ballast water report - · pdf fileeffective new technologies for the assessment of compliance...

EFFECTIVE NEW TECHNOLOGIES FOR THE ASSESSMENT OF COMPLIANCE WITH THE BALLAST WATER MANAGEMENT CONVENTION FINAL REPORT

1/90

EFFECTIVE NEW TECHNOLOGIES FOR THE ASSESSMENT OF COMPLIANCE WITH THE

BALLAST WATER MANAGEMENT CONVENTION FINAL REPORT

0014-00007/ R0

14/11/2014 Prepared by Prepared for

SGS Institut Fresenius GmbH

Federal German Hydrographic and Maritime Agency

Im Maisel 14 65323 Taunusstein/Germany

Bernhardt-Nocht-Strasse 78 20359 Hamburg/Deutschland

This report is approved by -------------------------------------------------------- Dr. Lothar Schillak Senior Marine Biologist SGS Institut Fresenius GmbH

-------------------------------------------------------- Hans-Peter Heuser Director Environmental Services SGS Institut Fresenius GmbH

2/90

0014-00007 R0

REPORT CONTENT

EXECUTIVE SUMMARY .............................................................................................. 5

LIST OF ABBREVIATIONS .......................................................................................... 6

1. PROJECT BACKGROUND ................................................................................ 7

2. HISTORY OF THE PROJECT 2012 TO 2014 ..................................................... 7

3. RELEVANT IMO REGULATIONS ....................................................................... 8

3.1 IMO GUIDELINES ............................................................................................... 8 3.1.1 Guidelines for ballast water sampling (G2) ................................................. 9 3.1.2 Guidelines for approval of ballast water management systems (G8) .......... 9 3.1.3 Guidelines for port state control (MEPC 67/20) .......................................... 9

3.2 IMO REQUIREMENTS ........................................................................................ 10

4. GENERAL ASPECTS OF BALLAST WATER .................................................. 11

5. CONSIDERATIONS REGARDING PROJECT OBJECTIVES........................... 11

5.1 BALLAST WATER SAMPLING ............................................................................... 12 5.1.1 Representativeness of ballast water sampling ......................................... 12 5.1.2 The isokinetic principle of ballast water sampling ..................................... 12

5.2 BALLAST WATER ANALYSIS ............................................................................... 13

6. REFINEMENT OF PROJECT OBJECTIVES .................................................... 14

7. THE PROJECT LOCATION .............................................................................. 15

8. PROJECT ACHIEVEMENTS AND RESULTS .................................................. 16

8.1 BALLAST WATER SAMPLING ............................................................................... 16 8.1.1 Isokinetic sampling................................................................................... 16 8.1.2 The SGS ballast water sampling system .................................................. 22

8.2 BALLAST WATER ANALYSIS ............................................................................... 45 8.2.1 ATP - Adenosine Triphosphate Fluorometry ............................................ 46 8.2.2 FISH - Fluorescence-in-situ-Hybridization ................................................ 50 8.2.3 PAM - Assessment of Chlorophyll a by Pulse Amplitude Modulation Fluorometry 55 8.2.4 FDA - Fluorescein-Diacetate Fluorometry ................................................ 59

8.3 BALLAST WATER TRAINING SEMINARS ................................................................ 68 8.3.1 International ballast water training seminar Flensburg ............................. 69 8.3.2 National ballast water training seminar Singapore ................................... 69 8.3.3 National ballast water training seminar Flensburg .................................... 70

8.4 ON-BOARD TRIALS ............................................................................................ 70

9. EXPERIENCES, LESSONS LEARNT AND DISCUSSION ............................... 71

9.1 THE DEVELOPMENT OF THE SAMPLING SYSTEM .................................................. 71

3/90

0014-00007 R0

9.2 THE DEVELOPMENT OF ANALYTICAL METHODS ................................................... 72 9.2.1 Execution of the FDA method .................................................................. 72

9.3 THE TRAINING SEMINARS .................................................................................. 73 9.3.1 The Flensburg seminar ............................................................................ 73 9.3.2 The Singapore seminar ............................................................................ 74 9.3.3 The second Flensburg seminar ................................................................ 74

9.4 THE ON-BOARD COMPLIANCE TESTS IN SINGAPORE HARBOR ............................... 75

10. BALLAST WATER SAMPLING AND ACTUAL IMO REGULATIONS .............. 77

11. CONCLUSIONS AND FUTURE PERSPECTIVES ............................................ 78

12. ANNEX .............................................................................................................. 79

12.1 LIST OF PROJECT COOPERATION PARTNERS ................................................... 79 12.2 LIST OF SEPARATE DOCUMENTS APPENDED TO THIS REPORT ........................... 80 12.3 BALLAST WATER SAMPLING AND ANALYSIS TRAINING SEMINAR: PROGRAM ....... 82 12.4 PAMAS DATA .............................................................................................. 83 12.5 PHOTOGRAPHIC DOCUMENTATION ................................................................. 84 12.6 LITERATURE LIST .......................................................................................... 87

12.6.1 IMO documents .................................................................................... 87 12.6.2 Plankton sampling and separation ........................................................ 87 12.6.3 PAM – Variable Chlorophyll a fluorescence .......................................... 88 12.6.4 FDA - Fluorescein-Diacetate Fluorometry ............................................ 88 12.6.5 ATP – Adenosin-Triphosphate Fluorometry .......................................... 89 12.6.6 FISH – Fluorescence-In-Situ-Hybridization Microscopy ........................ 89 12.6.7 Ballast Water General ........................................................................... 89

This document is issued by the Company under its General Conditions of Service accessible at http://www.sgs.com/en/Terms-and-Conditions.aspx. Attention is drawn to the limitation of liability, indemnification and jurisdiction issues defined therein.

Any holder of this document is advised that information contained hereon reflects the Company’s findings at the time of its intervention only and within the limits of Client’s instructions, if any. The Company’s sole responsibility is to its Client and this document does not exonerate parties to a transaction from exercising all their rights and obligations under the transaction documents. Any unauthorized alteration, forgery or falsification of the content or appearance of this document is unlawful and offenders may be prosecuted to the fullest extent of the law.

4/90

http://www.sgs.com/en/Terms-and-Conditions.aspx

0014-00007 R0

EXECUTIVE SUMMARY

Within the frame of this project

The SGS ballast water sampling system 02 has been developed, a modular designed sampling system, which allows representative ballast water sampling on-board all types of ships and which can be operated in a close circuit modus

as well as in an open circuit modus.

Several rapid, indicative scientific methods for the analysis of ballast water on-board ships have been developed, which allow for the fast, indicative assessment of compliance with the regulations of the Ballast Water

Management Convention.

Three ballast water sampling and analysis training seminars have been executed, one in Germany and one in Singapore.

On-board compliance test cycles have been executed on six vessels.

5/90

0014-00007 R0

LIST OF ABBREVIATIONS

Abbr. Definition

ATP Adenosin-tri-phosphate

cATP cellular ATP

dATP dissolved, extracellular ATP

tATP toral ATP

BBT co-operation partner to SGS : Blue BioTec, Büsum, Germany

BSH Federal Maritime and Hydrographic Agency, Hamburg, Germany

BW ballast water

BWSS ballast water sampling system

BWTS ballast water treatment system

EPA Environmental Protection Agency, USA

ETV Environmental Technology Verification Program, USA

FDA Fluorescein-di-acetate

FFR Fast Repetion Rate

FISH Fluorescence-in-situ-hybridization

GHS Globally Harmonized System of Classification, Labelling and Packaging of Chemicals

IfMG co-operation pertner to SGS : Institute for Marine Science-GEOMAR, Kiel, Germany

IHFS co-operation partner to SGS : Institute for Hydrobiology and Fisheries Science

INMTco-operation partner to SGS : Institute of Nautics and Maritime Technologies, University of Flensburg, Flensburg, Germany

MEA-NL Marine Eco Analytics, Netherlands

MLML Moss Landings Marine Laboratories, Moss Landings California, USA

MMB co-operation partner to SGS : Microbi Maris Biotec, Kiel, Germany

MSDS Material Data Safety Sheet

NIOZ Royal Netherlands Institute for Sea Research, Netherlands

NIVA Norwegian Institute for water research, Norway

PAM Pulse Amplitude Modulation

PSC Port State Controls

SGSSGS Institute Fresenius GmbH, Taunusstein, Germany and SGS S.A., Environmental Services, Geneva, Switzerland

SOP standard operating procedure

USCG United Sates Coast Guard

6/90

0014-00007 R0

1. PROJECT BACKGROUND

This final project report presents the facts and findings of the research and development project “Effective new Technologies for the Assessment of Compliance with the Ballast Water Management Convention”. The detailed description of the project results are headed by general considerations regarding ballast water, ballast water pipe systems, ballast water sampling and ballast water analysis. In early 2012 the SGS Institute Fresenius GmbH, Taunusstein, Germany, was charged by the BSH to execute in co-operation with the Moss Landings Marine Laboratories, Moss Landings, California, USA, the project “Effective new Technologies for the Assessment of Compliance with the Ballast Water Management Convention”. The overall project management was given to SGS and in line with the overall project objectives the different tasks to be completed within the frame of this project were distributed to both of the project partners. On the side of SGS the project was managed by Dr. Lothar Schillak, marine biologist, and on the side of the MLML, Prof. Dr. Nick Welschmeyer was responsible for the execution of the project tasks. The major objectives of the project are twofold:

• The development of an on-board ballast water sampling system, which allows for representative sampling from BW pipe systems installed on ships.

• The development of rapid, analytical on-board methods, which generate reliable data that indicatively define “Compliance” or “gross NON-Compliance” of ballast water treatment systems in line with the “International Convention for the Control and Management of Ship’s Ballast Water and Sediments (BWM)” set up by the IMO in 2004.

Following the complex structure of the project and seen to the manifold impacts the facts and findings of the project would possibly generate on the international sector of ballast water management, it was decided to select institutes, companies and experts as co-operation partners covering project aspects, which stand outside the SGS field of expertise. Para 11.1 of this report presents the list of co-operation partners that contributed to this project.

2. HISTORY OF THE PROJECT 2012 TO 2014

Immediately after the start of the BSH project in 2012 SGS established a network of companies and experts, which would support SGS in achieving the project objectives. For the development of a new on-board ballast water sampling system SGS chose the Institute for Nautics and Maritime Technologies-INMT, University of Flensburg, Germany. The INMT runs a test site for maritime technologies close to the harbor of Flensburg, Germany. For the development of the adequate on-board ballast water analysis SGS collaborated with two companies for the ATP method, Luminultra, Canada, and Aqua Tools, France. Both companies

7/90

0014-00007 R0

play a leading role on the international level in the development of ATP methods for the analysis of water samples. For the development of the FISH method, SGS cooperated with the German company “Mikrobiologie vermicon AG”. Vermicon, too, plays a leading role in the application of this method in the field of environmental testing and clinical applications on the international level. The FISH method was developed in the laboratories of Vermicon by Vermicon personnel. All technical developments were executed in the INMT until the development of the SGS ballast water sampling system prototype 01. This prototype 01 was further transformed into the SGS ballast water sampling system 02 by the Italian engineering company TAKIN. The ATP method was partly developed in the INMT with further developments in the laboratories of Aqua Tools, France, and in the laboratories of the German company BlueBioTech. The ballast water sampling system prototype 01 as well as the developed analytical methods were tested within the frame of two research and testing stays at the marine biological station Solbergstrand, Oslo Fjord, Norway, which belongs to the “National Norwegian Institute for Water Research, NIVA”. After the finalization of the SGS ballast water sampling system 02, the sampling system was tested at the INMT and within the frame of shipboard tests on six different ships in total in Singapore harbor. In addition SGS executed two ballast water training seminars at the INMT, Germany and in the laboratories of SGS Testing & Control Services Singapore Pte. Ltd. Between 2012 and 2014 the facts and findings of this project were continuously presented on major ballast water conferences in several countries. Additional documents annexed to this report are:

• Standard Operating Procedures of all analytical methods described in this report • Protocols of all analytical methods described in this report • Material Safety Data Sheet of all chemicals used by the analytical methods described in

this report • Training documents for seminars in ballast water sampling and analysis • Validation reports elaborated by external laboratories for the SGS ballast water sampling

system and the analytical methods

3. RELEVANT IMO REGULATIONS

3.1 IMO GUIDELINES

To prevent ecological damage and economic losses generated by invasive species introduced by discharged ballast water originating from different marine areas the performance standard D-2 is set out in the “International Convention for the Control and Management of Ship’s Ballast Water and Sediments”. To date the main approach to achieve the requirements is the installation of ballast water management systems on-board ships. Some of the manifold regulations of this convention and its related documents clearly address the sampling and analysis of ballast water on-board ships and define a discharge standard for ballast

8/90

0014-00007 R0

water regarding, among others, the density of viable organisms in the ballast water to be discharged. Essential information on sampling/analysis and ballast water discharge quality can be found in:

Guidelines for Ballast Water Sampling (G2), October 10th, 2008 (source: MEPC 58/23, Annex 3, Resolution MEPC.173/58)) Guidelines for Approval of Ballast Water Management Systems (G8), October 10th, 2008 (source: MEPC 58/23, Annex 4, Resolution MEPC.174(58)) Guidelines on Port State Control under the 2004 BWM Convention, April 10th, 2014 (source: MEPC 67/20)

3.1.1 Guidelines for ballast water sampling (G2)

This document concentrates on recommendations for on-board ballast water sampling and describes the necessary, obligatory frame, in which on-board ballast water sampling has to be executed. This frame strongly demands isokinetic, “L”-shaped sampling ports, defines the location for sampling, also addresses the analysis of ballast water and indicates the necessary processing, compilation and documentation of the generated data.

