronic Effects of Bis(tri-n-Butyltin) Oxide to Several

       Species of Freshwater Organisms.   Report to Battelle Memorial Research  Institute,

       Columbus, Ohio, 20 pp.

Brooke, L.T., DJ. Call, S.H. Poirer,  T.P. Markee, C.A. Lindberg, DJ. McCauley and P.O.

       Simonsoa 1986b. Acute Toxicity and Chronic Effects of Bis(tri-n-Butyltin) Oxide to Several

       Species of Freshwater Organisms.   Center  for Lake  Superior Environmental Studies,

       University of Wisconsin-Superior, Superior, WI. 20 p.

Buccafusco, R_, 1976b. Acute Toxicity of Tri-n-butyhin oxide to channel catfish Ictaluruspunctatus,

       the freshwater clam Elliptic complanatus, the common mummichog Fundulus heteroclitus

       and the American oyster Crassostrea virginica. US EPA-OPP Registration Standard.

Buccafiasco, R., 1976a. Acute Toxicity of Tri-n-butyhin oxide to bluegill Lepomis macrochirus. U.S.

       EPA-OPP Registration Standard.
                                          27

-------
Bushong, S.J., L.W. Hall, Jr., W.S. Hall, W.E. Johnson and R. L. Herman. 1988. Acute toxicity of




       tributyltin to selected Chesapeake Bay fish and invertebrates.  Water Res., 22:1027-1032.




Bushong, S.J., M.C. Ziegenfuss, M.A. Unger and L.W. Hall, Jr. 1990. Chronic tributyltin toxicity




       experiments with the Chesapeake Bay copepod Acartia tonsa.  Environ. Toxicol. Chem.




       9:359-366.




Champ, M. A and P. F. Seligman. 1996. Organotin Environmental Fate and Effects. Chapman and




       Hall, London. 623 pp.




Chesapeake Bay Executive Council.   1988.  Chesapeake Bay Toxics Reduction Strategy.




       Chesapeake Bay Agreement Report. Chesapeake Bay Liaison, Annapolis, MD.




Clark, J.R., J.M. Patrick, Jr., J.C. Moore and E.M. Lores. 1987. Waterborne and sediment source




       toxicities of six organic chemicals to grass shrimp Palaemonetes pugio and amphioxus




       Branchiostoma caribaewn.  Arch. Environ. Con. Tox., 16:401-407.




Clark, E. A.,  K M. Sterritt and J. N. Lester. 1988. The fate of tributyltin in aquatic environment.




       Environ. Sci. Tech. 22: 600-604.




Cramm, G. 1979. Acute and Chronic Toxicity of Tributyltin oxide (TBTO) to Sheepshead Minnows




       (Cyprinodon variegatus).  U.S. EPA-OPP Registration Standard




Das, V.G.K., L.Y. Kuan, K.I. Suderuddin, C.K. Chang, V. Thomas, C.K. Yap, M.K. Lo, C.C. Ong.




       1984. The Toxic Effects of Triorganotin (IV) Compounds on the Culicine Mosquito Aedes




       Aegypti (L). Toxicology 32:57-66.




Davidson, B.M., AO. Valkirs and P.F. Seligman. 1986a. Acute and chronic effects of tributyltin on




       the mysid Acanthomysis sculpta (Crustacea, Mysidacea). NOSC-TR-1116 or AD-A175-294-




       8. National Technical Information Service, Springfield, VA






                                         28

-------
Davidson, B.M., A.O. Valkirs and P.F. Seligman. 1986b\ Acute and chronic effects of tributyltin on




       the mysid Acanihomysis sculpta (Crustacea, Mysidacea). In: Oceans 86, Vol 4. Proceedings




       International Organotin Symposium. Marine Technology Society, Washington, DC. pp. 1219-




       1225.




Dawson, P. H. et al. 1993. Depositional profiles and relationships between organotin compounds in




       freshwater and estuarine sediment cores. Environ. Monitor. Assess. 28: 145-161.




Delupis, G. D. D. and R. Miniero. 1989. Preliminary studies on the TBTO effects on freshwater biotic




       communities. Riv. Idrobiol. 28: 1-2.




De Mora, S. J. et al. 1989. Tributyltin and total tin in marine sediments: profiles and the apparent rate




       of degradation. Environ. Tech. 10: 901-908.




Floch, H. R. Deschiens and T. Floch. 1964. Sur les proprietes molluscicides de F oxyde et de F




       acetate de tributyl-etain (prophylaxie des bilharzioses). Bull. Soc. Pathol. Exotique, 57:454-




       465.




Foster, R.B. 1981. Use of Asiatic Clam Larvae in Aquatic Hazard Evaluation. In: Ecological




       Assessments of Effluent Impacts on Communities of Indigenous Aquatic Organisms, J.M.




       Bates and C.I. Weber (Eds).  American Society for Testing and Materials, Philadelphia,




       pp.280-288.




Giddings, J. M., L. W. Hall, Jr., K. R. Solomon, W. Adams, D. Vogel, L. Davis and R. Smith. 1997.




       An ecological risk assessment of diazinon in the  Sacramento and San Joaquin basins.




       Technical Report 11/97. Novartis Crop Protection, Greensboro, NC.




Giddings, J.M.   1992.  Aquatic mesocosm test for environmental data and ecological effects of




       diazinon. Report. No.  92-3-4155, Springborn Laboratories, Inc., Wareham, MA.






                                           29

-------
Goodman, L.R., G.M. Cripe, P.H. Moody and D. G. Halsell. 1988. Acute toxicity of malathion,




       tetrabromobisphenol-A and tributyltin chloride to mysids (Mysidopsis bahia) of three ages.




