HARMFUL AQUATIC ORGANISMS IN BALLAST WATER …...The complete dossier will be made available to the experts of the GESAMP-BWWG with the understanding ... 8.1.2 Chemical storage and

I:\MEPC\61\2-9.doc

E

MARINE ENVIRONMENT PROTECTION COMMITTEE 61st session Agenda item 2

MEPC 61/2/928 March 2010

Original: ENGLISH

HARMFUL AQUATIC ORGANISMS IN BALLAST WATER

Application for Final Approval of the Severn Trent De Nora BalPure®

Ballast Water Management System

Submitted by Germany

SUMMARY

Executive summary: This document contains the non-confidential information related to the application for Final Approval of the Severn Trent De Nora (STDN) BalPure® Ballast Water Management System under the "Procedure for approval of ballast water management systems that make use of Active Substances (G9)" adopted by resolution MEPC.169(57)

Strategic direction: 7.1

High-level action: 7.1.2

Planned output: 7.1.2.5

Action to be taken: Paragraph 5

Related documents: BWM/CONF/36; MEPC 57/21; BWM.2/Circ.13 and BWM.2/Circ.24

Introduction 1 Regulation D-3.2 of the International Convention for the Control and Management of Ships' Ballast Water and Sediments stipulates that ballast water management systems that make use of Active Substances to comply with the Convention shall be approved by the Organization. 2 The Procedure for approval of ballast water management systems that make use of Active Substances (G9) stipulates the required information (MEPC 57/21, annex 1, paragraph 4.2.1) and provisions for risk characterization and analysis (MEPC 57/21, annex 1, section 5), which, according to section 6 of Procedure (G9), should be evaluated by the Organization. 3 Basic Approval of the Severn Trent De Nora (STDN) BalPure® Ballast Water Management System was granted at MEPC 60.

MEPC 61/2/9 Page 2

I:\MEPC\61\2-9.doc

4 The receiving competent authority in Germany has verified the application dossier and believes it to satisfy the data requirements of Procedure (G9) adopted by resolution MEPC.169(57). In accordance with BWM.2/Circ.24, Germany therefore submits the non-confidential part of the manufacturer's application dossier in the annex. The complete dossier will be made available to the experts of the GESAMP-BWWG with the understanding of confidential treatment. Action requested of the Committee 5 The Committee is invited to consider the proposal for approval and decide as appropriate.

***

MEPC 61/2/9 Annex, page 1

I:\MEPC\61\2-9.doc

ANNEX

NON-CONFIDENTIAL INFORMATION ON THE SEVERN TRENT DE NORA BALPURE® BALLAST WATER MANAGEMENT SYSTEM

CONTENTS LIST OF ABBREVIATIONS 1 INTRODUCTION 1.1 Overview of the BalPure® technology 1.2 Response to the GESAMP-BWWG's comments on the application for Basic

Approval 2 IDENTIFICATION OF THE ACTIVE SUBSTANCES AND RELEVANT CHEMICALS

(G9: 4.1) 2.1 Introduction to the chemical basis of BalPure® Ballast Water Management System 2.2 Active Substances 2.2.1 Hypochlorous acid and hypobromous acid 2.3 Relevant Chemicals 2.3.1 Trihalomethanes (THMs) and haloacetic acids (HAAs) 2.3.2 2,4,6-Tribromophenol 2.3.3 Bromate, chlorate and monobromoacetonitrile 2.3.4 Hydrogen gas 2.4 Other chemicals 2.4.1 Sodium bisulfite and sodium bisulfate 3 DATA ON EFFECTS ON AQUATIC PLANTS, INVERTEBRATES AND FISH, AND

OTHER BIOTA, INCLUDING SENSITIVE AND REPRESENTATIVE ORGANISMS (G9: 4.2.1.1)

3.1 Acute aquatic ecotoxicity 3.2 Chronic aquatic ecotoxicity 3.3 Endocrine disruption 3.4 Sediment toxicity 3.5 Bioavailability/biomagnification/bioconcentration 3.6 Food web/population effects 4 DATA ON MAMMALIAN TOXICITY (G9: 4.2.1.2) 4.1 Acute mammalian toxicity 4.2 Effects on skin and eye 4.3 Repeated-dose toxicity 4.4 Chronic mammalian toxicity 4.5 Developmental and reproductive toxicity 4.6 Carcinogenicity 4.7 Mutagenicity and genotoxicity 4.8 Toxicokinetics


I:\MEPC\61\2-9.doc

5 DATA ON ENVIRONMENTAL FATE AND EFFECT UNDER AEROBIC AND ANAEROBIC CONDITIONS (G9: 4.2.1.3)

5.1 Modes of degradation (biotic; abiotic) 5.2 Bioaccumulation, partition coefficient, octanol/water partition coefficient 5.3 Persistence and identification of the main metabolites in the relevant media 5.4 Reaction with organic matter 5.5 Potential physical effects on wildlife and benthic habitats 5.6 Potential residues in seafood 6 PHYSICAL AND CHEMICAL PROPERTIES FOR THE ACTIVE SUBSTANCES,

RELEVANT CHEMICALS, AND TREATED BALLAST WATER (G9: 4.2.1.4) 7 ANALYTICAL METHODS AT ENVIRONMENTALLY RELEVANT CONCENTRATIONS (G9: 4.2.1.5) 7.1 Analysis of Total Residual Oxidants (TRO as Cl2) 7.2 Analysis of Disinfection By-products (DBPs) 8 USE OF ACTIVE SUBSTANCE 8.1 Manner of application 8.1.1 Process flow description 8.1.2 Chemical storage and handling 8.1.3 Various procedures and management measures 9 MATERIAL SAFETY DATA SHEETS (G9: 4.2.7) 10 RISK CHARACTERIZATION AND ANALYSIS 10.1 Screening for persistence, bioaccumulation, and toxicity (G9: 5.1) 10.1.1 Persistence (G9: 5.1.1.1) 10.1.2 Bioaccumulation (G9: 5.1.1.2) 10.1.3 Toxicity tests (G9: 5.1.2.3) 11 EVALUATION OF THE TREATED BALLAST WATER (G9: 5.2) 11.1 Total Residual Oxidants 11.2 Water quality parameters 11.3 Chemical analysis of disinfection by-products in treated ballast water 11.4 Ecotoxicity testing of treated ballast water, land-based testing 11.5 Determination of holding time 11.6 Reaction with organic matter 11.7 Characterization of degradation route and rate (G9: 5.3.5) 11.8 Prediction of Discharge and Environmental Concentrations (G9: 5.3.8) 11.8.1 Hydrodynamic modelling approach 11.8.2 Determination of substance concentrations and decay rates for modelling input 11.8.3 Predicted Environmental Concentration results from MAMPEC modelling 11.9 Effects on aquatic organisms 11.10 Assessment of potential for bioaccumulation 11.11 Effects on sediment 11.12 Effects assessment 11.13 Comparison of effect assessment with discharge toxicity


I:\MEPC\61\2-9.doc

12 RISK ASSESSMENT 12.1 Risk to safety of ship 12.1.1 Corrosion 12.2 Risks to human health 12.2.1 Introduction 12.2.2 Hazard identification/chemical of potential concern selection 12.2.3 Human Exposure Scenario 12.2.4 Health effects in humans 12.2.5 Risk characterization 12.2.6 Risk assessment conclusions 12.3 Risks to the aquatic environment 13 ASSESSMENT REPORT 14 REFERENCES APPENDICES (provided in confidential dossier) A.1 Key Data Table Summary A.2 Human Exposure Assessment A.3 BalPure® Process Flow Diagram A.4 Material Safety Data Sheets/International Chemical Safety Cards A.5 Analytical Method Information A.6 Laboratory Quality Assurance Documents A.7 Corrosion Information LIST OF TABLES Table 1 Overview of Chemical Identification Table 2 Acute Aquatic Ecotoxicity Data Table 3 Acute Aquatic Ecotoxicity Data – Land-based Testing Table 4 Overview of Chronic Aquatic Ecotoxicity Data Table 5 Chronic Aquatic Ecotoxicity Data – Land-based Testing Table 6 Koc Values for Relevant Chemicals Table 7 BCFs for Relevant Chemicals Table 8 Acute Mammalian Toxicity Data Table 9 Effects on Skin and Eye Table 10 Repeated-dose Toxicity Table 11 Chronic Mammalian Toxicity Table 12 Summary of Developmental and Reproductive Toxicity Data Table 13 Summary of Data on Carcinogenicity Table 14 Summary of Data on Mutagenicity and Genotoxicity Table 15 Summary of Uptake, Absorption, and Excretion of Chemicals Table 16 Fate and Mode of Degradation for Relevant Chemicals Table 17 Physical and Chemical Properties of Relevant Chemicals Table 18 Analytical Methods Table 19 PBT Criteria Evaluation Table 20 Average Oxidant Values (mg/L) in Treated Ballast Water Table 21 Water Quality Data Table 22 Post-Treatment Disinfection By-product Concentrations – Day 1 Table 23 Post-Treatment Disinfection By-product Concentrations – Day 5 Table 24 Median Disinfection By-product Concentrations


I:\MEPC\61\2-9.doc

Table 25 Summary of Relevant Chemical Concentration and Half-life for Modelling Table 26 PEC Summary for Environment A and Environment B Table 27 PNEC Derivation Summary Table 28 PEC/PNEC Calculation for MAMPEC Environment A Table 29 PEC/PNEC Calculation for MAMPEC Environment B Table 30 Comparison of Basic Approval and Land-based Ballast Water Residual COPC

Exposure Point Concentrations Table 31 Summary of Dermal Intake Factors Table 32 Summary of Oral Intake Factors Table 33 Summary of Toxicity Criteria Table 34 Summary of Cancer Risks and Non-cancer Hazards for BalPure® Technicians Table 35 Summary of Cancer Risks and Non-cancer Hazards for BalPure® Technicians

(or Others) – Ballast Water Discharge Sampling Table 36 Summary of Cancer Risks and Non-cancer Hazards for Ship's Crew/Dock

Workers Table 37 Summary of Cancer Risks for the General Public Table 38 Summary of Non-cancer Hazards for General Public Table 39 Estimated Concentrations of THMs in Vent Air and Comparisons to PELs LIST OF FIGURES Figure 1 Overview of the BalPure® Ballasting Process Figure 2 Overview of the BalPure® Deballasting Process Figure 3 Conceptual Exposure Model, BalPure® System


I:\MEPC\61\2-9.doc

LIST OF ABBREVIATIONS

Abbreviation Description Abbreviation Description

AIS Aquatic Invasive Species m Metres BCAA Bromochloroacetic acid MBAA Monobromoacetic acid BCF Bioconcentration Factor MCAA Monochloroacetic acid Br2 Bromine MEPC Marine Environment Protection Committee BWMS Ballast Water Management System mV Millivolts CAS Chemical Abstract Service μg/L Micrograms per litre Cl- Chloride ion

MAMPEC Marine Antifoulant Model to Predict Environmental Concentrations Cl2 Chlorine

CO2 Carbon dioxide mA Milliampere CO Carbon monoxide μM Micromolar COPC Chemicals of Potential Concern mg/L Milligrams per litre DBAA Dibromoacetic acid MSDS Material Safety Data Sheet DBCAA Dibromochloroacetic acid NaHSO3 Sodium bisulfite DBCM Dibromochloromethane NaOCl Sodium hypochlorite DBPs Disinfection By-products NIOZ Netherlands Institute for Sea Research DC Direct Current NOAEL No Observed Adverse Effect Level DCAA Dichloroacetic acid NOEC No Observed Effect Concentration DCBM Dichlorobromomethane

NOEL No Observed Effect Level (used interchangeably with NOAEL) DO Dissolved Oxygen

DOC Dissolved Organic Carbon OBr- Hypobromite ion DPD N,N-diethyl-p-phenylenediamine OCl- Hypochlorite ion

EC50 Effective concentration of substance that causes 50% of maximum response

OECD Organization for Economic Co-operation and Development

OIT Operator Interface Terminal EDCs Endocrine Disrupting Chemicals ORP Oxidation Reduction Potential EPC Exposure Point Concentrations PEC Predicted Environmental Concentrations EPI Estimation Programs Interface PBT Persistence, Bioaccumulation, Toxicity FAC Free Available Chlorine PFD Process Flow Diagram FAB Free Available Bromine pKa Dissociation Constant

GESAMP-BWWG

Joint Group of Experts on the Scientific Aspects of Marine Environmental Protection – Ballast Water Working Group

PLC Programmable Logic Controller PNEC Predicted No Effect Concentration POC Particulate Organic Carbon PPE Personal Protective Equipment

H+ Hydrogen ion ppm Parts per million H2 Hydrogen gas PSU Practical Salinity Units H2O2 Hydrogen peroxide SAR Structure Activity Relationship HSDB Hazardous Substances Data Bank SIDS Screening Information Data Set HSO-

3 Bisulfite ion STDN Severn Trent De Nora HSO-

4 Bisulfate ion TBAA Tribromoacetic acid HAA Haloacetic acid 2,4,6-TBP 2,4,6-Tribromophenol HOBr Hypobromous acid TCAA Trichloroacetic acid HOCl Hypochlorous acid THM Trihalomethane

IARC International Agency for Research on Cancer

TOC Total Organic Carbon TOXNET Toxicology Data Network

IMO International Maritime Organization TRO Total Residual Oxidants

IUPAC International Union of Pure and Applied Chemistry

TSS Total Suspended Solids

US EPA United States Environmental Protection Agency

Koc Organic Carbon Partition Coefficient

Kow Octanol/Water Partition Coefficient Kpa Kilopascal L/kg Litre per kilogram

LC50 Lethal Concentration for 50% of test population

LD50 Lethal Dose for 50% of test population

LEL Lower Explosive Limit

LOAEL Lowest Observable Adverse Effect Level


I:\MEPC\61\2-9.doc

1 INTRODUCTION Severn Trent De Nora, LLC (STDN) has developed the BalPure® Ballast Water Management System using seawater electrolysis combined with filtration and residual oxidant neutralization. STDN's experience with electro-chlorination in a variety of industrial and maritime settings has proven the effectiveness of the technology. Testing of the system in the laboratory, during land-based testing and onboard ship has demonstrated the technology's effectiveness for disinfection of aquatic invasive species in ballast water. Land-based testing of the BalPure® system was conducted at The Royal Netherlands Institute for Sea Research (NIOZ) during 2009. As required for Final Approval, an evaluation of aquatic ecotoxicity and Relevant Chemicals (e.g., disinfection by-products (DBPs)) was performed on samples drawn from the land-based test set-up. The quality assurance documents for the laboratories that performed ecotoxicity and DBP analyses are included in appendix A.6. Also presented are responses to comments of the GESAMP-BWWG (MEPC 60/2/16) during review of STDN's application dossier for Basic Approval (MEPC 60/2/9). STDN is submitting this application dossier to apply for Final Approval in accordance with Procedure (G9) (resolution MEPC.169(57)). 1.1 Overview of the BalPure® technology The BalPure® system provides a safe and economical electrolytic process for the on-site generation of a biocide solution from seawater to disinfect aquatic invasive species (AIS) in ballast water. Ballast water treatment is accomplished using a simple three-step process of filtration, injection of a biocide solution, and residual oxidant neutralization. The first phase is filtration using a 40 micron stainless steel mesh filter to remove organisms, large particles, and sediments. The second phase of the treatment process is electrochemical generation of the biocide solution. This involves passing a small supply (1/100 of total ballast flow) of seawater, either from the incoming ballast water line or sea cooling water, through electrolytic cells. The resulting disinfectant solution is injected directly into the incoming ballast water line where it will oxidize potential AIS. The third and final treatment process phase is residual oxidant neutralization to ensure environmental acceptability. When the treated ballast water is ready to be discharged, sodium bisulfite is injected directly into the ballast water discharge line. The sodium bisulfite (oxidant neutralization) addition is controlled with ORP and metering pump technology. The BalPure® system is designed to require minimal input from the ship's crew, with the system starting and stopping the necessary treatment steps based on electronic signals. This minimizes added duties to ship's crew and ensures proper ballast water treatment. More detailed information regarding operation of the BalPure® system is included in section 8. 1.2 Response to the GESAMP-BWWG's comments on the application for

Basic Approval Severn Trent De Nora submitted an application dossier for Basic Approval in August 2009 (MEPC 60/2/9). That dossier was reviewed during the 12th meeting of the GESAMP-BWWG in December 2009 and Basic Approval was recommended. The recommendations and comments by the GESAMP-BWWG (MEPC 60/2/16, annex 7), along with STDN's responses, are summarized below: .1 The Group recommended that a comprehensive evaluation of the corrosion

impacts on ship's structures be conducted using the maximum anticipated treatment concentration, 20 mg/L (TRO as Cl2) and that a realistic TRO decay profile may be considered if the total corrosion loss over the ship's


I:\MEPC\61\2-9.doc

life is to be projected. The Group noted however that such an analysis, while useful, is not necessary for the Group's consideration. The Group recommended that corrosion testing be conducted, and invited the applicant to consider the guidance provided in section 5.1 of the "Report of the eighth meeting of the GESAMP-BWWG", contained in document MEPC 59/2/16.

Response: Additional information regarding corrosion is presented in section 12.1 "Risk to safety of ship". Further, STDN has initiated a detailed corrosion study as recommended by the Group. The results of the study will be considered as part of the Type Approval process with the German Administration.

.2 The Group recommended to take into consideration the human exposure

during sampling of ballast water at discharge (inhalation, dermal contact), as well as during periodic sediment cleaning (inhalation, dermal contact). These scenarios should also be considered in the conceptual exposure model (CEM).

Response: See section 12.2 "Risks to human health" for consideration of human exposure during sampling of ballast water at discharge and during periodic sediment cleaning.

.3 The Group recommended that the data sets on the analysis of Active

Substance, Relevant Chemicals and Other Chemicals in treated ballast water should be further investigated during land-based and shipboard testing.

Response: STDN performed chemical analysis for Active Substances, Relevant Chemicals and Other Chemicals on samples drawn from the land-based test set-up. The human health risk assessment performed for Basic Approval has also been updated to reflect the land-based testing results. See section 11 "Evaluation of the treated ballast water" and section 12.2 "Risks to human health".

.4 The Group recommended that confirmation by further testing should be performed, that there will be no unacceptable risks (to ship safety, human health, or the environment) due to the chemical composition of the ballast water treated by this BWMS.

Response: See section 12.1 "Risk to safety of ship", section 12.2 "Risks to human health", and section 12.3 "Risks to the aquatic environment".

.5 The Group recommended that all missing information concerning the

environmental exposure assessment using an appropriate model be provided.

Response: See section 11.8 "Prediction of discharge and environmental concentrations".

.6 The Group recommended verification that the residual oxidant

neutralization with sodium bisulfate on discharge is effective at all ballast discharges with MADC < 0.20 mg/L TRO as Cl2.

Response: See section 11.1 "Total Residual Oxidants".


I:\MEPC\61\2-9.doc

.7 The Group recommended that both acute and chronic ecotoxicity testing of treated ballast water be performed in accordance with the full requirements of Procedure (G9) at the maximum proposed treatment dose.

Response: As required by Procedure (G9), STDN performed ecotoxicity testing of treated/neutralized ballast water on samples drawn from the land-based test set-up. See sections 3.1 and 3.2 for Acute and chronic aquatic ecotoxicity data (Tables 3 and 5) and section 11.4 "Ecotoxicity testing of treated ballast water, land-based testing" for discussion of the results.

2 IDENTIFICATION OF THE ACTIVE SUBSTANCES AND RELEVANT CHEMICALS

(G9: 4.1) 2.1 Introduction to the Chemical Basis of BalPure® Ballast Water Management

System The BalPure® system uses electrolysis of seawater to safely generate sodium hypochlorite on board, and immediately injects the solution directly into ballast water while it is pumped on board. Seawater is passed through electrolytic cells (electrolyzers) where it is subjected to medium amperage, low voltage direct current. Electrolytic cells are comprised of a cathode and an anode. At the outlet of the electrochemical generator, the seawater contains a mixture of chemicals produced at both the anode and cathode. Because seawater contains bromide ions, the electrochemical process results in the formation of hypochlorous acid (HOCl) in equilibrium with hypochlorite ion (OCl-), hypobromous acid (HOBr) in equilibrium with hypobromite ion (OBr-), and hydrogen gas (H2) as a by-product. During the electro-chemical process a small amount of caustic (OH-) is also produced, which is effectively neutralized within the cell and causes no apparent increase in the pH of treated ballast water. The measurable oxidant species present during chlorine-based water disinfection, and the terminology used, is often a point of confusion. By industry standards and convention, all chlorine present in a water sample, regardless of form, is referred to as Free Available Chlorine (FAC). FAC includes Cl2, HOCl, and OCl-. When including chloramines, the term Combined (or Total) Chlorine is used. Similarly, all bromine species present in a water sample are referred to Free Available Bromine (FAB). When including bromamine the term Combined (or Total) Bromine is used. The common factor in all of these chemical species is that they are considered oxidants, and due to the difficulty in measuring each chemical species individually in a complex water sample, the term Total Residual Oxidants (TRO) is used to include all of these oxidants when present in a sample. To clarify: Free Available Chlorine (FAC) = Cl2 + HOCl + OCl- Combined (or Total) Chlorine = FAC + chloramine Free Available Bromine (FAB) = Br2 + HOBr + OBr- Combined (or Total) Bromine = FAB + bromamine Total Residual Oxidants (TRO) = FAC + chloramine + FAB + bromamine Therefore, in this dossier, TRO is used to describe the total of all oxidants present in treated water. The chemical methods used for analyzing TRO often state that the measurement is calculated as mg Cl2/L. However, these methods include all of the oxidants present in a water sample as mentioned above, not just Cl2. Severn Trent De Nora classifies the substances associated with the BalPure® system using the definitions of Active Substances, Relevant Chemicals, and Other Components as provided in Procedure (G9) Methodology (23 May 2008) and presents information on each in the sections below.


I:\MEPC\61\2-9.doc

2.2 Active Substances 2.2.1 Hypochlorous acid and hypobromous acid To avoid storing Active Substances on board, the BalPure® system safely generates sodium hypochlorite by electrolysis of seawater. Seawater is passed through electrolytic cells (electrolyzers) where it is subjected to medium amperage, low voltage direct current. The electrolytic solution generated by the BalPure® system will consist of a near equal mixture of hypochlorous acid (HOCl) and hypochlorite ion (OCl-) at the pH of seawater (typical range of 7.5-8.4). As discussed above in section 2.1, the combination of Cl2, HOCl, and OCl- is commonly referred to as free available chlorine (FAC). Because seawater also contains bromide ion, it is important to note that reactions similar to those occurring with chloride ion can also be occurring with bromide ion in the electrolytic cell. For example, bromide ion will form Br2. At the pH of seawater the electrolytic solution generated by the BalPure® system will also consist of a mixture of hypobromous acid (HOBr) and hypobromite ion (OBr-), with a slightly greater proportion of the mixture existing as HOBr. As discussed in section 2.1, the combination of Br2, HOBr, and OBr- are commonly referred to as free available bromine (FAB). In the solution generated, hypochlorous acid (HOCl) in equilibrium with hypochlorite ion (OCl-) and hypobromous acid (HOBr) in equilibrium with hypobromite ion (OBr-) are the actual disinfecting agents. In this dossier, these Active Substances are considered together and referenced as Total Residual Oxidants (TRO as Cl2). Details on the primary cell reactions occurring in the electro-chemical generation process in seawater and the effect of pH on the distribution of chemical species formed were presented in STDN's application dossier for Basic Approval (MEPC 60/2/9, section 2.2.1). 2.3 Relevant Chemicals Disinfection of water with chlorine-based disinfectants can result in the formation of disinfection by-products (DBPs). Due to the neutralization step, the Relevant Chemical sodium bisulfite may also be present in ballast water discharge. Data on the substances that had been analysed during initial pilot studies were presented in STDN's application dossier for Basic Approval (MEPC 60/2/9). During land-based testing of the BalPure® system at NIOZ the same substances, along with additional DBPs were evaluated. The substances that were analysed for and presented in the Basic Approval dossier, and the substances evaluated for Final Approval are summarized in the columns below. Substances denoted with an "*" are those with a measurable concentration in samples from land-based testing. Substances with a measurable concentration in treated/neutralized ballast water are presented in this dossier. Substances with no measurable concentration (e.g., chlorate, bromate) are not considered further in this dossier. Basic Approval Final Approval THMs THMs

Chloroform Chloroform Bromoform Bromoform* Dibromochloromethane Dibromochloromethane* Dichlorobromomethane Dichlorobromomethane*


I:\MEPC\61\2-9.doc

HAAs HAAs Monochloroacetic acid Monochloroacetic acid* Dichloroacetic acid Dichloroacetic acid* Trichloroacetic acid Trichloroacetic acid Monobromoacetic acid Monobromoacetic acid* Dibromoacetic acid Dibromoacetic acid* Bromochloroacetic acid Tribromoacetic acid* Bromochloroacetic acid* Dibromochloroacetic acid* Dichlorobromoacetic acid Other Substances Other Substances Bromate 2,4,6-Tribromophenol* Monobromoacetonitrile Chlorate Bromate Sodium Bisulfite* Sodium Bisulfate* For the treated/neutralized ballast water samples evaluated during land-based testing, chemical analysis was performed by Eurofins/Analytico (The Netherlands) and NovaChem Laboratories (USA). The Relevant Chemicals are discussed in more detail in the following sections. Additionally, during the electrolysis process hydrogen gas is produced. For this reason, hydrogen gas is considered as a Relevant Chemical and is discussed in more detail in section 2.3.4. Table 1 provides a summary of the substances potentially associated with the BalPure® system.