3.1.2 Guidelines for approval of ballast water management systems (G8)

These Guidelines are referred to in many other IMO documents in the context of ballast water sampling. G8 stipulates that the BWMS should be provided with sampling facilities so arranged in order to collect representative samples of the ship’s ballast water. Sampling facilities should in any case be located on the BWMS intake, before the discharging points, and any other points necessary for sampling to ascertain the proper functioning of the equipment as may be determined by the Administration.

3.1.3 Guidelines for port state control (MEPC 67/20)

This is the most recent IMO document which concentrates on testing of ballast water by Port State Control on-board ships. It describes a practicable way forward by setting a four-staged on-board testing procedure without limiting the rights of port States in verifying compliance with the BWMC. Ballast water sampling and analysis is integrated in step 3 and 4. It gives definitions of indicative and detailed analysis of ballast water, presents lists of materials necessary to adequately perform ballast water sampling and analysis and addresses safety issues of on-board testing.

9/90

0014-00007 R0

3.2 IMO REQUIREMENTS

According to the documents listed above the major obligations for ballast water sampling are:

• The on-board ballast water sampling has to be executed via an “L”-shaped, isokinetic sampling port.

• The sampling ports have to be located as near to the on-board discharge points for ballast water as possible.

• The ballast water of all on-board discharge points has to be sampled and analyzed.

• The ballast water samples should have a large volume as indicated in IMO Guideline G2 annex, para 6.2.5 and IMO Guideline G8 annex 4, para 2.3.31.

• The on-board ballast water sampling has to be representative of the whole discharge of ballast water as indicated in IMO Guideline G2 annex, para 6.2.2.

In case of filtration steps as integral parts of the sampling procedures for ballast waters adequate filter materials have to be used. The mesh size should not exceed 50µm in diameter or, if the mesh is of rectangular shape, diagonal for the IMO target organism size class of plankton organisms ≥50µm (MEPC 58/23, Annex 4, paragraph 2.2.2.6.3.1) and should not exceed 10µm in diameter or, if the mesh is of rectangular shape, diagonal for the IMO target organism size class of plankton organisms >10µm<50µm (MEPC 58/23, Annex 4, paragraph 2.2.2.6.3.2). The regulations within the frame of the Ballast Water Management Convention define three organism classes (D-2 discharge standard):

• Organisms with a size of ≥50µm • Organisms with a size between >10µm and <50µm • Indicator microbes (Escherichia coli, Enterococci and Vibrio cholerae)

For each of the three target organism classes the IMO regulations set up maximal admissible concentrations of viable organism in the ballast water to be discharged.

Table 1: IMO discharge standard for ballast water

IMO Class IMO Standard Plankton ≥50µm <10 viable organisms/m³ Plankton >10µm<50µm

<10 viable organisms/ml

Escherichia coli <250 cfu/100ml Enterococci <100 cfu/100ml Vibrio cholerae <1 cfu/100ml or <1cfu/gr wet weight

(cfu : colony forming unit)

10/90

0014-00007 R0

4. GENERAL ASPECTS OF BALLAST WATER

Natural seawater is taken into the ship via coarse filters (mesh size 2mm) by seawater pumps and a very complex system of pipes flushes the seawater into a number of tanks with various volume and shape, distributed all around the ship’s hull. Entering the tanks the natural seawater is turned into an artificial water body, in which most of the natural regulatory biological, physical, chemical and microbiological processes are severely inhibited or even simply cease to exist. Disregarding the alterations in the natural ecological processes in the ballast water tanks, many marine organisms prove to be able and establish populations of considerable size under these sub-optimal ecological conditions. When re-discharged into a marine environment geographically separated from the original environment these populations in the ballast water might invade these regions and by establishing new populations: they might shift the local ecological equilibrium by stepping into a spatial and food competition with the autochthonous species, sometimes causing enormous economic and ecological damage to entire areas and ecosystems. By adequate treatment technologies applied during uptake and discharge of the ballast water the detrimental impacts from invasive species on the local marine environment can be minimized, if not eliminated. The main function of ballast water on ships is to balance the ship during loading and un-loading processes as well as to stabilize the ship during sailing, particularly in rough weather conditions. It is the character of a ship that decides on the necessary dimensions of the seawater pump, the number of BW tanks and on the dimensions of the ballast water pipes, too. Since large cruise liners do not load and unload large volumes with heavy weight, these ships have small ballast water tanks, small pipes and seawater pumps with a low capacity. In contrast, e.g. liquid gas tankers quickly load and unload large volumes with heavy weight and as a consequence these ships have many ballast water tanks, large pipes and seawater pumps with a very high capacity. The total volume of ballast water tanks on-board ships ranges from 5.000m³ to 110.000m³, the dimension (inner diameter) of ballast water pipes ranges from 10cm to 80cm and the capacity of ballast water pumps spreads from 200m³/h to 3.000m³/h (Anwar 2011, vom Baur 2013). With these constellations the total time needed to take up ballast water or to de-ballast, largely depends on the type of ship and can range from a few hours to several days (Anwar 2011).

5. CONSIDERATIONS REGARDING PROJECT OBJECTIVES

The BSH project objectives described above have to be regarded in light of the IMO regulations, principle engineering conditions as well as the real conditions on-board ships. To approach a refinement of the project objectives a set of important aspects have to be reflected.

11/90

0014-00007 R0

5.1 BALLAST WATER SAMPLING

5.1.1 Representativeness of ballast water sampling

To execute on-board ballast water sampling, the ballast water pipe system installed on-board has to provide an access to the ballast water flowing through the pipes by bypassing the ballast water through small pipes, tubes or similar installations. The ballast water taken via these sampling ports has to be of the same quality in terms of total suspended solids and concentrations of marine organisms as the ballast water in the main ballast water pipe system: the ballast water sample has to be representative. This representativeness is twofold. The ballast water sample does not only have to be of the same quality as the ballast water in the main ballast water pipe system in terms of particulate matter (organisms, organic and inorganic matter) as expressed above. It should also reflect alterations in the quality of the ballast water across the entire period of de-ballasting: the ballast water sample has to be representative in quality and in time.

5.1.2 The isokinetic principle of ballast water sampling

The optimal design of a small pipe inserted into a main ballast water pipe system, which ensures the highest representativeness for ballast water samples in terms of quality and time, was intensively investigated by the aid of computational fluid dynamics (U.S. Coast Guard Research and Development Center 2008). Although marine plankton organisms are naturally buoyant the isokinetic sampling of ballast water represents the optimal way to generate representative samples. With isokinetic sampling the flow velocity in the sampling (= isokinetic) pipe is the same as in the main ballast water pipe. The adequate diameter of the isokinetic sampling pipe to successfully sample ballast water is determined by the equation below: E1 (with Diso=diameter of isokinetic pipe; Qiso=flow rate in the isokinetic pipe; DM=diameter of main pipe; QM=flow rate in the main pipe) In search for the optimal design of the isokinetic sampling pipe, the tests of numerous variants defined a pipe placed concentrically in the centre of the main ballast water pipe with its opening facing upstream as optimal, since it generates the best sampling flow. The isokinetic sampling pipe may either by inserted as an “L” shaped bow or as a straight pipe fixed by a flange through a bow of the main ballast water pipe (cf. figures 1, 2).

12/90

0014-00007 R0

Figure 1: “L”-shaped isokinetic sampling port, d1= inner diameter of the main ballast water pipe and d2=inner diameter of the isokinetic pipe.

Figure 2: Straight isokinetic sampling port

5.2 BALLAST WATER ANALYSIS

The detection of viable plankton organisms of the two IMO target organism size classes listed above can be executed by traditional methods like microscopic counts, if necessary preceded by vital staining to distinguish between viable and dead organisms. For all IMO target organism size classes it might possibly be necessary to condense the ballast water by adequate filtration steps, also in respect to the IMO regulations for ballast water sampling. Indicator microbes in ballast water samples (e.g. Escherichia coli, Enterococci, Vibrio cholerae) are detected by the traditional incubation method on species or group specific media (agar) over the period of 24/48 hrs. However, these traditional methods are time consuming and for some methods require the application of eco-toxic substances (i.e. stains). Especially in combination with the IMO regulation for ballast water sampling the on-board test of the ballast water for compliance with the International Ballast Water Management Convention might demand between

ballast water main pipe

isokinetic sampling pipe length


isokinetic sampling pipe ("L"- shape)

length

d1

d2

13/90

0014-00007 R0

one to three days days from sample to result, mostly because of the incubation times for the indicator microbes as mentioned above. These time constraints combined with the traditional analytical methods currently available set the demand for new, analytical methods, reducing the time from sample to result down to preferably a few minutes. The new analytical methods which on the one hand generate rapid results, might on the other hand possibly not be able to distinguish between 9 and 10 individuals for the respective target plankton size class. However, by statistical considerations and in order to keep to the short time from sample to result, these new rapid methods might indicate “gross exceedance” in respect to the discharge standard set up by the IMO. The indicative analysis detects the concentration of a chemical indicator substance in the ballast water sample, from which a numerical value for aquatic organisms in the sample can be derived based on empirical data. In contrast to the indicative analysis, the detailed analysis uses traditional analytical methods, which directly assess the number of live aquatic organisms in the ballast water sample.

6. REFINEMENT OF PROJECT OBJECTIVES

The ensemble of IMO regulations, engineering and scientific principles and the “real life conditions” on-board the ships subsequently demands the further differentiation of the project objectives. The technical development of an on-board ballast water sampling system was orientated along a set of criteria that were found essential for this project:

• it can be fitted to the piping conditions on all ships • all IMO target organism groups can be sampled • the IMO target organisms are not impacted by the sampling procedure itself • the sample volume is variable • the sampling system does not generate large waste water volumes on-board the

ship • the dimensions and the weight of the sampling system allow an easy transportation • no special skill is needed to operate the sampling system.

The development of rapid, indicative on-board ballast water analysis methods, too, were orientated along a set of further criteria that were found to be essential for this project

• reliable data are generated within a minimum of time • the execution of the analysis method does not require special skills • the analysis method does not use harmful or dangerous substances • the volume and number of necessary material and supplies are kept to a minimum • the space on-board the ship needed to perform the analysis is kept to a minimum.

14/90

0014-00007 R0

7. THE PROJECT LOCATION

The project´s location can be described as follows: The Institute for Nautics and Maritime Technologies belongs to the University of Flensburg, Germany, and runs a large testing site in the harbor of Flensburg, Baltic Sea. The premises of the INMT harbor testing site provide in-house systems of running seawater, which is taken from the nearby Flensburg Bight of the Baltic Sea. Before entering the in-house pipe system the seawater is filtered by a commercial seawater filter with a mesh size of 2.0 mm. In agreement with the administration of the University of Flensburg the project´s location was established in the INMT harbor test site on a contractual basis, covering the unlimited access to the workshops, to the large construction halls as well as to the laboratory. In addition the personnel of the INMT harbor testing site (naval engineers, technical assistants, head of laboratory) cooperated for the completion of the various project tasks. It was mutually decided to install, exclusively for the project, a new seawater pipe system, which simulates an on-board pipe system comprising DN250 pipes with a capacity of maximal 300.0 m³/h and two commercial ballast water filters as well as 5 hydrocyclones.

Figure 3: Technical installations at the INMT project´s basic location: ballast water filter

Necessary modifications of the seawater systems installed for the project, which occurred during the course of the project, were executed by the naval engineers of the INMT themselves or by external co-operation partners (cf. para 12.1).