       Bull. Environ. Contam. Toxicol. 41:746-753.




Hall, L.W. Jr. and S.J. Bushong. 1996. A Review of Acuute Effects of Tributyltin Compounds on




       Aquatic Biota In: Organotin Environmental Fate and Effects M. A Champ and P.P. Seligman




       (Eds.) Chapman & Hall,  London, pp. 157-190.




Hall, L.W. Jr.,  S.J. Bushong, W.S. Hall and W.E. Johnson. 1988. Acute and chronic effects of




       tributyltin on a Chesapeake bay copepod.  Environ. Toxicol.. Chem. 7:41-46.




Hall, L. W. Jr., S. J. Bushong, W. D. Johnson, and W. S. Hall.  1988. Spatial and temporal distribution




       of butyltin compounds in a northern Chesapeake Bay marina and river system. Environ.




       Monitor. Assess. 10: 229-244.




Hall. L. W. Jr., M. J. Lenkevich, W. S. Hall, A. E. Pinkney and S. J. Bushong. 1987. Evaluation of




       butyltin compounds in Maryland waters of Chesapeake Bay. Mar. Pollut. Bull. 18: 78-83.




Hall, L.W. Jr., and A.E. Pinkney.  1985. Acute and Sublethal Effects of Organotin Compounds on




       Aquatic Biota: An Interpretive Literature Evaluation. CRC Critical Reviews in Toxicology




       14:159-209.




Hall, L. W., M. C. Scott and W. A. Killen. in press. An ecological risk assessment of copper and




       cadmium in surface waters of the Chesapeake Bay watershed. Environ. Tox. Chem.




Hall, L. W. Jr.,  M. A. linger, S. J. Bushong and M. C. Ziegenfuss. 1989. Butyltin monitoring in




       Northern Chesapeake Bay marina and river system in 1988: An assessment of tributyltin




       legislation. Report. The Johns Hopkins University, Applied Physics Laboratory, Aquatic




       Ecology Section, Shady  Side, MD.






                                          30

-------
Hall, L. W. Jr., M. A. Unger, M. C. Ziegenfuss, J. A. Sullivan and S. J. Bushong. 1992. Butyltin and




       copper monitoring in a northern Chesapeake Bay marina and river system in 1989:  An




       assessment of tributyltin legislation. Environmental Monitoring and Assessment 22: 15-38.




Harris, J. R. W., J. J. Cleary and A. 0. Valkirs. 1996. Particle water partitioning and the role of




       sediments as a sink and secondary source of TBT. In: Organotin Environmental Fate and




       Effects, M. A. Champ and P. F. Seligman (eds.) Chapman and Hall, London, pp 459-473.




Henderson, R S. 1985. Effects of tributyltin antifouling paint leachates on Peral Harbor organisms.




       Naval Ocean Systems Center Technical Report #1079, San Deigo, CA.




Henderson, R. S. 1986. Effects of organotin paint leachates on Pearl Harbor organisms:  A site




       specific flowthrough bioassay. In: Proceeding, Oceans 86 Conference - Organtin Symposium,




       Washington, D. C. pp 1126-1233.




Henderson, R. S. 1988. Marine micrososm experiments on effects of copper and tributyltin-based




       antifouling paint leachates. Naval Oceans System Center Technical Report #1060, San Deigo,




       CA.




Ho, S.K. 1984. Evaluation of Carbon-14 Uptake Algal Toxicity Assay and its Application in Field




       Assessment of Tributyltin Chloride and Chlorinated Sewage Toxicities. Masters Thesis, The




       College of William and Mary, Williamsburg, Virginia, 108pp.




Hooftman, R.N., D.M.N. Adema and J, Kauffinan-Van Bommel. 1989. Developing a Set of Test




       Methods for the Toxicological Analysis of the Pollution Degree of Waterbottoms Rep. No.




       16105, Netherlands Organization for Applied Scientific Research:68p. (DUT)




Huggett, R. J., D. A. Evans, W.  G. Maclntyre, M. Unger, P. F. Seligman and L. W. Hall, Jr. 1996.




       Tributyltin concentrations in waters of Chesapeake Bay. In: Organotin Environmental Fate





                                          31

-------
       and Effects. M. A Champ and P. F. Seligman (Eds.) Chapman and Hall, London, pp. 485-501.




Jandel Corporation. 1992. SigmaPlot Scientific Graphing System, Version 5. San Rafael, CA.




Khan, A.T., J.S. Weis, C.E. Saharig  and A.E. Polo.  1993. Effect of Tributyltin on Mortality and




       Telson Regeneration of Grass Shrimp, Palaemonetespugio.  Bull. Environ. Contam Toxicol.




       50:152-157.




Langston, W. J. andN. D. Pope. 1995. Determinants of TBT adsorption and desorption in estuarine




       sediments. Mar. Pollut Bull. 31: 32-43.




Lapota, D., D.E. Rosenberger, M.F. Platter-Rieger and P.P. Seligman. 1993. Growth and Survival




       of Mytilus edulis larvae  exposed to  low levels of dibutyltin and tributyltin. Mar. Biol.




       115:413-419.




Laughlin, R. B. 1996. Bioaccumlation of TBT by aquatic organisms. In: Organotin Environmental




       Fate and Effects. M. A. Champ and P. F. Seligman (eds), Chapman and Hall, London, pp 331-




       355.




Laughlin, R.B., O. Linden  and  H.E. Guard. 1982.  Acute toxicity of tributyltins and tributyltin




       leachates from marine antifouling paints.  Bull. Liaison Comite Int. Permanent Rech.