Table 1: Overview of Chemical Identification

Substance (IUPAC Name)

Identification Numbers

Chemical Formula

Structural Formula

Molecular Weight

Dose/Classification in Treated Ballast

Water

Active Substances in Ballast Water

Hypochlorous acid (Sodium Hypochlorite)

CAS # 7681-52-9

HOCl 52.46 <15 mg/L TRO as Cl2

Hypobromous acid (Sodium Hypobromite)

CAS # 13824-96-9

HOBr

96.91 <15 mg/L TRO as Cl2

Relevant Chemicals

Bromate (as sodium bromate)

CAS # 7789-38-0

NaBrO3 150.89 By-product

Chlorate CAS # 14866-68-3

ClO3-

83.451 By-product

Hydrogen CAS # 133-74-0

H2 H---H 2.016 By-product

Monobromoacetonitrile CAS # 590-17-0

C2H2BrN

119.947 By-product


I:\MEPC\61\2-9.doc



Chemical Formula

Structural Formula

Molecular Weight


Water

2,4,6-Tribromophenol CAS # 118-79-6

C6H3Br3O

330.80 By-product

THMs

Dibromochloromethane CAS # 124-48-1

CHBr2Cl

208.29 By-product

Dichlorobromomethane CAS # 75-27-4 CHBrCl2

163.8 By-product

Tribromomethane (bromoform)

CAS # 75-25-2 CHBr3

252.77 By-product

Trichloromethane (chloroform)

CAS # 67-66-3 CHCl3

119.38 By-product

HAAs

Bromochloroacetic acid CAS # 5589-96-8

C2H2BrClO2

173.39 By-product

Monochloroacetic acid CAS # 79-11-8 C2H3ClO2

94.50 By-product

Dibromoacetic acid CAS # 631-64-1

Br2CHCOOH 217.84 By-product

Dichloroacetic acid CAS # 79-43-6 CHCl2COOH 128.9 By-product

Dibromochloroacetic acid

CAS # 5278-95-5

C2HBr2ClO2

252.29 By-product

Dichlorobromoacetic acid

CAS # 71133-14-7

C2HBrCl2O2

207.84 By-product

Trichloroacetic acid CAS # 76-03-9 CCl3COOH

163.4 By-product

Tribromoacetic acid CAS # 75-96-7 C2HBr3O2 296.74 By-product

Monobromoacetic acid CAS # 79-08-3 C2H3BrO2

138.95 By-product


I:\MEPC\61\2-9.doc



Chemical Formula

Structural Formula

Molecular Weight


Water

Other Chemicals

Sodium Bisulfate CAS # 7681-38-1 HNaO4S 120.06 By-product

Sodium Bisulfite CAS # 7631-90-5 NaHSO3

104.06 Neutralizing agent

2.3.1 Trihalomethanes (THMs) and haloacetic acids (HAAs) During the disinfection process, reactions with organic materials present in water may result in the formation of THMs and HAAs. The amount of THMs and/or HAAs formed as DBPs will vary depending on water quality. The following THMs had measurable concentrations in treated ballast water from the land-based test set-up: dichlorobromomethane (DCBM), dibromochloromethane (DBCM), and bromoform. The data is presented in section 11.3, Tables 22 and 23. Bromoform was also present at low levels in untreated (control) water, and has been documented to occur naturally in seawater as it is produced by various algal species (HSDB/TOXNET Tribromomethane, 2009; Goodwin et al., 1997; Ohsawa et al., 2001). The following HAAs had measurable concentrations in treated ballast water from the land-based test set-up: bromochloroacetic acid (BCAA), monochloroacetic acid (MCAA), dichloroacetic acid (DCAA), monobromoacetic acid (MBAA), dibromochloroacetic acid (DBCAA), tribromoacetic acid (TBAA), and dibromoacetic acid (DBAA). The data is presented in section 11.3, Tables 22 and 23. 2.3.2 2,4,6-Tribromophenol 2,4,6-Tribromophenol (TBP) did not have measurable concentrations (<0.1 µg/L detection limit) in ballast water discharge from the high salinity test cycles. 2,4,6-TBP was detected at low levels (maximum = 1.3 µg/L) in ballast water discharge from the low salinity test cycles (section 11.3, Tables 22 and 23). 2,4,6-TBP was also detected in the untreated (control) samples at the same, or higher, concentrations. 2,4,6-TBP has been documented to occur in natural environments, either as a pollutant from anthropogenic sources (wood industry, antifungal use) or from natural production by marine benthic organisms (OECD SIDS, 2003). As presented in Table 4 below, the lowest chronic effect concentration identified for 2,4,6-TBP is 0.10 mg/L (Daphnia magna). The 2,4,6-TBP concentration measured in ballast water discharge is well below this effect concentration in both high and low salinity test waters. Therefore, 2,4,6-TBP is not considered to be present at concentrations of environmental concern in ballast water discharge. However, because 2,4,6-TBP did have measurable concentrations in the low salinity tests, data to allow for a full risk assessment is included in this dossier. 2,4,6-TBP was also included in the hydrodynamic modelling evaluation to verify low potential aquatic environmental risk. 2.3.3 Bromate, chlorate and monobromoacetonitrile The presence of bromate ions in treated ballast water was evaluated during land-based testing at NIOZ and concentrations were not measurable (<10 µg/L detection limit) (Tables 22 and 23). Further, the lowest acute aquatic toxicity effect concentration located for bromate is 13.6 mg/L (Glenodinium halli) and the lowest chronic effect concentration located


I:\MEPC\61\2-9.doc

is 16 mg/L (Thalassiosira pseudonana) (Tables 2 and 4, respectively). As bromate in ballast water discharge is non-detectable and well below these effect concentrations, it is not present at environmentally relevant concentrations. As such, bromate was not included in the hydrodynamic modelling assessment and further data on bromate are not considered in this dossier for Final Approval. Chlorate was not an anticipated DBP but was analysed for verification. Samples from the land-based test set-up demonstrated that chlorate was not present in ballast water discharge in measurable concentrations (<50 µg/L detection limit) (Tables 22 and 23). Therefore, chlorate is not addressed further in this dossier. Monobromoacetonitrile was also a DBP evaluated in ballast water discharge samples from the land-based test set-up at NIOZ. Concentrations in both low and high salinity test cycles were not measurable (<1.0 µg/L or <10.0 µg/L detection limits). Therefore, monobromoacetonitrile is not addressed further in this dossier. 2.3.4 Hydrogen gas Hydrogen (H2) is formed as a by-product of seawater electrolysis. Due to the explosive properties of hydrogen gas, allowing it to accumulate in the ballast tanks would create an unsafe situation. Therefore, the hydrogen produced during the electro-chlorination process is separated immediately upon exiting from the electrolytic cell using a patented cyclone separation device and is not allowed to enter the ballast water piping. The separated hydrogen gas is diluted to less than 1% hydrogen by forced air blowers and vented to the atmosphere outside of the ship. The lower explosive limit (LEL) for hydrogen is 4% and the operating limitations of the system provide a four-fold safety factor. Relevant information pertaining to potential risks associated with hydrogen gas was provided in STDN's application dossier for Basic Approval (MEPC 60/2/9). The GESAMP-BWWG considered the assessment and safety measures presented in Basic Approval with regard to H2 acceptable (MEPC 60/2/16, annex 7, sections 3.2.5 and 3.3.2). 2.4 Other chemicals 2.4.1 Sodium bisulfite and sodium bisulfate During discharge of treated water (deballasting), sodium bisulfite (NaHSO3) is injected into the ballast water discharge line to neutralize residual oxidants that may be present. The bisulfite chemically reduces any free chlorine and/or hypochlorous acid to chloride ion, and any free bromine and/or hypobromous acid to bromide ion. During land-based testing the highest residual sodium bisulfite concentration measured in the whole effluent after injection was 8.0 mg/L. Since the lowest sodium bisulfite LC50 located was 81 mg/L (Daphnia magna) (Table 2), which is significantly higher than the highest residual sodium bisulfite measured in ballast water discharge (8.0 mg/L), aquatic environmental risks would not be expected. Further, sodium bisulfite does not bioaccumulate, is highly soluble in water, and not expected to adsorb to sediments. However, to ensure that sodium bisulfite that may be present in neutralized discharge does not present environmental risk, data to allow for a full risk assessment was presented in STDN's application dossier for Basic Approval (MEPC 60/2/9) and is evaluated further in this Final Approval dossier. Sodium bisulfite was also included in the hydrodynamic modelling evaluation. When sodium bisulfite reduces the chlorine and/or bromine species present as residual oxidants in treated water, bisulfate ion can be formed. Bisulfate ion dissociates into sulfate and hydrogen ions. Because the pKa for the bisulfate ion/sulfate ion equilibrium is approximately 2,


I:\MEPC\61\2-9.doc

bisulfate ion that is formed during the neutralization step is expected to be in the form of sulfate ion. Natural seawater typically contains approximately 3 g/L (3,000 mg/L) of sulfate ion. Sodium bisulfate (as sulfate ion) was measured in samples collected from the land-based test set-up to verify that the neutralizing agent does not cause a significant increase in the level of sulfate ion present. The data show that the concentrations of sulfate ion in neutralized ballast water compared to untreated (control) water is not significantly different. The data are presented in section 11.3, Tables 22 and 23. Sulfate is not considered to pose unacceptable environmental risks and is not evaluated further in this dossier. Based on the GESAMP-BWWG report (section 3.1.1.4), this approach is accepted by the Group. 3 DATA ON EFFECTS ON AQUATIC PLANTS, INVERTEBRATES AND FISH, AND

OTHER BIOTA, INCLUDING SENSITIVE AND REPRESENTATIVE ORGANISMS (G9: 4.2.1.1)

A literature search was conducted to locate existing aquatic ecotoxicity data for the substances associated with the BalPure® system. The data summarized in Tables 2 and 4 present the acute and chronic toxicity data that was identified. A complete base-set for fish, crustaceans, and algae was not located for all substances and in some cases the data do not meet validity criteria. However, the data has been provided as supporting information and is discussed in relation to the Predicted No Effect Concentration (PNEC) derivation in section 11.9. Tables 3 and 5 summarize acute and chronic ecotoxicity data on whole effluent from land-based testing studies of the BalPure® system. Toxicity testing was done in conjunction with studies performed at the Royal Netherlands Institute for Sea Research (NIOZ), Texel (the Netherlands). Toxicity testing was performed by Grontmij/AquaSense (the Netherlands) and Chemex Environmental International Ltd. (UK). Toxicity tests were conducted with six taxonomic groups including bacteria, algae, crustaceans, rotifers, mollusks and fish in two water types, i.e. brackish water (low salinity) and marine water (high salinity). The samples for ecotoxicity testing were collected directly from the discharge pipe on Day 1 and on Day 5 after ballast water treatment and neutralization. Samples for analysis were stored in proper sample containers at 4°C and transported to the designated laboratory within 24 hours of collection. Analysis of the samples occurred within the time specified under standard laboratory procedures. All tests were performed according to laboratory protocols based on internationally recognized ISO, OECD, ASTM, and Parcom/OSPAR guidelines. Further, water quality data from the land-based test cycles at NIOZ are presented in section 11.2 "Evaluation of treated ballast water (Table 21)". 3.1 Acute aquatic ecotoxicity The following table summarizes the data found in literature concerning acute aquatic ecotoxicity for the substances associated with the BalPure® system. In some cases, data could not be located for all three taxonomic groups (fish, crustaceans and algae). Literature concerning acute aquatic toxicity for bromochloroacetic acid, dibromochloroacetic acid, and tribromoacetic acid could not be located for fish, crustaceans or algae. Similar data gaps are noted in other dossiers submitted for approval under Procedure (G9), suggesting that other literature searches have had similar findings. However, when available, data for other taxonomic groups (amphibians and ciliates) or alternative endpoints (LC100 or EC03) are included as supporting information. To complete the data set for sodium bisulfite, STDN commissioned a 96-hour growth study with Skeletonema costatum (Nautilus Environmental, 2009) that investigated potential toxic effects for several concentrations of sodium bisulfite. The study resulted in an EC50 of 224 mg/L.


I:\MEPC\61\2-9.doc

Table 2: Acute Aquatic Ecotoxicity Data

Substance

F C A Am Ci

Species Test Type E(L)C50 mg/L

NOEC mg/L

Reference/ Comments

TRO as Cl2 (Fresh water)

F Oncorhynchus mykiss 15-d survival 0.059 N/A Fisher et al. (1999) C Daphnia magna 48-h survival 0.032 N/A Fisher et al. (1999) A Natural phytoplankton 47-h change in Chla <0.1 N/A Brooks and Liptak (1978)

TRO as Cl2 (Saltwater)

F Menidia beryllina 96-h survival 0.143 N/A Fisher et al. (1999) C Mysidopsis bahia 96-h survival 0.062 N/A Fisher et al. (1999) A Dunaliella primolecta 24-h growth 0.075 N/A Videau et al. (1979)

Active Substance

Hypochlorous acid

F Oncorhynchus mykiss 96-h survival 0.032 N/A US EPA ECOTOX (Thatcher, 1978)

C Americamysis bahia 7-d survival 0.144 N/A Fisher et al. (1994)

A Porphyra yezoensis 10-d growth 1.4 N/A US EPA ECOTOX (Maruyama et al., 1988)

A Porphyra yezoensis 10-d mortality 2.3 N/A US EPA ECOTOX (Maruyama et al., 1988)

Hypobromous acid

A Chlorella sorokiniana 18-h survival 1 N/A US EPA ECOTOX (Kott and Edlis, 1969)

Relevant Chemicals

Bromate

F Morone saxatilis 96-h survival 30.8 N/A Richardson et al. (1981)

C Neomysis awatschensis

24-h survival 176 N/A US EPA ECOTOX (Crecelius, 1979)

A Glenodinium halli Cell growth; exposure period not specified

13.6 N/A Hutchinson et al. (1997)

2,4,6-Tribromophenol

F Oryzias latipes 96-h survival 1.5 N/A OECD SIDS (2003) F Pimephales promelas 96-h survival 6.25 N/A OECD SIDS (2003) F Cyprinus carpio 96-h survival 1.1 N/A OECD SIDS (2003) C Daphnia magna 48-h immobilization 2.2 N/A OECD SIDS (2003) C Daphnia magna 48-h immobilization 0.26 N/A OECD SIDS (2003)

A Selenastrum capricornutum

72-h biomass 0.76 N/A OECD SIDS (2003)


24-48-h growth 1.1 N/A OECD SIDS (2003)


24-72-h growth 1.6 N/A OECD SIDS (2003)

THMs

Chloroform

F Lepomis macrochirus 96-h survival 14 N/A US EPA ECOTOX (Anderson and Lusty, 1980)

F Cyprinus carpio 72-120-h egg fertilization to hatch

97 N/A Mattice et al. (1981)

C Artemia salina 24-h survival 30 N/A US EPA ECOTOX (Foster and Tullis, 1984)

A Scenedesmus subspicatus

48-h growth 950 N/A Kuhn and Pattard (1990)

Bromoform

F Cyprinodon variegates

72-h survival 18 N/A US EPA ECOTOX (Heitmueller et al., 1981)


52.3 N/A Mattice et al. (1981)

C Daphnia magna 48-h immobilization 46 N/A US EPA ECOTOX (LeBlanc, 1980)

A Skeletonema costatum

96-h growth inhibition

12.3 N/A US EPA ECOTOX (1978)

Chlorodibromo-methane


34 N/A Mattice et al. (1981)

Ci Tetrahymena pyriformis

24-h growth 650 N/A US EPA ECOTOX (Yoshioka et al., 1985)


I:\MEPC\61\2-9.doc

Substance

F C A Am Ci


NOEC mg/L

Reference/ Comments

Dichlorobromo-methane

F Oryzias latipes 96 h survival 72 N/A Toussaint et al. (2001)


67.4 N/A Mattice et al. (1981)

Ci Tetrahymena pyriformis

24-h growth 240 N/A US EPA ECOTOX (Yoshioka et al.,1985)

HAAs

Dichloroacetic acid

C Daphnia magna 24-h immobilization 106 N/A US EPA ECOTOX (Trenel and Kuhn, 1982)

C Nitocra spinipes 96-h survival 23 N/A US EPA ECOTOX (Linden et al, 1979)

Am Xenopus laevis 96-h development 3560 N/A US EPA ECOTOX (Fort et al., 1993)

Trichloroacetic acid

F Pimephales promelas 96-h survival 2000 N/A US EPA ECOTOX (Dennis et al., 1979)

C Daphnia magna 48-h immobilization 2000 N/A US EPA ECOTOX (Dennis et al., 1979)

A Chlorella pyrenoidosa 72-h growth N/A 500 US EPA ECOTOX (Huang and Gloyna, 1968)

Dibromoacetic acid

F Pimephales promelas 24-h survival 160 N/A US EPA ECOTOX (Mayes et al., 1985)




Dibromochloro-acetic acid

No data found in literature

Tribromoacetic acid


Bromochloro-acetic acid


Monobromo-acetic acid

F Cyprinus carpio 5-h survival 222 N/A US EPA ECOTOX (Loeb and Kelly, 1963)

C Daphnia magna 24-h survival 34 N/A US EPA ECOTOX (Kuhn, 1988)


48-h survival 0.34 N/A Kuhn and Pattard (1990)

Monochloro-acetic acid

F Cyprinus carpio 28-h survival 191

(LC100)N/A

US EPA ECOTOX (Loeb and Kelly, 1963)

C Daphnia magna 48-h mobility 96 N/A Kuhn and Pattard (1990)


48-h growth 0.028 N/A Kuhn and Pattard (1990)


48-h growth 0.07 N/A Kuhn and Pattard (1990)

Other Chemicals

Sodium bisulfite

F Carassius spp. 96-h survival 100 N/A MSDS Solvay Chemicals (2007)

F Gambusia affinis 96-h survival 240 N/A US EPA ECOTOX (Wallen et al., 1957)

C Daphnia magna 96-h survival 102 N/A US EPA ECOTOX (Freeman and Fowler, 1953)

C Daphnia magna 50-h survival 81 N/A US EPA ECOTOX (Dowden and Bennett, 1965)

C Daphnia magna 24-h survival 171 N/A US EPA ECOTOX (Dowden and Bennett, 1965)


96-h growth 224 128 Nautilus Environmental, 2009

F = fish, C = crustacean, A = algae, Am = amphibian, Ci = ciliate, N/A = not applicable.


I:\MEPC\61\2-9.doc

Table 3 summarizes acute aquatic toxicity testing results for treated discharge samples, post-neutralization, from the BalPure® system. Toxicity tests were conducted with six taxonomic groups including bacteria, algae, crustaceans, rotifers, mollusks and fish in two water types: brackish water (low salinity) and marine water (high salinity). The samples were collected for toxicity testing on Day 1 and Day 5 after ballast water treatment and neutralization. Toxicity was indicated in one acute test with Acartia tonsa one day after BalPure® treatment (NOEC of 18%). Discussion of the whole effluent toxicity data is located in section 11.4.

Table 3: Acute Aquatic Ecotoxicity Data - Land-based Testing

Substance

F C A

MB R

Species Test Type Days after

treatment

E(L)C50

% sample

NOEC %

sample

Reference/ Comments

Low Salinity (23.68 PSU)

BalPure® Treatment

Dose: 14.83 mg/L

(ave.)

C Acartia tonsa 48h mobility (ISO 14669)

1 >100 18 Grontmij|AquaSense, 20095 >100 100 Grontmij|AquaSense, 2009

M Crassostrea

gigas

48h larval development (ASTM E724)

1 >100 100 Grontmij|AquaSense, 2009


B Vibrio fischeri 30m inhibition of light emission (ISO 11348-3)

1 >90* N/R Grontmij|AquaSense, 2009


A Phaeodactylum

tricornutum

72h growth inhibition

(ISO 10253)



R Brachionus

plicatilis 24h survival

(MircoBioTests Inc.)1 >100 100 Grontmij|AquaSense, 20095 >100 100 Grontmij|AquaSense, 2009

F Scophthalmus

maximus 96h survival (OECD 203)


High Salinity (33.90 PSU)

BalPure® Treatment

Dose: 13.17 mg/L

(ave.)

C Acartia tonsa 48h mobility (ISO 14669)


M Crassostrea

gigas

48h larval development (ASTM E724)


5 >100 N/D Grontmij|AquaSense, 2009

B Vibrio fischeri 30m inhibition of

light emission (ISO 11348-3)



A Phaeodactylum

tricornutum

72h growth inhibition

(ISO 10253)



R Brachionus

plicatilis 24h survival

(MircoBioTests Inc.)1 >100 100 Grontmij|AquaSense, 20095 >100 100 Grontmij|AquaSense, 2009

F Scophthalmus

maximus 96h survival (OECD 203)


F = fish, C = crustacean, A = algae, M = mollusk, B = bacteria, R = rotifer. N/D = not determined; N/R = not reported. *Maximum tested concentration = 90% vol. of sample. 3.2 Chronic aquatic ecotoxicity Table 4 summarizes the data found in literature concerning chronic aquatic toxicity for the substances associated with the BalPure® system. Literature concerning chronic aquatic toxicity for several HAAs could not be located to compile a complete data set (fish, crustaceans, algae). Similar data gaps are also noted in other dossiers submitted for approval under Procedure (G9), suggesting that other literature searches have had similar findings. However, if available, data using alternative endpoints (EC03) is included as supporting information.


I:\MEPC\61\2-9.doc

Table 4: Overview of Chronic Aquatic Ecotoxicity Data

Substance F C A


NOEC mg/L

Reference/ Comments

Active Substance

TRO as Cl2 (Fresh water)

F Ictalurus punctatus 134-d growth N/A 0.005 Hermanutz et al. (1990)

C Gammarus pseudolimnaeus

70-d fecundity N/A 0.012 Arthur et al. (1975)

A Natural algae 24-d inhibition of biomass

N/A 0.079 Pratt et al. (1988)

TRO as Cl2 (Saltwater)

F Menidia peninsula 32-d egg to post-hatch N/A 0.04 Goodman et al. (1983) C Panaeus kerathurus 7-d growth/molting N/A 0.079 Saroglia and Scarano (1979)

A Natural phytoplankton

21-d reduction in cell density

N/A 0.001 Sanders et al. (1981)

Hypochlorous acid

F Oncorhynchus kisutch

23-d survival N/A 0.101US EPA ECOTOX (Holland et al., 1960)

C Americamysis bahia 7-d reproduction N/A 0.048 Fisher et al. (1994) Hypobromous acid


Relevant Chemicals

Bromate

F Morone saxatilis 10-d survival 92.6 -- Richardson et al. (1981)

F Leiostomus zanthurus

10-d survival 278.6 -- Richardson et al. (1981)

A Thalassiosira pseudonana

7-d growth inhibition 16 -- US EPA ECOTOX (Erickson and Freeman, 1978)

2,4,6-Tribromophenol

F Pimephales promelas

8-d survival 4.5 N/A US EPA ECOTOX (Phipps et al., 1981)

C Daphnia magna 21-d reproduction rate N/A 0.10 OECD SIDS (2003)


72-h biomass N/A 0.22 OECD SIDS (2003)


24-72-h growth N/A 1.0 OECD SIDS (2003)

THMs

Chloroform

F Oryzias latipes 9- month growth N/A 1.5 Toussaint et al. (2001)

C Daphnia magna 21-d reproduction N/A 13 US EPA ECOTOX (Kuhn et al., 1988)


5-d reduction in cell volume

477 216 Cowgill et al. (1989)

Bromoform F

Cyprinodon variegates

28d- growth N/A 8.5 US EPA ECOTOX (Ward et al., 1981)

A Pseudokirchneriella subcapitata

96-h biochemistry change

N/A 38.6 U.S. EPA ECOTOX (1978)





HAAs Dichloroacetic acid


7-d cell proliferation 1485

(EC03)N/A

US EPA ECOTOX (Trenel and Kuhn, 1982)

Dibromochloro-acetic acid


Dibromoacetic acid


Tribromoacetic acid


Trichloroacetic acid

A Scenedesmus quadricauda

7-d growth threshold N/A 200 OECD SIDS (2000)

A Chlorella vulgaris 4d-population change N/A 100 US EPA ECOTOX (Garten, 1990)

Bromochloro-acetic acid



I:\MEPC\61\2-9.doc

Substance F C A


NOEC mg/L

Reference/ Comments

Monobromo-acetic acid

C Daphnia magna 21d-NOEC N/A 3.2 US EPA ECOTOX (Kuhn, 1988)

C Daphnia magna 21d-NOEC on reproduction

N/A 1.6 US EPA ECOTOX (Kuehn et al, 1989)


3d-population change N/A 1.4 Kuhn and Pattard (1990)

Monochloro-acetic acid

C Daphnia magna 21d-reproduction change

N/A 32 Kuhn and Pattard (1990)


7d-histology 0.13

(EC03) N/A

US EPA ECOTOX (Trenel and Kuhn, 1982)

Other Chemicals Sodium bisulfite


F = fish, C = crustacean, A = algae, N/A = not applicable. Results for chronic whole effluent toxicity (WET) testing of treated/neutralized ballast water discharge are summarized in Table 5 below and discussed further in section 11 "Evaluation of treated ballast water". Samples were drawn from the land-based test set-up and no chronic toxicity was observed for any species tested. The highest concentration with no observed effect (NOEC) was 100 per cent sample (Grontmij|AquaSense, 2009). As agreed upon with the German Administration, chronic testing was performed only for organisms that had effects noted during acute testing. As acute effects were only observed during one of four acute tests with Acartia tonsa (NOEC 18%), STDN also performed chronic (sub-lethal) testing for fish. The data indicate that no chronic effects were noted for either species tested.

Table 5: Chronic Aquatic Ecotoxicity Data – Land-based Testing

Substance F C

Species Test Type Days after

treatment

E(L)C50

% sample

NOEC%

sample

Reference/ Comments

Low Salinity (23.68 PSU)BalPure®

Treatment Dose:

14.83 mg/L (ave.)

C Acartia tonsa 30d repro./devel. (ASTM STP 667)

1 N/D 100 Grontmij|AquaSense, 2009 5 N/D 100 Grontmij|AquaSense, 2009

F Scophthalmus

maximus 48h develop

(PARCOM 1994)



High Salinity (33.90 PSU) BalPure® Treatment

Dose: 13.17 mg/L

(ave.)

C Acartia tonsa 30d repro./devel. (ASTM STP 667)

1 N/D 100 Grontmij|AquaSense, 2009 5 N/D 100 Grontmij|AquaSense, 2009

F Scophthalmus

maximus 48h develop

(PARCOM 1994)

1 >100 N/D Grontmij|AquaSense, 2009


F = fish, C = crustacean, N/D = not determined. 3.3 Endocrine disruption Endocrine disrupting chemicals (EDCs) are defined as exogenous substances that cause adverse effects in an organism or its progeny, consequent to changes in endocrine functions. In a recently conducted study to investigate whether chlorinated by-products formed through waste water disinfection were estrogenic, no relationship was found between the formation of THMs and estrogenic activity (Schiliro, 2009). In fact, chlorination of surface water and effluents was found to decrease estrogenic activity, which is thought to be due to the oxidation effects of chlorine (Lee, 2004). A thorough review of the literature found no indications that either of the Active Substances in the TRO as Cl2 (HOCl/HOBr) or the chemically more complex THMs/HAAs were EDCs.


I:\MEPC\61\2-9.doc

Brominated phenols have been shown to cause a significant effect on the Ca2+ homeostasis in endocrine cells by causing an intracellular increase in Ca2+ (Hassenklöver et al., 2006). At concentrations of 300 µM or higher this effect may suggest a link to endocrine disruption by 2,4,6-tribromophenol (TBP). Due to the low concentration of 2,4,6-TBP (maximum concentration of 1.3 µg/L) (Tables 22 and 23) detected in treated ballast water, and the limited routes of exposure in environmental waters, no endocrine effects would be expected as a result of treated ballast water discharge. 3.4 Sediment toxicity The sediment toxicity of a chemical is a function of the ability of the chemical to be adsorbed to sediment as well as its persistence and toxicity. The organic carbon partition coefficient (Koc) is a measure of the tendency of an organic substance to be adsorbed by soil or sediment. STDN's application dossier for Basic Approval (MEPC 60/2/9, section 3.4) provided data for THMs/HAAs measured in previous studies of the BalPure® system. During land-based testing, additional DBPs were analysed and three were found in measurable concentrations. Data for these three additional substances is provided below. The Koc for 2,4,6-TBP (1,186 L/kg) indicates moderate partitioning into sediment. When 2,4,6-TBP is released to water, 93% is expected to stay in the water compartment and 7% is transported to the sediment compartment (OECD SIDS, 2003). 2,4,6-Tribromophenol is reported to dehalogenate rapidly in anaerobic sediments, with a reported half-life of approximately 4 days (CICADS 66, 2005). This is much more rapid than the sediment persistence criteria of 180 days in marine sediment and 120 days in fresh water sediment (Table 19 and the Methodology for information gathering and the conduct work of the GESAMP-BWWG (Methodology), section 6.1.4). Considering the low concentration of 2,4,6-TBP (maximum = 1.3 µg/L) in ballast water discharge, a moderate potential for sediment adsorption, and the rapid sediment degradation, no effects on sediment are expected as a result of the BalPure® system. Based on the low Koc values for dibromochloroacetic acid (DBCAA) and tribromoacetic acid (TBAA) (3.23 L/kg and 5.3 L/kg, respectively), these substances are not expected to partition into sediment. Therefore, no sediment toxicity impacts are anticipated as a result of BalPure® ballast water discharge.

Table 6: Koc Values for Relevant Chemicals

Substance Koc (L/kg) Reference/CommentsRelevant Chemicals

2,4,6-Tribromophenol (TBP) 1,186 OECD SIDS (2003) HAAs Dibromochloroacetic acid (DBCAA) 3.23 EPI Suite v4.0 Tribromoacetic acid (TBAA) 5.3 HSDB/TOXNET (2009)

3.5 Bioavailability/biomagnification/bioconcentration The bioconcentration factor (BCF) is the concentration of a particular chemical in a tissue per concentration of chemical in water (reported as L/kg). This physical property characterizes the accumulation of pollutants through chemical partitioning from the aqueous phase into an organic phase, such as fish or other aquatic organism tissues. The BCF is the best measure of bioavailability, biomagnification, and bioconcentration. Typical BCFs for organic chemicals in fish and most aquatic invertebrates are in the 500-1,000 L/kg range. STDN's application dossier for Basic Approval (MEPC 60/2/9, section 3.5) provided data for THMs/HAAs measured in previous studies of the BalPure® system. During land-based testing, additional DBPs were analysed and three were found in measurable concentrations. Data for these three additional substances is provided below.


I:\MEPC\61\2-9.doc

The BCFs summarized in Table 7 below are all well below 2,000 L/kg (Methodology, section 6.1.4 and Table 19 of this document), indicating that they are unlikely to bioconcentrate/bioaccumulate in aquatic organisms or present a food chain risk.