15/90

0014-00007 R0

8. PROJECT ACHIEVEMENTS AND RESULTS

8.1 BALLAST WATER SAMPLING

8.1.1 Isokinetic sampling

At present isokinetic sampling ports are still not a standard installation of main ballast water pipe systems onboard ships and isokinetic sampling ports which are installed onboard ships do not have a uniform design. Therefore standard isokinetic sampling ports with a uniform design as integral part of all main ballast water pipe systems on all ships are highly desirable for sampling and analysis of ballast water onboard ships.

8.1.1.1 Sampling Ports

The seawater pipe system at the INMT was equipped with two different types of sampling ports: (i) the “L”-shaped type and (ii) the straight type.

Figure 4: “L”-shaped, isokinetic sampling port, installed in the seawater pipe system at the INMT project workbase

Both sampling installations allowed the testing of different types of isokinetic pipes in regard to diameter and length. For “L”-shaped isokinetic pipes length is defined as the distance from the entrance of the isokinetic pipe until the bow of the “L”-shaped pipe. For straight isokinetic pipes length is the distance from the entrance of the isokinetic pipe until the bow of the main ballast water pipe.

16/90

0014-00007 R0

Figure 5: Straight isokinetic pipe sampling port, installed in the seawater pipe system at the INMT project workbase

Figure 6: “L” - shaped isokinetic pipe sampling port: d1= inner diameter of the main ballast water pipe and d2=inner diameter of the isokinetic pipe, definition of length as indicated in the sketch

Figure 7: Straight isokinetic pipe sampling port: definition of length


isokinetic sampling pipe length


isokinetic sampling pipe ("L"- shape)

length

d1

d2

isokinetic pipe

sampling port

main pipe

17/90

0014-00007 R0

Table 2: Dimensions of isokinetic pipes tested at both types of sampling ports. For the definition of length please see figures 6 and 7.

Tested Isokinetic Pipes

Lengths: 0cm, 25cm, 50cm

Diameters: 1/2 inch, 1 inch, 2 inch

Figure 8: Different types of isokinetic pipes for sampling ports

As described earlier, any sampling port, through which ballast water is sampled from a main ballast water pipe onboard a ship should ensure that the sample taken represents the conditions of the ballast water in the main ballast water pipe as accurately and realistically as possible. Therefore any design of a sampling port should be tested as to the representativeness of the samples taken. A sample is regarded as ‘representative’ if it displays the same characteristics, as the water in the main ballast water pipe. In this respect the representativeness of a sample can only be assessed by the comparison of analytical results from the water in the main ballast water pipe with the analytical results from the samples. The analytical data from the samples taken can be related to the analytical data from the main ballast water pipe and the representativeness of the sample is expressed as percentage then. The testing of the representativeness should not be restricted to the organisms contained but should be extended to the analysis of all particles in the water as requested in IMO Guidelines G2 (annex 3, para 6.2.2) and G8 (annex 4, para 7.1). To assess the representativeness of samples collected by the various combinations of different isokinetic sampling pipes installed in the two types of sampling ports, a particle count device was used: PAMAS S 4031. The PAMAS S 4031 has an inbuilt optical sensor and volumetric cell design and allows the counting of particles, including plankton organisms in a natural seawater or ballast water sample,

18/90

0014-00007 R0

where the upper and lower detection size can be chosen at different intervals within a rather wide range between 1µm and 400µm1. The cell design allows for an exact, highly accurate volumetric analysis and the results can be reported according to common standards NAS 1638 (U.S. National Aerospace Standard –NAS, a particulate contamination coding system), SAE AS 4059 (successor to NAS 1638) and ISO 4406 (Hydraulic fluid power - Fluids - Method for coding level of contamination by solid particles; International Organization for Standardization, Geneva, Switzerland, 1999). However, the PAMAS S 4031 only provides an indication of size of all particles, it does not determine minimum dimensions and does not distinguish between organisms and particles and also does not determine viability. The two major designs of sampling ports, “L”-shaped and “straight” have been equipped with the different isokinetic pipes as displayed in table 2. The tests series to assess the representativeness of isokinetic sampling were executed with natural seawater, since the particle load, sediments and organisms, of natural seawater is highly divers compared to treated ballast water. Since the majority of ballast water treatment systems already provide a filtration step as integral part of their treatment technology with mesh sizes between 20µm and 50µm (Anwar 2011) the detection range of the PAMAS S 4031 particle count was set to 10µm -100µm. The seawater from the main pipes used for these test series was prefiltered by a 2mm sieve. In addition, since natural seawater contains a higher particle load than treated ballast water a concentration of the tested seawater was not necessary and the sample volume was set to 10ml. To determine the particle load of the seawater samples were taken from the main ballast water pipe, approximately 2 meters in front of the installed isokinetic sampling ports. Sampling was performed by stopping the pump and draining the water from the ballast water pipe, this way it was ensured that these samples were represented 100% of the particles in the water. The samples taken from the isokinetic ports were analyzed under the same conditions as for the seawater samples taken directly from the main pipe. For the analysis of the 100% value as well as for each of the described isokinetic sampling port combinations a set of n=12 samples were analyzed. Since the natural plankton populations are subject to seasonal changes, the test series were executed in late spring/early summer and in autumn. As described above in total 18 different combinations of isokinetic sampling ports were assessed, 9 different combination for each of the two different shapes of isokinetic sampling pipe: “L”-shaped pipe and “straight” isokinetic pipe. All combinations were tested with the same method described here. An example of a data set for the tests of representativeness is displayed in the annex (cf. para 12.4).

8.1.1.2 Results

“L” shaped isokinetic sampling port

1 http://www.pamas.de/en/PARTICLE-COUNTERS/PORTABLE/PAMAS-S4031; http://www.pamas.de/de; accessed/downloaded on 23.11.2013

19/90

http://www.pamas.de/en/PARTICLE-COUNTERS/PORTABLE/PAMAS-S4031

http://www.pamas.de/de

0014-00007 R0

The results of the tests with different lengths and different diameters are displayed in figure 9.

Figure 9: Results of tests for representativeness of different isokinetic sampling ports installation with “L”-shaped isokinetic pipes. For definition of length see figure 6. Raw data is shown in Annex 12.4.

Each of the bars (with indication of standard deviation) displayed in figure 9 represent a set of mean data from 12 seawater samples upstream the isokinetic sampling port installation, regarded as 100%, in relation to the mean data of 12 samples taken through an isokinetic sampling port installation as indicated below the x-axis. The values scatter significantly from a representativeness of 85.8% (min) to 123.6% (max) (∆ 37.8%) with a standard deviation from ±4.15% (min) to ±15.2% (max). Straight isokinetic pipe sampling port The values obtained from test series with the “straight” isokinetic pipe combinations vary from a representativeness of 99.2% (min) to 113.5% (max) (∆ 14.3%) with a standard deviation from ±2.4% (min) to ±13.5% (max), cf. figure 10.

0,0

20,0

40,0

60,0

80,0

100,0

120,0

140,0

% of main flow

Representativeness of Sampling"L" shaped isokinetic sampling pipe

0cm 25cm 50cm0,5 inch

0cm 25cm 50cm1 inch

0cm 25cm 50cm2 inch

LengthDiameter

0,0

20,0

40,0

60,0

80,0

100,0

120,0

140,0% of main flow

Representativeness of Samplingstraight isokinetic sampling pipe

0cm 25cm 50cm0,5 inch

0cm 25cm 50cm1 inch

0cm 25cm 50cm2 inch

LengthDiameter

20/90

0014-00007 R0

Figure 10: Results of tests for representativeness of different isokinetic sampling ports installation with “straight”-shaped isokinetic pipes. For definition of length see figure 7. Raw data is shown in Annex 12.4.

8.1.1.3 Discussion of results and best combinations

The results above clearly show that the “straight” design version of sampling ports yield a much better representativeness for all tested isokinetic pipe variations (cf. figure 9, figure 10) than the “L”-shaped version (representativeness of 99.2 % – 113.5 % for straight pipes, 85.5 % – 123.6 % for “L”-shape).However, since IMO Guidelines ask for “L”-shaped isokinetic pipes (IMO Guidelines G2, annex 3, part 1, para 4.3, document page 7), this model was selected for further development. During both tests values of more than 100% occur for both sampling port versions. These exceeding values are due to partly hypo-kinetic situations directly at the entrance of the isokinetic sampling pipe, generated by the differential pressure regime of an open port and due to normal fluctuations in the capacity of the seawater pumps, which maintain the volume flow through the main ballast water pipe. As on the one hand the IMO regulations for the Ballast Water Management Convention clearly demand an isokinetic sampling method with the installation of an “L”- shaped isokinetic sampling pipe and on the other hand the installation of such an “L” - shaped isokinetic sampling should be as easy and simple as possible for retrofit and new built ships, isokinetic pipes with the same length, i.e. 0.0 cm (0.0 cm of pipe extends after the “L” shaped bend, see figure 6), and different diameter have been compared.

Table 3: The representativeness of samples taken through isokinetic sampling ports equipped with “L”shaped isokinetic pipes for different particle size fractions. Example: a sample taken through an isokinetic “L” shaped pipe with a diameter of 2.8cm and a length of 0.0cm is to 96.0 % representative for the fraction <2µm, to 97.8 % representative for the fraction of >10µm<50µm and to 100.8 % representative for the fraction of >50µm<100µm.

Isokinetic Pipes, Type “L”-shaped

Diameter: 1.4cm

Diameter: 2.8cm

Diameter: 5.6cm

Length: 0.0cm

Particle size fraction

Representativeness (%)

<2µm 112.0 96.0 103.9

>10µm<50µm 98.2 97.8 97.8

>50µm<100µm 101.2 100.8 100.9

Table 3 summarizes those combinations, which are in line with the Ballast Water Management Convention, associated Guidelines and related documents that at the same time ensure a very high representativeness of the samples taken.

21/90

0014-00007 R0

8.1.2 The SGS ballast water sampling system

In principle the on-board sampling of ballast water via an isokinetic sampling port followed by filtration can be executed with two alternative methods:

• Open circuit sampling: after the filtration step the filtered ballast water is directed into a recipient container or pumped into the ship’s bilge or over board (A)

• Closed circuit sampling: after the filtration step the filtered ballast water is re-directed into the on-board ballast water pipe system (B)

Figure 11: Alternatives for on-board sampling; A: open circuit sampling, B: closed circuit sampling

The open circuit sampling demands additional technical installations representing a disadvantage, since the filtered ballast water has to be stored in recipient containers or has to be pumped from the technical decks of a ship into the bilge or overboard by the aid of additionally installed, adequate pumps. The use of an open circuit sampling system will follow general hydraulic principles in the moment when the sampling port is opened. The situation can be compared to a water tap: in the moment the tap is opened the pressure in the water pipe system shoots the water out through the tap. The same situation is created in the moment when the isokinetic sampling port is opened due to the considerably high system pressure in the main ballast water pipes (values between 2.6 and 5.3 bar have been measured by SGS during practical on-board tests). The closed circuit sampling, in contrary, needs a backflush port to be installed downstream the sampling port, through which the filtered ballast water is re-directed to the main ballast water pipe. The closed circuit sampling does not create a differential pressure regime and meets the respective criteria listed above, cf. para 6 of this report.

8.1.2.1 The SGS Ballast water sampling system prototype 01

Methodology Based on the obtained results from the preceeding tests with different designs of sampling ports and the determination of major criteria for the design of an on-board sampling system, a mobile

sampling system


A

sampling system


B

22/90

0014-00007 R0

system for the sampling of ballast water on-board ships was designed and built (cf. figures 12, 13). From the isokinetic sampling port the ballast water enters a filter housing and runs through a 36µm filter mesh (diagonale 50µm) to retain the IMO organism size class >50µm, as described in D-2 of the Annex of the Ballast Water Management Convention. From the filter housing the ballast water passes along a straight section with an inductive flowmeter, which records the volume filtered. Through an isokinetic sampling bypass (straight isokinetic sampling pipe) additional samples are taken for the IMO organism size classes >10µm<50µm and for the bacteria. It has to be explicitely noted, that this bypass also allows for the generation of continuous drip samples. The outflow is connected to a backflush port installed downstream the isokinetic sampling port. Two pressure gauges installed before and behind the filter housing allow for the monitoring of the pressure regime within the filter housing to possibly indicate the clogging of the filter mesh. It has to be regarded as very unlikely that the filter clogs during ballast water sampling since it is supposed that treated ballast water does not contain a high sediment load. The installation of the differential pressure gauges has to be regarded as a control during ballast water sampling. In case, for some unforeseeable events the differential pressure is increasing the sampling procedure has to be stopped. The filter housing has to be opened and the filter has to be visually controlled. In case the filter is clogged, the filter has to be replaced and the sampling procedure has to be re-started.