       Preservation Matoriaux Milieu Mar. (COIPM), 13:3-20.




Leblanc,  G., 1976.  Acute Toxicity of Tributyltin  oxide to Daphnia magna  U.S. EPA-OPP




       Registration Standard.




Lee, R. F., A O. Valkirs and P. F. Seligman. 1989. Importance of microalgae in the biodegradation




       of tributyltin in esturaine  waters. Environ. Sci. Technol. 23: 1515-1518.




Lee, R. F. 1996. Metabolsim of tributyltin by aquatic organisms. In: Organotin Environemental Fate




       and Effects. M. A Champ and P. F. Seligman (eds), Chapman and Hall, London, pp 369- 382.






                                           32

-------
MacBean, E. A. and F. A. Rovers. 1984. Alternatives for handling detection limit data in impact




       assessments. Groundwater Monitoring Review, Spring 1984.




Maguire, R. J., J. H. Carey and E. J. Hale. 1983. Degradation of the tri-n-butylitn species in water.




       J. Agric. Food Chem. 31: 1060-1065.




M & T Chemicals Company. 1981. Comparative Toxicity of Tri-n-butyltin Oxide TBTO) Produced




       by two Different Chemical Processes to Pink Shrimp Penaeus duorarum. Report No. BP-81-




       4-55 submitted by EG & G Bionomics to M & T Chemicals Co., Rahway, New Jersey.




Matthijssen- Spiekman, E.A.M., J.H. Canton and C.J. Roghair. 1989. Research after the Toxicity of




       TBTO for a Number of Freshwater Organisms  Rep. No.  668118-00\ Natl. List. Public




       Health and Environ. Hyg.:48p. (DUT).




Meador, J.P. 1993. The Effect of Laboratory Holding on the Toxicity Response of Marine Infaunal




       Amphipods to Cadmium and Tributyltin. J. Exp. Mar. Biol. Ecol. 174:227-242.




Meador, J.P., U.  Varanasi  and C.A. Krone. 1993.  Differential Sensitivity of Marine Infaunal




       Amphipods to Tributyltin. Mar. Biol. 116:231-239.




Miniero, R. and G. D. D. Delupis. 1991. Effects of TBT (tributyltin) on aquatic microcosms. Toxicol.




       Environ. Chem. 31: 425-431.




Naval Oceans Systems Center (NOSC),  1981. unpublished data, San Diego, CA.




Parkhurst, B.R., W. Warren-Hicks, T. Etchison, J.B. Butcher, R.D. Cardwell and J. Volison.  1994.




       Methodology for aquatic risk assessment. Draft final report RP91-AER-1 prepared for the




       Water Environment Research Foundation, Alexandria, VA.




Pinkney, A.E., D.A. Wright and G.M.  Hughes.  1989. A morphometric study of the effects of




       tributyltin compounds on the gills of the mummichog Fundulus heteroclitus.  J. Fish. Biol.






                                         33

-------
       34:665-677.




Robert, R. and E. His. 1981. Action de L'acetate de Tributyl-etain sur les oeufs et les larves de deux




       mollusques d'interet commercial: Crassostrea gigas (Thunberg) et Mytilus gaUoprovincialis




       (Lamarck). Internal.  Council Explor. Sea Paper, CM1981:F41, 16 pp.




Roberts, M.H. Jr. 1987, Acute toxicity of tributyltin chloride to embryos and larvae of two bivalve




       molluscs, Crassostrea virginica and Mercenaria mercenaria.  Bull,  Environ.  Contam.




       Toxicol., 39:1012-1019.




Salazar, S. M., B. M. Davidson, M. H. Salazar, P. M. Stang and K. Meyers-Schulte. 1987. Effects




       of TBT  on marine organisms: Field assessment of a new site-specific bioassay system. In:




       Proceedings, Oceans 1987 - Organotin Symposium. Halifax, Nova Scotia, Canada, pp 1461-




       1470.




Salazar, M.H. and S.M. Salazar. 1989. Acute effects of (Bis)tributyltin Oxide on Marine Organisms,




       Summary of Work Performed 1981-1983. Draft Report, Naval Ocean Systems Center, San




       Diego, California, 88 pp.




Seinen, W., T. Helder, H. Vermij, A. Penninks and P. Leeuwangh. 1981. Short-term toxicity of tri-n-




       butyltinchloride in rainbow trout (Salmo gairdneri Richardson) yolk sac fry.  Sci. Total




       Environ., 19:155-166.




Short,  J.W.  and P.P.  Thrower. 1986. Tri-n-butyltin  caused mortality of Chinook  Salmon,




       Oncorhynchus tshawytscha, on transfer to a TBT treated marine net pen. In: Proceedings of




       the Qceans '86 Organotin Symposium, Vol. 4, Washington DC. pp. 1201-1205.




Sloof, W., J.A.M. van Oers and  D. De Zwart.  1986.  Margins of uncertainty in ecotoxicological




       hazard assessment. Environ. Toxicol. Chem. 5:841-852






                                          34

-------
Smith, P.J., A.J. Crowe, V.G. Kumar Das and J. Duncan. 1979. Structure-activity relationships for




       some organotin molluscicides. Pesticide Sci., 10:419-422.




Snell, T.W., B.D. Moffat, C. Janssen and G. Persoone. 1991a. Acute Toxicity Tests Using Rotifers




       IV. Effects of Cyst Age,  Temperature and Salinity  on the Sensitivity of Brachionus




       cafyciflorus.  Ecotoxicol. Environ. Saf. 21:308-317.