Table 7: BCFs for Relevant Chemicals

Substance BCF (L/kg) Reference/CommentsRelevant Chemicals

2,4,6-Tribromophenol (TBP) 83 - 513 HSDB/TOXNET (2009) HAAs Dibromochloroacetic acid (DBCAA) 3.16 EPI Suite v4.0 Tribromoacetic acid (TBAA) 0.63 HSDB/TOXNET (2009)

3.6 Food web/population effects Population and food web effects can occur if substances have properties that tend to bioaccumulate and/or persist in the environment. Little biomagnification and persistence in aquatic and mammalian food webs is anticipated from the Relevant Chemicals. This conclusion is based on the low Koc values for DBCAA and TBAA. Additionally, the BCFs for DBCAA and TBAA are well below 500 L/kg indicating that bioconcentration/bioaccumulation is not expected to occur in aquatic organisms. Although the Koc for 2,4,6-TBP indicates a moderate potential to be transported to the sediment compartment, and a BCF range of 83-513 L/kg indicates moderate bioconcentration/ bioaccumulation potential, the concentration of 2,4,6-TBP measured in treated/neutralized ballast water was very low (maximum = 1.3 µg/L). In one case, 2,4,6-TBP had higher measurable concentration in the control than in the treated/neutralized ballast water sample (Table 23). Therefore, no food web and/or population effects can be expected as a result of these Relevant Chemicals in BalPure® ballast water discharge. 4 DATA ON MAMMALIAN TOXICITY (G9: 4.2.1.2) STDN's application dossier for Basic Approval (MEPC 60/2/9, section 3.4) provided data for DBPs measured during previous studies of the BalPure® system. During land-based testing at NIOZ, additional DBPs were analysed for and three substances were found in measurable concentrations. Mammalian toxicity data for these three additional substances is provided in Tables 8 to 16 below. Information on all previously measured substances was presented in the Basic Approval dossier (MEPC 60/2/9). In the mammalian toxicity studies reviewed, the chemicals were tested for various exposure routes (oral, dermal, etc.) and/or chemical forms, which do not necessarily reflect the actual potential exposure routes or chemical forms associated with the BalPure® system. In some cases, data could not be located in available literature. 4.1 Acute mammalian toxicity

Table 8: Acute Mammalian Toxicity Data

Substance Exposure Route SpeciesValueRange

Reference/Comments

Relevant Chemicals

2,4,6-Tribromophenol (TBP) a. Acute Oral LD50 b. Inhalation LC50 c. Acute Dermal LD50

a. Rat b. Rat c. Rabbit

a. >5000 mg/kg b. >50 mg/L/4hr c. >8000 mg/kg

a. CICADS 66 (2005) b. CICADS 66 (2005) c. CICADS 66 (2005)

HAAs

Dibromochloroacetic acid (DBCAA)

Limited toxicology data available. -- -- Richardson et al. (2007)

Tribromoacetic acid (TBAA) Limited toxicology data available. -- -- Richardson et al. (2007)


I:\MEPC\61\2-9.doc

4.2 Effects on skin and eye

Table 9: Effects on Skin and Eye

Substance Skin Eye Sensitization Reference/Comments

Relevant Chemicals 2,4,6-Tribromophenol (TBP)

Not irritating to rabbits Moderately irritating to rabbits

Skin sensitizer in guinea-pigs

CICADS 66 (2005)

HAAs


May be harmful if absorbed through skin; causes skin burns

Causes eye burnsNo data found in literature

Sigma Aldrich MSDS (2009)

Tribromoacetic acid (TBAA)

a) Corrosive; causes burning sensation, and is destructive to tissue of mucous membranes of skin

a) Destructive to tissue of the mucous membranes of eyes

b) No sensitizing effect known.

a) HSDB/TOXNET (2009)

b) Alfa Aesar MSDS (2009)

4.3 Repeated-dose toxicity

Table 10: Repeated-dose Toxicity

Substance Exposure Route Species Value Range Reference/Comments

Relevant Chemicals

2,4,6-Tribromophenol (TBP)

Oral, 48 day subchronic reproduction/developmental

Rat NOAEL 100 mg/kg body weight/day

CICADS 66 (2005)

HAAs Dibromochloroacetic acid (DBCAA)

Limited toxicology data available.

-- -- Richardson et al. (2007)


Limited toxicology data available.

-- -- Richardson et al. (2007)

4.4 Chronic mammalian toxicity

Table 11: Chronic Mammalian Toxicity

Substance Exposure Route Species Value Range Reference/Comments

Relevant Chemicals


No long term exposure studies on brominated phenols were identified.

-- -- CICADS 66 (2005)


-- -- -- No data found in literature


-- -- -- No data found in literature


I:\MEPC\61\2-9.doc

4.5 Developmental and reproductive toxicity

Table 12: Summary of Developmental and Reproductive Toxicity Data

Substance Exposure Route Species Effects Reference/Comments

Relevant Chemicals


Oral gavage, doses from 10- 3000 mg/kg/day, on gestation days 6-15

Rats

No effect on fertility, growth, or survival at doses of 200 mg/kg/d or less; Slight decrease in body weight and decrease in number of viable fetuses in 1000 mg/kg/d; NOAEL 300 mg/kg/d

CICADS 66 (2005)

HAAs


26- hour Embryo exposure; Concentrations 1500-2500 µM DBCAA

Mice

≥1500 µM increased the number of dysmorphic embryos with prosensephalic hypoplasia and eye defects were observed; ≥2000 µM caused heart dysmorphology and abnormal yolk sac morphology; ≥2500 µM caused failure to develop

Hunter et al. (2006)


a) Oral, tribromoacetic acid in drinking water, 14 days, doses up to 400 ppm TBAA b) 24-26 hours Embryo exposure; Concentrations from 0-5,000 µM TBAA

Rats

a) No effects on reproductive function, as a general toxicant, or as reproductive toxicant in males b) ≥3,000 µM TBAA induced malformation effects; nonclosure of the cranial neural tube and reduction in development of bulbus cordis

a) NTP (1998)

b) Hunter

et al. (1996)

4.6 Carcinogenicity

Table 13: Summary of Data on Carcinogenicity

Substance Carcinogen Classification Description Reference/Comments

Relevant Chemicals


a. Not classified by NTP b. Not classified by IARC c. No carcinogenicity studies

on brominated phenols were identified.

-- a. NTP (2009) b. IARC (2009) c. CICADS 66 (2005)

HAAs


a. Not classified by NTP b. Not classified by IARC

Oral exposure in drinking water; 2 years; shown to induce tumours in the livers of mice; limited details on test doses; references c. and d. identified DBCAA has cancer potential and is currently under study

a. NTP (2009) b. IARC (2009) c. Richardson et al.

(2007) d. Woo et al. (2002)


a. Not classified by NTP b. Not classified by IARC

c. Likely to be weakly carcinogenic; or carcinogenic toward a single species/target at relatively high doses based on structure-activity relationship (SAR) analysis

a. NTP (2009) b. IARC (2009) c. Woo et al. (2002)


I:\MEPC\61\2-9.doc

4.7 Mutagenicity and genotoxicity

Table 14: Summary of Data on Mutagenicity and Genotoxicity

Substance Study Results Reference/Comments

Relevant Chemicals


Tested in Salmonella/microsome preincubation assay with doses 0-333 µg/plate in Salmonella typhimurium strains TA98, TA100, TA1535, and TA1537 in presence and absence of Aroclor-induced rat/hamster liver S9; highest nonmutagenic test dose = 100 µg/plate causing a slight clearing in background lawn above this dose level

HSDB/TOXNET (2009)


Limited mutagenicity / genotoxicity studies on DBCAA available. Richardson et al. (2007)


Studies show TBAA causes mutagenic and genotoxic responses in Salmonella typhimurium strain TA100 in a) fluctuation and microsuspension assays and b) in SOS chromotest

a. Nelson et al. (2001) b. Giller et al. (1997)

4.8 Toxicokinetics

Table 15: Summary of Uptake, Absorption, and Excretion of Chemicals

Substance Study Results Reference/Comments

Relevant Chemicals


After administration of a single oral dose (4.04-5.34 mg/kg) to rats the substance was rapidly absorbed from gastro-intestinal tract; 77% excreted via urine, 2-14% eliminated in feces within 48 h; blood t1/2 was 2.03 h

OECD SIDS (2003)

HAAs


After administration of a single dose (25 µmol/kg) by iv rats, plasma elimination of DBCAA was rapid with t1/2 of 1.6 h; dose urinary recovery was 33%; Same dose given by gavage yielded plasma elimination t1/2 of 4.59 h; 40-70% of total body clearance due to metabolic clearance

Saghir & Schultz (2005)


After administration of a single dose (25 µmol/kg) by to rats, elimination of TBAA was rapid with t1/2 of 46 min; dose urinary recovery was 8%; Same dose given by gavage yielded plasma elimination t1/2 of 2.11h; 40-70% of total body clearance due to metabolic clearance

Saghir & Schultz (2005)

5 DATA ON ENVIRONMENTAL FATE AND EFFECT UNDER AEROBIC AND

ANAEROBIC CONDITIONS (G9: 4.2.1.3) 5.1 Modes of degradation (biotic; abiotic) STDN's application dossier for Basic Approval (MEPC 60/2/9, section 3.4) provided degradation data for DBPs measured during previous studies of the BalPure® system. During land-based testing at NIOZ, additional DBPs were analysed and three substances were found in measurable concentrations. Degradation data for these three additional substances is provided in Table 16 below.


I:\MEPC\61\2-9.doc

Table 16: Fate and Mode of Degradation for Relevant Chemicals

Substance Fate/Effect Under

Aerobic/Anaerobic Conditions

Modes of Degradation Reference/ Comments

Relevant Chemicals


Anaerobic: reported to dehalogenate rapidly in anaerobic sediments, half-life of approx. 4 days (CICADS 66, 2005)

Biotic: BOD 49% after 28 days in activated sludge; 82% degraded in 3-day biodegradation test in river water. Abiotic: Stable in water and not hydrolyzed regardless of pH; Direct photolysis by UV in air indicates a half life of 4.6 hours

CICADS 66 (2005)

HAAs


No data located. Biotic and Abiotic: demonstrated in study with and without microbial inhibitors in which degradation occurred in both experiments.

Hashimoto et al. (1998)


Aerobic: Used as carbon source by Pseudomonas and Nocardia when measured for halide release over 20 days (30ºC)

Biotic: microbial degradation demonstrated (Hashimoto et al., 1998) Abiotic: Atmospheric half-life = 30.9 days; not expected to undergo hydrolysis or susceptible to direct photolysis by UV

HSDB/TOXNET (2009)

5.2 Bioaccumulation, partition coefficient, octanol/water partition coefficient Based on the data presented for degradation in Table 16 above, the BCFs presented in Table 7, and Koc values in Table 6, these additional substances associated with the STDN BalPure® system have a low likelihood of bioaccumulation and partitioning to aquatic sediments. 5.3 Persistence and identification of the main metabolites in the relevant media Higher tier simulation tests are not required; this section is not applicable. 5.4 Reaction with organic matter The formation of disinfection by-products (DBPs) when using chlorine-based water disinfection methods is well known. DBP formation is a result of disinfectant reactions with natural organic matter, often measured as total organic carbon (TOC), which serves as the organic precursor (IPCS, 2000). Two major classes of DBPs that are commonly formed in chlorine treated water include THMs and HAAs (Westerhoff, et al., 2003). Both of these DBP classes have been measured in water treated with the BalPure® system. Water quality (e.g., pH, temperature, bromide level, TOC) and treatment conditions (e.g., hypochlorite dose, contact time, removal of organic matter prior to treatment) influences the formation of DBPs as a result of organic matter reactions. Further, the distribution of DBP species (up to four THM species, up to nine HAA species) is affected by the amounts of TOC, bromide and chlorine present (IPCS, 2000). Analysis to quantify DBP concentrations in ballast water discharge was conducted during land-based testing of the BalPure® system at NIOZ in 2009. Different water types (brackish, seawater) were treated/neutralized and then evaluated for DBPs. The data are presented in section 11, Tables 22 and 23. 5.5 Potential physical effects on wildlife and benthic habitats When evaluated using GESAMP Reports and Studies No. 64, the chemical and physical properties of the Active Substances and Relevant Chemicals associated with the BalPure® system as presented in STDN's application dossier for Basic Approval (MEPC 60/2/9, Tables 17 to 21) and in Table 17 below indicate that physical effects on wildlife and benthic habitats as described are not expected to occur.


I:\MEPC\61\2-9.doc

5.6 Potential residues in seafood Chemicals measured in treated ballast water discharge may pose a threat to seafood consumers only if they tend to bioaccumulate (within a trophic level, an increase in concentration of a substance in tissues due to uptake from food and sediments) or biomagnify (an increase in the concentration of a substance in the food web). The bioaccumulation criteria per the Methodology (MEPC 58/2/7, annex 4) include an octanol-water partition coefficient (Log Kow) of ≥3 or a bioconcentration factor (BCF) >2000. With the exception of 2,4,6-tribromophenol, the chemicals presented in STDN's application dossier for Basic Approval (MEPC 60/2/9) and identified in this risk assessment all have Log Kow values less than 3. However, based on the discussion for 2,4,6-TBP in section 3.4 above, and the fact that the BCFs for all substances are less than 2000, the substances are not expected to bioaccumulate, persist in the food web, or cause tainting of seafood. 6 PHYSICAL AND CHEMICAL PROPERTIES FOR THE ACTIVE SUBSTANCES,

RELEVANT CHEMICALS, AND TREATED BALLAST WATER (G9: 4.2.1.4) STDN's application dossier for Basic Approval (MEPC 60/2/9, section 3.4) provided chemical and physical property data for Active Substances and relevant DBPs measured in treated ballast water during previous studies of the BalPure® system. The Basic Approval dossier also presented the chemical and physical properties of treated ballast water. During land-based testing at NIOZ, additional DBPs were analysed and three substances were found in measurable concentrations. Available chemical property data for these three additional substances is provided in Table 17 below.

Table 17: Physical and Chemical Properties of Relevant Chemicals Physical/Chemical Properties

2,4,6-Tribromophenol Dibromochloroacetic

acid Tribromoacetic acid

Melting Point (ºC) 93.9 (CICADS 66, 2005) 151.68 (EPI Suite, 2009) 129-135 (HSDB/TOXNET,

2009) Boiling point (ºC) 244 (CICADS 66, 2005) 263.59 (EPI Suite, 2009) 245 (HSDB/TOXNET, 2009) Flammability (flash point for liquids; ºC)

Not available Not available Not available

Density (20ºC; kg/m3) 2550

(Alfa Aesar MSDS, 2009) Not available Not available

Vapour pressure/Vapour density (air=1)

4.2 x 10-2 Pa (CICADS 66, 2005)

0.692 Pa (EPI Suite, 2009)

3.73 x 10-2 Pa (HSDB/TOXNET, 2009)

Water solubility (temp; effect of pH; mg/L)/ Dissociation constant (pKa)

59 mg/L (25ºC) / pka = 6.08

(CICADS 66, 2005)

2353 mg/L (25ºC) (EPI Suite, 2009)

2.0 x 105 (25ºC) / pka = 0.72 (HSDB/TOXNET, 2009)

Oxidation/Reduction potential

ORP of natural seawater ORP of natural seawater ORP of natural seawater

Corrosivity to materials Not available Not available Not available Auto-ignition temperature (ºC)

Not available Not available Not available

Explosive properties Not explosive

(Alfa Aesar MSDS, 2009) Not available

Not explosive (Alfa Aesar MSDS, 2009)

Oxidizing properties No available Not available Not available Surface tension Not available Not available Not available Viscosity (20ºC) Not available Not available Not available Thermal stability Stable Stable Stable Reactivity to container material

Not stored; produced in situ

Not stored; Produced in situ

Not stored; produced in situ

Other knowns None None None


I:\MEPC\61\2-9.doc

7 ANALYTICAL METHODS AT ENVIRONMENTALLY RELEVANT CONCENTRATIONS (G9: 4.2.1.5)

7.1 Analysis of Total Residual Oxidants (TRO as Cl2) TRO (as Cl2) is measured at two concentration ranges (high and low) using two different analytical procedures. An iodometric titration procedure is used to measure higher range TRO (as Cl2) in the product stream exiting the BalPure® unit. Concentrations are typically 1,000 mg/L and this procedure is accepted by the American Water Works Association (AWWA) for examination of water and waste water. The procedure is only used if needed to perform diagnostics on the BalPure® system, and is included in appendix A.5. A second method is typically used for measuring lower range TRO (as Cl2) in potable water and waste water streams. A Hach colourimeter can be used to measure the TRO (as Cl2) in ballast tanks and in the deballast stream pre- and post-neutralization. The TRO (as Cl2) oxidizes triiodide ion (I3

-) to iodine (I2). The iodine and free chlorine react with DPD (N,N-diethyl-p-phenylenediamine) to form a red solution. The colour intensity is proportional to the TRO concentration. This method is also provided in appendix A.5. During land-based testing at NIOZ, TRO (as Cl2) measurements were performed with a Hach Pocket Colorimeter using Hach Method 8167 (equivalent to EPA Method 330.5). For TRO (as Cl2) in the 0.1 mg/L to 8.0 mg/L range the method has a detection limit of 0.1 mg/L. For TRO (as Cl2) in the 0.02 mg/L to 2 mg/L range the method has a detection limit of 0.02 mg/L. Residual sodium bisulfite in post-neutralization discharge samples was measured using the Hach SU-5 test kit with a titration/iodometric method. Information on the above methods is provided in Table 18 and appendix A.5. 7.2 Analysis of Disinfection By-products (DBPs) Treated ballast water samples were collected immediately after neutralization at discharge. Samples for chemical analysis were stored in proper sample containers at 4°C and transported to the designated laboratory within 24 hours of collection. Analysis of the samples occurred within the time specified under standard laboratory procedures. Qualified and accredited laboratories were utilized for analysis of treated water samples. Acceptable field-testing analytical methods were also used where applicable (e.g., TRO (as Cl2) and bisulfite). The methods used to determine DBP concentrations are listed in Table 18 below and more details are provided in appendix A.5. Information regarding quality assurance from the analytical laboratories is included in appendix A.6.

Table 18: Analytical Methods

Substance/Parameter Method Description Detection Limit TRO (as Cl2) High Range: 0.1-8.0 mg/L Low range: 0.02-2.0 mg/L

Hach Method 8167 (Equivalent to

US EPA Method 30.5) DPD, Colorimetric

High Range: 0.1 mg/L1 Low Range: 0.02 mg/L1

Bisulfate measured as sulfate ion NEN-EN-ISO 10304-1 Ion chromatography 1.0 mg/L1 Bisulfite measured as sulfite ion Hach Test Kit SU-5 Titration/Iodometric 1 mg/L1 THMs NEN-EN-ISO 15680 Gas chromatography 0.1 μg/L HAAs US EPA Method 552.2 Gas chromatography 1.0 -10 μg/L3 2,4,6-Tribromophenol US EPA Method 528 Gas chromatography 0.1-0.2 μg/L3 Monobromoacetonitrile US EPA Method 524.1 Gas chromatography 1-10 μg/L3 Bromate US EPA Method 317 Ion chromatography 10 µg/L2 Chlorate US EPA Method 300.1 Ion chromatography 50 µg/L2

1 = Method Detection Limit (MDL). 2 = Minimum Reporting Level (MRL). 3 = Limit of Detection (LOD).


I:\MEPC\61\2-9.doc

8 USE OF ACTIVE SUBSTANCE 8.1 Manner of application The BalPure® system can service ships with ballast water flow requirements from 100 to 10,000 cubic metres per hour. The BalPure® system size (electrolytic cell size) is defined by the ship's designed ballast water flow rate, but because the system has a 100% turn down capability the hypochlorite production rate is matched to the actual measured ballast flow rate. Treatment is done during ballasting operations allowing the biocide solution to be generated anywhere on the ship and injected as needed. This would apply to ships with one or multiple ballast systems or ships with multiple ballast pumps mounted within separate ballast tanks. The system can be delivered as a complete system on one skid or divided into seven separate smaller skids based on specific unit operation (pumping, hypochlorite generation, hydrogen-hypochlorite separation, bisulfite storage, bisulfite metered addition, instrument and controls, and power rectification). This allows multiple skid locations as space is available throughout the vessel. The multiple skid design is perfect for retrofit applications and the one skid system is best for new buildings. Operational control of the BalPure® system is accomplished by the main control panel attached to the hypochlorite generation unit or by a remote control panel that can be placed where most appropriate for a specific vessel. Together, the control panel and the transformer/rectifier unit comprise the main Programmable Logic Controller (PLC) control and rectification components. The control panel provides a visual display of the status of the rectifier DC voltage, current, and can alert the operator of any system alarms via the operator interface terminal (OIT). The main control panel also has a local emergency stop push button. All major equipment such as booster pumps, bisulfite metering pumps, and hydrogen blowers can be controlled from the main control panel. All monitoring equipment such as flow meters and ORP probes send the output in the form of 4-20 mA signals. All system control screens are password protected to limit operator modification to existing control logic. Only properly trained ship's crew members and BalPure® technicians will be allowed to access password-protected portions of the PLC. Other system monitoring and control devices include a magnetic flow transmitter to monitor the flow of seawater at the electrolyzer inlet, transformer temperature switches that monitor the internal temperature of the rectifier, and airflow switches that monitor the hydrogen gas dilution blowers. 8.1.1 Process flow description The BalPure® system is designed to operate automatically and has two operation processes: ballasting and deballasting. During ballasting, the biocide solution is produced and injected into the ballast water intake piping (maximum treatment dose of 15 mg/L TRO as Cl2); during deballasting sodium bisulfite is added directly to the ballast water discharge piping. The ballasting process is discussed first, followed by the deballasting process. A complete process flow diagram (PFD) of the system is included in appendix A.3. Figures 1 and 2 presented below provide a simplified overview of each phase of the treatment process. 8.1.1.1 Ballasting process – ballast water intake Very generally, during the automatic ballasting process the system will sense that adequate seawater flow is available using a flow meter in the main ballast line. Once the system is primed with sufficient water (an approximate 1-minute cycle), the transformer/rectifier will supply DC current to the electrolyzer and initiate the electro-chemical generation process.


I:\MEPC\61\2-9.doc

The hypochlorite production rate is primarily tied to the ballast flow rate and achieves the pre-set required oxidant concentration by modulating the rectifier output amperage to coincide with the ballast flow rate. This is considered the large or gross adjustment. The produced oxidants are then injected into the main ballast line. If necessary, the system may also be operated manually. Figure 1 below provides an overview of the BalPure® ballasting process.

Figure 1: Overview of the BalPure® Ballasting Process

The BalPure® system remains in a stand-by mode until the flow meter in the main ballast line detects ballast water flow and activates the BalPure® system. In the ballasting process, seawater from the sea chest will be filtered with a 40 micron stainless steel mesh filtration unit. Filtration is only required during ballast uptake. If an installation requires that seawater be taken from a source other than filtered ballast water (e.g., seawater cooling) a duplex strainer filters the seawater to remove any particles larger than 800 microns. Using an inline booster pump, the BalPure® system typically receives <1% of the total incoming ballast water flow as a side- stream off the main ballast intake line. The booster pump increases the line pressure to provide proper flow through the hypochlorite generation unit and allow solution injection into the main ballast line. An inlet flow transmitter, designed to protect the treatment system, will automatically shut down the electrical current to the electrolyzer should seawater flow through the electrolyzer drop below 80% of the design flow rate. The filtered seawater passes through a heat sink where it absorbs excess heat generated by the silicon-controlled rectifiers. The seawater then passes through a series of electrolytic cells which are connected to the transformer/rectifier. The electrolyzer units then generate the Active Substance along with hydrogen gas as a by-product. Immediately after exit from the electrolytic cells, the combined seawater/biocide solution (liquid) and hydrogen (gas) is delivered to a patented cyclone separation device where the hydrogen is separated from the biocide/seawater solution. Hydrogen gas does not enter the ballast water piping or tanks. The hydrogen gas is directed to a vent stack where it is diluted by forced air blowers and vented to the atmosphere outside of the ship. To ensure proper ventilation the vent line is serviced by two redundant blowers coupled to sail switches.


I:\MEPC\61\2-9.doc

The sail switches ensure proper air flow to dilute the air stream to <1% hydrogen. The LEL for hydrogen is 4% and the operating limitations on the system provide a four-fold safety factor. The PLC monitors proper air flow and will shut down the system, resulting in an alarm condition if proper ventilation air flow is not maintained. Both blower motors and sail switches are tied to an automatic emergency shutdown for the production unit in the unlikely event all four devices fail. For additional safety, a hydrogen sensor with alarm will be provided in the immediate proximity of the hypochlorite generation unit. For optimum hypochlorite generator performance (i.e. to produce 90% of theoretical values), the inlet seawater temperature should be at least 15°C and have a sodium chloride concentration of at least 19,000 mg/L (19 PSU). However, the BalPure® system can operate at temperatures as low as 5°C and salinities of 10,000 mg/L (10 PSU). Less optimal conditions result in a slightly decreased hypochlorite production rate to 20% less than optimum. The hypochlorite production rate of the electrolytic cells is directly related to chloride ion availability and mobility. The salinity (PSU) of the source water determines the chloride ion availability, while temperature affects chloride ion mobility. These parameters impact the operating voltage of the electrolytic cells. For instance, when chloride ions are abundantly available and mobile, the operating voltage and hypochlorite production within the electrolytic cells will be optimum. If chloride ions and/or temperature are low, the operating voltage increases proportionally and hypochlorite production will be reduced. The BalPure® system is designed to adjust the hypochlorite production rate by monitoring the operating voltage of the electrolytic cells and increasing the rectifier operating amperage to sustain the TRO (as Cl2) concentration required to treat incoming ballast water (maximum treatment dose of 15 mg/L TRO (as Cl2)). These adjustments are small and used to control the hypochlorite production rate to a narrow range (fine tuning). The BalPure® system can also operate at higher salinities, such as ≥30,000 mg/L (≥30 PSU). In this case where chloride ions are highly abundant, the generator will only produce the design hypochlorite concentration. This is because higher salinity does not increase the efficiency or the production rate of the electrolytic cell beyond the optimum. As mentioned above, the BalPure® unit is designed to operate in seawater with salinities ≥10 PSU. However, with minor process adjustments, the BalPure® unit can operate in fresh water. For operation in fresh water, an alternative supply of saltwater input to the BalPure® unit is required. Because the BalPure® unit requires only a small amount (<1% of total incoming ballast flow), existing onboard alternative sources of seawater can be used. By monitoring the operating voltage of the electrolytic cells and adjusting the amperage applied, the PLC programming of the BalPure® system guarantees effective operation over a wide temperature and/or PSU range. 8.1.1.2 Deballasting and neutralization process – ballast water discharge In the deballasting process, the final phase of treatment, residual chlorine present in the ballast tank is neutralized (de-chlorinated) with sodium bisulfite just prior to discharge from the ship. The holding time that treated ballast water needs to be retained before discharge is discussed further in section 11.5. Neutralization is accomplished using ORP technology and bisulfite metering pumps. Two ORP probes are used; one confirms the presence of residual oxidant exiting the ballast tanks, while the second probe monitors and controls the addition of sodium bisulfite to neutralize TRO (as Cl2) during ballast water discharge. The first ORP probe is located upstream of the bisulfite injection point, and the second ORP probe is located downstream of the bisulfite injection point. During deballasting, the main ballast line filter used during the ballast uptake process is not required and is by-passed automatically.


I:\MEPC\61\2-9.doc

Figure 2: Overview of the BalPure® Deballasting Process ORP technology is based on the potential measurement between two dissimilar metals within the ORP probe. ORP is not a quantitative measure of oxidant or bisulfite concentration but rather a qualitative measure that confirms the presence of oxidizing or reducing agents in the water. ORP technology allows for an in-line rapid determination of the amount of oxidation (or reduction) potential of the water. ORP is measured in millivolts (mV) and increases or decreases in direct relation with the oxidant residual in the water (Kelley, 2004). For example, natural waters have an ORP reading in the range of 100-150 mV. When hypochlorite is added to water, the ORP reading increases with an increase in hypochlorite dosage. Similarly, when a de-chlorinating agent such as sodium bisulfite is added to the water, the ORP reading will drop to a low mV level (Kelley, 2004). In this way, ORP measurements provide a practical and efficient method of optimizing water disinfection and controlling neutralizing agent dosage. The BalPure® BWMS uses ORP readings as a general indicator of the presence/absence of residual oxidants, but STDN believes that ORP measurements are not sufficient to accurately quantify TRO levels. The first BalPure® ORP probe confirms that there are residual oxidants in the treated water being discharged from the ballast tank. In principle, this confirmation of oxidants translates to affirmation that there are no living organisms in the discharge. The ORP reading at the first probe is typically between 700 and 800 mV and is used as an indicator that residual oxidant is present. The second ORP probe measures the ORP value of the deballast stream after bisulfite has been added to the discharge piping. This ORP value is typically <200 mV (similar to natural water) and is used to control the addition of bisulfite solution to ensure complete neutralization of residual oxidants. The low ORP value is an indicator that the residual oxidant has been neutralized. It is important to note the data from land-based testing confirms that the TRO (as Cl2) concentration is below detection levels post neutralization with sodium bisulfite. This verifies that residual oxidant neutralization with sodium bisulfite at discharge was effective at all ballast discharges and ensures that the maximum allowable discharge concentration (MADC) of <0.20 mg/L TRO (as Cl2), as recommended by the GESAMP-BWWG, is not exceeded (see section 11.1).