Figure 12: On-board ballast water sampling system (scheme)

in

bypass sampling

isokineticbypass

filterhousing

out

pressure gauge

pressure gauge

flow meter Sampling of organisms

Sampling of organisms >10µm<50µm and bacteria

23/90

0014-00007 R0

In this constellation the ballast water is filtered in the housing and the organisms >50µm are retained. The filtrate is then directed to the bypass sampling point to collect samples for the analysis of the organisms >10µm<50µm and the bacteria (<2µm). The effectiveness of filter material for the separation of plankton size fractions has been discussed since long (Evans & Sell 1985, Taguchi & Laws 1988, Logan 1993, Chavez et al 1995, Seda & Dostalkova 1996, Harris et al eds. 2000, Postel et al. 2000, USEPA 2003, USEPA 2005, Hwang et al. 2007, Wu et al. 2011). The fact that not all plankton organisms are spherical renders a total, complete, 100% separation impossible. Spiny species long >50µm, but <20µm thick may pass through a 50µm mesh filter material and these species occur in the size fraction of >10mm<50µm then. Only a few authors dare to specify a general scale, e.g. Seda & Dostalkova 1996 state: that a mesh size of ¾ of the body length will ensure a 90% retention of the targeted organism. In addition to this impossibility to precisely and totally separate plankton organisms size fractions it much depends on the ecological status of the coastal area in which the ballast water is pumped into the tank. In general tropical areas show a reduced abundance of plankton populations, whereas the boreal marine areas tend to have high plankton abundances due to the naturally available nutrients. Also, coastal marine areas under severe human impact tend to eutrophicate and hence show different patterns of the plankton populations.(cf. publications on general marine biology/ecology: Sommer 2005). In this respect the concept of the sampling system presented here just follows what has been described in the various publications cited above. The Prototype 01 has a small footprint (40x60x90cm) and with a weight of less than 20 kg the system is easy to transport and the on-board handling is simple. On-board the Prototype 01 is connected to the isokinetic sampling port and to the back flush port by flexible, pressure resistant hoses which are part of the complete sampling system. The major pipe elements of Prototype 01 have been fixed to diameter of 2 inch, a dimension, which is regarded to be the upper limit for a transportable system. The functional scheme displayed in figure 12 has been developed to the “SGS Ballast Water Sampling System Prototype 01” shown in figure 13.

Figure 13: The SGS Ballast Water Sampling system Prototype 01

24/90

0014-00007 R0

Dominant parameters and basic values for prototype 01 The adequate filtration, i.e. the separation and concentration of the plankton organisms from the ballast water is regarded as the pre-dominant criterion for a ballast water sampling system, since it has by all means to be avoided that the live plankton organisms in the ballast water are impacted by the filtration process itself. Subsequently this criterion mainly addresses the flow velocity within the filter housing of Prototype 01, which depends on (i) the diameter of the isokinetic pipe installed in the sampling port and (ii) the flow velocity in the main ballast water pipe system of the ship. The IMO regulations for the ballast water convention state an approximate filtering velocity of 0.5 m/s, when samples are taken from within the ballast water tank by plankton net samplers (IMO Guidelines G2, annex 3, part 2, para 2.5.2, page 8). This value was confirmed by various national and international institutions for marine research (University of Hamburg, Germany; Helmholtz Center for Ocean Research Kiel, Germany; Glosten Associates inc., USA). Experimental tests with plankton nets show that a differential pressure of >0.2 bar disintegrates the plankton organisms. In 2009 Boll Filtersystem, Germany, producer of ballast water filtersystems, executed a study, in which, by computational fluid dynamics, the critical flow velocity should be determined, with which plankton organisms are disintegrated directly at the filter material. The larvae of bivalves were taken as test organisms. Figures 14 and 15 display the flow velocity directly at the disc filter (stainless steel screen).

Figure 14: Computational fluid dynamics: flow velocity at a stainless steel disc filter (100µm). Grey: filtermaterial. Different colors: flow velocities according to the scale, left side top. (picture courtesy of Boll Filtersystems)

25/90

0014-00007 R0

Figure 15: Computational fluid dynamics: flow velocity at a stainless steel disc filter (100µm). Grey: filtermaterial. Different colors: flow velocities according to the scale, left side top. (picture courtesy of Boll Filtersystems). The picture shows the pathways of particles (bivalve larvae) when approaching and passing the filter material. The red colored areas indicate flow velocities of >0.45 m/s.

Although the aim of the experiment shown above was to increase the flow velocity in the filter systems in order to disintegrate the plankton organisms and thus prevent the filters from being clogged and reduce the frequency of backflush events, the result of the study can also be interpreted in a different way: to safely sample live plankton organisms by filtration the flow velocity should not exceed 0.5 m/s directly at the filter material.

Hydraulic tests with Prototype 01

For the hydraulic tests Prototype 01 was connected to a straight isokinetic pipe sampling port of the main ballast water treatment system at the INMT. For these tests with Prototype 01 the ballast water filter of the INMT ballast water treatment system (MAHLE ballast water filter, sieve type with concentrical backflush rotator) was switched on.

26/90

0014-00007 R0

Figure 16: Straight isokinetic pipe sampling port in the ballast water treatment system of the INMT

The tests were executed in a quite broad range of values for the basic parameters of the main ballast water pipe system such as volume flow, system pressure and pipe diameter (cf. table 4 and 5, page 24).

Table 4: Basic parameters of the main ballast water pipe system (volume flow, system pressure) preset for hydraulic tests with the sampling system prototype 01. At 80 m3/h the flange in the main ballast water pipe was partly closed to increase pressure in the system.

Volume Flow (m³/h)

Pressure (bar)

30 0.2

80 < 0.2

80 0.8 – 0.9

80 1.6 – 1.8, max 3.2

160 1.0 , max 2.6

198 1.1 – 1.2 After pressure tests under various conditions further hydraulic tests with Prototype 01 were executed (i) as an open circuit sampling system and (ii) as a closed circuit sampling system. The volume flow and the flow velocities were recorded from volume meters and flow meters installed in the different pipe sections of the prototype.

27/90

0014-00007 R0

Table 5: Basic parameters for hydraulic tests with Prototype 01, diameter of pipes

Pipe Inner Diameter

(mm) BW system at INMT 100.0 Isokinetic pipe in sampling port 25.4 Tube connecting Prototype 01 with isokinetic sampling port 52.6

Inlet at filterhousing 50.8 Filterhousing 123.7

Filtration tests with Prototype 01 The filtration tests with Prototype 01 were executed under the hydraulic conditions described in table 4 using two different types of filter displayed in figure 17 and 18. The filter shown in figure 17 represents a technical filter used in the beverage industries. The filter presented in figure 18 was designed and constructed especially for the filter housing within the frame of this project. During filtration the plankton organisms are collected in the small beaker at the lower end of the filter. The beaker can easily be screwed off for the recovery of the trapped plankton organisms.

Figure 17: Technical Nylon filterbag, mesh size 36µm

28/90

0014-00007 R0

Figure 18: Plankton Nylon filter, mesh size 36µm, recipient beaker with ∅ 30mm, height 70mm

The filtration tests with the two types of filter were executed with different sample volumes at different flow velocities in the filter housing. After filtration the filters were taken out from the filter housing and examined under the dissecting microscope regarding criteria such as (i) detection of live organisms, (ii) occurrence of smashed plankton organisms, (iii) clogging of the filter mesh and (iv) easy rinsing and cleaning of the filter mesh for re-use.

After the first initial test had been executed with Nylon as filter material the continuation of these test were stopped because of three major disadvantages of this filtermaterial:

• In general Nylon has to be sewed to the desired form and to fit the filter into any housing. This method of sewing creates numerous small gaps and niches into which the plankton organisms escape any retrieval.

• Nylon is a textile (cf. figure 33) and because of this characteristics it flexible. It has to be assumed, that the shape of the under free from tension conditions rectangular meshes change their shape, when coming under tension

• The sampling of ballast water demands the disinfection of all parts of the sampling device prior the execution of another on-board compliance test. Nylon is not an inert substance.

After the filtration tests with Nylon filters were stopped another filtermaterial, a laser perforated stainless steel screen was used, described in the next section.

29/90

0014-00007 R0

Results The pressure tests with the Prototype 01 using natural water from the Flensburg Bight revealed that the sampling system is compression-proof to system pressures of >3.2 bar. The results for the flow velocity measurements from both test cycles, (i) open circuit sampling system and (ii) closed circuit sampling system were cross checked with the results of the relevant mathematical calculations based on the parameters ‘volume flow’ and ‘pipe diameter’ in the different sections with the following equation: V (m/s) = (Q (m³/h) / A (m²))/3600

(with V= flow velocity, Q=flow volume and A=cross-section area of the pipe)

The test results with the open circuit sampling system revealed that the flow velocities in the different pipe sections measured during the tests largely differ from the expected values derived from the mathematical calculations as shown below.

Figure 19: Open circuit sampling system, calculated and measured flow velocities in the filter-housing

To further detect the reasons for the discrepancy between the calculated flow velocities in the filter housing and the flow velocities measured, additional tests with different system pressures were executed. With the constellation as displayed in figure 20, i.e. a volume flow in the BW main pipe system of 80.0m³/h and an isokinetic pipe diameter of 25.4mm the flow velocity in the filter housing expected from mathematical calculations would be 0.24m/s.

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,00 1,00 2,00 3,00 4,00 5,00 6,00 7,00 8,00

measured calculatedvfilterhousing

(m/s)

Ballast Water Sampling System Prototype 01Flow Velocity (v) with open Circuit for disokinetic=2,54cm

vmain pipe (m/s)

30/90

0014-00007 R0

Figure 20: Open circuit sampling system, calculated and measured flow velocities in the filter-housing under different system pressures

In contrast to the results from mathematical calculations the flow velocity in the filter housing was measured as 0.44 m/s with a system pressure of 0.8 bar and as 0.68 m/s with a system pressure of 1.8 bar, indicating that the velocity in the isokinetic pipe is triggered by the system pressure. The test results with the closed circuit sampling system revealed that the flow velocities in the different pipe sections measured during the tests do not significantly differ from the expected values derived from the mathematical calculations (cf. figure 21).

Figure 21: Closed sampling system, expected and measured flow velocities in the filter-housing

0

0,2

0,4

0,6

0,8

1

Diagrammtitel

vfilterhousing (m/s)

Ballast Water Sampling System Prototype 01Flow velocity (v) with open circuit for disokinetic= 2,54cm and Qmain pipe= 80,0 m³/h

calculatedP = 0,8 bar P = 1,8 bar

measured at system pressure

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,70

0,00 1,00 2,00 3,00 4,00 5,00 6,00 7,00 8,00

measured calculatedvfilterhousing

(m/s)

Ballast Water Sampling System Prototype 01Flow Velocity (v) with Backflush for disokinetic=2,54cm

vmain pipe (m/s)

31/90

0014-00007 R0

Figure 22: Technical Nylon filter after sampling 6.0 m³ of ballast water at 1.3 m/s at filter material

The filtration tests executed with two different types of filters as displayed in figures 17 and 18 revealed similar results. With flow velocities in the filter housing exceeding 0.5 m/s the live plankton organisms in the ballast water are killed during filtration, see table 6. Figure 22 displays the technical filter after sampling a ballast water volume of 6m³ at a flow velocity in the filter housing of 1.3 m/s. Figure 23 displays the technical filter after sampling a ballast water volume of 1.0 m³ at a flow velocity in the filter housing of 0.6 m/s.

Figure 23: Technical Nylon filter after sampling of 1.0 m³ at 0.6 m/s at filter material

32/90

0014-00007 R0

The results from the filtrations tests are summarized in table 6.

Table 6: Summary of results from filtrations tests with prototype 01

Criteria Flow Velocity in Filter Housing (m/s)

0.16 0.44 0.55 0.67 Detection of live Organisms YES YES YES NO Occurrence of smashed Organisms

NO NO YES YES

Clogging of Filter Mesh NO NO YES YES Rinsing of Filter Mesh possible YES YES YES NO

At flow velocities in the filter housing of 0.16 m/s and 0.44 m/s live organisms are collected from within the filter mesh, live organisms are not killed by the filtration process, the filter mesh does not clog and can easily be rinsed. With increasing flow velocity in the filter housing (0.55m/s) the plankton organisms are impacted by the increased sheer forces occurring at the surface of the filter mesh, the filters mesh clogs (0.67 m/s) and for high sample volumes (>1m³) taken at higher flow velocities (>0.67 m/s) the debris of smashed plankton organisms cannot be removed anymore from the filter mesh by simple rinsing. The results of the filtration tests as shown in table 6 confirm that the flow velocity for sampling, i.e. filtration of ballast water should not exceed 0.5m/s.