Snell, T.W., B.D. Moffat, C. Janssen and G. Persoone. 1991b. Acute Toxicity Tests Using Rotifers




       III. Effects of Temperature Strain and Exposure Time on the Sensitivity of Brachionus




       plicatitis. Environ. Toxicol. Water Qual. 6:63-75.




Society of Environmental Toxicology and Chemistry (SETAC).  1994.  Aquatic Risk Assessment




       and Mitigation Dialogue Group.   Final Report.  SETAC Foundation for Environmental




       Education, Inc., Pensacola, FL




Solomon, K.R. et al.  1996.  Ecological risk assessment of atrazine in North American surface




       waters. Environ. Tox. Chem. 15:31-76.




Stang, P. M. and P. F. Seligman. 1986. Distribution and fate of butyltin compounds in the sediment




       of San Diego Bay. In: Ocean 86 Conference Record:  Science-Engineering-Adventure.




       Volume 4 Organotin Symposium. Oceans 86. Washington, D. C., pp 1256-1261.




Stephan, C.E., D. I. Mount, D. J. Hansen, J. H. Gentile, G. A. Chapman and W. A. Brungs. 1985.




       Guidelines for deriving numerical national water quality criteria for the protection of aquatic




       organisms and their uses. Report # PB85-227049, National Technical Information Service,




       Springfield, VA.




Suter, G.W. n.  1990. Endpoints for regional ecological risk assessment.  Environ. Mange. 14:19-




       23.






                                          35

-------
TAI Environmental Sciences Inc. 1989a. Toxicity of Tri-butyl tin oxide to two species of Hydra.




       Mobile, AL, 15 August. 31pp.




TAI Environmental Sciences Inc. 1989b. Toxicity of Tri-butyl tin oxide to Hydra littoralis and




       Chlorohydra viridissima. Mobile, AL, 13 September. 25pp.




Thain, J.E. 1983. The Acute Toxicity of Bis (Tributyltin) Oxide to the Adults and Larvae of Some




       Marine Organisms, Intdernational Council for the Exploration of the Sea, Copenhagen, Paper




       CM1983/E:13, 5pp.




Tillman, D. 1996. Biodiversity: Population versus ecosystem stability. Ecology 77: 350-363.




Tillman, D., D. Wedlin and J. Knops. 1996. Productivity  and sustainability influenced by biodiversity




       in grassland ecoystems. Nature 379: 718-720.




Unger, M. A. Personal communication. Raw data from VIMS TBT monitoring.




Unger, M A., W. G. Maclntyre, J. Greaves and R. J. Huggett. 1986. GC determination of butyltins




       in natural waters by flame photometric detection of hexyl derivatives with mass spectrometric




       confirmation. Chemosphere 15: 461-470.




ITren, S.C. 1983. Acute toxicity of bis(tributyltin)oxide to a marine copepod. Marine Pollut. Bull.,




       14:303-306.




U.S. Environmental Protection Agency (U.S. EPA). 1991. Chesapeake Bay Toxics of Concern List




       Information Sheets. Report prepared by the Chesapeake Bay Programs Toxics Subcommittee




       Living Resources Subcommittee's Joint Criteria and Standards Workgroup. Annapolis, MD.




U.S.  Environmental Protection Agency (U.S. EPA).  1992.   Framework for ecological risk




       assessment.  Risk Assessment Forum, EPA 630/R92/001.  Washington, DC.




U.S. Environmental Protection Agency (U.S. EPA). 19%. Chesapeake Bay Toxics of Concern List.





                                          36

-------
       Prepared by  the  Chesapeake Bay Program Toxics Subcommittee's Living Resources



       Subcommittee's Joint Criteria and Standards Workgroup.  Annapolis, MD.



U.S. Environmental Protection Agency. 1997. Ambient Aquatic Life Water Quality Criteria for



       Tributyltin.   USEPA  Office of Research and  Development,  Environmental Research



       Laboratories. Duluth, Minnesota Narragansett, Rhode Island



U. S. Navy. 1997. Navy program to monitor ecological effects of organotin. Report to congress.



       Naval Command, Control and Ocean Surveillance Center, San Diego, CA.



Valkirs, A. O. et al., 1995. U. S Navy monitoring of tributyltin in selected U.  S harbors. Final



       report/March 1995, Naval Command, Control and Ocean Surveillance Center, San Diego,



       CA.



Valkirs, A. O., P. F. Seligman, and R. F. Lee. 1986. Butyhin partitioning in marine waters and
                                                      j


       sediment. In: Proceedings, Oceans 86 Conference - Qrganotin Symposium, Washington, D.



       C., pp 1165-1170.



Valkirs, A. O., M. O. Stallard and P. F. Seligman. 1987. Butyltin partitioning in marine waters. In:



       Proceedings, Ocean 86 Conference - Organotin Symposium, Nova Scotia, Canada, pp 1375-



       1380.



Walsh,  G.E., L.L. McLaughlan, E.M. Lores, M.K. Louie and C.H. Deans. 1985. Effects of organotins



       on growth and survival of two marine diatoms, Skeletonema costatum and Thalassiosira



       pseudonana.  Chemosphere,  14:383-392.



Walsh, G.E., M.K. Louie, L.L. McLaughlan and E.M. Lores. 1986. Lugworm Arenicola cristata



       larvae in toxicity tests: survival and development when exposed to organotins.  Environ.



       Toxicol. Chem., 5:749-754.




                                          37

-------
Walsh,  G.E., L.L. McLaughlan, N.J. Yoder, P.H. Moody, E.M.  Lores, J. Forestet and P.B.




       Wessinger-Duval. 1988. Minutocelluspolymorphus; A new marine diatom for use in algal




       toxicity tests. Environ. Toxicol Chem. 7:925-929.