I:\MEPC\61\2-9.doc

With the mixing effect of the deballast water pump and the discharge pipe, the bisulfite will easily react with any free oxidants that may be present. As a fail-safe measure, the BalPure® system is fitted with two redundant bisulfite metering pumps; one pump is the primary dosing pump and the second is an on-line ready spare. In the unlikely event of an equipment failure, the neutralization process can also be operated manually to ensure that residual oxidants are neutralized during discharge. This is accomplished by the operator manually setting the bisulfite metering pump rate in the PLC. The delivery rate of bisulfite can be set based on the stoichiometric ratio of oxidant to neutralizing agent at 1:2 parts. Thus the sulfite metering pumps are set based on the maximum discharge ballast water flow rate and hypochlorite concentration. 8.1.2 Chemical storage and handling The only chemical that will require onboard storage and handling is the neutralizing agent, sodium bisulfite. The only potential activities that may require "handling" of bisulfite is either system maintenance or chemical re-supply operations. During the previous evaluation undertaken for STDN's application dossier for Basic Approval (MEPC 60/2/9, section 8.1.3) the Group recognized all risk evaluations of the ship and personnel including consideration of the storage, handling, and application of sodium bisulfite as sufficient and considered that this BWMS will not pose unacceptable risk regarding the exposure to bisulfite and treated ballast water (MEPC 60/2/16, sections 7.1 and 7.2). No changes have been made in the full-scale operation of the BalPure® system since the submittal of STDN's application dossier for Basic Approval. Therefore the previous evaluations made by the Group remain valid. 8.1.3 Various procedures and management measures Technicians and ship's crew will be sufficiently trained in procedures and management measures concerning fire, accidental release, marine environmental release, and controlled release according to methods outlined in STDN's application dossier for Basic Approval (MEPC 60/2/9, section 8.1.3). During the previous evaluation undertaken for STDN's application dossier for Basic Approval, the Group recognized all working principles of the BalPure® system as sufficient and considered that this BWMS will not pose unacceptable risk to the health of the crew and the general public (MEPC 60/2/16, sections 7.1 and 7.2). No changes have been made in the full-scale operation of the BalPure® system since the submission of STDN's application dossier for Basic Approval. Therefore the previous evaluations made by the Group remain valid, and no change to procedures and management measures are presented in this dossier. 9 MATERIAL SAFETY DATA SHEETS (G9: 4.2.7) Information on the hazard classifications of Active Substances and Relevant Chemicals was presented in sections 9.1, 9.2 and 9.3 of STDN's application dossier for Basic Approval (MEPC 60/2/9) and all Material Data Safety Sheet (MSDS) were provided. In this Final Approval dossier, all MSDSs, including the three additional DBPs (dibromochloroacetic acid, tribromoacetic acid, and 2,4,6-tribromophenol) that were measured in treated ballast water during land-based testing are provided in appendix A.4. It is important to note that these DBPs may be formed in situ at low concentrations as a result of ballast water treatment. Concentrated solutions of these DBPs will never be transported or utilized on board ship. As such, typical MSDSs for these low level (µg/L range) substances do not accurately represent the potential hazards in treated ballast water. Determining the hazards for these substances at µg/L levels is not feasible when the MSDSs reflect hazards for much higher concentrations.


I:\MEPC\61\2-9.doc

10 RISK CHARACTERIZATION AND ANALYSIS 10.1 Screening for persistence, bioaccumulation, and toxicity (G9: 5.1)

Table 19: PBT Criteria Evaluation

Substance

Persistence (P) Half-life: >60 Days in Marine Water (MW), or >40 Days in Fresh Water (FW), or >180 Days in Marine Sediment (MS), or >120 Days in Fresh Water Sediment (FWS) (YES or NO)

Bioaccumulation (B) BCF>2,000 Log Kow ≥ 3 (YES or NO)

Toxicity (T) Chronic NOEC < 0.01 mg/L (YES or NO)

Active Substances

Hypochlorous acid No <2 hours (EU RAR Sodium hypochlorite, 2007) FW

No Log Kow = -0.87

No Lowest NOEC = 0.048 mg/L (Crustacean)

Hypobromous acid

No<2 hours (No specific data located, based on hypochlorous acid) FW

NoBCF = 3.2 Log Kow =-0.78 Koc = 13.2

No Lowest EC50 = 1.0 mg/L (Algae)

Relevant Chemicals

Bromate No 15 days (EPI Suite, 2009) Water type not specified; Temperature = 25°C;

NoBCF = 3.2 Log Kow = -4.6 Koc = 31.8

No Lowest EC50 = 13.6 mg/L (Algae)

2,4,6-Tribromophenol No 1.21 days (CICADS 66, 2005) FW

NoBCF = 83 - 513 Log Kow =3.89 Koc = 1,186

No Lowest NOEC = 0.10 mg/L (Crustacean)

THMs

Chloroform

No0.25 days (Yang, 2001); Not expected to adsorb into sediment (HSDB/TOXNET 2009) MW

No BCF = 2.9-10.35 Koc = 153-196

No Lowest NOEC = 1.5 mg/L (Fish)

Bromoform No 0.3 days (EU RAR Sodium hypochlorite, 2007) FW

NoBCF = 14 Log Kow 2.4 Koc = 35

No Lowest NOEC = 8.5 mg/L (Fish)

Dichlorobromomethane No 0.08 days (EU RAR Sodium hypochlorite, 2007) FW

NoBCF = 7 Log Kow = 2 Koc = 35-251

No Lowest LC50 = 67.4 mg/L (Fish)

Dibromochloromethane No 0.11 days (EU RAR Sodium hypochlorite, 2007) FW

No BCF = 9 Koc = 84

No Lowest LC50 = 34mg/L (Fish)

HAAs

Bromochloroacetic acid

No2.7 days (Hashimoto et al., 1998); Not expected to adsorb to suspended solids or sediment. (HSDB/TOXNET 2009) FW; Temperature = 20°C

No BCF = 3.2 Log Kow = 0.61 Koc = 1.9

Data not located.

Monochloroacetic acid

No3.58 Days (Hanson et al., 2002) FW; Temperature = 18.7-23.8°C; pH = 7.7-8; DO= 92-11.3 mg/L; Alkalinity = 158-171 mg/L

NoBCF = 3.2 Log Kow = 0.22 Koc = 31

No Lowest NOEC = 32 mg/L (Crustacean)

Monobromoacetic acid

No3.2 days (Hashimoto et al., 1998); Not expected to adsorb to sediment. (HSDB/TOXNET 2009) FW; Temperature 20°C

NoBCF = 3.2 Log Kow = 0.41 Koc = 39.8

No Lowest NOEC = 1.4 mg/L (Algae)


I:\MEPC\61\2-9.doc

Substance

Persistence (P) Half-life: >60 Days in Marine Water (MW), or >40 Days in Fresh Water (FW), or >180 Days in Marine Sediment (MS), or >120 Days in Fresh Water Sediment (FWS) (YES or NO)

Bioaccumulation (B) BCF>2,000 Log Kow ≥ 3 (YES or NO)

Toxicity (T) Chronic NOEC < 0.01 mg/L (YES or NO)

Dichloroacetic acid No 5.4 days (Hashimoto et al., 1998) FW; Temperature 20°C

NoBCF = 0.3 Log Kow = 0.92 Koc = 75

No Lowest EC50 = 23 mg/L (Crustacean)

Dibromochloroacetic acid No 3.67 days (Hashimoto et al., 1998) FW; Temperature 20°C

NoBCF = 3.16 Log Kow = 1.62 Loc = 3.23

Data not located.

Dibromoacetic acid No3.2 days (Hashimoto et al., 1998) FW; Temperature 20°C

NoBCF = 0.17 Koc = 1.5

No Lowest LC50 = 69 mg/L (Fish)

Tribromoacetic acid No 4.2 days (Hashimoto et al., 1998) FW; Temperature 20°C

No BCF = 0.63 Koc = 5.3

Data not located.

Trichloroacetic acid No 14.37 days (Ellis et al., 2001) FW; Temperature 20°C

No BCF = 0.1-1.7 Log Kow = 1.33 Koc = 130

No Lowest NOEC = 100 mg/L (Algae)

Other Chemicals

Sodium bisulfite

No0.000747 days (Yan et al., 2007) Salinity not specified; Temperature = 25°C; pH = 4; DO = 7.86 mg/L

NoBCF = 3.2 Log Kow = -6.85 Koc = 2.21

No Lowest LC50 = 81 mg/L (Crustacean)

Based on an evaluation of the available data presented in the Table, none of the substances meet all three criteria to be considered as PBT substances. 10.1.1 Persistence (G9: 5.1.1.1) Using the half-life criteria specified in Procedure (G9) (resolution MEPC.169(57)) of more than 40 days in fresh water, 60 days in marine water, marine sediment persistence more than 180 days, or fresh water sediment persistence greater than 120 days, an evaluation of the environmental persistence data for Active Substances, Relevant Chemicals, and other chemicals indicates that none of the substances are likely to be persistent in the environment. 10.1.2 Bioaccumulation (G9: 5.1.1.2) Using the criteria specified in Procedure (G9) (resolution MEPC.169(57)) of either BCF >2000 or Log Kow ≥3, an evaluation of the available data for Active Substances, Relevant Chemicals, and other chemicals indicates that the substances have low potential for bioaccumulation. Although 2,4,6-tribromophenol has a reported Log Kow of 3.89 (OECD, 2003) which is slightly above 3, the BCF is well below the criteria of 2000. 10.1.3 Toxicity tests (G9: 5.1.2.3) Due to the limited availability of chronic aquatic toxicity data for some substances, the evaluation for toxicity was based on all available data. In some cases, acute endpoint data was utilized when available. For substances where no toxicity data points could be located, the data with respect to the other two PBT criteria is available and therefore, a PBT determination can still be made.


I:\MEPC\61\2-9.doc

With the exception of substances having limited toxicity data, all other substances did not have chronic NOEC or E(L)C50 values <0.01 mg/L. Therefore, the criterion to be considered a toxic substance is not met. 11 EVALUATION OF THE TREATED BALLAST WATER (G9: 5.2) Numerous studies have been conducted on the BalPure® system to evaluate various aspects of treated/neutralized water. In addition to the aquatic ecotoxicity data presented in section 3, data on TRO (as Cl2), water quality parameters (e.g., salinity, pH, particulate organic carbon (POC), dissolved organic carbon (DOC), total suspended solids (TSS)), and disinfection by-products (DBPs) have been measured and recorded. Treated ballast water samples were collected immediately after neutralization at discharge. Samples were stored in proper sample containers at 4°C and transported to the designated lab within 24 hours of collection. Analysis of the samples occurred within the time specified under standard laboratory procedures. 11.1 Total Residual Oxidants During testing of the BalPure® system conducted at the NIOZ land-based testing facility in 2009, hypochlorite treatment was applied and TRO (as Cl2) was measured at different time intervals after treatment. TRO (as Cl2) was evaluated immediately after treatment (T0), 1 day after treatment (T1), 5 days after treatment (T5), and after neutralization on day 5 (post neutralization). The average TRO (as Cl2) at each time interval for 6 low salinity test cycles and 6 high salinity test cycles are shown below in Table 20.

Table 20: Average Oxidant Values (mg/L) in Treated Ballast Water

Test Cycle Salinity

BalPure® Treatment Dose

(Set Point)

TRO (as Cl2) @ T0

TRO (as Cl2) @ T1

TRO (as Cl2) @ T5

TRO (as Cl2) Post Neutralization

Low Salinity (23.68 PSU)

14.83 14.80 7.19 4.46 Not Detected

High Salinity (33.90 PSU)

13.17 12.36 4.97 2.98 Not Detected

Not Detected = <0.02 mg/L (Hach Method 8167)

The data indicate that the applied hypochlorite treatment dose is confirmed following treatment by the TRO (as Cl2) measurement at T0. As expected, the TRO (as Cl2) concentration decayed quickly (TRO (as Cl2) at T1) following treatment, and then more slowly after the demand of the water was met (TRO (as Cl2) at T5). Lastly, the data confirm that the TRO (as Cl2) concentration is below detection levels post neutralization with sodium bisulfite. This verifies that residual oxidant neutralization with sodium bisulfite at discharge was effective at all ballast discharges and ensures that the maximum allowable discharge concentration (MADC) of <0.20 mg/L TRO (as Cl2), as recommended by the GESAMP-BWWG, is not exceeded. Also, the features of the BalPure® system pertaining to neutralization and ORP control (section 8.1.1.2) ensure sodium bisulfite addition for effective neutralization. 11.2 Water quality parameters Water quality data was collected by the researchers at NIOZ during testing of the BalPure® system in 2009. Land-based tests were done using natural waters taken from the Wadden Sea, with minimal amendments to alter water quality. Variation in water quality is caused by tidal and wind influences, which results in variable concentrations of TSS and POC. The average values for brackish water samples (low salinity) and marine water samples (high salinity) are presented in Table 21 below. Data is as provided to STDN by NIOZ.


I:\MEPC\61\2-9.doc

Table 21: Water Quality Data

Parameter Control Treatment

Brackish Water (low salinity)

T0 T5 T0 T5

pH 8.33 8.25 8.52 7.90

Temperature (˚C) 11.45 11.35 11.05 11.23

Salinity (PSU) 24.05 24.05 23.68 23.68

POC (mg/L) 16.34 25.79 15.31 5.62

DOC (mg/L) 2.85 2.69 3.00 3.50

TOC (mg/L)* 19.19 28.48 18.31 9.12

TSS (mg/L) 46.83 16.20 41.91 12.58

DO (%) 111.05 91.87 138.08 119.60

Marine Water (high salinity)

T0 T5 T0 T5

pH 8.10 7.92 8.33 7.50

Temperature (˚C) 14.56 14.30 14.66 14.48

Salinity (PSU) 33.40 33.40 33.90 33.90

POC (mg/L) 7.63 4.96 8.07 3.68

DOC (mg/L) 3.04 2.77 3.31 3.69

TOC (mg/L)* 10.67 7.73 11.38 7.37

TSS (mg/L) 18.55 10.66 19.60 8.38

DO (%) 92.25 68.11 126.94 106.02 * TOC = DOC +POC

The physical and chemical properties of treated ballast water were provided in STDN's application dossier for Basic Approval (section 6 and Table 21). 11.3 Chemical analysis of disinfection by-products in treated ballast water During land-based testing of the full commercial BalPure® system conducted at NIOZ in 2009, treated ballast water discharge was evaluated for disinfection by-products (DBPs). DBP concentrations from three low salinity and three high salinity test cycles were quantified. During each test cycle, samples were collected on Day 1 (24 hours post-treatment) and Day 5 (5 days post-treatment), resulting in 6 separate analyses for each water type. All samples were neutralized at discharge to ensure that the data represent full-scale employment (hypochlorite treatment + sodium bisulfite neutralization) of the ballast water system. Table 22 presents DBP data for samples collected on Day 1 (24 hours post-treatment) for low salinity (23.68 PSU) and high salinity (33.90 PSU) test cycles. Table 23 presents DBP data for samples collected on Day 5 (5 days post-treatment) for low salinity (23.68 PSU) and high-salinity (33.90 PSU) test cycles.

Table 22: Post-Treatment Disinfection By-product Concentrations – Day 1

(Table 22 provided in confidential dossier.)


I:\MEPC\61\2-9.doc

Table 23: Post-Treatment Disinfection By-product Concentrations – Day 5


For THM analysis, three of the four substances tested in both high and low salinity cycles indicate that bromoform, DBCM and DCBM had measurable concentrations in treated/ neutralized discharge. Chloroform was not detected in any sample analysed. Low levels (0.15-2.7 µg/L) of bromoform were also present in untreated (control) water for both high and low salinity test cycles. As previously mentioned in section 2.3.1, bromoform is a chemical naturally produced by many algal species such as Macrocystis pyrifera and Corallina pilulifera (HSDB/TOXNET Tribromomethane, 2009; Goodwin et al., 1997; Ohsawa et al., 2001). For HAA analysis, MBAA, DBAA, TBAA, MCAA, DBCAA and BCAA had measurable concentrations in low salinity test cycles. In high salinity test cycles, MBAA, DBAA, TBAA, DCAA, DBCAA and BCAA were present with measurable concentrations. In brackish water test cycles, 2,4,6-TBP had measurable concentrations in treated water (0.67-1.3 µg/L), as well as in untreated (control) water (0.44-1.80 µg/L). The background concentration was consistently significant (see discussion on modelling 2,4,6 TBP; section 11.8.2). The substance, 2,4,6-TBP, has been documented to occur in natural environments, either as a pollutant from anthropogenic sources (wood industry, antifungal use) or from natural production by marine benthic organisms (OECD SIDS, 2003). Therefore, the presence of 2,4,6-TBP in control samples is consistent with the findings of other studies. Sodium bisulfite is added in excess to ensure complete neutralization of remaining oxidants in ballast water discharge. In one sample, a maximum concentration of 8 mg/L sulfite ion was measured in ballast water discharge. The median concentration measured for all samples was 5 mg/L. Bisulfite was measured immediately after injection into the discharge pipe to ensure that an excess is present, which the data confirms. The bisulfite concentration ranges in low and high salinity test cycles was 2.0-8.0 mg/L and 2.0-3.0 mg/L, respectively. As discussed above in section 2.4.1, bisulfite is converted to bisulfate, and ultimately to sulfate ion during chemical reduction. Both the control and treated waters were analysed to verify that the excess of bisulfite does not significantly increase the sulfate ion. The data in Tables 22 and 23 show the difference between the measured sulfate ion in control and treated samples is insignificant (<10%). Considering the typical concentration of sulfate ion in natural waters, the increase of sulfate ion in ballast water discharge is not considered environmentally relevant. GESAMP-BWWG agrees that sulfate ion is ubiquitous in the marine environment and considered that the discharge or environmental concentrations of sodium bisulfite should not pose unacceptable environmental risks (MEPC 60/2/16, section 3.1.1.4). As discussed in section 5.4, formation of DBPs is dependent upon several variables such as Active Substance dose, contact time, pH, and organic matter content. Data from the BalPure® discharge reflects those trends. For instance, the highest THM and HAA formation occurred in the brackish water treatment where the hypochlorite dose, organic matter, and suspended solids were the highest. Similarly, the lowest THM and HAA formation occurred in the seawater study where the hypochlorite dose, organic matter and suspended solids were at lower levels.


I:\MEPC\61\2-9.doc

11.4 Ecotoxicity testing of treated ballast water, land-based testing Samples of water treated with the BalPure® system during land-based testing at NIOZ were evaluated by Grontmij|AquaSense (The Netherlands) and Chemex Environmental International Ltd. (UK). Ecotoxicity data is presented in section 3 and Tables 3 and 5. Toxicity tests were conducted with six taxonomic groups including bacteria, algae, crustaceans, rotifers, mollusks and fish in two water types; low salinity (23.68 PSU) and high salinity (33.90 PSU). The samples for ecotoxicity testing were collected on Day 1 and on Day 5 after ballast water treatment/neutralization. All tests were performed according to laboratory protocols based on internationally recognized ISO, OECD, ASTM, and Parcom/OSPAR guidelines. Documentation of quality assurance for Grontmij|AquaSense and Chemex Environmental International Ltd. are included in appendix A.6. The acute testing data indicate that 100% whole effluent did not have toxic effects on five of the six species tested (Table 3). One of four acute tests with the crustacean, Acartia tonsa, in the low salinity sample collected 24 hours after treatment (Day 1) indicated there was a moderate effect in the two highest test concentrations (100% and 32% sample). This is reflected by a NOEC of 18% volume (Grontmij|AquaSense, 2009). It is important to note that during tests with Acartia moderate effects were also observed in a control (untreated NIOZ harbour water) sample (EC20 of 58% sample) (Grontmij|AquaSense, 2009). A possible explanation for the moderate toxic effect in the treated and control samples is the presence of Phaeocystis, an algal species known to be abundant in the NIOZ harbour and to excrete allelochemicals. Although not a highly toxic algal species, the massive blooms often result in ecosystem effects. Residual grazing deterrents and growth retarding compounds were likely to be the main cause of the effects on Acartia tonsa in the control and treated ballast water test cycles (Veldhuis and Kools, 2009). Phaeocystis was present in the NIOZ harbour during the whole season of testing but was particularly abundant during the first low salinity test cycles (Veldhuis and Kools, 2009), which corresponds with when effects were observed in the Acartia tests. All other acute (Table 3) and chronic (Table 5) tests with Acartia, as well as the five other species tested, demonstrate that 100% whole effluent sample had no toxic effect, indicating that the NOEC of 18% (Table 3) is an outlying data point. 11.5 Determination of holding time Because the BalPure® system employs an oxidant neutralization step at discharge, there is no specified holding time required before treated water can be safely discharged to the environment. Using ORP technology (discussed in section 8.1.1.2) sodium bisulfite can be added to the ballast water discharge line at the concentration needed to ensure that all residual oxidants are neutralized, regardless of when deballasting needs to occur. The neutralization reaction between sodium bisulfite and TRO (as Cl2) is extremely quick and occurs well before water is discharged from the ballast tank. Therefore, the holding time that treated ballast water needs to be retained is determined by the amount of contact time required to properly disinfect and inactivate AIS. Studies performed at the University of Washington determined that at low (<2 mg/L) TOC concentrations 98-100% of organisms were inactivated 5 hours after hypochlorite treatments of 2.95-3.71 mg/L (Herwig et al., 2006). Studies performed at the Naval Research Laboratory determined BalPure® hypochlorite treatments of 14-16 mg/L effectively inactivated >99% of organisms 24 hours after treatment in seawater with 10 mg/L TOC (NRL, 2007). These studies indicate that even if a ship had to discharge treated ballast water shortly after hypochlorite treatment, the number of live organisms would be very limited and no aquatic toxicity could be expected from neutralized discharge.


I:\MEPC\61\2-9.doc

11.6 Reaction with organic matter As discussed in STDN's application dossier for Basic Approval (MEPC 60/2/9) and in section 5.4, chlorine-based water disinfection can result in the formation of THMs, HAAs or other DBPs. The concentration of these DBPs resulting from disinfection is based on the Active Substance dose as well as the parameters of the water being treated (e.g., organic matter, pH, temperature), with the amount of organic matter being of significant influence. Lower organic matter levels will reduce the amount of DBPs formed. The BalPure® system filters incoming ballast water, which will reduce the amount of organic matter in treated water, and thereby, the DBPs formed during disinfection. The levels of DBPs in treated/neutralized ballast water are presented in section 11.3, Tables 22 and 23. Because TRO (as Cl2) is neutralized during discharge, and also considering that DBPs are present in the µg/L range in ballast water discharge, no additional reactions with organic matter are expected to occur once ballast water is released into the environment. 11.7 Characterization of degradation route and rate (G9: 5.3.5) Table 16 above includes degradation data for substances not previously analysed for or presented in STDN's application dossier for Basic Approval (MEPC 60/2/9) and which were measured in treated/neutralized ballast water during land-based testing. Degradation information on all previously measured substances was presented in the Basic Approval dossier (MEPC 60/2/9, Table 16). The data indicate that both biotic and abiotic modes of degradation are possible for 2,4,6-TBP, and degradation in anaerobic sediments is rapid (half-life of ~4 days (CICADS 66, (2005)). In an activated sludge study, 2,4,6-TBP reached 49% of the biological oxygen demand (BOD) in 28 days (CICADS 66, 2005). However, in 3-day biodegradation tests in river water, 2,4,6-TBP (10 mg/L) was degraded by 82% (CICADS 66, 2005). Based on the low concentrations (maximum = 1.3 µg/L) of 2,4,6-TBP present in ballast water discharge and the rapid degradation of 2,4,6-TBP documented in water and sediment, there is low potential for persistence and impacts to the aquatic environment from 2,4,6-TBP. Degradation data for DBCAA and TBAA is limited. However, the stability of HAAs in water was evaluated in Tokyo Bay (Hashimoto, et al, 1998). In river water, DBCAA (20 µg/L) degraded to <0.069 µg/L after a 30-day incubation. Similarly, TBAA (10 µg/L) degraded to <0.070 µg/L after a 30-day incubation. The studies also showed that abiotic and biotic degradation occurred for HAAs. This was demonstrated by samples that were allowed to degrade with and without microbial inhibitors. Prior work with HAA stability also indicates that bacterial enzymes (haloacid dehalogenases) decompose HAAs (Hashimoto et al., 1998). As such, these substances are not expected to persist in the environment. 11.8 Prediction of discharge and environmental concentrations (G9: 5.3.8) To determine the predicted environmental concentrations (PECs) of substances associated with STDN BalPure® system, the MAMPEC modelling program was utilized. Deltares, located in the Netherlands, developed the MAMPEC model and was commissioned by STDN to derive PEC values. Appropriate emission scenarios were developed in consultation with Deltares' hydrodynamic modelling experts. Due to Deltares' continued research and development of the MAMPEC model, it was determined that a research version of MAMPEC was the best to utilize for this assessment. This research version, not currently available on the internet, incorporates the latest developments and is based on the last officially released MAMPEC version (2.5.04) (Deltares, 2010).


I:\MEPC\61\2-9.doc

11.8.1 Hydrodynamic modelling approach It is important to note that the MAMPEC model was designed as a rapid assessment tool to evaluate the continuous leaching of anti-fouling products from ships' hulls. To determine PEC values, MAMPEC employs a "steady state" approach, which implies that an equilibrium concentration of a substance in a harbour is calculated under constant environmental conditions as a result of continuous and constant emission (Deltares, 2010). This is very different from the intermittent and non-continuous nature of ballast water discharges. Therefore, the steady state approach in MAMPEC simulations likely overestimates PEC values and provides a highly conservative, worst-case scenario. In an effort to predict conservative environmental concentrations of substances present in BalPure® treated/neutralized ballast water discharge, two different environmental schematizations were employed. These two scenarios are labelled "Environment A" and "Environment B". Environment A applied a simulation environment defined by the "OECD-EU commercial harbour" configuration which is the recommended scenario for risk assessment of antifoulants in all OECD regulatory frameworks. The "OECD-EU commercial harbour" is a downsized version of the whole Port of Rotterdam being four times smaller than the Rotterdam harbour, with a surface area of 5 km2 (Deltares, 2010). To determine the amount of ballast water discharge in the harbour, the ballast water discharge estimated by Van Niekerk (2008) was used. According to the Van Niekerk study, a monthly estimate of ballast water discharge amounted to 2.76 million m3 for the entire Port of Rotterdam. The flow rate specified in the MAMPEC model was derived using the 2.76 million m3 ballast water discharge volume, which equalled 1.10 m3/s. Given that the "OECD-EU commercial harbour" configuration is four times smaller than the whole Port of Rotterdam, using this ballast water discharge flow rate representative of the entire Rotterdam harbour constitutes a highly conservative worst case scenario. Environment A also used the standard exchange flow of 5.1x107 m3 per tidal cycle, which equals 68% of the total harbour water volume. Environment B modelled substances present in BalPure® treated/neutralized ballast water discharge using the same "OECD-EU commercial harbour" configuration and ballast discharge flow rate as Environment A, with the exception that a lower tidal exchange flow rate was used. This alternative environment had smaller density driven water exchange in the harbour. The harbour water exchange resulting from density differences was assigned a value of 0.02 kg/m3 in Environment B compared to the value of 0.4 kg/m3 assigned in Environment A. The exchange flow rate for Environment B was 2.1x107 m3 per tidal cycle, which equals only 28% of the total harbour water volume. Therefore, the Environment B emission scenario and the resulting PEC values can be considered extremely conservative. The harbour is smaller, the amount of water exchange (flushing) is greatly reduced, yet the ballast water discharge rate is based on statistics for the whole Port of Rotterdam. 11.8.2 Determination of substance concentrations and decay rates for modelling input Due to the oxidant neutralization step during ballast water discharge, the TRO (as Cl2) in treated water discharge will not be present in measurable concentrations. Therefore, the discharge and predicted environmental concentration (PEC) for TRO (as Cl2) is 0 mg/L. The relevant substance concentrations used for MAMPEC modelling were based on analysis of treated/neutralized ballast water samples drawn from the full-scale land-based test set-up of the BalPure® system. For the purposes of modelling, the highest median concentration for each Relevant Chemical with a measurable concentration (as presented in the Table 24), regardless of the water type (high or low salinity water), was entered into the model.