8.1.2.2 The SGS Ballast water sampling system 02

After the successful test of the SGS ballast water sampling system prototype 01, this first version of a ballast water sampling system was further optimized in terms of dimensions, geometry and footprint. The footprint of the sampling system 02 is approximately 45cmx25 and the weight of the sampling system excluding the other modular components has been reduced 6.5 kg only. The SGS ballast water sampling system 02 comprises the same functional elements, which, however, have been re-arranged.

33/90

0014-00007 R0

Figure 24: The SGS ballast water sampling system 02

Figure 24 shows the SGS ballast water sampling system 02 with the large filterhousing, two pressure gauges, the connected isokinetic pipe and the volume count at the backflush hose. The ballast water enters the sampling system via the isokinetic pipe, is directed into the large filterhousing, in which the plankton organisms ≥50µm are retained. It leaves the filterhousing through a straight pipe section, in which a small isokinetic bypass sampling port allows for the sampling of ballast water for analysis of plankton organisms >10µm<50µm and bacteria. The entire ballast water sampling system can easily be dismantled for disinfection of the single functional subunits. The different parts, like hoses, volume count, isokinetic pipe are connected by a simple system of clamps.

Figure 25: The isokinetic bypass sampling port, dismantled (left) and installed (right)

34/90

0014-00007 R0

Figure 26: The connecting system and additional ball valves

The flange, which fixes the isokinetic pipe into the ship’s main ballast water pipe is installed to a corresponding flange installed in the ballast water pipe. To ensure that the SGS ballast water sampling system 02 can be applied to a broad variety of ballast water piping system on-board many, if not all types of ships the flange, which fixes the isokinetic pipe, a special cartouche has been developed which allows the use of isokinetic pipes of different diameters (2.54cm, 1.27cm and 0.64cm). Based on hydraulic principles the diameter of the isokinetic pipe will decide on the flow volume through the SGS ballast water sampling system 02 and subsequently the diameter of the isokinetic pipe will finally decide on the flow velocity directly at the filter material in the large filter housing, which filters the plankton organisms >50µm. As described above this flow velocity must not exceed 0.5 m/s. Figure 27 presents the functional parts of the cartouche.

35/90

0014-00007 R0

Figure 27: The cartouche to fix the isokinetic pipes.

The isokinetic pipe with the seal is inserted into the cartouche and fixed with the screw, which compresses the seal to water tightness.

Figure 28: The cartouche mounted to the flange, arrows indicate the possibilities to adjust the position of the isokinetic pipe

Once the flange with the cartouche and the isokinetic pipe are fixed to the main ballast water pipe, the position of the isokinetic pipe inside the main ballast water pipe can easily be adjusted. Figure 28 presents the mounted cartouche with yellow arrows indicating the possibilities to adjust the isokinetic pipe inside the main ballast water pipe, i.e. the opening of the isokinetic pipe inside the main ballast water pipe is positioned concentrically and faces upstream. The isokinetic pipe is equipped with a scale for easy orientation of the isokinetic pipe inside the main ballast water pipe.

cartouche screw

Isokinetic pipe seal

flange

36/90

0014-00007 R0

Figure 29 displays the schematic view of an SGS ballast water sampling system 02 connected to a main ballast water pipe, with isokinetic flange, sampling unit, volume count and backflush flange

Figure 29: Schematic view of SGS ballast water sampling system 02 installed to a ballast water main pipe, arrows indicating the direction of the water flow

Figure 30: Isokinetic pipes with different diameter, which can be used with the cartouche

The cartouche has been dimensioned to allow for the fixation of isokinetic pipes with different diameter. Figure 30 presents the different isokinetic pipes with diameters of 0.25, 0.5 and 1.0 inch, which can be used with the cartouche. It is essential and in fact highly advisable to assess the range of ballast water pipes and volume flows for which these isokinetic pipes can be used.

37/90

0014-00007 R0

Figure 31: Flow velocities in ballast water main pipe systems (x-axis) and resulting, mathematically determined flow velocities in the SGS sampling system 02 (y-axis) for isokinetic pipes with different diameter (see legend).

Figure 31 shows the theoretical calculation of the effect that isokinetic pipes with different diameter have on the flow velocity in the filter housing of the SGS ballast water sampling system 02. The grey shaded bar indicates the range of a critical flow velocity of 0.5 m/s, which should not be exceeded. The graph verifies that any diameter of the isokinetic pipe ≤1.0 inch is adequate to be used in main ballast water pipe systems up to a flow velocity of 7.0 m/s. Since the SGS ballast water sampling system 02 makes use of isokinetic pipes with diameters of 0.25, 0.5 and 1.0 inch, this ballast water sampling system can be applied for a very broad range of ballast water pipe systems on-board all types of ships. Experiences with the SGS ballast water sampling system 02 at the INMT in Flensburg, Germany, revealed that the simple installation of a flange equipped with a cartouche is likely to be a problem on-board ships, since the installation of the simple flange carrying cartouche and isokinetic pipe demands the emptying of the main ballast water pipe system on-board the ships. This procedure of emptying and refilling of the on-board ballast water main pipe system for sampling demands additional time for on-board compliance testing and in fact represents a burden for the technical personnel on-board the ships. For these reasons the isokinetic flange as well as the backflush flange had to be further developed to allow for the installation of the SGS ballast water sampling system without the necessity to empty the main ballast water pipes on-board ships.

38/90

0014-00007 R0

Figure 32: Installation of the isokinetic pipe into main ballast water pipes on-board ships without emptying the main ballast water pipe

Figure 32 displays a schematic view of the further developed sampling port. A 3 inch ball valve is installed to the main ballast water pipe. This 3 inch ball valve is closed when the flange carrying the cartouche and the isokinetic pipe is installed to the 3 inch valve, cf. figure 32, left. As the cartouche is water tightly sealed the 3 inch valve can be opened as soon as the flange carrying the cartouche and the isokinetic pipe is fixed and as the isokinetic pipe can be moved inside the tightly fixed cartouche the isokinetic pipe is gently pushed into the correct position inside the main ballast water pipe, cf. figure 32, right. The new connection for sampling port and backflush port were tested with the ballast water system at the INMT for their functionality. The additional tests revealed that this system for the installation of the SGS ballast water sampling system 02 to ballast water main pipes without the necessity to empty the main ballast water pipe is fully functional.


Sampling Port

to sampling system

3 inch ball valve

isokinetic pipe

flange with cartouche


Sampling Port

to sampling system

3 inch ball valve

isokinetic pipe

flange with cartouche

39/90

0014-00007 R0

Figure 33: Test arrangement of the SGS ballast water sampling system 02 (left) with the new sampling port and backflush ports (right) installed at the ballast water system of the INMT

The test of filter material with the SGS ballast water sampling system prototype 01 (cf. para 8.1.2) revealed, that nylon mesh might possibly not represent the adequate filter material. A stainless steel material, used both in food industry and medical applications, was selected for further filtration test (see table 9 and accompanying text). The stainless steel screen filters have a thickness of 0.45-0.55 mm and are laser perforated to a mesh size of exactly 50µm. In the experience of SGS, the nylon filters are relatively fragile, since the holes are square the size selectivity depends on the orientation of the organism as is passes the filter. Additionally, when there is pressure on the filter material, the material will stretch. This increases the diameter of the mesh allowing larger organisms to pass through the filter. The material is also difficult to clean. The steel filter on the other hand is extremely sturdy; the holes are round and the material not flexible. In addition it is easy to clean, the use of disinfectants does not affect the filter material.

40/90

0014-00007 R0

Figure 34: Scanning Electron Microscopy pictures of different filter material for ballast water sampling : stainless steel screen (above), nylon mesh (below)

Statistical measurements of the two different filter materials, a nylon mesh and a stainless steel screen, have been compared by the aid of direct length measurements under the scanning electron microscope (cf. figure 34). Table 7 displays the results.

Table 7: Comparison of filter material for mean mesh size and filter ratio

Filter material n Mean mesh size

(diagonal/diameter, µm) Filterratio

(%)

Nylon mesh size 36µm 38 48.68 ± 2.24 3.2

Stainless steel screen laser perforated 50µm

90 51.9 ± 2.39 5.1

These results show that the stainless steel filter screen has a better filter ratio, also, seen to the fact that the filter material has to be disinfected prior to the execution of the next on-board test and the resistance to physical damage the steel filter is to be preferred. As a consequence the SGS ballast water sampling system 02 was equipped with a new filter inlet, which is displayed in figure 35.

41/90

0014-00007 R0

Figure 35: The stainless steel screen filter inlet of the SGS ballast water sampling system 02

The stainless steel screen filter inlet is shaped as a tube-like filter with a rubber seal at its upper opening to prevent ballast water from being bypassed and a small ball valve at the bottom to sample the retained plankton organisms from the inside of the filter tube. Since the SGS ballast water sampling system 01 and 02 have been equipped with a new filter material, the execution of additional tests was necessary to verify the maximum allowable flow velocity directly at the filter material, which has been determined to be 0.5 m/s for the SGS ballast water sampling system prototype 01. To execute these tests the SGS ballast water sampling system 02 was connected to the ballast water system at the INMT and filtration tests were run with different flow velocities in the main ballast water system subsequently triggering different flow velocities at the new filter inlet of the SGS ballast water sampling system 02. The samples were assessed as to the detection of damaged marine plankton organisms.

Table 8: Test cycles for the verification of maximum admissible flow velocity at the stainless steel filter inlet of SGS ballast water sampling system 02

Test Cycle Flow volume main pipe (m³/h)

Flow velocity at filter material (m/s)

1 50 0.32

2 120 0.49

3 190 0.77

As natural seawater was used during these test cycles, the assessment of damages to plankton organisms in relation to the flow velocity at the filter material used a selection of different organism types which were present in the plankton populations of the natural seawater during times of investigation (early summer). By selecting different taxa from the natural plankton, with

42/90

0014-00007 R0

particular focus on selecting fragile organisms which were likely to show effects of filtration, impact of flow velocity on wide variety of plankton organisms could be assessed (Table 9).

Table 9: Marine target taxa for the assessment of individual damages induced by the flow velocity at the stainless steel filter material.

Acartia sp. Zooplankton, copepod

Ceratium tripos Phytoplankton, dinoflagellate

Chaetoceros sp. Phytoplankton, diatom

Leptodora kindtii Zooplankton, cladoceran

Marenzelleria viridis Zooplankton, polychaete

Protoperidinium sp. Phytoplankton, dinoflagellate

Pygospio elegans Zooplankton, polychaete

Stokesia vernatis Zooplankton, ciliate

Tigriopus sp. Zooplankton, copepod

As a unified result for all flow velocities the majority of the observed organisms appeared intact and exhibited their species specific movement. Copepods, copepodites, copepod nauplia, ciliates, bivalve larvae, balanid nauplia and polychaete larvae including very fragile species with long, thin body appendices like spines and antennae showed no signs of damage (Figure 36).

Figure 36. Chaetoceros sp. (left) and Acartia sp. (right). Both the long spines of Chaetoceros and the antennae of Acartia were undamaged after being collected from the steel filter.

The results clearly demonstrate that the maximum admissible flow velocity at the stainless steel screen filter material ranges around 0.7 m/s, but should be kept at a level of less than 0.7 m/s during sampling of ballast water. Additional pressure tests of the SGS ballast water sampling system 02, executed by the Italian company TAKIN, who developed the SGS ballast water sampling system prototype 01 into the sampling system 02, showed that the SGS ballast water sampling system 02 withstands pressure of up to 5.5 bar.

43/90

0014-00007 R0

Since the SGS ballast water sampling system 02 can be operated in the open circuit modus and in the closed circuit modus (cf. para 8.1.2) additional test have been executed to assess the flow velocity in the sampling system when operated in the two different modi. Figure 37 displays the results of these flow velocity tests.