Webbe, G. and R.F. Sturrock. 1964. Laboratory tests of some new molluscicides in Tanganyika.




       Ann. Trop. Med. Parasitol., 58:234-239.




Wong, P.T.S., Y.K. Chau, O. Kramer and G.A Bengert. 1982. Structure-toxicity relationship of tin




       compounds on algae. Can. J. Fish. Aquat. Sci. 39:483-488.
                                          38

-------
TABLES
   39

-------
  Table 1. Summary of three data sources used for this risk assessment.
Reference
Data ID       Total #    Sample Period
              samples
Detection Limit
    (ng/L)
Hall etal., 1987

Hall etal., 1988

Hall etal., 1989

Hall etal., 1992

VaDcirs etal., 1995

MA. linger, pers. comm.
Hall Data 85        %   monthly July 1985-June 1986

Hall Data 86        63   June-September 1986

Hall Data 88        63   June-September 1988

Hall Data 89        63   June-September 1989

Navy            1,027   summer 1986; quarterly 1988-1992

VIMS            2,307   monthly 1986-1996
             20

              5

              2

              2

            0.2

              1
                                                40

-------
Table 2.  Summary of TBT exposure data for all basins and stations. Maximum concentrations and 90th percentile values
(minimum of four detected concentrations) are presented by station and basin.
Drainage
Data ID
James Basin
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
TBT Concentration (ng/L)
Station
.Elizabeth River
Elizabeth River Station 1 5
Elizabeth River Station 17A
Elizabeth River Station 19
Elizabeth River Station 2 1
Elizabeth River Station 32
Elizabeth River Station 13 A
Elizabeth River Station 1 1
Elizabeth River Station 10
Lafayette River Station 37
Naval Station 9
Naval Station 4
Naval Station 3
Elizabeth River all stations combined
James Basin
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
James mainstem/Norfolk Harbor
Hampton Roads Station 29
Hampton Roads Station 35
Hampton Roads Station 23
Hampton Roads Station 3A
Hampton Roads Station 34
Hampton Roads Station 1
Hampton Roads Station 25
Hampton Roads Station 25 A
Hampton Roads Station 25B
Hampton Roads Station 36
Norfolk Harbor all stations combined
James Basin
VMS
VMS
NAVY
VMS
VMS
VMS
. Hampton River
OPC
HRM2
Station 33
HRM1
HYC
CD
Hampton River all stations combined
York Basin.
VMS
VMS
VMS
VMS
Sarah Creek
Potomac
Hall Data
Choptank
Hall Data
West
Sarah Creek
A
B
C
D
all stations combined

Potomac River

Choptank River
Hartge Marina
# Samples

7
10
82
10
85
7
84
82
6
79
80
22
611

79
4
5
74
3
79
78
14
14
A
354

258
258
86
257
256
251
1372

256
255
256
254
1021

12

12
12
# Detections

4
5
80
8
84
6
79
78
0
72
74
22
562

71
0
0
70
0
70
69
13
13
SL
306

252
257
84
257
256
257
1363

251
255
253
122
881

2

0
10
Maximum

8.9
10.7
41
13.4
29.4
9.8
45.4
14.3
BLD
8.9
9.8
7-2
45.4

9.8
BLD
BLD
10.7
BLD
6.1
5.1
5.3
3.2
BLD
10.7

42
1,300
38
180
340
25
1,300

120
72
23
16
120

24

BLD
186
90th percentile

11.1
10.2
19.7
28.4
18.6
12.8
16.0
10.0
-
6.4
5.7
5.5
14.2

4.9
-
-
3.3
-
4.5
5.4
4.5
3.0
_;
4.1

7.2
337
10
41
54
26
77

27
31
14
16.
23

-

-
138
                                                        41

-------
Drainage
Data ID
Severn
Mid-Bay
Chester
Patapsco
C&D Canal
TBT Concentration (ng/L)
Station
six Back Creek stations
Severn River
Pier 1 Marina
Piney Narrows Marina
Baltimore Harbor
C&D Canal
# Samples
174
27
12
12
12
12
# Detections
174
27
10
11
8
0
Maximum
1,801
89
307
338
112
BLD
90* percentile
351
53
387
354
129
.
42

-------
Table 3. Summary of TBT sample preparation procedures and analytical methods for the various monitoring
studies.
Database ID Detection Limits
ng/L
Hall Data 85
Hall Data 86
Hall Data 88
Hall Data 89
Navy
VIMS
20
5
2
2
0.2
1
Filtered/
Unfiltered
unfiltered
unfiltered
unfiltered
unfiltered
unfiltered
unfiltered
Preservation Sample Type Analytical Method
iced
iced
iced
iced
iced
iced
grab
grab
grab
grab
grab
grab
GC-FPD
GC-FPD
GC-FPD
GC-FPD
AAS
GC-FPD
                                              43

-------
.c."... 3
Refe
1
o
«r>
3
        0)
        I
        en
        O
        O.