I:\MEPC\61\2-9.doc

This approach ensured a representative evaluation of potential environmental risk as a result of DBPs in ballast water discharge. For each substance analysed, the median concentration was calculated by using all available data to derive the most representative concentration for each salinity. For example, six separate analyses for bromoform were performed in the low salinity test cycles, and the median result of these 6 data points was derived. Because the data were not normally distributed and the data set was relatively small (n<10) for each substance, median values are the most representative and appropriate. Additionally, due to the layers of conservatism already built into the environmental risk assessment (discussed in section 12.3) the use of median values allowed for a more realistic assessment of discharge DBP concentrations. The median result, as well as the minimum and maximum concentration, for each Relevant Chemical is presented in Table 24 below. In the case of TBAA, MCAA (low salinity cycles) and DCAA (high salinity cycles) the average concentrations, rather than medians, were calculated and used for MAMPEC modelling. This was done because a median calculation would not have been representative of all data points. For these substances, the average is a more representative result. The results are presented in Table 24 below. Residual sodium bisulfite present in neutralized discharge was also measured; the data is presented in Table 24 below.

Table 24: Median Disinfection By-product Concentrations


As discussed in sections 2.3.2 and 11.3, 2,4,6-TBP has been documented to occur naturally. The median 2,4,6-TBP concentration present in control samples was significant when compared to the median concentration in treated water samples. The difference between background 2,4,6-TBP levels in the control water and treated water is 0.21 µg/L. As such, the concentration modelled in MAMPEC was derived by subtracting the median control concentration from the median treated ballast water concentration (Table 24). This approach achieves a more realistic 2,4,6-TBP concentration in ballast water discharge as a result of the BalPure® system and excludes 2,4,6-TBP apparently present in natural water. As presented in section 2.3.1 above, bromoform is also a naturally produced chemical. Because median bromoform concentrations in control samples were at low levels and not considered significant when compared to treated sample median concentrations, the control concentration was not subtracted from the treated ballast water samples for the purposes of MAMPEC modelling. Sodium bisulfite is added to ballast water discharge at an excess (median = 5 mg/L; maximum = 8 mg/L) to ensure that all residual oxidants have been reacted (Tables 23 and 24). As such, ballast water discharge may contain sodium bisulfite. To demonstrate that any excess sodium bisulfite that may be present in ballast water discharge does not present environmental risk, the median bisulfite concentration measured was evaluated with MAMPEC. Table 25 below summarizes the Relevant Chemical concentrations and the associated half-lives in water used for MAMPEC modelling.


I:\MEPC\61\2-9.doc

Table 25: Summary of Relevant Chemical Concentration and Half-life for Modelling

Substance Highest Median

Discharge Concentration

Half-life in Water

Reference/Comments

Disinfection By-products (DBPs)

Bromoform 500 µg/L 0.3 days EU RAR Sodium Hypochlorite (2007)

Dichlorobromomethane 0.98 µg/L 0.08 days EU RAR Sodium Hypochlorite (2007)

Dibromochloromethane 21 µg/L 0.11 days EU RAR Sodium Hypochlorite (2007)

Bromochloroacetic acid 13 µg/L 2.7 days Hashimoto et al. (1998)

Monochloroacetic acid 1.08 µg/L 3.58 days Hanson et al. (2002)

Monobromoacetic acid 14.5 µg/L 3.2 days Hashimoto et al. (1998)

Dichloroacetic acid 1.15 µg/L 5.4 days Hashimoto et al. (1998)

Dibromoacetic acid 50 µg/L 3.2 days Hashimoto et al. (1998)

Tribromoacetic acid 13.66 µg/L1 4.2 days Hashimoto et al. (1998)

Dibromochloroacetic acid 8.8 µg/L1 3.67 days Hashimoto et al. (1998)

2,4,6-Tribromophenol 0.21 µg/L2 1.21 days CICADS 66 (2005)

Neutralization Chemical

Sodium bisulfite 5.0 mg/L 0.000747 days3 Yan et al. (2007)

1 = average concentration. See section 11.3 for calculation explanation. 2 = concentration in ballast water discharge after concentration in control sample subtracted. 3 = derived using a zero-order rate constant (Deltares, 2010).

11.8.3 Predicted Environmental Concentration results from MAMPEC modelling The PEC values as a result of the two different MAMPEC modelling approaches used by Deltares are presented below. Table 26 includes the PEC values for both Environment A and the more conservative Environment B.

Table 26: PEC Summary for Environment A and Environment B

Substance PEC (µg/L)

Environment A PEC (µg/L)

Environment B Bromoform 0.3190 0.3810 Dichlorobromomethane 0.0002 0.0002 Dibromochloromethane 0.0057 0.0059

Total THMs 0.3249 0.3871 Bromochloroacetic acid 0.0236 0.0646 Monochloroacetic acid 0.0021 0.0064 Monobromoacetic acid 0.0272 0.0799 Dichloroacetic acid 0.0023 0.0084 Dibromoacetic acid 0.0939 0.2757 Tribromoacetic acid 0.0269 0.0877 Dibromochloroacetic acid 0.0169 0.0525 2,4,6-Tribromophenol 0.0003 0.0006 Sodium bisulfite 0.0133 0.0133

The simulated environmental concentrations primarily depend on three aspects of the modelling input: the substance emission (i.e. discharge concentration), the substance characteristics (i.e. half-life), and the environmental conditions (harbour configuration, flushing, etc.). For instance, the role of substance decay rates is evident when evaluating TBAA and DBCM (Tables 25 and 26). The emissions for these two substances are in the same order, but as the data indicate, Environment A PEC values differ by an order of magnitude due to the difference in the decay rates of these substances (Deltares, 2010).


I:\MEPC\61\2-9.doc

The effect of varying environmental conditions can be seen when comparing the PEC values for Environment A with Environment B. The "reduced flushing" in Environment B produces larger PEC values for most substances, primarily for those with slower decay rates. This is because removal of these substances from a harbour is more dependent upon flushing, and flushing is reduced in the Environment B simulation. For unstable substances with rapid decay rates, the reduced flushing in Environment B is less influential. This is because the substances decay rapidly and disappear before the effect of flushing can be realized (Deltares, 2010). To ensure that the results obtained with the MAMPEC research version could be considered sufficiently conservative, Deltares also generated PEC values with the last officially released version of MAMPEC (version 2.5.04). Appendix B of the Deltares report provides PEC values generated with MAMPEC version 2.5.04. For the most conservative environment in this assessment (Environment B), the simulations with the research version produced PEC values consistently higher than those obtained with version 2.5.04 (appendix B, Deltares, 2010). Therefore, the research version of MAMPEC used in this assessment of the BalPure® system can be considered more conservative (Environment B) and comparable to the standard version of MAMPEC. 11.9 Effects on aquatic organisms In this section, the available ecotoxicity data for each substance that had measurable concentrations in ballast water discharge is evaluated and appropriate Predicted No Effect Concentration (PNEC) values are derived. PNEC values for substances that were analysed for, but not found with measurable concentrations in treated/neutralized ballast water discharge, are not derived or considered. Tables 22 and 23 present the concentrations of substances measured in ballast water discharge. A thorough literature review of available acute and chronic aquatic ecotoxicity data for the substances associated with the STDN system was conducted; the data are presented in Tables 2 and 4. However, the data set is limited for many of the substances. In some cases, data located in literature did not meet validity criteria and could only be used as supporting information. This resulted in incomplete data sets for fish, crustaceans, and algae that prevented individual PNEC derivation for all substances based on the guidance in the Methodology (23 May 2008 version) as well as the EU Technical Guidance Document on Risk Assessment (2003). Similar data gaps are also noted in other dossiers submitted for Procedure (G9) approval, suggesting that other literature searches have had similar findings. When a complete valid data set was located in literature, and/or any supporting data could be identified, the reported toxicity values have been used for PNEC derivation. Where available data is limited, the PNEC derivation approach used in the European Union Risk Assessment Report (EU RAR) for Sodium Hypochlorite (2007) was followed. Because the STDN system employs a neutralization step during deballasting, no measureable concentration of TRO (as Cl2) is present in ballast water discharge. As TRO (as Cl2) includes HOCl/OCl- and HOBr/OBr- these substances are therefore not considered relevant with respect to PNEC derivation. A PNEC value of 2.0 µg/L for 2,4,6-TBP was derived based on a complete chronic aquatic ecotoxicity data set. The lowest chronic effect (NOEC) concentration located was 0.10 mg/L, and because the chronic and acute data set was complete an assessment factor of 50 was applied. This assessment factor and PNEC value is also reported in the CICADS 66 (2005) document for 2,4,6-TBP.


I:\MEPC\61\2-9.doc

With respect to THMs, a complete valid acute toxicity data set was available for bromoform, and two chronic endpoints (one fish, one alga) were available as supporting information. Using the lowest acute effect concentration (12.3 mg/L) and an assessment factor of 1000, results in a bromoform PNEC of 12.3 µg/L. The data sets for dibromochloromethane (DBCM) and dichlorobromomethane (DCBM) were incomplete. One valid acute endpoint for fish and one acute test with a ciliate was located for DBCM. No chronic studies for DBCM were located. For DCBM, two acute endpoint studies with fish and one additional acute test with a ciliate were located. No chronic endpoint studies for DCBM were located. It is beyond the scope of the application process under Procedure (G9) to conduct chemical-specific testing with these substances on a variety of organisms to complete the data set available in open literature. This is particularly true considering that whole effluent toxicity (WET) testing results of the ballast water discharge are available. Therefore, as an alternative to having PNEC values that are non-derivable for DBCM and DCBM, the PNEC derivation approach for THMs used in the EU RAR for Sodium Hypochlorite (2007) was utilized. For chloroform, the EU RAR references an aquatic PNEC of 146 µg/L. The EU RAR presents that the data collected in connection with an assessment for seawater chlorination "... suggest that the ecotoxicities of the brominated THMs are not markedly different from chloroform" (EU RAR, Sodium Hypochlorite, 2007). The EU RAR concludes that for the purposes of a broad assessment, THMs can be regarded as having similar toxicities and applies the chloroform PNEC to total trihalomethanes. Therefore, utilizing the EU RAR suggested PNEC value of 146 µg/L for total THMs ensures that all THMs are considered, even in cases where toxicity data is limited and individual PNEC values cannot be derived. To maintain a conservative approach, the lower PNEC value for bromoform (12.3 µg/L) derived from the available data can still be used for PEC/PNEC ratio evaluation. For all HAAs, the database is incomplete with respect to chronic ecotoxicity endpoints. With the exception one HAA, monobromoacetic acid (MBAA), the acute aquatic toxicity database is also incomplete and did not allow for direct PNEC calculation for each individual HAA. Again, it is beyond the scope of the application process under Procedure (G9) to conduct the extensive amount of toxicity testing that would be required to complete the HAA data set. This is particularly true considering that whole effluent toxicity (WET) testing results of ballast water discharge are available. Therefore, the approach utilized in the Sodium Hypochlorite EU RAR (2007) for PNEC derivation of HAAs was utilized. That is, the EU RAR considered it "... conservative to treat all haloacetic acids simplistically as being TCAA ..." in regards to PNECs and all HAAs could be considered to have a PNEC of 0.85 µg/L. Since this PNEC is far lower than the PNEC of 500 µg/L that would be derived for TCAA based on the data found in literature (acute data set in Table 2 and assessment factor of 1000), 0.85 µg/L is very protective and seems to be a reasonable approach for this assessment. For sodium bisulfite, the available aquatic toxicity data in literature was also limited. No chronic endpoint studies were located, which is likely due to the rapid degradation (minutes time scale) of sodium bisulfite. Two acute endpoints for fish were located along with three acute endpoints for crustaceans. Since the submission of the Basic Approval dossier, STDN commissioned a study (Nautilus, 2009) to determine the ecotoxicity of sodium bisulfite to algae to complete the data set. Evaluating the completed data set, the lowest acute effect was 81 mg/L sodium bisulfite (Daphnia magna). Using an assessment factor of 1000, a PNEC value of 81 µg/L was derived. Table 27 below provides a summary of the PNEC values derived.


I:\MEPC\61\2-9.doc

Table 27: PNEC Derivation Summary

Substance Lowest Effect Concentration

(mg/L)

Assessment Factor

PNEC Value (µg/L)

Reference/Comments

Relevant Chemicals

2,4,6-Tribromophenol 0.1 50 2.0 21-day NOEC for Daphnia magna; supported in CICADS 66 (2005)

THMs

Bromoform 12.3 1000 12.3 / 146 146 µg/L to be used as PNEC for Total THMs

Dibromochloromethane (DBCM)

Insufficient Data Located

-- Not Derivable Use EU Sodium Hypochlorite RAR suggested PNEC for Total THMs

Dichlorobromomethane (DCBM)


-- Not Derivable Use EU Sodium Hypochlorite RAR suggested PNEC for Total THMs

Total THMs -- -- 146 Based on EU Sodium Hypochlorite RAR suggested PNEC for Total THMs

HAAs

Bromochloroacetic acid Insufficient Data

Located -- 0.85

Based on EU Sodium Hypochlorite RAR suggested PNEC for HAAs

Monochloroacetic acid Insufficient Data

Located -- 0.85


Dichloroacetic acid Insufficient Data

Located -- 0.85


Dibromochloroacetic acid


-- 0.85 Based on EU Sodium Hypochlorite RAR suggested PNEC for HAAs

Monobromoacetic acid Insufficient Data

Located -- 0.85


Dibromoacetic acid Insufficient Data

Located -- 0.85


Tribromoacetic acid Insufficient Data

Located -- 0.85


Other Chemicals

Sodium bisulfite 81 1000 81 Based on lowest acute effect; no chronic data available.

Utilizing the PEC values obtained with MAMPEC modelling for two different environmental configurations and the PNEC values derived from available toxicity data enables calculation of PEC/PNEC ratios. The tables below show each PEC and PNEC value and the resulting ratios for the two different emission scenarios modelled with MAMPEC.


I:\MEPC\61\2-9.doc

Table 28: PEC/PNEC Calculation for MAMPEC Environment A

Substance PEC µg/L (Environ. A) PNEC (µg/L) PEC/PNEC

Relevant Chemicals

2,4,6-Tribromophenol 0.0003 2.0 0.00015

THMs

Bromoform 0.3190 12.3 0.02593

Dichlorobromomethane 0.0002 Not Derivable --

Dibromochloromethane 0.0057 Not Derivable --

Total THMs 0.3249 146 0.00222

HAAs

Bromochloroacetic acid 0.0236 0.85 0.02776

Monochloroacetic acid 0.0020 0.85 0.00243

Monobromoacetic acid 0.0272 0.85 0.03200

Dichloroacetic acid 0.0023 0.85 0.00275

Dibromoacetic acid 0.0939 0.85 0.11047

Tribromoacetic acid 0.0269 0.85 0.03164

Dibromochloroacetic acid 0.0169 0.85 0.01988

Other Chemicals

Sodium bisulfite 0.0133 81 0.00016

Table 29: PEC/PNEC Calculation for MAMPEC Environment B

Substance PEC µg/L (Environ. B) PNEC (µg/L) PEC/PNEC

Relevant Chemicals

2,4,6-Tribromophenol 0.0006 2.0 0.00029

THMs

Bromoform 0.3810 12.3 0.03097

Dichlorobromomethane 0.0002 Not Derivable --

Dibromochloromethane 0.0059 Not Derivable --

Total THMs 0.3871 146 0.00265

HAAs

Bromochloroacetic acid 0.0646 0.85 0.07600

Monochloroacetic acid 0.0063 0.85 0.00747

Monobromoacetic acid 0.0799 0.85 0.09400

Dichloroacetic acid 0.0084 0.85 0.00982

Dibromoacetic acid 0.2757 0.85 0.32435

Tribromoacetic acid 0.0877 0.85 0.10317

Dibromochloroacetic acid 0.0525 0.85 0.06176

Other Chemicals

Sodium bisulfite 0.0133 81 0.00016


I:\MEPC\61\2-9.doc

For both Environment A and B, the highest PEC/PNEC ratio result was for DBAA (0.11047 and 0.32435, respectively). All other ratios were well below 1, even when using PEC values derived from the very conservative Environment B. Therefore, the data indicate that all PEC/PNEC ratios are <1. According to section 7.3.2 of the Methodology (23 May 2008), when no aquatic toxicity of the ballast water discharge is found through direct testing, or the PEC/PNEC ratios are <1, no further assessment of direct toxic effects to the aquatic environment is necessary. 11.10 Assessment of potential for bioaccumulation Based on the data presented in this Final Approval dossier, as well as in the Basic Approval dossier (MEPC 60/2/9) all of the substances associated with the BalPure® system have low BCFs (<2000), indicating that they are unlikely to bioaccumulate in aquatic organisms or present a significant food chain risk. With the exception of the THMs and 2,4,6-TBP, the Active Substances and Relevant Chemicals have very low log Kows, which indicates the compounds are hydrophilic. The THMs and 2,4,6-TBP, as would be expected, are moderately hydrophobic. However, none of the THM log Kow values estimated and/or located in the literature are ≥3 and none of the BCFs are ≥2000 (Table 19). For 2,4,6-TBP, the reported log Kow is slightly above 3 (3.89); however, the BCF is less than 2000. Therefore, bioaccumulation potential is low. 11.11 Effects on sediment Based on the chemical and physical properties of all substances associated with the BalPure® system presented in this Final Approval dossier, as well as in the Basic Approval dossier (MEPC 60/2/9), no effects on sediment are anticipated (sections 3.4 and 5.2). The only exception is 2,4,6-TBP with a Koc value of 1,186 L/kg (Table 6), which indicates a moderate potential for partitioning into sediment. When 2,4,6-TBP is released to water, 93% is expected to stay in the water compartment and 7% is transported to the sediment compartment (OECD SIDS, 2003). This substance is also reported to dehalogenate rapidly in anaerobic sediments, with a reported half-life of approximately 4 days (CICADS 66, 2005). This is much more rapid than the sediment persistence criteria of 180 days in marine sediment or 120 days in fresh water sediment (Table 19 and the Methodology, section 6.1.4). Considering the low concentration of 2,4,6-TBP in ballast water discharge, a moderate potential for sediment adsorption, and the rapid sediment degradation, no effects on sediment are expected as a result of the BalPure® system. 11.12 Effects assessment The data presented in this Final Approval dossier, as well as in the Basic Approval dossier (MEPC 60/2/9), for the substances associated with the BalPure® system indicate that there is a low potential for bioaccumulation and persistence in the aquatic environment, and moderate potential for sediment adsorption for one substance (2,4,6 TBP). Moderate sediment adsorption for 2,4,6-TBP poses low potential risk because the 2,4,6-TBP median concentration discharged by the BalPure® system was only slightly more (0.21 µg/L) than the environmental background level measured in the control (untreated water) samples. Further, 2,4,6-TBP is readily biodegradable in anaerobic sediment (Table 16) and the 2,4,6-TBP concentration in ballast water discharge is well below the lowest effect level for Daphnia magna (21-day NOEC of 0.10 mg/L, Table 4). No effects or risks in the form of secondary (food chain) poisoning or to sediment species are anticipated. The information reviewed for degradation and bioaccumulation of the substances related to the BalPure® system suggest that potential effects from these mechanisms cannot be reasonably anticipated. As such, aquatic toxicity presents the most likely potential risk for aquatic organisms.


I:\MEPC\61\2-9.doc

Tables 2 and 4 present aquatic toxicity data located in literature for a variety of species, endpoints, and exposure periods. It is important to note that the effect concentrations for all DBPs are notably higher than the measured concentrations of DBPs in ballast water discharge. For instance, the overall lowest effect concentration located for all DBP substances is an acute value of 0.028 mg/L (28 µg/L) for MCAA (Table 2). The highest concentration of MCAA measured in ballast water discharge was 1.20 µg/L. As such, although aquatic toxicity presents the most likely potential risk, one can reasonably conclude that the risk is low. Further, as discussed in section 11.4 and below in section 11.13, the whole effluent toxicity data confirms a low potential aquatic toxicity risk. 11.13 Comparison of effect assessment with discharge toxicity As discussed in section 11.12 above, the effects assessment establishes that aquatic toxicity, rather than bioaccumulation or sediment toxicity, presents the most likely potential risk to aquatic organisms. However, the aquatic toxicity data for DBPs, when compared the measured DBP levels in treated/neutralized ballast water discharge, suggests that aquatic toxicity risks are low. The low potential aquatic toxicity identified in the effects assessment is confirmed when compared to whole effluent toxicity (WET) data for ballast water discharge from the BalPure® system. The concentrations of DBP substances (µg/L range) in treated/neutralized discharge are substantially lower than the aquatic effect concentrations found in literature (mg/L range). Further, with the exception of testing with one species (Acartia tonsa), all WET tests after neutralization with bisulfite resulted in E(L)C50 and/or NOEC values ≥100% ballast water sample. The acute aquatic toxicity data for ballast water discharge presented in Table 3 suggest no apparent ecotoxicity for all end points in five of six species tested (E(L)C50 and/or NOEC values of ≥100% ballast water sample). For one species, Acartia tonsa, an effect was observed in one of four acute tests with a low salinity sample (Day 1). In this sample, a moderate effect was observed in the two highest test concentrations (100% and 32% sample), resulting in a NOEC of 18% volume (Grontmij|AquaSense, 2009). This acute test is considered an anomalous data point for several reasons. First, a total of eight ecotoxicity tests (4 acute and 4 chronic) were performed with Acartia; only one acute test resulted in a moderate effect. Second, during tests with Acartia moderate effects were also noted in a control (untreated, NIOZ harbour water) sample (EC20 of 58% sample) (Grontmij|AquaSense, 2009). A possible explanation for the moderate toxic effect in the treated and control samples is the presence of Phaeocystis, an algal species known to be abundant in the NIOZ harbour and which excretes allelochemicals. Although not a highly toxic algal species, the massive blooms often result in ecosystem effects (Veldhuis and Kools, 2009). Phaeocystis was present in the NIOZ harbour during the whole season of testing but was particularly abundant during the first low salinity test cycles (Veldhuis and Kools, 2009), which corresponds with when effects were observed in the Acartia tests. In addition to the above points, it should be noted that STDN performed aquatic toxicity testing with six taxonomic groups, rather than just the three required groups (algae, crustacean, and fish). Both acute and chronic (sub-lethal) tests were performed. Further, discharge toxicity was evaluated for two different salinities (high and low salinity) and at two different time periods after treatment/neutralization (1 and 5 days). Treated/neutralized ballast water discharge tests with all of the five other taxonomic groups under these variable conditions were unaffected at 100% sample. These facts allow for increased confidence in the conclusion that the 18% NOEC result for Acartia is an anomaly data point among 8 tests with this species and the only one of six taxonomic groups having a moderate effect. No toxic effects were observed in all other acute (Table 3) or chronic (Table 5) tests.


I:\MEPC\61\2-9.doc

Lastly, previous chronic ecotoxicity studies carried out for Basic Approval at the treatment concentration of 15.0 mg/L HOCl resulted in no ecotoxicity (NOEC = 100% sample) in any of the four species tested (MEPC 60/2/9, Table 5). Considering all of the above information, although aquatic toxicity is identified as the most likely potential for risk to aquatic organisms, discharge toxicity testing results suggest the potential risk is extremely low. 12 RISK ASSESSMENT The potential risks presented by having the BalPure® Ballast Water Management System (BWMS) on board are discussed in section 12.1 and the potential risks to human health are presented in section 12.2. Section 12.3 addresses risks to the aquatic environment. 12.1 Risk to safety of ship Additional information regarding corrosion to that presented in the dossier for Basic Approval submitted in August 2009 (MEPC 60/2/9) is presented in this section to address recommendations and comments by the GESAMP-BWWG (MEPC 60/2/16, annex 7). The additional data presented in this Final Approval dossier for sodium hypochlorite (or TRO as Cl2) produced by the BalPure® system indicate that there is a low potential for corrosion of the ballast water tanks. 12.1.1 Corrosion The following information and evaluations making use of the GESAMP-BWWG recommendations support that there is a low potential for corrosion of ballast water tanks due to the effect of sodium hypochlorite (or TRO as Cl2). An extensive search was conducted for published literature on the topic of sodium hypochlorite's effect on corrosion in seawater. A review of these studies was conducted and the results are discussed in the paper, "Effect of hypochlorite on the corrosion of carbon steel in ballast water tanks" by Dr. Kenneth Hardee, Ph.D. and Prof. Giuseppe Faita, Ph.D. (appendix A.7, Hardee and Faita, 2010). The authors generally confer that the corrosion risk would not be significantly in excess of that for seawater alone in an existing corroded ballast water tank. Dr. Hardee explains that by extrapolating data from Song, et al. (2009) for 5 mg/L chlorine, the corrosion rate in the non-closed (flowing) system would be 0.104 mm/yr or only a 17% increase from that of seawater alone. The corrosion rate will be much lower where there is existing rusted steel and/or the system is closed (semi-stagnant) and has a specific chlorine demand similar to applications with the STDN BalPure® BWMS. Kim, et al. (2009) further supports the calculations that a ballast tank with 5 mg/L chlorine will have no more than a small effect on the corrosion rate (appendix A.7, Hardee and Faita, 2010). Also the literature indicates that the corrosion rate will slow as there is a buildup of surface rust on the steel, which restricts diffusion of the hypochlorite to the metal surface, although the effect is very difficult to quantify. In addition, the Hardee and Faita paper (appendix A.7) draws on corrosion and hypochlorite chemistry, findings from the relevant published literature, and directly applies this to operating data from land-based tests of the STDN BalPure® BWMS conducted at the Royal Netherlands Institute for Sea Research (NIOZ), Texel, The Netherlands, and Maritime Environmental Resource Centre (MERC), University of Maryland. The Hardee and Faita paper (appendix A.7) presents and compares the corrosion rate of a carbon steel ballast water tank exposed to hypochlorite as a closed system to the typical value for carbon steel structures fully immersed in free flowing seawater. The paper indicates that the corrosion risk is not significantly in excess of that for seawater alone. To further support the opinions regarding corrosion, STDN has submitted the Hardee and Faita paper to BUREAU VERITAS, seeking third party verification.


I:\MEPC\61\2-9.doc

Ballast tank and pipe coating suppliers have conducted independent tests relating to the impact of seawater and treated seawater on their coatings. Ameron, as one of these coatings vendors, sees no effect with 5 mg/L chlorine in seawater on their coating as stated in a letter to STDN (appendix A.7). Per the German Administration's request, STDN has included Table 1 in appendix A.7, which outlines point-by-point the recommendations for corrosion testing contained in document MEPC 59/2/16 in column one and a response to what extent these guidelines were met in the second column by the BMT (2004) study presented in the Basic Approval dossier. We acknowledge that the information provided therein, though informative and germane to corrosion of steel and coated steel in ballast water applications with chlorine, do not fully meet the recommendations outlined in document MEPC 59/2/16. In consequence, and in support of the conclusions of the paper by Dr. Kenneth Hardee and Prof. Giuseppe Faita that corrosion risk is not significantly in excess of that for seawater alone, we have voluntarily contracted Corrosion Testing Laboratories (CTL), Inc. to complete six months of corrosion tests to fully meet the guidance contained in document MEPC 59/2/16 as recommended by the Group. The results of the study will be considered as part of the Type Approval process with the German Administration. 12.2 Risks to human health As discussed in section 1.2, Severn Trent De Nora submitted an application dossier for Basic Approval in August 2009 (MEPC 60/2/9). That dossier was reviewed during the 12th meeting of the GESAMP-BWWG in December 2009 and Basic Approval was recommended. The Group considered the human health risk assessment information "sufficient to allow the Group to recommend Basic Approval" (comment 0.10) and "that the human exposure assessment submitted by the Applicant makes possible to conclude that the operation of this BWMS poses no unacceptable risk to ships' crew, technicians and to the general public" (comment 4.4.2). However, as noted in the comments listed in section 1.2, a number of recommendations were made as follows:

.1 The Group recommended to take into consideration the human exposure during sampling of ballast water at discharge (inhalation, dermal contact), as well as during periodic sediment cleaning (inhalation, dermal contact). These scenarios should also be considered in the conceptual exposure model (CEM).

.2 The Group recommended that the data sets on the analysis of Active

Substance, Relevant Chemicals and Other Chemicals in treated ballast water should be further investigated during land-based and shipboard testing.

.3 The Group recommended that confirmation by further testing should be

performed, that there will be no unacceptable risks (to ship safety, human health, or the environment) due to the chemical composition of the ballast water treated by this BWMS.

The following subsections from the dossier for Basic Approval have been revised to consider additional potentially affected receptor groups (ballast water samplers or persons removing tank sediment). Differences in chemical composition of ballast water discharge based on land-based testing results are compared to the previously reported level of sodium bisulfite and maximum levels of bromate ion, THMs, and HAAs measured during studies of the BalPure® system conducted by the University of Washington (Sosik and Herwig, 2009; as reported in Tables 27 to 29 of the application dossier for Basic Approval). Risks to receptors potentially exposed to treated ballast water discharge are re-quantified using the land-based testing results.