Fi

gure 37: Results of flow velocity tests run with the SGS ballast water sampling system 02 in the open and closed circuit modus

The results clearly demonstrate that the flow velocity in an open circuit sampling system connected to a ballast water main pipe (yellow bars) increases considerably, whereas the flow velocity in a closed circuit sampling system connected to a ballast water main pipe (dark blue bars) matches the values of the expected flow velocity derived from hydraulic calculations (light blue bars). Ballast water pipes systems on-board ships are operated under high pressure, which might range at considerable high level of 4.0 bar. Any open circuit sampling system, which is connected to this high pressure ballast water pipe will immediately create a differential pressure regime, when operated: the high pressure within the ballast water main pipe is released through the isokinetic pipe and presses the ballast water through the sampling port into the sampling system, which is at ambient pressure level. The flow velocity in the sampling system will increase with increasing differential pressure. Especially when ballast water sampling procedures with open circuit sampling systems are executed on-board ships for compliance testing, this uncontrollable increase of flow velocity has to be compensated by throttling the corresponding valve at the sampling port to secure that the flow velocity directly at the filter material of the sampling system does not exceed 0.7 m/s for the SGS ballast water sampling system 02 and 0.5 m/s for any other open circuit sampling systems.

calculated

closed system

open system

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

flow velocity sampling system

(m/s)

Flow Velocities in Open and Closed Sampling Systems(diameter main pipe : 100mm; diameter isokinetic pipe : 25,4 mm)

30 80 160 200volume f low main pipe (m³/h)

44/90

0014-00007 R0

8.2 BALLAST WATER ANALYSIS

The detection of live plankton organisms is executed under the dissecting microscope or with the inverted microscope, following classical methods of marine plankton research. For larger plankton (the ≥50 µm size class) the sample is poured into a plankton count chamber (e.g. Bogorov chamber; Bogorov 19272, Harris et al. 20003), which simplifies the counting of the organisms. Smaller plankton organisms (the ≥10 and <50 µm size class), e.g. microalgae, may be counted by aid of the “Neubauer Chamber” (haemocytometer4) or the “Sedgewick Rafter Cell”, both of which allow the counting of small particles on a microscopic slide with laser engraved quadrates. Microalgae may also be quantitatively detected by fluorometric particle count methods (flow cytometry). The concentration of marine bacteria in the seawater is classically assessed by cultivation methods on group or species specific agars and 24/48 hour incubation. All of these methods are either very time consuming (plankton counts, incubation of bacteria), their accuracy depend on the skill of the executing person (plankton counts) or demand bulk material (particle count methods). On-board compliance testing should make use of methods that enable the executing person to generate reliable data under a set of arduous conditions on-board a ship under operation or in the harbor:

• Technical decks on ships do not leave much space for analytical apparatus and material.

• Special on-board safety regulations have to be respected. • Sampling has to respect on-board operations for de-ballasting. • Restricted time available for the execution. • Analytical data have to be generated as rapid as possible to enable port state

controls to clearly define adequate consequences in case of non-compliance. Under these pre-conditions the traditional methods to analyze ballast water for the content of viable plankton organisms are not recommended for the on-board compliance testing. However, the need to respect the time constraints and at the same time generate reliable results requires a complex solution. New analytical methods, applicable for ballast water analysis on-board ships under operation, are needed to accurately and rapidly generate data to indicatively assess the “gross exceedance” of the IMO standards.

2 Bogorov, B.G. 1927: Zur Methodik der Bearbeitung des Planktons. (Eine neue Kammer zur Bearbeitung des Zooplanktons) Russ.gidrobiol. Zh. 6:193-198

3 Harris, R., Wiebe, P., Lenz, J., Skjoldal, H.R., Huntley, M. eds. 2000: ICES Zooplankton Methodology Manual; monogr. Academic Press 2000, pp684

4 Swamy, P.M. 2008: Laboratory Manual on Biotechnology; monogr. Pastogy Publications 2008, pp617

45/90

0014-00007 R0

LH2 + O2 + ATP Luciferase

Mg 2+oxy-L + CO2 + AMP + hν

8.2.1 ATP - Adenosine Triphosphate Fluorometry

Adenosine Triphosphate - ATP, is the universal energy carrier used in all living cells. It is of central importance for all physiological processes in living cells. Within the cell compartments the ATP molecule is charged with chemical energy, which is passed on to other intracellular carrier molecules then. Within a water sample containing microorganisms, there are two types of ATP: (i) intra-cellular ATP representing ATP contained within living biological cells and (ii) extra-cellular ATP representing ATP contained outside of living biological cells. Thus the total concentration of ATP (tATP) in a water sample comprises intra-cellular ATP (cATP) and extra cellular or dissolved ATP (dATP). tATP = cATP + dATP E2 Accurate measurement of these types of ATP is critical in the application of ATP-based measurements especially for assessments of concentration of the living biomass in a sample. ATP is an important co-factor for the Luciferin-Luciferase reaction, in which Luciferin (LH) is transferred by Luciferase in presence of oxygen and ATP: E3

Luciferin emits light at λmax= 537nm with the amount of light (hν) being directly proportional to the

amount of ATP molecules. The luminescence is measured as relative light units (RLU) using a luminometer. The assessment of viable biomass in a water sample through the concentration of cATP in the sample is a common method in a wide range of applications since years: potable water, sanitary water, cooling water, industrial process water, waste water, petrol and bio-fuel. The Luminultra Quench Gone Aqueous QGATM distributed by Aqua-Tools is a highly sensitive test kit for the detection of bacteria in liquid samples. The second generation of this kit line reflects the overlapping of cATP and tATP and physically separates the dATP from the tATP (cf. E2) by filtration and thus clearly addresses the cATP in the sample. In cooperation with Luminultra, Canada and Aqua Tools, France, the QGATM test kit line was taken as a base for the further development of a protocol for ballast water analysis. Initial test series with marine microalgae revealed that the extraction technique to mobilize the ATP from within the cells is crucial for accurate analysis of cATP in ballast water samples. Various extraction tests have been performed with cultured microalgae, with cultured Artemia salina individuals as well as with natural plankton samples from the Baltic Sea and the North Sea. The final protocol uses the technique of grinding with beating beads to extract the ATP from the plankton organisms after which the sample was put in a buffer solution and analyzed for ATP content. The time needed from sample to result is 12-15 minutes. The evaluation of this method regarding the correlation between the concentration of cATP in a ballast water sample and the density of viable organisms in the sample was performed with seawater samples from the Baltic Sea and the North Sea. The organism concentration of the sample was determined using

46/90

0014-00007 R0

microscopy; the sample was then diluted using an iso-osmotic solution to create a dilution series. The results of this evaluation are displayed in figures 38 and 39.

Figure 38: Correlation between cATP concentration and number of natural plankton organisms ≥50µm in diluted sea water samples from the harbor of Büsum, all values based on triplicate measurements

Figure 39: Correlation between cATP concentration and number of natural plankton organisms ≥10µm and <50µm in diluted sea water samples from the harbor of Büsum, all values based on triplicate measurements.

y = 0,1273xR² = 0,9931

0

5

10

15

20

25

30

35

0 50 100 150 200 250

cATP in marine Plankton Organisms >50µmcATP (ng/ml)

number of organisms

y = 0,0922xR² = 0,9961

0

5

10

15

20

25

30

35

0 50 100 150 200 250

cATP in marine plankton organisms >10µm<50µmcATP (ng/ml)

number of organisms

47/90

0014-00007 R0

Based on the experiments described above, Aqua-Tools, France (a cooperation partner to this project) produced an ATP “Ballast Water Test Kit” comprising all material and supplies necessary to perform analysis of ballast water. With this “Ballast Water Test Kit” additional tests were performed to assess the functionality of the test kit and to further assess the content of ATP in single marine plankton organisms with a size of ≥50µm and a size of ≥10µm and <50µm.

Figure 40: Test of the ATP method for the size class ≥10 and <50 µm, content of cATP using the cultured marine plankton organism: Tetraselmis suecica (marine microalgae)

Figure 41: Test of the ATP method for the size class ≥50 µm, content of cATP in marine plankton organisms: Copepods ≥50µm from the Oslo Fjord

y = 0,0784xR² = 0,9765

0

50

100

150

200

250

300

350

400

450

0 1000 2000 3000 4000 5000 6000

Content of ATP in Marine Plankton OrganismsTetraselmis sp.

cATP (pg/ml)

organism concentration (cells/ml)

y = 654,38xR² = 0,8181

0

2000

4000

6000

8000

10000

12000

0 5 10 15 20

Content of ATP in Marine Plankton OrganismsCopepods >50µm (wild populations)

cATP (pg/ml)

number of organisms

48/90

0014-00007 R0

Figure 42: Test of the ATP method for the size class ≥50 µm, content of cATP in the marine plankton organism: Artemia salina Larvae (nauplii)

Figures 40 to 42 display the three test series for different groups of marine plankton executed at the marine biological station Solbergstrand, Oslo Fjord, Norway. Two of the test series use cultured organisms, these are the same organisms used by NIVA to increase plankton density during land-based testing of ballast water management systems5. For the ≥10 and <50 µm size fraction Tetraselmis suecica (minimum dimension around 10 µm), for the ≥50 µm size fraction nauplia (larvae) of Artemia salina (minimum dimension around 200 µm) were used. The third test series used the natural population of copepods and their different life stages found in the Oslo Fjord (Calanus sp. and Acartia sp., not identified to species level, minimum dimensions varied from 100 – 700 µm). A fourth test series using natural plankton ≥10 and <50 µm was planned, but there was insufficient plankton of this size class present in the water of the Oslo Fjord. The generated data from the test program resulted in the identification of a basic value for the content of pg (picogram) cATP in marine plankton organisms of the two IMO target organism size classes ≥50µm and >10µm<50µm. These basic values allow to transform a value for the concentration of cATP in a ballast water sample into a numerical value of viable plankton organisms per ml (size class >10µm<50µm) or per m³ (size class ≥50µm). Aqua-Tools and Luminultra already had an ATP method for the detection of total heterotropic bacteria in sea water. No further development was needed on this method, although it was tested for applicability on sea water samples. A comparison was made in the lab of MicrobiMaris Biotec

5 NIVA (2010) Land based testing of the CleanBallast ballast water management system of RWO. http://www.bsh.de/de/Meeresdaten/Umweltschutz/Ballastwasser/RWO/RWO_NIVA.pdf

y = 938,56xR² = 0,9603

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

0 5 10 15 20

Content of ATP in Marine Plankton OrganismsArtemia sp. larvae (nauplii)

cATP (pg/ml)

number of organisms

49/90

0014-00007 R0

in Kiel between the results of the ATP method and classical microbiological plating methods (Figure 43), which found a good correlation. While the ATP bacteria method is suitable for detecting total heterotrophic bacteria, it is not a species-specific method and can therefore not be used to detect the IMO D-2 indicator microbes directly.

Figure 43 Relation of number of bacteria (in colony forming units/100 mL) and cATP content in samples of marine water. Note that the x-axis is in log-scale.

Additional documents accompany this report (Standard Operating Procedures, protocols, training documents and validation reports by an independent party), in which the ATP method is described in detail including the final calculations of the desired numerical value for the concentration of viable organisms in ballast water.

8.2.2 FISH - Fluorescence-in-situ-Hybridization

Fluorescence in situ hybridization – FISH is a technique in Cytogenetics, used to detect and localize the presence or absence of specific DNA sequences. FISH uses fluorescent gene probes binding to only those parts of the chromosome with which they show a high degree of sequence complementarities. By means of fluorescence microscopy the fluorescent probe can be detected and its location where it is bound to the chromosomes. FISH is applied to assess specific features in DNA in the field of genetic counseling, medicine, and species identification. The ScanVIT E.coli/Coliforms test kit produced by Vermicon, Munich, Germany, represents the fastest FISH test kit yielding real quantitative data of the concentration of viable bacteria in fresh water samples within a time frame of 8 hours. The ScanVIT E.coli/Coliforms test kit formed the basis for the further development of a rapid FISH test kit for the qualitative and quantitative detection of the three bacteria groups in ballast water samples defined by the IMO as target organisms within the frame of the Ballast Water Management Convention. The aim of the further

0

100

200

300

400

500

600

700

800

900

1,00E+00 1,00E+01 1,00E+02 1,00E+03cfu/100 mL

cATP

pg/

mL Sample 1

Sample 2

Sample 3

Sample 4

Sample 5

50/90

0014-00007 R0

development of the ScanVIT E.coli/Coliforms test kit was to combine the qualitative and quantitative detection of all of the three bacterial target groups possibly in just one assay with a minimum of time requirements. Specific gene probes were designed in silico for the detection of Enterococcus spp. and for Vibrio cholerae. The gene probes are specifically designed for the detection of viable bacteria of Enterococcus spp. and for Vibrio cholerae in ballast water samples. Both gene probes were tested in silico and in situ on target and non target strains and results confirmed the specificity on the probes: only target bacteria were detected with the probes whereas non-target cells could be discriminated, giving negative results with the probes. This development led to two new ScanVIT test kits and matching analysis protocols, the ScanVIT E.coli/Enterococci test kit and the ScanVIT V.cholerae test kit. The main steps of the analysis protocol are:

1. Filtration. The sample is concentrated on a filter. 2. Incubation. The filter is incubated on an agar plate. 3. Hybridization. The filter is placed in the ScanVIT reactor (Figure 44) to let the fluorescent

gene probes bind to the DNA or RNA. 4. Preparation. The filter is washed and prepared for analysis. 5. Analysis. The filter is placed under a fluorescence microscope where the fluorescent

colonies of the target bacteria can be easily counted (Figure 45).