        £
        •o
        .8
        «
        1/3

        •s
        
                                           C
                                        O\ 00
                   •"-en •*= O
                   ••a- t-~ «s o
                   CS ON t~ —
I
^3

t




                                      S
                                  I
                                  .s;
                                  S
                      ,1
                   ,1

                   i!
                   Si
,

H
S3
                     i
I
                                    <<
it
It

I
^

i

i
                     *
                                l
,

l!
S3

                         I

                            44

-------
        00
        oo
        ON
 «



 I
T3
 O
             00
             00
             ON

                        oo
                        ON
 
O
_)
i  §^
        •** o
        vo —
        ON ~
                                                                       O *
                                                                    vO ON
                                                                    ON <—'
                                                                               Os
                                                                               O
                                                                               O
                                                                               O
                                                                               O
                                                                               WN
                                                                               O
                                                                                u,  e
                                                                               •c oo
                                                                               o i^-
                                                                               CS '"^
                                                                               — 1
i §?
1 '§
S »
Sheepshead
Cyprinodon

1

§
li
'•M
co U



a
1
a> §
•3 "o
CO Co

1
•S
Polychaete
Arenicola cr
§
!
?
Alga
Prorocentrui
bebouriae

-------
                                                                      00
                                                                      ON
         oo
         oo
         Os
                  00
                  oo
                  ON
                                                                               00
                                                                               oo
                                                                               ON
         .8
         CO
                                                    00
                                                    ON
                                    I/I

                                   ffl
                                                                                                ro
                                                                                                oo
                                                             00
                                                             ON

£«
o»o
«^iJ
t?U
*1  &
                                            o»
                                   vo«0    
-------
00
ON
•fi
O

                          ON
                          ON
                          •a
                          t

                          1
                          CO
                                  00
                                  oo
                                  ON
                                  •a
                                                                     •S
                                                                     1
        s
        J  0
vo r-    o  t-
ON «••    «—'  ON
                                                    VO ON
                                                    ON -*
        V)
                          CO
                                     i
Amphioxus
Branchiostoma caribi
Amphipod
Rhepoxynius abroniu:
Amphipod
Orchestia traskiana
Poacher
Agonus catophractus
Mummichog
Fundulus heteroclitui
Clam
Protothaca staminea
European Oyster
Ostrea edulis
Rotifer
Brachionus plicatilis
Alga
Minutocellus polymot
                47

-------
Table 5.  The 10* percentile intercepts for freshwater and saltwater tributyltin toxicity data by test
duration and trophic group. These values represent protection of 90% of the test species.
Water type
Test type
Trophic Group
n   10th Percentile (ng/L)
Freshwater



Freshwater
Saltwater




Saltwater
acute



chronic
acute




chronic
All species
zooplankton
benthos
fish
All species
All species
phytoplankton/algae
zooplankton
benthos
fish
All species
23
3
11
5
3
43
5
5
24
9
4
103
400
44
849
102
320
124
5
312
1,009
5
                                            48

-------

1

0)

c
8
1
•c
W5 _tf
§^
CA
^^ 11
^•5 •
C/3 flj •*-*
.8 e g
•+3 it .,
2 JL "
s ^
e co *o
 13
^J ^^
•o •§ «JJ
^ « "^
"S £ "
^5 * nJ
£ 1? 4
CO M W)
r^ H3
a SI
"3 -a tj
rt 3 O
Table 6. Freshwater
from reported compo
R= renewal, FT= fl







1
fi
fl)
<*:
^




O yj
VTi ^^
r^ OQ
« c
h^ ^^^


1
s
5s
o
0.

s
T3
^
«
S




CA
1
Q.
V)
t/f
-J c >-l c
fa ^^ fa V}
•^ ^^ *^ ^5
vO ^^ vO C^
ON "-1 ON •— i










CO^ p^
S" Z


^
g a
Hydra
Hydra oligactis
Three spine Sticklebi
Gasterostelts aculeai
of
4>
O
V
^ "3 -^ —
ON r ** ** y>
00 _, VO °°
v^5 v^ ^ v^o \^s *«•'
r*** ^^ £ (T\ •*. ^*^ ^
C .fa ON u" -g "* «=
e £ CON-« a) ^ g
"S °5o'"« § o °
•J ac H£ * °*
siSiSiaiai^iai
i—  RJO g
in ijiiiHiij
^q w a; ffi O on c^ £ a, j S 2u
49

-------
                                                               vO
00
ON
"a
 O
 2
CQ
         00
         ON
00
Os
          O
          8
         CQ
 O
 S
CQ
 K)


ON




 I



!
ON
r-
ON
CO
                                             .o
                                             vO
                                             00
Brooke
^
o
3
to
•o
o
o
2
CO
•o

tN
CO
ON
*~^
"73

cs
CO
ON
"""^
~3
1
                           A
                           CU


                          *
                                                                       1
                                                                                          00
                                                                                          ON

                                                      1
                                              a
                                              F-M
                                              ON
                                              ON
                                                                                                   c
                                                                                co
                                                                           TT        IT


                                                                            »   <5  w
ON ro   ON
                                                      (S ~   <* —
                                                                           «n
                                                                           VO
                                                                           cs
         CO

         2
CO

z
                                             CO
CO

z"
                                                              CO
                                                                       CO
                                                                                CO
                                                                                         CO
                                                                                                  co
l

is macrochims
Blueg

Lepo
                                       !
                                       1
                                       I
Snail
Biomp
                     i   i
                                             2  d,
                                             cr
                                             CO
1
s:
-§
Snail
Biomphalaria su>



Snail
Bulinus nasutus


-------
\o
ON
•s
o
      oo
      ON
en

O
             ON
             u
             en
O
o


ffl
         *



otc4  ° g*
V> 00  »/-> C
U c  Up
VO —

ON ~
      £ 00

      ^> 2
      ON ^
VI
             c/a
         III
         ,3  U &
 of
 a^s
         a
             •SS
              0) £3
                                      51

-------
§
S
            U
  '   flr
s--
o
I
en
a

            CA
            4>
            CO
       o
       a
                    txo

                   1
                    V)

                   ffl
                           m
                           ON
                           ON
                                  OO
                                  OO
                                  ON
                      1
                      *
                           •o  c
                                  a
edu
                                    i
ra
Cope