I:\MEPC\61\2-9.doc

ExposureAssessment

Review &Assessment ofToxicological

Data

RiskQuantification

UncertaintiesEvaluation

Risk AssessmentConclusions

HazardIdentification

(Data Review/COPC

Selection )

The evaluation presented herein supplements the evaluation presented in the application dossier for Basic Approval in two ways: (1) risks for technicians or others who may be exposed during ballast water discharge sampling are evaluated quantitatively, and (2) risks to receptors previously evaluated for potential exposure to treated ballast water discharge have been revised using the land-based testing results to estimate exposure. 12.2.1 Introduction Quantitative human health risk assessments generally follow the process outlined below:

Step 1 Step 2 Step 4 Step 5 Step 6

Step 3

With respect to the potential human health risks associated with the BalPure® BWMS, Step 1 is the hazard identification process. The BalPure® operational system is reviewed, and chemicals stored on the ship and added to the system or produced in situ at concentrations potentially harmful to human health are carried forward in the quantitative risk assessment as the chemicals of potential concern (COPCs). In Step 2, COPC-specific toxicity values for use in the quantitative risk analysis are compiled via an exhaustive literature search. If no toxicity data exists, information from appropriate surrogates and/or structurally similar chemicals may be used. For COPCs without toxicity information, quantitative risk evaluation is not possible. In Step 3, exposure scenarios are developed (1) to describe the potential exposures on board during routine vessel operations including operation, monitoring and maintenance of the BalPure® BWMS and (2) to describe the potential exposures on and off board during equipment failure/and or accidental release of the COPCs, and (3) to provide a basis for quantifying those exposures. Each exposure scenario addresses the COPCs, the potential route or mechanism of exposure, and potentially exposed human populations (known as "receptors"). When operation-specific data for scenario development are unavailable, conservative values found in the appropriate regulatory guidance are used. In Step 4, the toxicity and exposure assessments are integrated into quantitative expressions of risks. This includes COPC-specific, multi-pathway risks for each of the potential receptors. The risk values presented in a risk assessment are conditional estimates derived from a considerable number of conservative, health-protective assumptions about exposure and toxicity. Thus, to place the risk estimates in proper perspective, it is important to specify the assumptions and uncertainties inherent in the risk assessment. This process is conducted in Step 5. This step may also involve the reevaluation of data or the identification of additional data requirements to decrease uncertainty. Step 6 involves the development and presentation of conclusions that can be inferred from the findings of the risk assessment. This step provides risk managers with insight into the interpretation of the risk assessment results.


I:\MEPC\61\2-9.doc

12.2.2 Hazard identification/chemical of potential concern selection The chemicals of potential concern (COPCs) associated with the use of BalPure® BWMS were selected by determining which chemicals, either stored and added to the system or formed in situ, have the potential to adversely impact human health under either routine operating conditions, maintenance operations, or during an emergency resulting in an unexpected release. As described in sections 1 and 2, the BalPure® BWMS is a nearly closed system that produces the disinfection agent, sodium hypochlorite (as a mixture of hypochlorous acid/hypobromous acid), via electrolysis of seawater. Hydrogen gas, a secondary by-product of the in situ electro-chemical process, is vented to the atmosphere (using blowers) at concentrations of less than 1%. However, exposure to atmospheric gases typically are not included in risk assessment, and the safety features inherent in the BalPure® BWMS as well as the physical properties of hydrogen gas suggest it is unlikely to accumulate to levels of concern (e.g., above the LEL). Therefore, as indicated in section 2.3.3, hydrogen gas is not considered relevant for the assessment of human health risks. Sodium bisulfite is added to de-chlorinate the ballast water as it is discharged from the ship. UPDATED TEXT PERTINENT TO FINAL APPROVAL APPLICATION In the application dossier for Basic Approval, the final list of COPCs carried forward in the risk assessment included total residual oxidants (TRO) evaluated as hypochlorous acid, total THMs (including chloroform, bromoform, bromodichloromethane, and chlorodibromomethane), total HAAs (including bromochloroacetic acid, mono-, di- and trichloroacetic acids, and mono- and dibromoacetic acids), bromate and sodium bisulfite. With respect to chemicals detected in treated ballast water, this risk assessment for Final Approval is revised to evaluate only chemicals detected during land-based testing. As described in section 2.3, chemicals included for Basic Approval but not detected in the land-based testing of the BalPure® system at NIOZ include chloroform, trichloroacetic acid, and bromate ion. Chemicals detected during land-based testing of the BalPure® system at NIOZ but not previously considered in the risk assessment are the HAAs tribromoacetic acid and dibromochloroacetic acid, and 2,4,6-tribromophenol (TBP). The re-quantification of risks presented in section 12.2.5 includes tribromoacetic and dibromochloroacetic acids in the total HAA calculation. As discussed in section 2.3.2, 2,4,6-TBP was not detected in measurable concentrations in land-based tests using high salinity water (section 11.3, Tables 22 and 23). 2,4,6-Tribromophenol was detected at low levels (maximum = 1.3 µg/L) in ballast water discharged during the low salinity test cycles, but also was detected in the untreated (control) samples at the same, or higher, concentrations. 2,4,6-tribromophenol has been documented to occur in natural environments, either as a pollutant from anthropogenic sources (wood industry, antifungal use) or from natural production by marine benthic organisms (OECD SIDS, 2003). Despite its detection in source (control) water, 2,4,6-TBP is added to the list of COPCs carried forward in this risk assessment. Chloroform was not detected in land-based discharge samples, and risks from exposure to THMs will be evaluated based on the total concentration of all other THMs detected. Bromate ion was not detected during land-based testing, and therefore has not been considered as a COPC in this revised risk assessment.


I:\MEPC\61\2-9.doc

12.2.3 Human Exposure Scenario The US EPA (1989) defines an exposure pathway as "the course a chemical or pollutant takes from the source to the organism exposed". An exposure route is "the way a chemical or pollutant enters an organism after contact" (US EPA 1989). A complete exposure pathway requires four key elements: chemical sources; migration routes (i.e. environmental transport); potentially exposed human receptors; and routes of exposure to impacted media (e.g., ingestion of chemicals in water). All four factors are required for a complete exposure pathway; if any one factor is missing, the pathway is considered incomplete. Because an incomplete pathway does not pose a potential health hazard, incomplete exposure pathways are not included in this health risk assessment. Potential human receptor groups and exposure routes are described below. UPDATED TEXT PERTINENT TO FINAL APPROVAL APPLICATION The exposure scenarios evaluated quantitatively in the application dossier for Basic Approval were: (1) STDN BalPure® BWMS Technicians exposed (via dermal contact and incidental ingestion) to hypochlorous acid during maintenance activities and/or sodium bisulfite during chemical resupply, (2) ship's crew/dock workers exposed (via dermal contact) to COPCs in treated ballast water spray drift, and (3) the general public who may swim in the vicinity of (and incidentally ingests and has dermal contact with) recently discharged ballast water associated with the BalPure® system. However, the Group recommended the consideration of exposure during sampling of ballast water at discharge (inhalation, dermal contact), as well as during periodic sediment cleaning (inhalation, dermal contact), which has been added to the discussion below. 12.2.3.1 Potential exposure pathways The conceptual exposure model (CEM) provides the basis for a comprehensive evaluation of the risks to human health by identifying the mechanisms through which people may be exposed to COPCs. The CEM traces the COPCs in a logical migration from their sources through various release mechanisms and exposure routes to potentially affected receptor groups. The CEM identifies the exposure routes that are potentially complete under the given use(s). These pathways are evaluated in the quantitative risk assessment for each receptor. The CEM also facilitates the analysis and screening of exposure pathways not likely to pose significant risks. Primary, secondary, and tertiary sources of COPCs associated with the BalPure® BWMS are listed in the CEM (Figure 3). The primary sources of COPCs include the in situ production of sodium hypochlorite, the sodium bisulfite storage tanks and associated pipelines, and the treated ballast water itself. Chemicals associated with the BalPure® BWMS are expected to remain contained, unless a storage or pipeline failure occurs. Such a release would allow COPCs to migrate to secondary and tertiary sources. For example, accidental spills and pipe or tank failure could result in the release of COPCs onboard the vessel or to seawater. From these secondary sources, the COPCs may migrate to tertiary sources such as volatizing into air within enclosed spaces on the vessel or in outdoor air, or migration to shorelines and/or docking areas. As shown in Figure 3, only exposure to COPCs resulting from leaks of storage tanks, pipelines and ballast tank discharges are evaluated quantitatively in this risk assessment. This is because no exposure is anticipated during normal operating conditions, because chemicals used in the BalPure® BWMS are contained within a closed system (hypochlorous acid is produced in situ and is injected directly into the ballast system; sodium bisulfite is injected into the ballast water as it is discharged from the ship), While both the chemical


I:\MEPC\61\2-9.doc

storage tanks and ballast tanks are vented, vent lines are directed to the atmosphere or to spaces within the ship with proper ventilation. Significant volatilization of COPCs from solution (from either storage tanks or treated ballast water) into indoor or outdoor air is extremely unlikely, and is not evaluated quantitatively. It is noted, however, that the trihalomethanes produced as disinfectant by-products are volatile and hydrophobic, and as such, could be inhaled near ballast water or vent air outlets. While this inhalation pathway is not evaluated in the quantitative risk assessment, potential exposures and risks associated with vent air during ballasting and deballasting were discussed in section 12.2.5.5 of the application dossier for Basic Approval, and have been revised herein (now section 12.2.5.6) to reflect maximum concentrations of THMs detected during land-based testing. UPDATED TEXT PERTINENT TO FINAL APPROVAL APPLICATION As noted by the Group, exposure to treated ballast water also could occur during discharge sampling. In addition, workers could be exposed to treated ballast water when cleaning sediment from the tanks, although this activity would be conducted following established and standard confined space safe entry practices, and persons engaged in this activity would be wearing personal protective equipment (PPE). These additional potential exposure scenarios are considered in the risk assessment, as discussed below. 12.2.3.2 Identification of potentially exposed populations The BalPure® system is designed to operate automatically, with the system processes during ballasting and deballasting started and stopped based on electronic signals. Therefore, under normal operating conditions, neither STDN BalPure® technicians nor the ship's crew is expected to have contact with the BWMS-related COPCs. Because the BalPure® BWMS is closed system, exposure to the Active Substance, disinfection by-products, or sodium bisulfite is highly unlikely. As shown in Figure 3, people with the greatest potential for exposure to BalPure® BWMS-related COPCs are STDN BalPure® technicians involved in routine maintenance and repair, or chemical resupply activities. These technicians, or others, could be exposed to COPCs in treated ballast water during routine discharge sampling. Although unlikely, ship's crew on board and in the spaces where the system components are present could also be exposed to COPCs during a pipeline or chemical tank failure; it is anticipated that these potential exposures would be less than those for STDN technicians conducting routine maintenance or chemical resupply. Because ballast water tanks remain closed at all times (with the exception of planned inspections/repairs of empty tanks) STDN technicians and the ship's crew are not expected to have direct contact with treated ballast water, although this potential exposure pathway is evaluated below. Further, either the ship's crew or dock workers could have dermal contact with released ballast water as a result of spray drift. As shown in Figure 3, while STDN technicians, ship's crew, or dock workers potentially could inhale chemicals volatilizing from treated ballast water, these source-exposure pathways are considered highly unlikely and are not evaluated quantitatively. Others with potential for COPC contact include the general public recreating at beaches near areas where treated ballast water has been discharged from the ship, although this exposure scenario is highly unlikely. These receptors and the potential exposure pathways are summarized below.


I:\MEPC\61\2-9.doc

SOURCES

Primary Secondary Tertiary

Vessel Surface

Seawater

Dock/Shoreline

• Pathway is or might be complete; sufficient data available for quantitative evaluation

X Pathway is or might be complete, but judged to be minor; not evaluated quantitatively1STDN Technician exposure considered worst-case scenario for anyone working onboard vessel.2Not evaluated quantiatively in risk assessment; semi-qualitatively evaluation provided.

Pathway is not complete; no evaluation required

STDN Technician/Other

- Discharge Sampling

X

•

Ship's Crew - Accidental Release

X1

X

X2

•

On Board

X

Dermal Contact/Incidental Ingestion

Inhalation

X

Exposure Routes

•

X

Ship's Crew - Ballast Tank

Entry

X

Release fromBallast Tanks

Dermal Contact/Incidental Ingestion

X

Inhalation X

General Public

Off Board

Dock Workers

•

X

Spills/Leaks from Storage Tanks,

Pipelines or BWTSVapors/Indoor Air

Vapors/Outdoor Air

•

STDN Technician - Maintenance

Figure 3: Conceptual Exposure Model, BalPure® System

.1 STDN BalPure® BWMS Technicians: STDN technicians or properly trained BalPure® personnel are the only receptors who would work directly with the BWMS, including routine maintenance and chemical resupply of sodium bisulfite, and therefore have the greatest potential for actual exposure. Although proper chemical storage and handling, safety training, and use of appropriate PPE would prevent direct contact with COPCs, technicians could be exposed to hypochlorous acid, sodium bisulfite, or treated ballast water during maintenance activities via:

- incidental ingestion of COPCs during routine maintenance and

chemical resupply activities; - dermal contact with COPCs during routine maintenance and

chemical resupply activities; and - inhalation of COPC vapors or mists in air during routine

maintenance and chemical resupply activities. The potentially complete and quantifiable exposure pathways for STDN BalPure® technicians include: a) incidental ingestion of and dermal contact with hypochlorous acid at maximum concentrations of 1,000 mg/L, and b) incidental ingestion of and dermal contact with sodium bisulfite during chemical resupply at maximum concentrations of 380,000 mg/L. For hypochlorous acid, potential exposure is further mitigated by a pipeline flushing cycle that is engineered into the system. For instance, after generation of the Active Substances and the ballasting operation is complete, a 10 minute cycle is initiated by the PLC to the electrolyzer cells and injection pipelines with seawater. The same can be done for the sodium bisulfite delivery piping when a deballasting operation is complete, if needed to enhance the safety of an installation. These flush cycles ensure that pipelines do not remain filled with chemicals when the BWMS is not in operation.


I:\MEPC\61\2-9.doc

.2 STDN BalPure® BWMS Technicians (or others) engaged in sampling of treated ballast water: BalPure® technicians and/or ship's crew will likely be responsible for sampling treated/neutralized ballast water at the point of discharge. Although use of appropriate PPE would prevent direct contact with COPCs, technicians could be exposed to COPCs in treated ballast water during sampling via:

- dermal contact with COPCs during discharge sampling; and - inhalation of COPC vapours or mists in air during discharge

sampling.

However, as discussed previously, during deballasting operations, residual chlorine present in the ballast tank is neutralized (de-chlorinated) with sodium bisulfite just prior to discharge from the ship. The trihalomethanes are the only disinfectant by-products identified as volatile (see Table 17 of this dossier and 18 in the dossier for Procedure (G9) Basic Approval), and therefore could be inhaled near the ballast water discharge outlet. However, it is anticipated that concentrations in air would be low and immediately mixed into the surrounding atmosphere. Therefore, the only potentially complete and quantifiable exposure pathway for ballast water samplers is dermal contact with COPCs in discharged ballast water, including hypochlorous acid, THMs, HAAs, 2,4,6-tribromophenol, and sodium bisulfite at the maximum residual ballast water concentrations detected during land-based testing: <20, 0.845, 0.375, 0.0013, and 8.0 mg/L, respectively (see section 11.3 of Basic Approval dossier and section 11.3 herein). It is also important to note that treated ballast water discharge will be neutralized with sodium bisulfite, so the maximum residual ballast water concentration for hypochlorous acid is overly conservative because it does not consider neutralization. This exposure scenario is similar to the ship's crew/dock worker scenario included in the Basic Approval dossier (and described below), but would not be accidental and is evaluated with possible greater frequency.

.3 Ship's crew: Under normal operating conditions, the ship's crew has limited

to no interaction with the BalPure® BWMS. The system and piping are located in areas where the crew may be present, but normal ballasting and deballasting modes of the BalPure® system are controlled automatically and do not require chemical contact. However, this group has the potential to be exposed to sodium bisulfite while chemical-resupply activities take place, or if a pipeline or storage tank fails, or to any COPCs during a failure of the BalPure® system. Exposure to treated ballast water is highly unlikely, since the tanks are opened for required inspections only when empty, although the potential for exposure to ballast water spray drift exists. The actual exposure potential for this group is extremely low, but the possible exposure routes include:

- incidental ingestion of the COPCs during an accidental release; - dermal contact with COPCs during an accidental release; and - inhalation of COPC vapors or mists in air during an accidental

release.

As previously indicated, accidental release of hypochlorous acid and/or sodium bisulfite is expected to result in exposures smaller than those for


I:\MEPC\61\2-9.doc

STDN BalPure® technicians engaged in routine maintenance and chemical resupply. Moreover, as described in section 8.1.3.2 of the application dossier for Basic Approval, STDN technicians and ship's crew will be trained to respond to sodium bisulfite spills; therefore these exposure scenarios are not included in the quantitative risk assessment. The only potentially complete and quantifiable exposure pathway for ship's crew is dermal contact with COPCs in discharged ballast water, including hypochlorous acid, THMs, HAAs, 2,4,6-tribromophenol, and sodium bisulfite at the maximum residual ballast water concentrations, as described above.

.4 Ship's crew (or others) responsible for ballast tank sediment cleaning:

Workers engaged in tank cleaning could be exposed to residual treated ballast water remaining in the tanks (via dermal contact or incidental ingestion, or to any vapours that remain in the tank once it has been opened). However, because standard confined space entry protocols that are required by applicable regulations will be implemented prior to tank entry, this route of exposure is not considered complete. For instance, protocols such as, but not limited to, those outlined by the International Association of Classification Societies Ltd. (IACS) Confined Space Safe Practice (2007) are performed as standard operating procedure prior to ballast tank entry. For the purposes of example, these practices may include:

- a safety meeting should be held prior to tank entry/survey; - Entry Permit should be obtained for the space to be entered; - identify the hazards and assess the risks; - evaluate ventilation of the space; - ensure that a standby and/or rescue team is in place; - check and evaluate gas measurements taken − as a

minimum, oxygen measurements should be carried out before entry into the enclosed space;

- use of personal gas measuring equipment during the

entry/survey; and - evaluate if special clothing and/or equipment are required.

Given the standard practices and mitigation measures implemented prior to any ballast tank entry, the possibility of exposure to ballast water related chemicals is very low. Therefore, as illustrated in the revised conceptual exposure model shown in Figure 3, only the ballast water sampling scenario has been added for quantitative evaluation in this risk assessment.

.5 Dock workers: Under normal operating conditions, dock workers will have

no interaction with the BalPure® BWMS or chemical resupply activities. While exposure to treated ballast water is highly unlikely, the potential for exposure to ballast water spray drift exists. The actual exposure potential for this group is extremely low, but the possible exposure routes include:

- dermal contact with COPCs in ballast water spray drift.


I:\MEPC\61\2-9.doc

The only potentially complete and quantifiable exposure pathway for dock workers is dermal contact with COPCs in discharged ballast water, including hypochlorous acid, THMs, HAAs, 2,4,6-tribromophenol, and sodium bisulfite at the maximum residual ballast water concentrations. This quantitative evaluation is identical to that for ship's crew.

.6 General public: Although it is unlikely that the general public would

recreate near docked ships, it is possible that recreation areas could be located downstream of shipping docks. While any COPC releases from the ship would be immediately diluted into surrounding seawater, the general public (adults or children) could be exposed to COPCs via:

- incidental ingestion of ballast water COPCs discharged to

seawater near the shoreline; - dermal contact with ballast water COPCs discharged to

seawater near the shoreline; and - inhalation of ballast water COPC vapors or mists in outdoor air

near the shoreline.

The potentially complete and quantifiable exposure pathways for swimmers include: incidental ingestion of and dermal contact with hypochlorous acid, THMs, HAAs, 2,4,6-tribromophenol, and sodium bisulfite in recently discharged ballast water. It is conservatively assumed that maximum residual ballast water COPC concentrations are diluted 100-fold prior to contact with beachgoers.

In the application dossier for Basic Approval, the chemical composition of treated ballast water was based on maximum levels of THMs and HAAs measured during the studies of the BalPure® system conducted by the University of Washington (Sosik and Herwig, 2009; as reported in Tables 27 to 29 of the Basic Approval dossier). Since that time, land-based testing has been conducted, with the maximum concentrations shown in Tables 22 and 23. In regards to hypochlorous acid, the maximum treatment dose (20 mg/L TRO as Cl2) presented in STDN's application dossier for Basic Approval was also used in this human risk assessment to maintain a high level of conservatism. However, the maximum treatment dose applied during land-based testing was 15 mg/L TRO as Cl2. Because the human risk assessment is performed using a 20 mg/L dose any risks due to exposure of hypochlorous acid can be considered less than presented here. Table 30 provides a summary of the exposure point concentrations (EPCs) discussed above and used this risk assessment, and compares them to the values used in the dossier for Basic Approval. Overall, there is very little difference in the maximum concentrations of THMs and HAAs between the two tests.


I:\MEPC\61\2-9.doc

Table 30: Comparison of Basic Approval and Land-based Ballast Water Residual COPC Exposure Point Concentrations

Basic and Final Approval

Dossiers STDN Technicians:

Maximum Concentration in

System

Basic Approval Dossier

Maximum Residual Ballast

Water Concentrations

Land-based Maximum

Residual Ballast Water

Concentrations

Land-based Diluted (100X)

Residual Ballast Water

Concentrations

COPC mg/L mg/L mg/L mg/L

Hypochlorous Acid 1.00E+03 2.00E+01* 2.00E+01* 2.00E-01*

Bromate NA 4.20E-02 Not detected Not detected

THMs NA 1.10E+00 8.45E-01 8.45E-03

HAAs NA 2.42E-01 3.75E-01 3.75E-03

2,4,6-Tribromophenol NA NA 1.30E-03 1.30E-05

Sodium bisulfite 3.80E+05 2.00E+00 8.00E+00 8.00E-02

NA = Not applicable * = Assumed as worst case residual; maximum treatment dose concentration

12.2.3.3 Quantitative exposure assessment As described above, although highly unlikely, STDN BalPure® technicians could have direct contact (ingestion and dermal contact) with COPCs within the BWMS during maintenance activities and to sodium bisulfite during chemical resupply activities. STDN BalPure® technicians or others could have dermal exposure to treated ballast water during discharge sampling and ship's crew or dock workers could have dermal exposure to ballast water spray drift. Beachgoers may recreate in the vicinity of ballast water discharges. Intake factors are used to determine the amount of COPC that each receptor is potentially exposed to via each exposure route, and are used to evaluate both cancer risk and non-cancer hazard. The following equation was used to quantify intake from the dermal contact pathway:

Iw = (Cw)(SA)(PC)(CF)(ET)(EF)(ED) / (BW)(AT) Where:

Iw = intake from dermal contact with COPC in ballast water/solution (mg/kg-d) Cw = maximum concentration of a COPC in ballast water/solution (mg/L) SA = skin surface area in contact with ballast water/solution (cm2/d) PC = dermal permeability coefficient (cm/hr), chemical-specific CF = conversion factor 10-3 (L/cm3) ET = exposure time (h/d) EF = event frequency (d/y) ED = exposure duration (y) BW = body weight (kg) AT = averaging time (d), ED x 365d/y (non-carcinogens), 70y x 365d/y (carcinogens)


I:\MEPC\61\2-9.doc

The following equation was used to quantify intake from the incidental ingestion pathway:

Iw = (Cw)(IngR)(EF)(ED) / (BW)(AT)

Where:

Iw = intake from incidental ingestion of COPC in ballast water/solution (mg/kg-d) Cw = maximum concentration of COPC in ballast water/solution (mg/L) IngR = ingestion rate (L/day) EF = event frequency (d/y) ED = exposure duration (y) BW = body weight (kg) AT = averaging time (d) - ED x 365d/y (non-carcinogens),

70y x 365d/y (carcinogens) The receptor-specific exposure parameters used to estimate COPC intake based on the equations above are summarized below and the resulting intake factors summarized in Tables 31 and 32. As shown in Tables 31 and 32, the intake equations described above for the general public (beachgoers) are age-adjusted to account for six years of exposure as a child and 24 years of exposure as an adult. This approach is consistent with guidance (US EPA 1989 and 1991e), and accounts for any differences in exposure assumptions between children and adults. .1 STDN Technicians

The adult exposure factors used to quantify potential exposure to hypochlorous acid during maintenance of the BalPure® BWMS or to sodium bisulfite during chemical resupply are:

- Ingestion Rate: While incidental ingestion is likely to be only

a fraction of the ingestion rate for swimming, 0.05 L/hour is conservative and therefore used (US EPA 1989).

- Skin Surface Area Available for Contact: During an

accidental spill, technicians or workers may have significant contact with hands and arms, which results in a skin surface area of 3,300 cm2 (US EPA 2004).

- Permeability Constant: Chemical-specific; for inorganics

without a permeability coefficient, and especially for ionized chemicals, guidance recommends using 0.001 cm/hour (US EPA 2004).

- Body Weight: The adult body weight typically applied in risk

assessment is 70 kg (US EPA 2002). - Exposure Time: The exposure time is assumed to

be 0.25 hours per event, is based on best professional judgment. - Event Frequency: Based on best professional judgment,

an exposure frequency of 1 spill event per year is used. - Exposure Duration: Based best professional judgment, the

exposure duration is 10 years.


I:\MEPC\61\2-9.doc

- Averaging Time: The life expectancy of 70 years (25,550 days) was used as the averaging time for exposure to carcinogenic contaminants (US EPA 2002). The averaging time for non-carcinogenic effects is equal to the exposure duration of 10 years (3,650 days).

.2 STDN BalPure® BWMS Technicians (or others) responsible for ballast

water discharge sampling

The adult exposure factors used to quantify potential exposure to discharged ballast water during sampling are expected to be similar to those used to quantify ship's crew/dock worker exposure to spray drift, with the exception of a higher event frequency:

- Skin Surface Area Available for Contact: During sampling

of ballast water at discharge, crew or workers may have limited contact with hands, arms, and face, which results in a skin surface area of 1,800 cm2 which is based on best professional judgment (10% of total body; US EPA 2004).


without a permeability coefficient, and especially for ionized chemicals, guidance recommends using 0.001 cm/hour (US EPA, 2004). The value for chloroform, 0.0068 cm/hour, is used as a surrogate for THMs, HAAs, and 2,4,6-tribromophenol (US EPA 2004).


assessment is 70 kg (US EPA, 2002). - Exposure Time: The exposure time is assumed to

be 0.25 hours per event, is based on best professional judgment. - Event Frequency: Based on best professional judgment, an

exposure frequency of 52 ballast water discharge sampling events per year is used.

- Exposure Duration: Based on best professional judgment,

the exposure duration is 10 years. - Averaging Time: The life expectancy of 70 years (25,550 days)

was used as the averaging time for exposure to carcinogenic contaminants (US EPA, 2002). The averaging time for non-carcinogenic effects is equal to the exposure duration of 10 years (3,650 days).

.3 Ship's Crew/Dock Workers

The adult exposure factors used to quantify potential exposure to discharged ballast water spray drift are:

- Skin Surface Area Available for Contact: During a spray drift

incident, crew or workers may have limited contact with hands, arms, and face, which results in a skin surface area of 1,800 cm2 which is based on best professional judgment (10% of total body; US EPA 2004).


I:\MEPC\61\2-9.doc

- Permeability Constant: Chemical-specific; for inorganics without a permeability coefficient, and especially for ionized chemicals, guidance recommends using 0.001 cm/hour (US EPA 2004). The value for chloroform, 0.0068 cm/hour, is used as a surrogate for THMs, HAAs, and 2,4,6-tribromophenol (US EPA 2004).


assessment is 70 kg (US EPA 2002). - Exposure Time: The exposure time is assumed to

be 0.25 hours per event, is based on best professional judgment. - Event Frequency: Based on best professional judgment,

an exposure frequency of 1 spray drift event per year is used. - Exposure Duration: Based on best professional judgment,

the exposure duration is 10 years. - Averaging Time: The life expectancy of 70 years (25,550 days)

was used as the averaging time for exposure to carcinogenic contaminants (US EPA 2002). The averaging time for non-carcinogenic effects is equal to the exposure duration of 10 years (3,650 days).

.4 General Public (Beachgoers)

The exposure factors used to quantify potential exposure to COPCs in discharged ballast water to swimming beachgoers (adults and children) are described below. As discussed above and shown in Tables 31 and 32, the intake factors used to evaluate cancer risk for beachgoers are age-adjusted to account for differences in exposure assumptions between children and adults.