Figure 44. The ScanVIT reactor (copyright Vermicon AG, used with permission).

The analysis protocols of the two ScanVIT test kits are almost identical, there are three important differences:

1. The fluorescent probe. The ScanVIT E.coli/Enterococci protocol requires a different set of fluorescent probes than the ScanVIT V.cholerae protocol. The bottle with the fluorescent probes is colour-coded te prevent confusion.

2. The agar on which the filters should be incubated. The ScanVIT E.coli/Enterococci protocol requires M1 agar, the ScanVIT V.cholerae protocol requires TCBS agar.

51/90

0014-00007 R0

3. The incubation time. The ScanVIT E.coli/Enterococci protocol requires a minimum incubation time of 8.5 hours, the ScanVIT V.cholerae protocol requires a minimum incubation time of 8 hours.

Figure 45. Colonies of Escherichia coli (red fluorescent) and Enterococcus (green fluorescent) (copyright Vermicon AG, used with permission).

The ScanVIT E.coli/Enterococci test kit and the ScanVIT V.cholerae test kit and analysis protocols were tested by an independent party (MicrobiMaris Biotec, Kiel) on sea water samples of various origins, enriched with the target bacteria (Table 10). This confirmed both that the ScanVIT approach is applicable for seawater samples and that it produced comparable results to classical incubation methods. Since the ScanVIT methods use an incubation step, the result is expressed in cfu and can be directly compared to classical incubation methods. The time from sample to result was found to be 10 to 12 hours.

52/90

0014-00007 R0

Table 10. Performance of both ScanVIT methods in comparison to classical bacterial incubation methods on sea water samples of various origins.

53/90

0014-00007 R0

Figure 46: FISH work place in SGS laboratory (SGS Testing & Control Services Singapore Pte. Ltd)

Additional documents on the ScanVIT methods accompany this report (Standard Operating Procedures, protocols, training documents and validation report by an independent party) in which the FISH method is described in more detail. Uncertainty calculations: Based on the independent validation reports for ATP and FISH, uncertainty calculations have been made by SGS for these methods. A report detailing these calculations and the results accompanies this report.

54/90

0014-00007 R0

8.2.3 PAM - Assessment of Chlorophyll a by Pulse Amplitude Modulation Fluorometry

This paragraph has been contributed by Prof.Dr.Nick Welschmeyer, Moss Landings Marine Laboratories, California, USA The determination of variable chlorophyll a (chl) fluorescence from natural, whole-cell phytoplankton serves as a rapid method to assess relative ballast water compliance. The protocol overview and its associated Standard Operating Procedure (SOP) provide background and instructions for shipboard ballast water compliance test methods utilizing variable chl a fluorescence technology. Variable chl fluorescence has been reviewed extensively (Schreiber et al. 2000; Baker 2008) for its use as a rapid indicator of photoautotrophic physiological state. The essential elements of variable chl fluorescence include the determination of minimum fluorescence levels, Fo, representing conditions of ‘open’ photosynthetic electron acceptors in autotrophic organisms and maximum fluorescence levels, Fm, representing conditions of ‘closed’ electron acceptors (Schreiber et al., 2000). The energy of light that is absorbed by photoautotrophic algal cells is dissipated by three competing processes: 1) emission as fluorescence at wavelengths longer than the incident light, 2) initiation of electron transport in the light reactions of photosynthesis and 3) radiationless decay into heat. Over short time scales of µseconds to milliseconds, radiationless decay is relatively constant; however, at time scales of µseconds to milliseconds, fluorescence and photosynthesis rapidly responsive and they are mutually competitive processes. That is, a reduction in one will result in increased energy flux through the other. For instance under dark conditions, when photosynthetic electron acceptors are ‘open’ for electron transport, the fluorescence emissions will be at a minimum, Fo. Conversely, under high irradiance, when electron acceptors are saturated and therefore ‘closed’ for electron transport, the fluorescence emission will rise to a maximum, Fm. The measured difference, Fm - Fo, termed Fv (variable fluorescence) represents the total ‘active’ chlorophyll in a natural water sample; it is a density-dependent term. That is, the measured parameter, Fv, scales proportionally with ambient chl a concentration. The popular term, Fv/Fm, defines the Photosystem II quantum efficiency (the fraction of absorbed photons successfully utilized in photosynthetic electron transport); it is a density-independent ratio, scaling from 0 to 0.7, that has been used often to evaluate physiological conditions in all forms of photoautotrophs (Genty et al. 1989). Note, however, that Fv/Fm, is independent of ambient chl a concentration, and thus, does not relate to the actual concentration of live cells in a given ballast water sample. In ballast water compliance evaluation, considerable focus is given to two characteristics of ballast water organism viability, 1) the relative health or metabolic status of the organisms and 2) the absolute number of living cells (numeric live concentrations). From a practical position, the latter characteristic, live cell concentrations, receives most attention in ballast water work since it conforms most readily to the stated ballast water discharge standards which are specifically

55/90

0014-00007 R0

defined in terms of acceptable live concentrations for any given ballast water organism size category. The protocol here and the associated Standard Operating Procedure (SOP) provide details on the application of variable chl a fluorometry to ballast water compliance testing. The method is specifically configured to yield indirect data that scales proportionally with the total living biomass within the 10 - 50 µm size class; a calibrated conversion factor allows a simple estimate of the equivalent numerical live count for purposes of compliance evaluation against published ballast water discharge standards. In the protocol outlined here, the analysis of variable chl a fluorescence has been reduced to its simplest element, Fv, in an effort to achieve the fastest, most convenient and most direct evaluation of phytoplankton living biomass possible. The method does have its limitations for direct comparison to ballast water discharge standards because the detection response is restricted to autotrophic, chl-containing organisms only, thus ignoring heterotrophs. The trade-off, however, lies in the 1) speed (1-2 min), 2) simplicity (no reagents) and convenience (portable, pocket-sized instrumentation) of the protocol; these are considered ideal traits for rapid ballast water compliance monitoring. To our knowledge, the method, as described here, could be executed by any of the currently marketed variable chl fluorometers capable of delivering stable measurements of Fv at levels of chlorophyll approximating those concentrations expected for successfully treated ballast water. Below is a list of vendors (alphabetically) and some examples of instruments (Figure 47, Figure 43).

BBE Moldaenke, Schwentinental, Germany Chelsea Instruments, Ltd., West Molesey, United Kingdom Hach Corp, Loveland, CO USA Hansatech, Ltd., Pentney, United Kindgom Heinz Walz GmbH, Effeltrich, Germany Opti-Sciences, Hudson, NH USA Photon Systems, Inc., Brno, Czech Republic Qubit Systems, Inc., Kingston Ontario, Canada Satlantic Instruments, Dalhousie, Canada Turner Designs, Inc., Sunnyvale CA USA.

Figure 47 The Turner Designs Ballast-Check hand-held fluorometer (left) and the BBE Moldaenke 10cells portable fluorometer (right) (images are copyright of their respected companies, used with permission).

56/90

0014-00007 R0

In the SGS example protocol we describe the use of the pocket-sized, battery operated BW680 fluorometer, originally developed by PSI Inc. (Czech Republic) and now modified and marketed for ballast water work by Hach Corporation (USA). The instrument is preprogrammed with repetitive flash sequences, utilizing both blue and red excitation energy (to optimally cover fluorescence excitation of all algal taxa, including cyanophytes) yielding a single averaged variable fluorescence response, termed the Ballast Water Index (BWI). The BWI response shares all the characteristics of Fv described above. The convenient features of the instrument for rapid ballast water testing relate to its simplicity: 1) no instrument blanking is required, 2) the wide dynamic range of the detector requires no gain adjustments from its minimum detection level of 0.05 µg Chl a/L to over 40 µg Chl a/L (figure 48), and 3) no user-based programming manipulations are required (nor offered); it operates on the push of a single button. The utility of using the density-dependent variable fluorescence ‘BWI’ parameter (and by analogy, Fv) in tracking live cell numbers, and hence ballast water compliance, is shown in figure 42 (lower) where BWI accurately tracked the absolute number of live cells, even in widely varying lab mixtures of live and dead cells (heat killed cells, but still bearing a fluorescent Fo signal comparable to that of the living cells). Again, this same tracking ability would be evident in all modulated fluorometers capable of delivering sensitive measures of Fv (see manufacturer’s list above). Figure 49 shows the relationship between BWI (or Fv) and cell size for several freshwater and marine phytoplankton species. figure 43 confirms the central assumption that BWI is a biomass-dependent parameter. That is, large cells show larger BWI responses than small cells, in relation to their relative cell volume (biomass). This inherent relationship is often ignored when research is focused primarily on the popular, density-independent parameter Fv/Fm. The data in figure 49 show that BWI (or Fv) can be reasonably predicted for autotrophic cells of any given cell-size within, and below, the 10-50 µm ballast water discharge size category. This relationship can be used to estimate numeric concentrations of live photoautotrophs from simple measures of BWI or Fv (see below). The method described here is given in two forms: 1) size-fractionated analysis of the 10-50 µm size category (Protocol 1) and 2) analysis of active chl a for the ‘whole-water’ sample, with no size fractionation steps (Protocol 2). Protocol 1 strives to mimic characteristics appropriate to the IMO 10-50 µm part of the ballast water discharge standards as defined in D-2. Protocol 2 yields the simplest, most conveniently executed assay possible with minimum effort on the side of the user.

57/90

0014-00007 R0

Figure 48: Upper: Variable fluorescence parameter, BWI (ballast water index), as a function of chlorophyll a (extracted). Chlorophyll was measured in seven species of lab cultured phytoplankton. Lower: Laboratory example of live cell tracking using the modulated fluorometric Ballast Water Index (BWI) from the BW680 fluorometer. Live cells and heat-killed dead cells (Prorocentrum micans, approximately 26 µm diameter) were combined quantitatively to yield varying proportions of live and dead cells; equal cell concentrations on the far right of the graph and predominantly dead cells far left (15:1, dead:live, left side). Dead cells had equal steady state fluorescence (Fo) to live cells. Note the linear tracking of BWI (red line) with live cell numbers (x-axis), even in the dominant presence of dead cells.

58/90

0014-00007 R0

Figure 49: Size-based measurements of variable Chl fluorescence (BWI/(cell/mL)). Cell volume was measured with a Coulter Z2 particle analyzer; cell numbers enumerated flow cytometrically (Accuri C6 cytometer, red fluorescence). All phytoplankton cultures were grown under identical day/night incubation conditions. Taxa included: (marine) Gymnodinium, Phaeodactylum, Nitzschia, Porphyridium, Tetraselmis, Dunaliella, Thalassiosira weissflogii, Thalassiosira pseudonana, Pleurochrysis, Prorocentrum, Helicotheca; (freshwater): Peridinium, Chlorella, Ankistrodesmus, Haematococcus, Scenedesmus. Vertical dotted lines mark 10 µm and 50 µm size limits (equivalent spherical diameter, ESD). Red arrows show cell-specific calibration responses calculated for 15 µm ESD from regression equations. On the right the Hach BW680 hand-held fluorometer is shown.

8.2.4 FDA - Fluorescein-Diacetate Fluorometry

This paragraph has been contributed by Prof.Dr.Nick Welschmeyer, Moss Landings Marine Laboratories, California, USA. Fluorescein diacetate (FDA) has a long history of use as a visual marker for cell-specific determination of viability (Rotman and Papermaster, 1966). FDA is a non-fluorescent compound which, when hydrolyzed by biological enzyme activity, yields fluorescein, a highly fluorescent compound that clearly marks ‘live’ cells with optically-induced green fluorescent emission (Fig. 50). FDA cell-specific viability analysis is commonly quantified by numeric counts made either by epifluorescence microscopy or flow cytometry (Dorsey et al. 1989; Brussaard et al. 2001; Garvey et al. 2007). It should be understood, however, that the FDA technique is not a staining procedure per se, e.g., the reaction product, fluorescein, is not chemically bound to specific targets within the cell. FDA and fluorescein readily pass diffusively through cell membranes making the loading of FDA into the interior of cell tissue a simple task in aqueous cellular suspensions. Unfortunately, the subsequent efflux of fluorescein out of the cell’s interior results in rapid fading of the optical cellular signature. FDA cell-specific viability assays have been criticized often for this apparent inconvenience since immediate numerical counts must be made with no opportunity for storage of samples for later analysis (Garvey et al. 2007).