Acart
Musse

Myti
Cope

Eury
                                         oo
                                         ON
 i
T3
1
                             t
                             r^   \*t w   c** c**
       •p  -c      ^  "5 S      S

       I-S   -SJ3   II  -ol
Mysid
                                            52

-------
               «
               o
               g
60
                V
               •i
Chronic

                o
                o.
•o
J
3  s
                u
                a.
                CO
         a)
        VO
        00
        ON
                        1
                        8
                        CQ
                                 oo
                  4>


                  I
                  •5
                  t/3
                                          of
                                          ^O
                                          CO
15
3
 o
 2
m
                        O
                                 UH
                      •c    I
o

r
ales
d

h
                  £3
Fathea

Pi
imep
                                                       53

-------
Table 9.  The percent probability of exceeding the TBT acute and chronic saltwater 10th percentiles for all species.
Drainage
James Basin. Elizabeth












Elizabeth River
Station
River
Elizabeth River Station 15
Elizabeth River Station 17A
Elizabeth River Station 1 9
Elizabeth River Station 21
Elizabeth River Station 32
Elizabeth River Station 13A
Elizabeth River Station 1 1
Elizabeth River Station 10
Lafayette River Station 37
Naval Station 9
Naval Station 4
Naval Station 3
all stations combined
acute saltwater tests -
all species (320 ng/L)

0.01
O.01
O.01
0.12
O.01
O.01
O.01
O.01
-
- O.01
O.01
<0.01
<0.01
chronic saltwater
tests (5 ng/L)

30
74
61
49
70
60
50
35
-
15
13
12
33
James Basin. Jai^ies jnainstem/Norfolk Harbor










Norfolk Harbor
James Basin. Hampton






Hampton River
Hampton Roads Station 29
Hampton Roads Station 35
Hampton Roads Station 23
Hampton Roads Station 3A
Hampton Roads Station 34
Hampton Roads Station 1
Hampton Roads Station 25
Hampton Roads Station 25A
Hampton Roads Station 25B
Hampton Roads Station 36
all stations combined
River
OPC
HRM2
Station 33
HRM1
HYC
m
all stations combined
<0.01
-
-
O.01
-
<0.01
O.01
0.01
O.01
-
O.01

O.01
12.0
O.01
0.02
0.13
0.01
1.12
10
-
-
3
-
8
11
8
3
.
7

26
99
26
83
85
8Q
73
York Basin. Sarah Creek




Sarah Creek
Potomac
Choptank River
West
A
B
C
U
all stations combined
Potomac River
Choptank River
Hartge Marina
O.01
O.01
O.01
0.01
0.08
-
-
0.99
69
88
59
12
52
-
-
97
                                                           54

-------
Drainage
Severn
Mid-Bay
Chester
Patapsco
C&D Canal
Station
six Back Creek stations
Severn River
Pier 1 Marina
Piney Narrows Marina
Baltimore Harbor
C&D Canal
acute saltwater tests
- all species
(320 ng/L)
12.0
0.01
10.0
8.40
0.12
m
chronic saltwater
tests (5 ng/L)
>99
97
97
99
>99
—
55

-------
FIGURES
   56

-------
Figure 1. Ecological risk assessment approach




                PROBLEM FORMULATION

  •     Stressor Characteristics

  •     Ecosystems at Risk

  •     Endpoints

  •     Cumulative Stressors Impacting Aquatic Communities

  •     Conceptual Model
                          I
                        ANALYSIS

  Characterization of Exposure: Water column monitoring data on tributyltin
  in the Chesapeake Bay watershed.

  Characterization of Ecological Effects: Laboratory toxicity studies.
               RISK CHARACTERIZATION

        Probabilistic comparison of species sensitivity and

        surface water exposure distributions.
                           57

-------
Figure 2.  Location of 41 stations where TBT was measured from
          1985 to 1996.  See  key to map where stations are
          ^J *\ r* *% •**« V^ yt 4*4
          described.
                        BALTIMORE
                                                  41
                                          9
                                                         N
                                  1-8
                            58

-------
Figure!, continued.
Station number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
Drainage Basin
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
James
York
York
York
York
Potomac
Choptank
West
Severn
Severn
Mid-Bay Mainstem
Chester
Patapsco
C&D Canal
Dataset
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
NAVY
VIMS
VIMS
NAVY
VIMS
VIMS
VIMS
VIMS
VIMS
VIMS
VIMS
Hall Data
Hall Data
Hall Data
Hall Data
Hall Data
Hall Data
Hall Data
Hall Data
Hall Data
Station
Elizabeth River station 15
Elizabeth River station 17A
Elizabeth River station 19
Elizabeth River station 21
Elizabeth River station 32
Elizabeth River station 13A
Elizabeth River station 1 1
Elizabeth River station 10
Lafayette River station 37
Naval Station 9
Naval Station 4
Naval Station 3
Hampton Roads station 29
Hampton Roads station 35
Hampton Roads station 23
Hampton Roads station 3A
Hampton Roads station 34
Hampton Roads station 1
James River station 25
James River station 25A
James River station 25B
James River station 36
Hampton River station OPC
Hampton River station HRM2
Hampton River station 33
Hampton River station HRM1
Hampton River station HYC
Hampton River station CD
Sarah Creek station D
Sarah Creek station A
Sarah Creek station C
Sarah Creek station B
Potomac River
Choptank River
Hartge Marina
Back Creek (6 stations)
Severn River
Pier 1 Marina
Piney Narrows Marina
Baltimore Harbor
C&D Canal
Latitude
36.7755
36.7973
36.8031
36.8216
36.8323
36.8398
36.8472
36.8700
36.9065
36.9159
36.9491
36.9611
36.9453
36.9610
36.9613
36.9781
36.9849
36.9928
36.9988
37.0051
37.0142
37.0242
37.0005
37.0163
37.0164
37.0170
37.0205
37.0228
37.2458
37.2553
37.2597
37.2628
Longitude
76.2953
76.2938
76.2949
76.2920
76.2961
76.2757
76.3000
76.3295
76.3072
76.3416
76.3343
76.3322
76.3913
76.4365
76.4108
76.3497
76.3750
76.3017
76.4744
76.4925
76.4785
76.5252
76.3138
76.3442
76.3411
76.3417
76.3442
76.3438
76.5005
76.4797
76.4675
76.4850
see map for remaining locations
