- Ingestion Rate: The ingestion rate for swimming, 0.05 L/hour,

is used (US EPA 1989). - Skin Surface Area Available for Contact: The average of

the 50th percentile total skin surface area for adults is 18,000 cm2 and for children is 6,600 cm2 (US EPA 2004).


without a permeability coefficient, and especially for ionized chemicals, guidance recommends using 0.001 cm/hour (US EPA 2004). The value for chloroform, 0.0068 cm/hour, is used as a surrogate for THMs, HAAs, and 2,4,6-tribromophenol (US EPA 2004).

- Body Weight: The adult and child body weights typically

applied in risk assessment are 70 and 15 kg, respectively (US EPA 2002).

- Exposure Time: The US EPA's recommended swimming

duration of 60 minutes per event, based on the 50th percentile value (US EPA 1997).


I:\MEPC\61\2-9.doc

- Event Frequency: Based on best professional judgment, an upper-end exposure frequency of 10 events per year is used.

- Exposure Duration: Based on US EPA guidance, the

exposure duration is 30 years, which includes six years as a child and 24 years as an adult (US EPA 2002).

- Averaging Time: The life expectancy of 70 years (25,550 days)

was used as the averaging time for exposure to carcinogenic contaminants (US EPA 2002). The averaging time for non-carcinogenic effects is equal to the exposure duration; 24 years as an adult (8,760 days) and 6 years as a child (2,190 days).

Table 31: Summary of Dermal Intake Factors

IFdermal =

IFdermal/adj =

IFdermal = Dermal Intake Factor, L water/kg body weight-day

SA = Surface Area, cm2

PC = Dermal Permeability Constant, cm/hourCF = Volumetric Conversion Factor, 1 Liter/1000 cm3

ET = Exposure Time, hours/dayEF = Exposure Frequency, days/yearED = Exposure Duration, years

BW = Body Weight, kgAT = Averaging Time, days

ChildAdult (0-6 years)

SA 3300 1800 1800 18000 6600

CF 0.001 0.001 0.001 0.001 0.001

ET 0.25 0.25 0.25 1 1

EF 1 52 1 10 10

ED 10 10 10 24 6

BW 70 70 70 70 15

ATcarcinogens 25550 25550 25550 25550 25550ATnonarcinogens 3650 3650 3650 8760 2190

PATHWAY-SPECIFIC INTAKE FACTORS:

Carcinogens x PC 4.61E-06 1.31E-04 2.52E-06 3.45E-03 NA

Noncarcinogens x PC 3.23E-05 9.16E-04 1.76E-05 7.05E-03 1.21E-02

Exposure Variable

General PublicSTDN

TechnicianShip's Crew/Dock Worker

STDN Technician/Other -

Sampling

Chemical-Specific Intake Factors via Dermal Contact (IFdermal), L water/kg body weight-day

SA x PC x CF x ET x EF x EDBW x AT

SAchild x PC x CF x ETchild xEFchild x EDchild

BWchild x AT+ SAadult x PC x CF x ETadultx EFadult x EDadult

BWadult x AT

NA = Not applicable.


I:\MEPC\61\2-9.doc

Table 32: Summary of Oral Intake Factors

IForal =

IForal/adj =

IForal = Oral Intake Factor, L water/kg body weight-day

IR = Ingestion Rate, L/hourET = Exposure Time, hours/dayEF = Exposure Frequency, days/yearED = Exposure Duration, years

BW = Body Weight, kgAT = Averaging Time, days

ChildAdult (0-6 years)

IngR 0.05 NA NA 0.05 0.05

ET 0.25 NA NA 1 1

EF 1 NA NA 10 10

ED 10 NA NA 24 6

BW 70 NA NA 70 15

ATcarcinogens 25550 NA NA 25550 25550ATnonarcinogens 3650 NA NA 8760 2190

PATHWAY-SPECIFIC INTAKE FACTORS:Chemical-Specific Intake Factors via Oral Ingestion (IForal), L water/kg body weight-day

Carcinogens 6.99E-08 NA NA 1.45E-05 NA

Noncarcinogens 4.89E-07 NA NA 1.96E-05 9.13E-05

Exposure Variable

General Public

STDN Technician

Ship's Crew/Dock Worker

STDN Technician/Other -

Sampling

IngR x ET x EF x EDBW x AT

IngRchild x ETchild xEFchild x EDchild

BWchild x AT+ IngRadult x ETadultx EFadult x EDadult

BWadult x AT

NA = Not applicable

12.2.4 Health effects in humans The objective of this section is to provide information regarding the potential for health risks from exposure to chemicals potentially present in treated ballast water. Specifically, this section describes how dose-response, or toxicity values, are established and used for non-carcinogenic and carcinogenic COPCs. Because only the ingestion and dermal exposure routes are evaluated quantitatively, this section focuses exclusively on oral toxicity values (which are commonly used as surrogates for evaluation of dermal exposure). 12.2.4.1 Acute health effects With the exception of 2,4,6-tribromophenol, all of the COPCs identified can be skin and eye irritants as described in Table 9 of this dossier and the Basic Approval dossier. Unlike the COPCs previously evaluated, 2,4,6-tribromophenol can be an eye irritant but it has not been demonstrated to irritate the skin (see Table 9).


I:\MEPC\61\2-9.doc

With the exception of sodium bisulfite, all of the COPCs identified are generated in situ at aqueous concentrations in the µg/L to mg/L range. These in situ generated COPCs are at concentrations too low to cause serious eye and skin irritation. There may be some limited potential for minor eye irritation if there is direct contact with ballast water, due to a combination of the COPCs, salts, and other compounds which may be in the ballast/seawater prior to treatment. Skin irritation is considered unlikely given the very low concentrations of the COPCs. THMs are known to cause both skin and eye irritation when present at much high concentrations than those present in situ in ballast water (IPCS 2004). The one COPC which could represent an acute eye and skin hazard is the sodium bisulfite solution stored on the vessel. The MSDS (Table 9 of the dossier for Basic Approval) clearly states that contact with the 38% solution can cause eye and skin irritation. Appropriate PPE should be used when working with this solution to minimize the potential for these acute health effects, and only trained personnel should be handling the concentrated solution (Mallinckrodt, 2006a). The potential for acute irritation of the nose and throat upon inhalation is unlikely. The THMs which are the most likely COPCs to volatize were considered in a separate inhalation assessment in the dossier for Basic Approval (Table 39). The safe levels used in that inhalation assessment, and updated here (section 12.2.5.6) are considered protective of irritation of mucous membranes caused by volatile THMs (OSHA 2009). The more critical endpoint for potential exposure to ballast water COPCs at low concentrations are chronic health issues such as liver disease and cancer. These chronic effects are fully addressed in the risk assessment. 12.2.4.2 Non-carcinogenic adverse health effects For the non-carcinogenic effects of specific constituents, most regulatory agencies assume a dose exists below which no adverse health effects will be seen (US EPA 1989). Below this "threshold" it is believed that exposure to a chemical can be tolerated without adverse effects. Adverse effects manifest only when physiologic protective mechanisms are overcome by exposure to doses above the threshold. The reference dose (RfD), expressed in units of milligrams per kilogram-day (mg/kg-d), represents the daily intake of a constituent (averaged over a year) per kilogram of body weight that is below the effect threshold for the constituent. It is assumed that non-carcinogenic exposure doses are not cumulative from age group to age group over a lifetime of exposure (US EPA 1989). An RfD is specific to the constituent, route of exposure, and duration over which the exposure occurs. Agencies, most notably the US EPA, review all relevant human and animal studies for each constituent and select the studies pertinent to the derivation of specific RfDs. Each study is evaluated to determine the no-observable-adverse-effect level (NOAEL) or, if data are inadequate for such a determination, the lowest-observable-adverse-effect level (LOAEL). The NOAEL corresponds to the dose (mg/kg-d) that can be administered over a lifetime without inducing observable adverse effects. The LOAEL corresponds to the lowest daily dose (mg/kg-d) that can be administered over a lifetime that induces an observable adverse effect. The toxic effect characterized by the LOAEL is referred to as the "critical effect" (US EPA 1989). To derive an RfD, the NOAEL (or LOAEL) is divided by uncertainty factors (alternatively known as safety factors) to ensure protection of human health. Uncertainty factors are applied to account for: (1) extrapolation of data from laboratory animals to humans (interspecies extrapolation), (2) variation in human sensitivity to the toxic effects of a constituent (intraspecies differences), (3) derivation of a chronic RfD based on a subchronic


I:\MEPC\61\2-9.doc

rather than a chronic study, and (4) derivation of an RfD from the LOAEL rather than the NOAEL. A safety factor of 10 is typically applied for each of these uncertainties during the development of the RfD. Thus, the safety factor for an individual COPC could be as high as 10,000. In addition to these uncertainty factors, modifying factors between 0 and 10 may be applied to reflect additional qualitative considerations (US EPA 1989). 12.2.4.3 Carcinogenic adverse health effects The incremental lifetime cancer risk (ILCR) attributed to a carcinogen is calculated as a product of the daily intake (mg/kg-d) and the cancer slope factor (CSF). The US EPA's model of carcinogenesis assumes the relationship between exposure to a carcinogen and cancer risk is linear over the entire dose range, except at very high doses (US EPA 1989). This linearity assumes there is no threshold-of-exposure dose below which harmful effects will not occur. Because of this, carcinogenic effects are considered to be cumulative across age groups when considering lifetime exposures. CSFs are upper-bound (95% upper confidence limit (UCL)) estimates of the increased cancer risk per unit dose, in which risk is expressed as the probability that an individual will develop cancer within his or her lifetime as the result of exposure to a given level of a carcinogen. All cancers or tumours are considered whether or not death results. This approach is inherently conservative because of the no-threshold assumption and the use of the 95% UCL of the estimated slope of dose versus cancer risk. To identify oral RfDs and slope factors, the studies and approaches described by the US EPA in its Integrated Risk Information System (IRIS) were compared to the toxicological information reported in section 4 (Tables 11 to 13) of both this dossier and the dossier for Basic Approval. When no toxicity values are reported in IRIS, best professional judgment was used to develop values based on the data reported in Tables 11 to 13 in both this dossier and the dossier for Basic Approval. 12.2.4.4 Hypochlorous acid and hypobromous acid The health effects associated with hypochlorite (salts of hypochlorous acids) have been studied primarily as it relates to its use as a disinfectant in drinking water. While various drinking water goals for disinfectants have been established by the US EPA, the World Health Organization (WHO), and the European Union (EEC), only the US EPA has identified an oral reference dose for chlorine: 0.1 mg/kg-day (US EPA 1994). This value, as reported in IRIS, is based on based on the NOAEL of 14 mg/kg-day for rats exposed to chlorine in drinking water for two years (NTP 1992a), as reported in Table 11 of the dossier for Basic Approval (Chronic Mammalian Toxicity). The US EPA (1994) applies a composite uncertainty factor of 100: 10 to account for interspecies extrapolation and 10 to account for the protection of sensitive subpopulations. No toxicological information was found for hypobromous acid, or the associated hypobromite salts. As previously reported, hypobromous acid is believed to be similar in toxicity to its chlorinated analogue, hypochlorous acid. Therefore, as indicated in the COPC selection (section 12.2.2), the Active Substance in the BalPure® BWMS is evaluated qualitatively as hypochlorous acid. 12.2.4.5 Total trihalomethanes During land-based testing, the trihalomethane (THM) by-products detected in ballast water included bromoform (tribromomethane), dichlorobromomethane, and chlorodibromomethane. Although not detected in samples from land-based testing, chloroform is another THM for


I:\MEPC\61\2-9.doc

which toxicity data is available. The US EPA has identified oral reference doses for four of these THMs in IRIS, with values ranging from 0.01 mg/kg-day (chloroform; US EPA 2001b) to 0.02 mg/kg-day (bromoform [US EPA 1991a], dichlorobromomethane [US EPA 1991b], and chlorodibromomethane [US EPA 1991c]). The lowest (and therefore most conservative) value is from the study of beagles administered oral doses (15 or 30 mg/kg-day) of chloroform in capsules for 5 days/week over 7.5 years for which a LOAEL of 15 mg/kg/day for liver effects, as is reported in Table 11 (Chronic Mammalian Toxicity) of the dossier for Basic Approval.1 The US EPA (2001b) derived the oral reference dose based on a modeled low benchmark dose (BMDL10) of 1.0 mg/kg-day, and included a composite uncertainty factor of 100: 10 to account for extrapolation from animal data to humans and 10 to account for intrahuman variability (US EPA 2009).2 As reported in Table 13 "Summary of Data on Carcinogenicity" of the Basic Approval dossier either IARC or the US EPA has classified each of the four THMs as "probable" or "possible" human carcinogens. Based on the studies identified in Table 13 of the Basic Approval dossier, US EPA, in IRIS, has derived oral slope factors for bromoform (7.3 x 10-3 per mg/kg-day; US EPA 1991a), dichlorobromomethane (6.2 x 10-2 per mg/kg-day; US EPA 1991b), and chlorodibromomethane (8.4 x 10-2 per mg/kg-day; 1991c)3. The highest value is the most conservative, therefore the oral slope factor for chlorodibromomethane is used to evaluate carcinogenic risks from THMs. 12.2.4.6 Total haloacetic acids As described in section 2, haloacetic acid (HAA) by-products detected in ballast water discharge during land-based testing included bromochloracetic acid, mono- and dichloroacetic acids, and mono- and di- and tribromoacetic acids, and dibromochloroacetic. Of these, the US EPA (2003) has identified an oral reference dose in IRIS for only dichloroacetic acid, however, this value is based on a subchronic (90 day study), which is not optimal for the evaluation of chronic effects. As described above, a chronic effects level can be developed based on the NOAELs or LOAELs reported for chronic toxicity studies in Table 11 of the Basic Approval. In section 11.3 of this dossier, no reported NOAELs or LOAELs could be located for the additional substances. The available NOAELs identified for HAAs in the Basic Approval tables range from 3.5 mg/kg/day for monochloroacetic acid (EU 2005) to 40 mg/kg/day for trichloroacetic acid (IPCS 2000a); all studies are based on effects on rodents, therefore a composite uncertainty factor of 100 (10 to account for extrapolation from animal data to humans and 10 to account for intrahuman variability) is appropriate for the determination of a reference dose from any of these studies. The lowest (and therefore most conservative) NOAEL-based calculated value is for monochloroacetic acid – 0.034 mg/kg/day. The LOAELs identified for HAAs range from 2 mg/kg/day for dibromoacetic acid (NTP 2007) to 40 mg/kg/day for dichloroacetic acid (IPCS 2000a); all studies are based on effects on rodents, therefore a composite uncertainty factor of 1000 (10 to account for the use of a LOAEL instead of a NOAEL, 10 to account for extrapolation from animal data to humans, and 10 to account for intrahuman variability) is appropriate for the determination of a reference dose from any of these studies. The lowest (and therefore most conservative) LOAEL-based value is for dibromoacetic acid – 0.002 mg/kg/day. Because the LOAEL-based value results in the lowest reference dose, it is used to evaluate non-carcinogenic risks from HAAs. 1 IRIS reference doses for bromoform and chlorodibromomethane are based on subchronic studies

associated with the same NTP studies (1989 and 1985, respectively) cited in Table 11 of the dossier for Basic Approval. The IRIS reference dose for dichlorobromomethane is based on the NTP (1987) chronic LOAEL of 25 mg/kg/d cited in Table 11 of the dossier for Basic Approval.

2 A similar oral reference dose resulted from the NOAEL/LOAEL approach. 3 USEPA has determined that while oral exposure to chloroform may cause cancer, the reference dose

of 0.01 mg/kg-day is considered to be protective against cancer risk.


I:\MEPC\61\2-9.doc

As reported in Table 13 (Summary of Data on Carcinogenicity) of this and the dossier for Basic Approval only dichloroacetic acid, dibromoacetic acid, and trichloroacetic acid have been classified as either "probable" or "possible" human carcinogens.4 However, IRIS includes oral slope factors for only dichloroacetic acid (0.05 per mg/kg-day; US EPA 2003), therefore this value is used to evaluate carcinogenic risks from HAAs. It is noted that while no cancer slope factor has been developed for the remaining HAAs, the effects levels reported in Table 13 of the dossier for Basic Approval are up to 20 times lower for dibromoacetic acid (≥ 4 mg/kg/day) than those reported for dichloroacetic acid (84 mg/kg/day). Therefore, some uncertainty exists in the carcinogenic evaluation of HAAs; specifically carcinogenic risks associated with dibromoacetic acid, the dominant residual HAA measured in ballast water (see Tables 22 and 23) may be underestimated. 12.2.4.7 2,4,6-Tribomophenol As shown in Tables 11 and 13, no long-term exposure or carcinogenic studies on brominated phenols have been identified (CICADS 66 2005). With respect to carcinogenicity, 2,4,6-tribromophenol has not been classified (NTP 2009, IARC 2009). However, given the very low levels of 2,4,6-TBP detected (two orders of magnitude lower than other COPCs; see Table 30) and the presence of this chemical in source water, any risks associated with this chemical as a result of BalPure® BWMS are expected to be very low. 12.2.4.8 Sodium bisulfite The US EPA has not developed an oral reference dose for sodium bisulfite inclusion in IRIS. However, as reported in Table 11 "Chronic Mammalian Toxicity" in the application dossier for Basic Approval, The Health Council of the Netherlands identified a NOAEL of 72 mg/kg-day for rats fed up to 20,000 mg/L sodium bisulfite, based on local effects (blood in feces; hyperplastic changes of gastric epithelium) (2005). In the absence of other information, an oral RfD can be calculated by applying a composite uncertainty factor of 100, 10 to account for interspecies extrapolation and 10 to account for intrahuman variability, to the NOAEL (72 mg/kg-day). This results in an oral RfD of 0.72 mg/kg-day. As reported in Table 13 "Summary of Data on Carcinogenicity" in the application dossier for Basic Approval, sodium bisulfite is not classifiable as to its carcinogenicity, therefore no oral cancer slope factor is available. As shown in Table 33, reference doses were identified for all COPCs except 2,4,6-tribromophenol. Cancer slope factors are available for only THMs and HAAs; therefore carcinogenic effects can be evaluated for only these three COPCs.

Table 33: Summary of Toxicity Criteria

Substance Oral Reference Dose (mg/kg/day) Oral Cancer Slope Factor

(per mg/kg-day)

Hypochlorous Acid 1.00E-01 (chlorine) NC

THMs 1.00E-02 (chloroform)* 8.40E-02 (chlorodibromomethane)

HAAs 2.00E-03 (dibromoacetic acid) 5.00E-02 (dichloroacetic acid)

2,4,6-Tribromophenol NC NC

Sodium bisulfite 7.20E-01 NC

* Although chloroform was not detected in the land-based testing of the Balpure® system, it was identified as having the most conservative oral reference dose and to maintain the health-protectiveness of the evaluation, is retained here.

NC = No criteria. 4 As indicated in Table 13, NTP (2009) reports “Clear evidence of carcinogenic activity” in animals for

bromochloroacetic acid.


I:\MEPC\61\2-9.doc

12.2.5 Risk characterization Risk characterization, the final step in the risk assessment process, combines data from the hazard identification, health effects assessment, and exposure assessment to estimate the potential carcinogenic and non-carcinogenic effects of COPCs over the applicable duration of exposure. Despite quantification, it is believed that the actual potential health risks are very low because of the low potential of exposure to chemicals associated with the BalPure® BWMS. Risks presented below should be considered worst-case estimates. Risks presented herein differ from the application dossier for Basic Approval in two ways: (1) risks for technicians or others who may be exposed during ballast water discharge sampling are evaluated quantitatively, and (2) risks to receptors previously evaluated for potential exposure to treated ballast water discharge have been revised using the land-based testing results to estimate exposure. 12.2.5.1 Acceptable risk levels The risk that is acceptable is very much dependent on site-specific characteristics that include: the number of people potentially exposed, the likelihood of exposure, the chemicals driving the risk, and the decisions of risk managers. The incremental lifetime cancer risk (ILCR) is compared to a range of acceptable probabilities to determine whether the potential risk poses an unacceptable health threat. The US EPA directive, Role of the Baseline Risk Evaluation in Superfund Remedy Selection Decisions (US EPA 1991a), states that action is generally not warranted at a site when the cumulative carcinogenic risk for current and future land use is less than 1 x 10-4 and the cumulative non-carcinogenic hazard index (HI) is less than 1.0. The level of non-cancer hazard concern increases as the HI increases above unity, although the two are not linearly related (US EPA 1989). The US EPA uses a potential excess individual lifetime cancer risk of 1 x 10-6 (1 in 1,000,000) as a point of departure for risk management actions. 12.2.5.2 Quantitative estimate of potential risks to STDN BalPure® technicians Table 34 summarizes the risks quantified for the BalPure® Technician who may be exposed to concentrated hypochlorous acid (generated concentration of 0.1%) during maintenance activities or to the 38% sodium bisulfite solution during chemical resupply. Neither hypochlorous acid nor sodium bisulfite is classifiable with respect to carcinogenicity, therefore cancer risks are not reported below. The non-cancer hazard of 0.3 is well below the level of concern (1.0); use of personal protective equipment (PPE) and safe chemical handling procedures will further reduce any potential exposures and associated effects.

Table 34: Summary of Cancer Risks and Non-cancer Hazards for BalPure® Technicians

Substance Cancer Risk Non-cancer Hazard

Ingestion Dermal Total Ingestion Dermal Total

Hypochlorous Acid NC NC NA 0.0049 0.00032 0.0052

Sodium Bisulfite NC NC NA 0.26 0.017 0.28 TOTAL NA 0.3

NA = not applicable, NC = no criteria.


I:\MEPC\61\2-9.doc

12.2.5.3 Quantitative estimate of potential risks to STDN BalPure® technicians (or others) during ballast water discharge sampling

Table 35 summarizes the risks quantified for the BalPure® Technician, or other worker, who may be exposed to COPCs in ballast water during discharge sampling, based on land-based testing maximum COPC concentrations. Hypochlorous acid, 2,4,6-tribromophenol, and sodium bisulfite are not classifiable with respect to carcinogenicity, therefore cancer risks are reported for only THMs and HAAs. The total estimated cancer risk of 8 x 10-8 (8 in 100,000,000) is far below the US EPA point of departure of 1 x 10-6 (1 in 1,000,000). The total non-cancer hazard, 0.002, is several orders of magnitude below the target level of 1.0.

Table 35: Summary of Cancer Risks and Non-cancer Hazards for BalPure® Technicians

(or others) – Ballast Water Discharge Sampling



Hypochlorous Acid NA NC NA NA 0.00018 0.00018

THMs NA 6.3E-08 6.3E-08 NA 0.00053 0.00053

HAAs NA 1.7E-08 1.7E-08 NA 0.0012 0.0012

2,4,6-Tribromophenol NA NC NC NA NC NC

Sodium Bisulfite NA NC NC NA 0.000010 0.000010 TOTAL 8E-08 0.002


12.2.5.4 Quantitative estimate of potential risks to ship's crew or dock workers Table 36 summarizes the risks quantified for ship's crew or dock workers who may be exposed to COPCs in ballast water from spray drift, based on land-based testing maximum COPC concentrations. Hypochlorous acid, 2,4,6-tribromophenol, and sodium bisulfite are not classifiable with respect to carcinogenicity, therefore cancer risks are reported for only THMs and HAAs. The total estimated cancer risk of 2 x 10-9 (2 in 1,000,000,000) remains essentially unchanged from that estimated in the Basic Approval dossier, and is far below the US EPA point of departure of 1 x 10-6. The total non-cancer hazard, 0.00004, is only slightly higher than reported in the previous risk assessment and remains many orders of magnitude below the target level of 1.0.

Table 36: Summary of Cancer Risks and Non-cancer Hazard for Ship's Crew/Dock Workers



Hypochlorous Acid NA NC NA NA 0.0000035 0.0000035

THMs NA 1.3E-09 1.2E-09 NA 0.000010 0.000010

HAAs NA 3.2E-10 3.2E-10 NA 0.000023 0.000023

2,4,6-Tribromophenol NA NC NC NA NC NC

Sodium Bisulfite NA NC NC NA 0.00000020 0.00000020 TOTAL 2E-09 0.00004



I:\MEPC\61\2-9.doc

12.2.5.5 Quantitative estimate of potential risks to the general public (beachgoers) Table 37 shows cancer risks to beachgoers who may swim in the vicinity of recently discharged ballast water associated with the BalPure® BWMS, based on land-based testing maximum COPC concentrations (assumed to be instantly diluted 100-fold). Hypochlorous acid, 2,4,6-tribromophenol, and sodium bisulfite are not classifiable with respect to carcinogenicity, therefore cancer risks are reported for only THMs and HAAs. The total estimated cancer risk is very low at 3 x 10-8 (3 in 100,000,000), is slightly lower than estimated in the Basic Approval dossier and is far below the US EPA point of departure of 1 x 10-6.

Table 37: Summary of Cancer Risks for the General Public

Substance Cancer Risk*

Ingestion Dermal Total

Hypochlorous Acid NC NC NA

THMs 1.0E-08 1.7E-08 2.7E-08

HAAs 2.7E-09 4.4E-09 7.1E-09

2,4,6-Tribromophenol NC NC NC

Sodium Bisulfite NC NC NC TOTAL 3E-08

Cancer risk is age-adjusted, assuming 6 years of exposure as a child and 24 years as an adult. NC = no criteria.

Non-cancer hazards for the general public, which are summarized in Table 38, also are very low; 0.0007 for children and 0.0002 for adults. Although slightly higher than reported in the Basic Approval dossier, these non-cancer hazards remain many orders of magnitude below the target level of 1.0.

Table 38: Summary of Non-cancer Hazards for General Public

Substance Adult Non-cancer Hazard Child Non-cancer Hazard


Hypochlorous Acid 0.000039 0.000014 0.000053 0.00018 0.000024 0.00021

THMs 0.000017 0.000040 0.000057 0.000077 0.000069 0.00015

HAAs 0.000037 0.000090 0.00013 0.00017 0.00015 0.00033

2,4,6-Tribromophenol NC NC NC NC NC NC

Sodium Bisulfite 0.0000022 0.00000078 0.0000030 0.000010 0.0000013 0.000012 TOTAL 0.0002 0.0007

NC = no criteria 12.2.5.6 Semi-quantitative evaluation of potential risks from inhalation The risk estimates reported in Tables 34 to 38 do not include inhalation of COPCs in indoor or outdoor air. While significant volatilization of COPCs from ballast water into indoor or outdoor air is not anticipated, some exposure to COPCs in air could occur. The most likely operation that could result in exposure to COPCs in air is during ballasting, when air in the ballast tanks is vented to the atmosphere as water is brought into the tanks. If the ship's crew is present during this operation, they could be exposed to residual THMs in vent air, since these disinfection by-products are both volatile and hydrophobic. Although concentrations of THMs in vent air have not been measured, worst-case estimates of air concentrations can be made based on maximum ballast water THM concentrations (see section 11.3) and the


I:\MEPC\61\2-9.doc

appropriate Henry's Law constants (the ratio of the aqueous-phase concentration of a chemical to its equilibrium partial pressure in the gas phase). Because vent air will be immediately mixed into the surrounding atmosphere, the air concentration estimates are also adjusted for dilution into ambient air by dividing by a factor 10. As a conservative evaluation of the inhalation exposure pathway, Table 39 summarizes these data (updated to reflect land-based test results), and compares the revised estimated air concentrations to U.S. OSHA Permissible Exposure Levels (PELs).

Table 39: Estimated Concentrations of THMs in Vent Air and Comparisons to PELs

Substance

Basic Approval Dossier

Maximum Residual

Ballast Water Concentration

(µg/L)

Final Approval Dossier

Maximum Residual

Ballast Water Concentration

(µg/L)

Henry's Law

Constant (unitless)

Final Approval Dossier

Estimated Conc. in

Air (mg/m3)*

Air Exposure

Limit (mg/m3)**

Air Concentration

Exceeds Exposure Limit?