                           59

-------
  Figure 3.  Temporal trend of 90th percentile concentrations of
            TBT for Sarah Creek from 1986-1996.
              Sarah Creek 90th percentiles by year
CD


0
8
8
40
   20 -
        1986 1987  1988 1989 1990 1991 1992  1993  1994 1995 1996

                             Year
                             60

-------
Figure 4.  Temporal trend of 90th percentile concentrations of
          TBT for Hampton Creek from 1986-1996.
           Hampton Creek 90th percentiles by year
      1986 1987 1988 1989  1990  1991  1992 1993 1994 1995 1996
                           Year
                            61

-------
 Figure 5.  Distribution of TBT acute saltwater toxicity data.
                     TBT - saltwater acute effects
   99 -r
   90 -
I 70
.*
i 50
or
I
(D
0- 10
    1 -_
      10°
1Q1     1Q2     103     104    10s    106

            TBT concentration (ng/L)
107
108
   n = 43
   b[0] = -4.3203754973
   b[1] = 1.2130417922
                                    •  benthos
                                    o  fish
                                    v  phytoplankton
                                    v  zooplankton
                                   	regression line
                                  62

-------
Figure 6.  Distribution of TBT acute freshwater toxicity data.
                  TBT - freshwater acute effects
 99 -f
   102
103
10«       105       106

TBT concentration (ng/L)
10s
 n = 23
 b[0] =-2.583918915
 b[1 ] = 0.6472125449
                                  •  amphibians
                                  O  cyanobacteria
                                  T  benthos
                                  v  fish
                                  •  phytoplankton
                                  a  zooplankton
                                 	regression line
                                63

-------
Figure 7.  Distribution of TBT chronic saltwater toxicity data.
                    TBT - saltwater chronic effects
   99 H
~ 90'

'to
® 70 -

i 50 -
cn

-------
 Figure 8.  Distribution of TBT chronic freshwater toxicity data.
                   TBT - freshwater chronic effects
   99 -f
   90 -
70 -
50 -
'55
®
CO

«
a:
I
Q- 10 -
    1 -I
       10
                             100

                     TBT concentration (ng/L)
1000
   n = 3
   b[0] =-11.3749917057
   b[11 = 5.0255763015
                                            •  zooplankton
                                            O  fish
                                           	regression line
                                 65

-------
                APPENDIX A




Tributyltin risk characterization by basin and station

-------
                           Elizabeth River basin
CO


§
CO
o
Q.
X
0)
T3

-------
                              James River
                             Norfolk Harbor
   99.9 -f
CO
O
CL
X
0)

•o
0)
o

-------
                         Hampton River drainage
§
CO
O
Q.
X
0)
•o

Q.
99.9 -


 99 -




 90 -


 70 -

 50 -

 30 -


 10 -



   1 -
       0.1
            Probability of
            exceedence = 73%
Probability of
exceedence
= 1%
                                           i	1—i—n-
                   1              10             100

                      TBT Concentration (ng/L)
    1000
                                   A-3

-------
co

2
D
CO
O
O.
X

                                                           0)
                                                           0.
                                                                    o

                                                                    
-------
                             Hartge Marina
    99
CO

9?
3
CO
O
Q.
X
0
T3
Q)
^


fi
M—
O
0
Q.
90 -




70 -


50 -


30 -




10 -
Probability of
exceedence = 97%
                                                           Probability of
                                                           exceedence
                                                           = 0.99%
                             10                  100

                             TBT Concentration (ng/L)
                                                                1000
                                      A-5

-------
                           Six Back Creek Stations
CO

2
13
CO
o
Q.
X
99%
Probability of
exceedence =12%
                       10
                              100
                               TBT Concentration (ng/L)
    1000
                                        A.-6

-------
                               Severn River
   99 -f
o
Q.
X
0)
T3
0)
   90 -
    70 -
C  50 -
(0


*0  30
0)
 c
 
-------
                                    Pier 1 Marina
CO
o
Q.
X
(D
•D
0)
J^
I
(D
Q.
                                                     Probability of
                                                     exceedence= 10%
Probability of
exceedence = 97%
                              10                  100

                              TBT Concentration (ng/L)
                                                             1000
                                        A-8

-------
                        Piney Narrows Marina
    99
to
o
Q.
X
(D



-------
                               Baltimore Harbor
    99
to


§
CO
o
Q.
T3
(U
    90 -
    70 -
to  50 -

*o
J  30 H



I
0  m
n    iU
         Probability of

         exceedence = >99%
                             'o

                             8.
                             M

                             "co
                             _»

                             1
                             u
                             'c
                             I

                             u
                                                              (A
                                                              _g>
                                                              "Q


                                                              8.
                                                              tf)


                                                              "5
                                                              0)
                                                              o
                                                              0)

                                                              •s
                                                              u
                                                              CD
                                 10                 100


                               TBT Concentration (ng/L)
                                                                       1000
                                    A-10

-------