THMs Bromoform 1050 810 0.0218 1.8 5 No

Chloroform 53 Not detected 0.150 Not

detected 240 No


157 33 0.0320 0.11 NA NA


90 1.5 0.0654 0.0098 NA NA

NA = Not Available. * Estimated based on Henry's Law constant at 25°C, and divided by a factor of 10 for dilution into ambient air. ** U.S. Department of Labor, 2009. Occupational Health and Safety Administration, Permissible Exposure

Limits (PEL). Actual air exposures will depend on proximity to vent outlets, atmospheric conditions (wind), as well as the duration of ballasting. Because the THMs in vent air would quickly dissipate into the atmosphere, it is anticipated that air exposures will decrease as distance to the ballast water increases; therefore potential air exposures for dock workers and the general public would be even less than those for the ship's crew. Based on this assessment of THMs in vent air as a result of concentrations in ballast water, there is no evidence for the potential of adverse health effects to the ship's crew, dock workers, or the general public. 12.2.6 Risk assessment conclusions Overall risk findings from the application dossier for Basic Approval remain unchanged: the health risks to all receptors from the BalPure® BWMS under normal operating conditions are very small and well within acceptable levels. Even an STDN technician, exposed to concentrated sodium bisulfite during a spill during chemical resupply, and who is wearing no PPE – an unlikely scenario – would have low health risk. Cancer risk and non-cancer hazards for the general public/beachgoers are well below levels of concern; actual risks to these receptors are expected to be even lower, as they are unlikely to be repeatedly exposed to treated ballast water as assumed in this risk evaluation. Since the risk assessment only evaluated direct exposure via ingestion and dermal contact with chemicals or ballast water, as these were considered the most relevant pathways, further semi-qualitative evaluation of the inhalation pathway was performed for the volatile THMs. This assessment demonstrated that conservative concentrations of airborne THMs generated by a release of ballast water were well below occupational exposure limits.


I:\MEPC\61\2-9.doc

12.3 Risks to the aquatic environment Potential risk to the aquatic environment as a result of the BalPure® BWMS was evaluated in several regards. The substances potentially present in treated/neutralized ballast water discharge were evaluated with respect to aquatic ecotoxicity, bioaccumulation potential and environmental persistence and the whole effluent was evaluated for discharge ecotoxicity. The Basic Approval dossier (MEPC 60/2/9) presented data from pilot scale studies, while this document presents data generated during land-based testing of the full commercial scale BalPure® system. Additionally, emission scenarios were simulated using the MAMPEC model to generate meaningful PEC values to allow for a risk assessment with the PEC/PNEC ratio method. Direct ecotoxicity evaluation of whole effluent from the BalPure® system indicates that the potential risks to the aquatic environment are very low. The acute and chronic WET testing data indicate that 100% whole effluent did not have toxic effects on five of the six species tested (Tables 3 and 5). For one species, Acartia tonsa, an effect was observed in one of four acute tests with a low salinity sample. In this sample, a moderate effect was observed in the two highest test concentrations (100% and 32% sample), resulting in a NOEC of 18% volume (Grontmij|AquaSense, 2009). It is important to note that during tests with Acartia moderate effects were also noted in a control (untreated, NIOZ harbour water) sample (EC20 of 58% sample) (Grontmij|AquaSense, 2009). As explained in sections 11.4 and 11.13 above, a possible explanation for the moderate toxic effect in the treated and control samples is the presence of Phaeocystis, an algal species known to be abundant in the NIOZ harbour and which excretes allelochemicals. All other acute and chronic tests with Acartia, as well as the five other species tested, demonstrate that 100% whole effluent sample had no toxic effect, indicating that the NOEC of 18% (Table 3) is an outlying data point. The conclusion of low potential risk to the aquatic environment is supported by the results of worst-case discharge scenarios simulated with the MAMPEC model. By employing the modelling approach described in section 11.8.1, it is important to note the layers of conservatism applied. First, MAMPEC assumes continuous discharges (leaching from antifoulants), rather than intermittent discharges (ballast water). Second, the physical dimensions of the "OECD Commercial Harbour" configuration are smaller and tidal flushing is less, yet the ballast water discharge rate for the whole Port of Rotterdam was used. Environment B builds onto the worst case emission scenario with a smaller tidal exchange flow rate, minimizing the harbour exchange percentage due to flushing. These elements result in overly conservative, rather than realistic, PEC values. Lastly, in derivation of PNEC values the lowest effect concentrations and highest appropriate assessment factors have been used (see section 11.9). All PEC/PNEC ratios, even when calculated using PEC values generated in the very conservative "reduced flushing" Environment B, are well below 1 and can be considered as protective of the aquatic environment. Therefore, no further assessment of direct toxic effects is necessary as risks to the aquatic environment are not expected from the use of the BalPure® system. 13 ASSESSMENT REPORT The substances associated with the BalPure® system were evaluated for a variety of endpoints, including ecotoxicity, bioaccumulation, and persistence in the environment. Additionally, STDN's application dossier for Basic Approval (MEPC 60/2/9) included a full assessment (including ship safety and human health) with respect to the onboard storage of sodium bisulfite and the onboard generation of sodium hypochlorite. During the assessment of potential risks related to the BalPure® system, a variety of reputable sources were consulted to gather test data and conduct the assessments required. These sources include, but are not limited to the WHO, IPCS, OECD and US EPA. Information was also gathered from published independent research, which in most cases is peer-reviewed by others with


I:\MEPC\61\2-9.doc

similar levels of expertise. Further, reputable laboratories that utilize recognized testing methods and appropriate quality control measures were commissioned for the toxicological and analytical evaluations of ballast water discharge. Based on these facts, the level of uncertainty associated with this assessment is low. Aquatic toxicity was established as being the most likely potential risk to the aquatic environment. Because TRO (as Cl2) is neutralized prior to discharge, the DBPs potentially present in treated/neutralized ballast water discharge are the most important to consider for this evaluation. In regards to aquatic ecotoxicity, the existing database for chronic endpoints is limited for many of the THMs and HAAs. As such, the acute database, which is more complete, was utilized for some substances. The evaluation of the chemical and physical properties of all chemicals relevant to the BalPure® system suggests that bioaccumulation, sediment adsorption, or environmental persistence is not expected to occur. Further, none of the chemicals met all three criteria to be classified as PBT substances. Whole effluent toxicity (WET) testing was performed with treated/neutralized samples drawn from the land-based test set-up. With the exception of one acute test with Acartia tonsa, all other acute and chronic discharge toxicity tests indicate no aquatic toxicity. Because only one in eight tests with Acartia indicated moderate toxicity, and toxic effects were also observed in a control sample during testing with this species, the NOEC of 18% sample was likely to have been caused by compounds excreted by the algal species Phaeocystis. All other tests for the five remaining species of aquatic organisms resulted in effect concentrations of ≥100% sample. Lastly, two different emission scenarios (Environments A and B) were simulated with MAMPEC using the median DBP concentrations in ballast discharge. The resulting PEC values allowed for calculation of PEC/PNEC ratios to estimate aquatic risk. All calculated ratios for the substances measured in ballast water discharge are well below 1. Based on this assessment, no potential risks as a result of ballast water discharge from the BalPure® system are anticipated. 14 REFERENCES Aguayo, J., Barra, R., Becerra, J., and Martinez, M. 2009. Degradation of 2,4,6-tribromophenol and 2,4,6-trichlorophenol by aerobic heterotrophic bacteria present in psychrophilic lakes. World J. Microbiol. Biotechnol. Vol. 25 (pp 553-560). Alfa Aesar Material Safety Data Sheet. 2009. Tribromoacetic acid. Alfa Aesar Material Safety Data Sheet. 2009. Tribromophenol. Arthur, J.W., Andrew, R.W., Mattson, V.R., Olson, D.T., Glass, G.E. and Halligan, B.J. 1975. Comparative Toxicology of Sewage-effluent disinfection to fresh water aquatic life. Environmental Research Laboratory, Office of Research and Development, U.S. EPA. Duluth, Minnesota. BMT Fleet Technology. 2004. Studies to Address the Issues Raised by the MESB (2002) Critical Review of a Ballast Water Biocide Treatment Demonstration Project Using Sodium Hypochlorite. State of Michigan, Office of the Great Lakes. Brooks, A.S. & Liptak, N.E. 1978. The Effect of Intermittent Chlorination on Fresh water Phytoplankton. Pergamon Press, Ltd., Great Britain. Water Research, Vol. 13 (pp 49-52). Concise International Chemical Assessment Document 66 (CICAD). 2005. 2,4,6-Tribromophenol and other simple brominated phenols. Cowgill, U.M., D.P. Milazzo, and B.D. Landenberger. 1989. Toxicity of nine benchmark chemicals to Skeletonema costatum, a marine diatom. Environ. Toxicol. Chem. Vol. 8 (pp 451-455).


I:\MEPC\61\2-9.doc

Deltares, Jos van Gils. 2010. Modelling of PECs due to treated ballast water discharges. (STDN proprietary information, non-public document). Ellis, D.A., Hanson, M.L., Sibley, P.K., Tazeen, S., Fineberg, N.A., Solomon, K.R., Muir, D.C.G., and Mabury, S.A. 2001. The fate and persistence of trifluoroacetic and chloroacetic acids in pond waters. Chemoshpere. Vol. 42 (pp 309-318). Estimation Programs Interface (EPI). 2009. Version 4.0. Bromate. Estimation Programs Interface (EPI). 2009. Version 4.0. Bromoacetic acid. Estimation Programs Interface (EPI). 2009. Version 4.0. Dibromochloroacetic acid. Estimation Programs Interface (EPI). 2009. Version 4.0. Sodium bisulfite. Eurofins-Analytico. 2009. Certificates of analysis. Severn Trent De Nora ballast water discharges. (STDN proprietary information, non-public document). European Union (EU), 2007. Risk Assessment Report for Chloroform. http://echa.europa.eu/doc/trd_substances/chloroform/rar/trd_rar_hh_france_chloroform.pdf European Union (EU), 2007. Risk Assessment Report for Sodium Hypochlorite. http://ecb.jrc.ec.europa.eu/documents/Existing-Chemicals/RISK_ASSESSMENT/SUMMARY/sodiumhypochloritesum045.pdf Fisher, D.J., D.T. Burton, L.T. Yonkos, S.D. Turley, B.S. Turley, G.P. Ziegler, and E.J. Zillioux. 1994. Acute and short-term chronic effects of continuous and intermittent chlorination on Mysidopsis bahia and Menidia beryllina. Environ. Toxicol. Chem. 13:1525-1534. Fisher, D.J., Burton, D.T., Yonkos, L.T., Turley, S.D., and Gregory, P.Z. 1999. The relative acute toxicity of continuous and intermittent exposures of chlorine and bromine to aquatic organisms in the presence and absence of ammonia. Wye Research and Education Center, The University of Maryland, P.O. Box 169, Queenstown, MD 21658, USA.

Giller, S., Le Curleux, F., Erb, F., and Marzin, D. 1997. Comparative genotoxicity of halogenated acetic acids found in drinking water. Mutagenesis. Vol 12, No. 5 (pp 321-328).

Goodman, L.R., Middaugh, D.P., Hansen, D.J., Hidgon, P.K, and Cripe, G.M. 1983. Early life-stage toxicity test with tidewater silversides (Menidia peninsulae) and chlorine produced oxidants. U.S. Environmental Protection Agency, Environmental Research Laboratory, Sabine Island, Gulf Breeze, FL.

Goodwin, K.D., Wheeler J.N., and Lidstrom, M.E. 1997. Production of Bromoform and Dibromomethane by Giant Kelp: Factors Affecting Release and Comparison to Anthropogenic Bromine Sources. Limnology and Oceanography. Vol. 42, No. 8 (pp 1725-1734).

Grontmij|AquaSense. 2010. Report of ecotoxicological studies. Samples from land-based testing NIOZ 2009. (STDN proprietary information, non-public document).

Hanson, M.L., Sibley, P.K., Ellis, D.A., Mabury, S.A., Muir, D.C.G., and Solomon, K.R. 2002. Evaluation of monochloroacetic acid (MCA) degradation and toxicity to Lemna gibba, Myriophyllum spicatum, and Myriophyllum sibiricum in aquatic microcosms. Aquatic Toxicology. Vol. 61 (pp 251-273).

Hashimoto, S., Tadashi, A., and Akira, O. 1998. Distribution, sources, and stability of haloacetic acids in Tokyo Bay, Japan. Envrionmental Toxicology and Chemistry. Vol. 17, No. 5 (pp 798-805).

Hassenklöver, T., Predehl, S., Pilli, J., Ledwolorz J., Assmann, M., and Bickmeyer, U. 2006. Bromophenols, both present in marine organisms and in industrial flame retardants, disturb cellular Ca2+ signalling in neuroendocrine cells (PC12). Aquatic Toxicology. Vol. 76 (pp 37-45).


I:\MEPC\61\2-9.doc

Hazardous Substances Databank (HSDB), Toxicology Data Network (TOXNET). 2009. Bromoacetic acid. National Library of Medicine. http://toxnet.nlm.nih.gov

Hazardous Substances Databank (HSDB), Toxicology Data Network (TOXNET). 2009. Bromochloroacetic acid. National Library of Medicine. http://toxnet.nlm.nih.gov/

Hazardous Substances Databank (HSDB), Toxicology Data Network (TOXNET). 2009. Chloroform. National Library of Medicine. http://toxnet.nlm.nih.gov/

Hazardous Substances Databank (HSDB), Toxicology Data Network (TOXNET). 2009. Tribromoacetic acid. National Library of Medicine. http://toxnet.nlm.nih.gov

Hazardous Substances Databank (HSDB), Toxicology Data Network (TOXNET). 2009. Tribromomethane. National Library of Medicine. http://toxnet.nlm.nih.gov

Hazardous Substances Databank (HSDB), Toxicology Data Network (TOXNET). 2009. Tribromophenol. National Library of Medicine. http://toxnet.nlm.nih.gov/cgi-bin/sis/search/f?./temp/~bLXIef:2

Hermanutz, R.O., Allen, K.N., and Hedtke, S.F. Toxicity and Fate of Total Residual Chlorine in Outdoor Experimental Streams. In Water Chlorination: Chemistry, Environmental Impact and Health Effects. R.L. Jolley et al. (Eds.), Lewis Publishers, Inc., Chelsea, Mich., 6, 463-477.

Herwig, R.P., Cordell, J.R., Nielsen, B.C., Ferm, N.S., Lawrence, D.J., Perrins, J.C., and A.C.E. Rhodes. 2006. Efficacy testing of the Severn Trent De Nora BalPure® System. School of Aquatic and Fishery Sciences. University of Washington, Seattle, WA. (STDN proprietary information, non-public document).

Hunter, E.S. III, Rogers, E., Blanton, M., Richard, A., and Chernoff, N. 2006. Bromochloro-haloacetic acids: Effects on mouse embryos in vitro and QSAR considerations. Reproductive Toxicology. Vol. 21 (pp 260-266).

Hunter, E.S. III, Rogers, E.H., Schmid, J.E., and Richard, A. 1996. Comparative Effects of Haloacetic Acids on Whole Embryo Culture. Teratology. Vol. 54 (pp 57-64).

Hutchinson, T.H., Hutchings, M.J. & Moore, K.W. 1997. A Review of the Effects of bromate on aquatic organisms and toxicity of bromate to oyster (Crassostrea gigas) embryos. Ecotoxicology and Environmental Safety. Vol. 38 (pp. 238-243).

International Agency for Research on Cancer (IARC). 2009. Agents Reviewed by the IARC Monographs, Vol. 1-99.

International Association of Classification Societies Ltd (IACS). 2007. Confined Space Safe Practice. Rec. No. 72.

International Programme on Chemical Safety (IPCS). 2000. Disinfectants and disinfectant by-products. Environmental Health Criteria 216. Kelley, D.G. 2004. Chlorine disinfectants and ORP control. Washington Tree Fruit Postharvest Conference. Yakima, WA.

Kim, S.J., and Jang, S.K. 2009. Corrosion characteristics of steel in seawater containing various chloride concentrations generated by electrochemical methods. Transactions of Nonferrous Metals Society of China. Vol. 19 (pp. 50-55). Kuhn, R. and Pattard, M. 1990. Results of the harmful effects of water pollutants to green algae (Scenedesmus subspicatus) in the cell multiplication inhibition test. Water Research. Vol. 24, No. 1 (pp. 31-38).

Lee, B.C., Kamata, M., Akatsuka, Y., Takeda, M., Ohno, K., Kamei, T., and Yasumoto, M. 2004. Effects of chlorine on the decrease of estrogenic chemicals. Water Research, Vol. 38 (pp. 733-739).


I:\MEPC\61\2-9.doc

Mattice, J.S., Tsai, S.C., Burch, M.B., and Beauchamp, J.J. 1981. Toxicity of Trihalomethanes to Common Carp Embryos. Transactions of the American Fisheries Society, 110:261-269. National Toxicology Program (NTP). 1998. Final Report on the Short Term Reproductive and Developmental Toxicity of Tribromoacetic Acid (CAS #75-96-7) Administered in Drinking Water to Srague-Dawley Rats. U.S. Department of Health and Human Services, Public Health Service. National Toxicology Program (NTP). 2009. Report on Carcinogens, Eleventh Edition; U.S. Department of Health and Human Services, Public Health Service. National Toxicology Program (NTP). 2007. Technical Report Series No. 537. Toxicology and Carcinogenesis Studies of Dibromoacetic Acid in F344/n Rats and B6C3F1 mice (drinking water studies) http://ntp.niehs.nih.gov/files/537_web.pdf Nautilus Environmental, 2009. Toxicity testing results for Severn Trent de Nora, Sodium Bisulfite. (STDN proprietary information, non-public document) Naval Research Laboratory (NRL). 2007. Pilot Environmental Technology Verification (ETV) Test Report of the Severn Trent De Nora BalPure® Ballast Water Treatment System. Environmental Standards Division (CG-5224), U.S. Coast Guard, Washington, DC. (STDN proprietary information, non-public document). Nelson, G.M., Swank, A.E., Brooks, L.R., Bailey, K.C., and George, S.E. 2001. Metabolism, Microflora Effects, and Genotoxicity in Haloacetic Acid-Treated Cultures of Rat Cecal Microbiota. Toxicological Sciences. Vol. 60 (pp 232-241). NovaChem Laboratories, Inc. 2009. Analysis Results for Bromate / Chlorate Ion Analysis NIOZ samples. (STDN proprietary information, non-public document).

Official Journal of the European Union (EC Directive). 2003. Directive 2003/10/EC of the European Parliament and of the Council. (38-44).

Ohsawa, N., Yasuhiro, O., Noriyuki O., and Nobuya, I. 2001. Physiological function of bromoperoxidase in the red marine alga, Corallina pilulifera: production of bromoform as an allelochemical and the simultaneous elimination of hydrogen peroxide. Phytochemistry. Vol. 58 (pp 683-692).

Organization for Economic Cooperation and Development (OECD) Screening Information Data Set (SIDS). 2003. Tribromophenol.

Organization for Economic Cooperation and Development (OECD) Screening Information Data Set (SIDS). 2000. Trichloroacetic acid.

Pratt, J.R., Bowers, N.J., Niederlehner B.R, and Cairns J. Effects of chlorine on microbial communities in naturally derived microcosms. 1988. University Center for Environmental and Hazardous Materials Studies, Virginia Polytechnic Institute and State University, Blacksburg, VA.

Richardson, L.B., D.T. Burton, and J.C. Rhoderick. 1981. Toxicity of bromate to striped bass (Saxatilis) and juvenile spot (Leiostomus xanthurus). J. Toxicol. Environ. Health Vol. 8 (pp 687-695)

Richardson, S.D., Plewa, M.J., Wagner, E.D., Schoeny, R., DeMarini, D.M. 2007. Occurrence, genotoxicity, and carcinogenicity of regulated and emerging disinfection by-products in drinking water: A review and roadmap for research. Mutation Research. Vol. 636 (pp 178-242).

Saghir, S.A. and Schultz, I.R. 2005. Toxicokinetics and Oral Bioavailability of Halogenated Acetic Acids Mixtures in Naïve and GST zeta-Depleted Rats. Toxicological Sciences. Vol. 84 (pp 214-224).

Sanders J.G., Ryther, J.H., and Batchelder, J.H. 1981. Effects of copper, chlorine, and thermal addition on the species composition of marine phytoplankton. J. exp. Mar. Biol. Ecol., Vol. 49 (pp 81-102) Elsevier/North Holland Biomedical Press.


I:\MEPC\61\2-9.doc

Saroglia, M.G. & Scarano G. 1979. Influence of Molting on the Sensitivity to toxics of the crustacean Penaeus kerathurus. Ecotoxicology and Environmental Safety. Vol. 3, No. 3 (pp 310-320). Schiliro, T., C. Pignata, R. Rovere, E. Fea, and G. Giorgio. 2009. The endocrine disrupting activity of surface waters and of wastewater treatment of plant effluents in relation to chlorination. Department of Public Health and Microbiology: University of Torino, Via Santena, Italy. Sigma Aldrich Material Safety Data Sheet. 2009. Chlorodibromoacetic acid. Solvay Chemicals. 2007. Material Safety Data Sheet. Sodium bisulfite. Song, Y., Dang, K., Chi, H., and Guan, D. 2009. Corrosion of marine carbon steel by electrochemically treated ballast water. Journal of Marine Engineering and Technology. No. A13. Sosik, E. and Herwig, R. 2009. Disinfection byproduct chemistry and aquatic toxicity. University of Washington School of Aquatic and Fishery Sciences, Fresh water Facility. Seattle, WA. (STDN proprietary information, non-public document). Toussaint, M.W., Brennan, L.M., Rosencrance, A.B., Dennis, W.E., Hoffman, F.J., and Gardner, H.S. 2001. Acute Toxicity of Four Drinking Water Disinfection By-products to Japanese Medaka Fish. Bull. Environ. Contam. Toxicol (2001) 66:255-262. U.S. Environmental Protection Agency (US EPA). 1991. Role of the Baseline Risk Evaluation in Superfund Remedy Selection Decisions. OSWER Directive 9355.0-30. U.S. Environmental Protection Agency (US EPA). 2002. Supplemental Guidance for Developing Soil Screening Levels at Superfund Sites. December. OSWER 9355.4-24. U.S. Environmental Protection Agency (US EPA). 2004. Risk Assessment Guidance for Superfund (RAGS), Vol., I, Human Health Evaluation Manual, Part E, Supplemental Guidance for Dermal Risk Assessment, Final. EPA/540/R/99/005. U.S. Environmental Protection Agency (US EPA). Aquatic Report on Bromate on the Ecotoxicology (ECOTOX) database (Erickson, S.J., and Freeman, A.E., 1978). http://cfpub.epa.gov/ecotox/report.cfm

U.S. Environmental Protection Agency (US EPA). Aquatic Report on Bromate on the Ecotoxicology (ECOTOX) database (Crecelius, E.A., 1979). http://cfpub.epa.gov/ecotox/report.cfm

U.S. Environmental Protection Agency (US EPA). Aquatic Report on Bromoacetic acid on the Ecotoxicology (ECOTOX) database (Kuehn, R., Pattard, M., Pernak, K.D., and Winter, A., 1989). http://cfpub.epa.gov/ecotox/report.cfm

U.S. Environmental Protection Agency (US EPA). Aquatic Report on Bromoacetic acid on the Ecotoxicology (ECOTOX) database (Kuhn, R., 1988). http://cfpub.epa.gov/ecotox/report.cfm

U.S. Environmental Protection Agency (US EPA). Aquatic Report on Bromoacetic acid on the Ecotoxicology (ECOTOX) database (Loeb, H.A., and Kelly, W.H., 1963). http://cfpub.epa.gov/ecotox/report.cfm

U.S. Environmental Protection Agency (US EPA). Aquatic Report on Bromoform on the Ecotoxicology (ECOTOX) database (Heitmueller, P.T., Hollister, T.A., and Parrish, P.R., 1981). http://cfpub.epa.gov/ecotox/report.cfm

U.S. Environmental Protection Agency (US EPA). Aquatic Report on Bromoform on the Ecotoxicology (ECOTOX) database (Ward, G.S., Parrish, P.R., and Rigby, R.A., 1981). http://cfpub.epa.gov/ecotox/report.cfm

U.S. Environmental Protection Agency (US EPA). 1978. Aquatic Report on Bromoform on the Ecotoxicology (ECOTOX) database. http://cfpub.epa.gov/ecotox/report.cfm


I:\MEPC\61\2-9.doc

U.S. Environmental Protection Agency (US EPA). Aquatic Report on Bromoform on the Ecotoxicology (ECOTOX) database (LeBlanc, G.A., 1980). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Chloroacetic acid on the Ecotoxicology (ECOTOX) database (Loeb, H.A., and Kelly, W.H., 1963). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Chloroacetic acid on the Ecotoxicology (ECOTOX) database (Trenel, J., and Kuhn, R., 1982). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Chlorodibromomethane acid on the Ecotoxicology (ECOTOX) database (Yoshioka, Y., Ose, Y., and Sato, T., 1985). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Chloroform on the Ecotoxicology (ECOTOX) database (Anderson, D.R., and Lusty, E.B., 1980). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Chloroform on the Ecotoxicology (ECOTOX) database (Foster, J.D., and Tullis, R.E., 1984). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Chloroform on the Ecotoxicology (ECOTOX) database (Kuhn, R., 1988). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Dibromoacetic acid on the Ecotoxicology (ECOTOX) database (Mayes, M.A., Blanchard, F.A., Hopkins, D.L., and Takahashi, I.T., 1985). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Dichloroacetic acid on the Ecotoxicology (ECOTOX) database (Linden, E., Bengtsson, B.E., Svanberg, O., and Sundstrom, G, 1979). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Dichloroacetic acid on the Ecotoxicology (ECOTOX) database (Trenel, J. and Kuhn, R., 1982). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Dichloroacetic acid on the Ecotoxicology (ECOTOX) database (Fort, D.J., Stover, E.L., Rayburn, J.R., Hull, M., and Bantle, J.A., 1993). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Dichlorobromomethane on the Ecotoxicology (ECOTOX) database (Yoshioka, Y., Ose, Y., and Sato, T., 1985). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Hypobromous acid on the Ecotoxicology (ECOTOX) database (Kott, Y., and Edlis, J., 1969). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Hypochlorous acid on the Ecotoxicology (ECOTOX) database (Holland, G.A., Lasater, J.E., Neumann, E.D., and Eldridge, W.E, 1960). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Hypochlorous acid on the Ecotoxicology (ECOTOX) database (Thatcher, 1978). http://cfpub.epa.gov/ecotox/report.cfm


I:\MEPC\61\2-9.doc

U.S. Environmental Protection Agency (US EPA). Aquatic Report on Hypochlorous acid on the Ecotoxicology (ECOTOX) database (Maruyama, T., Ochiai, A., and Yoshida, T., 1988). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Sodium bisulfite on the Ecotoxicology (ECOTOX) database (Freeman, L., and Fowler, I., 1953). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Sodium bisulfite on the Ecotoxicology (ECOTOX) database (Wallen, I.E., Greer, W.C., and Lasater, R., 1957). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Sodium bisulfite on the Ecotoxicology (ECOTOX) database (Dowden, B.F., and Bennet, H., 1965). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Tribromophenol on the Ecotoxicology (ECOTOX) database (Phipps, G.L., Holcombe, G.W., and Fiandt, J.T., 1981). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Trichloroacetic acid on the Ecotoxicology (ECOTOX) database (Huang, J.C., and Gloyna, E.F., 1968). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Trichloroacetic acid on the Ecotoxicology (ECOTOX) database (Dennis, W.H.J., Meier, E.P., Randall, W.F., Rosencrance, A.B., and Rosenblatt, D.H., 1979). http://cfpub.epa.gov/ecotox/report.cfm U.S. Environmental Protection Agency (US EPA). Aquatic Report on Trichloroacetic acid on the Ecotoxicology (ECOTOX) database (Garten, C.T.Jr., 1990). http://cfpub.epa.gov/ecotox/report.cfm Van Niekerk, H. 2008. Ballast water quantities Port of Rotterdam. Ballast water intake and discharge February 2008. Port of Rotterdam report no 9T2104. Velduis, M.J.W., Kools, S.A.E. 2009. Side effects associated to the presence of Phaeocystis in Ballast Water Treatment research. Videau, C., Khalanski, M, and Penot, M. 1979. Preliminary results concerning effects of chlorine on monospecific marine phytoplankton. J. exp. Mar. Biol. Ecol., Vol. 36 (pp. 111-123) Elsevier/North-Holland Biomedical Press. Westerhoff, P., P. Chao, and H. Mash. 2003. Reactivity of natural organic matter with aqueous chlorine and bromine. Water Research, Vol. 38 (pp. 1502-1513). Woo, Y., Lai, D., McLain, J.L., Manibusan, M.K., and Dellarco, V. 2002. Use of Mechanism-based Structure-Activity Relationships Analysis in Carcinogenic Potential Ranking for Drinking Water Disinfection By-products. Environmental Health Perspectives. Vol. 110, No.1 (pp 75-87). Yan, S., Xiancheng, Z., Lie, M., Limli, L., and Chengrong, L. 2007. Evaluation of antioxidants using oxidation reaction rate constants. Frontiers of Chemisty in China. Vol. 2, No.2 (pp 140-145). Yang, J.S. 2001. Bromoform in the effluents of a nuclear power plant: a potential tracer of coastal water masses. Hydrobiologia. Vol. 464 (pp 99-105).

___________