59/90

0014-00007 R0

Figure 50: Simplified summary of the use of fluorescein diacetate (FDA) as a tag for viable organism metabolism; esterase activity is present in all living organisms and is rapidly lost upon cell death.

In contrast to the classic internal cell-specific application of FDA in viability assays described above, a few reports have described the use of FDA as a bulk indicator of viable cell biomass, in this case, based on simple fluorometric analysis of the extracellular bulk fluid in which the cellular material is suspended. This procedure has been described for use with soil samples, leaf litter and industrial cell slurries (Breeuwer et al. 1995). Generally, the bulk approaches have enjoyed limited application due to a) well known non-biological conversion of FDA into fluorescein in the extracellular media (Clarke et al. 2001) and 2) uncertainties in the relation between fluorescein production and viable biomass concentrations, per se (Breeuwer et al. 1995). The protocol described here is a recent conversion of the cell-specific FDA viability assay into a bulk assay that yields simple, cuvet-based fluorometric readings, specifically tailored to ballast water compliance testing (Welschmeyer and Maurer, 2011). Reagent mixtures were developed to accommodate the wide range of salinities that can characterize ballast water organisms, e.g., 0-35 PSU. Reagents were developed to prevent non-biological conversion of FDA to fluorescein thus reducing the possibility of a ‘false-positive’ indication of viable cells. The protocol was optimized to yield the highest sensitivity in the shortest period of time. The method requires small volumes of sample water (<200 mL) to detect undesirable living planktonic biomass in ballast water at low levels commensurate with ballast water performance standards. Most importantly, the rate of production of fluorescein was shown to be predictable as a function of living cell biomass and thus the derived fluorescence values can be converted to equivalent numerical counts using reasonable assumptions for the average cell size of organisms within the 10-50 µm size class. Analytical Features of the bulk FDA Assay The bulk FDA assay involves three essential steps: 1) quantitative capture of plankton on appropriate pore-size filters, 2) incubation of captured cells in appropriate buffer fluid, with resultant production of fluorescein derived from enzymatic activity of live biomass and 3) simple

60/90

0014-00007 R0

fluorometric measurement of the final concentration of fluorescein in the incubation fluid. The final concentration of fluorescein will be a function of 1) incubation time, 2) biomass loading and 3) temperature. An ideal enzyme based analytical method would show linearity in parameters 1 and 2, and would display predictable effects of temperature, often parameterized by a constant Q10 effect (increase in reaction rate per 10 °C increase in temperature). Below we demonstrate adherence of the bulk FDA assay to the analytical ideals above. Time Time-based studies were initiated to define the time scales, and fate, of fluorescence derived from FDA-labeling of live phytoplankton cells. Flow cytometry was used to quantify the intensity of intracellular fluorescence that accumulates (and fades) after FDA inoculation. At least 2,000 cells were analyzed at each time-point to capture individual cellular levels of green fluorescence over the course of ca. one hour. The population mean fluorescence was computed for each time-point yielding the temporal changes in average fluorescence/cell (figure 51). Parallel measurements of the same sample were made simultaneously over the same time period in a conventional cuvet-based fluorometer (Spex Fluoromax 2) so that total fluorescence of the suspension could be monitored. At the end of the observation period the cells were removed by centrifugation and the extracellular fluid was measured for fluorescence. The results in figure 51 give an example of the FDA labeling characteristics generally found for all samples analyzed. The intracellular fluorescence rose quickly after FDA inoculation and, for the diatom Thalassiosira sp., peaked at ca. 5 min (figure 51). There was a slow, but continuous, drop in fluorescence/cell that eventually approached the initial time-zero level of fluorescence. All species tested showed the same trend, with maximum fluorescence found somewhere between 4-20 minutes. Interestingly, even though the cellular fluorescence was adequately bright for cytometric detection, the total fluorescence of the algal suspension was dominated entirely by extracellular fluorescence. That is, removal of the cells by centrifugation resulted in <1% reduction in the measured fluorescence of the fluid in the cuvet. As seen in figure 51 the resultant extracellular fluid fluorescence rose linearly over the time course of the experiment. The results suggested that a predictable analytical production rate of fluorescein was evident, e.g., the live cells exhibit a constant rate of fluorescein production, and at the FDA final concentrations used here (10 μM), the extracellular fluid fluorescence could be predicted over the course of at least 1h in this experiment; this was found to be true over 3h in corroborative experiments.

61/90

0014-00007 R0

Figure 51: Time series of FDA derived fluorescence measured intracellularly (w/ flow cytometry) and extracellularly (w/ conventional cuvet fluorometry) with living suspensions of the diatom, Thalassiosira sp. Note that the fluorescence response of both instruments are not directly comparable (relative units).

Biomass loading Experiments were conducted to test the fluorescence response of cell suspensions to relative biomass loading, under conditions of constant incubation conditions, e.g., identical temperature, FDA concentration and incubation time. The biomass levels were controlled by quantitative volumetric filtration of sample suspensions (cultures or natural seawater) onto filters of appropriate pore size. As shown in figure 52, under equivalent FDA substrate concentrations, the production of fluorescein was proportional to loaded biomass. Notice that in rich algal cultures (figure 46) only a fraction of a mL was required to generate measurable fluorescence; the FDA response of natural plankton communities could be assessed with filtration of as little as 100 mL. The biomass-dependent yield of FDA-derived fluorescence was observed for all experiments to date and constitutes the fundamental quantitative premise of the analytical method described here.

62/90

0014-00007 R0

Figure 52: FDA-derived fluorescence as a function biomass during a 2 hour experiment at 20°C. Relative biomass loading was controlled by quantitative volumetric filtration on appropriate pore size filters; <10um fractions captured on GF/F filters, nylon 10 and 50um filters were cut from sheets of Nitex screen.

Temperature The enzymatic hydrolysis of FDA to fluorescein is expected to be temperature dependent due to the catalytic enzyme activity of the living biological systems that are being tested. The temperature effect was quantified using three temperature controlled water baths set at 10, 20 and 30°C. The incubators were used to incubate replicate aliquots of the prasinophyte, Tetraselmis sp.; aliquots were identical in volume and derived from the same culture, originally grown at 20°C. Figure 53 shows the resulting linear Arrhenius plot with a calculated Q10 of 2.05, e.g., a 10°C increase in temperature yields a two-fold increase in fluorescence signal, all other factors held constant; this is the Q10 expected for typical enzyme based reactions.

Figure 53: Standard Arrhenius plot (inverse temperature (Kelvin) vs. natural logarithm of reaction response, showing expected temperature-dependent linearity with a Q10 of 2.05.

63/90

0014-00007 R0

Fluorescein production as a function of cell size Under standardized incubation conditions (2.5 mL incubation buffer, 10 µM final FDA inoculation, 20°C, 1 h incubation) the measured fluorescein production rate from living algal cultures of known cell volume and cell density was plotted as a function of cell volume in Fig. 54. Cell volumes and counts were determined from volume displacement measurements made on a Coulter Z2 particle sizer, a Coulter LS 13-320 laser diffraction sizer (for freshwater cultures) and an Accuri C-6 flow cytometer. The algal cultures plotted in Fig. 54 include both freshwater and marine organisms; the linear log-log plot confirms the central quantitative tenet of the bulk FDA viability assay. That is, the cellular fluorescein production rate scales quantitatively with cell biomass. These results are quite encouraging; for example, within the common IMO size class of 10-50 µm organisms, we expect the cellular biomass (approximated as cell volume) of 10 µm and 50 µm spherical cells to differ by 125-fold (vol of sphere = 4/3·π·r3); the empirical regression equation derived in Fig. 54 estimates a 145-fold increase in fluorescein production rates within the same cell size range (10-50 µm), remarkably close to the geometric volume prediction. The data in figure 48 can be used to predict the fluorescein production rate per cell for any organism if the volume of the organism is known. Conversely, if the mean cell size of a group of organisms is known, and if the production rate of fluorescein is measured for that same group of organisms, one could calculate the numeric concentration of cells. This is fundamentally significant in relation to ballast water compliance testing since all ballast water discharge standards (BWDS) are stated in numeric terms.

Figure 54: Log-log relationship of cellular fluorescein production (fg cell-1 h-1) as a function of cell volume (µm3). Each data point represents the results from a freshwater or marine algal species cultured at 20 C. Cell volumes and cell concentrations were determined by Coulter Counter volume displacement and flow cytometry, respectively. Dotted vertical lines represent the limits of the 10-50 µm size class; red arrows mark the results expected for a nominal 15 µm equivalent spherical diameter cell, computed from the regression equation given in figure.

64/90

0014-00007 R0

Do we know the mean size of cells in the 10-50 µm regulatory size class? Given the extensive analysis of organism size distributions that have been made in limnology and oceanography over the last 40 years we can easily constrain the characteristic size/abundance conditions that pertain to the ballast water 10-50 µm class with reasonable confidence (Welschmeyer in prep). That is, the scientific literature clearly shows that the likelihood of finding 10 µm vs. 50 µm cells is not equivalent. The early size distribution work of Sheldon et al (1972, 1977) led to the provocative hypothesis that the biomass of all organisms (from bacteria to whales) was nearly equivalent across the expected size range of ocean biota (now known as the ‘Sheldon-type’ distribution). In other words, within the ballast water 10-50 µm class, 10 µm cells are expected to be 125-fold more numerous than 50 µm cells given a Sheldon-like biomass size spectrum. Numerous recent size distribution analyses (by microscope, particle counter and flow cytometry) have found that the Sheldon-type distribution actually underestimates the observed numeric abundance of small particle size classes (Platt and Denman 1978, 1985; Cavender-Bares et al. 2001). As shown in figure 50, the estimated biomass concentration for any given cell size along a size distribution axis is often modeled as

biomass/L = aWc E4 where a is a constant (an appropriate assumption for similar protists within 10-50 µm), W is weight (or cell volume) and c is the exponent that defines the slope of the typical log-log plot of environmental biomass (y) vs. apparent cell volume (x) (Blanco et al, 1994; Vidondo et al, 1997). The log-log Sheldon-like distribution has a flat slope of zero when log biomass is plotted against log cell volume (c = 0). However, nearly all modern measurements have suggested that the observed planktonic size distribution is typically more negative; c ≈ -0.1 for coastal environments and c ≈ -0.2 for open ocean (reviewed in Cavender-Bares et al. 2001). Thus, the true numerical organism concentrations are heavily skewed toward the smallest size classes; e.g., the smaller the organism the more abundant they are. Mathematical integration of the ballast-related, 10-50 µm size distribution spectra described by E4 yields a mean characteristic diameter (ESD) of 16.52, 15.71, and 15.01 µm for c set to 0, -0.1 and -0.2, respectively. Thus, the mean cell size within the 10-50 µm class is likely to be significantly closer to 10 µm than to 50 µm. Corresponding probability calculations show that numeric counts of cells smaller than the mean diameters cited above should be observed with >80% likelihood; the 10-50 µm ballast water regulation organism class is expected to be dominated in biomass and in numbers by small cells.

65/90

0014-00007 R0

Figure 55: Expected relative cell concentration within the 10-50 µm equivalent spherical diameter (ESD) size range given a size distribution of biomass for coastal phytoplankton species modeled by the equation: biomass/L = aW-0.1, where a is a constant, W is cellular volume (biomass) and the exponent -0.1 represents the literature mean value characteristic of coastal species. The curve is normalized such that one cell is counted at 50 µm ESD; red arrow notes the resultant mean cell size. Note that the expected concentration of small 10 µm ESD cells is over 350x that of large 50 µm cells.

Empirical determination of mean cell size based on flow cytometry In an effort to validate the theoretical predictions given above, concerning the ‘average cell size’ with the 10-50 µm ballast regulation class, we analyzed uptake ballast water from six geographic locations using cell-specific sizing information derived from forward-scatter flow cytometry of red-fluorescing phytoplankton. All samples were collected during real-time ballast water efficacy tests; 50 samples were analyzed, yielding over 100,000 individual cell-size estimates, calibrated against a range of known bead-standards that straddled the 10-50 µm range. Results for the empirical, flow cytometric size distributions are given in figure 56. The observed, average cell size was 17.45 µm – remarkably close to the theoretical size calculated from the literature review summarized in figure 55. The mean size will of course vary as a function of community composition, but we can constrain the estimate reasonably well. For now, we have chosen 15 µm as a reasonable estimate of mean cell size within the 10-50 µm category.

66/90

0014-00007 R0

Figure 56: Flow cytometric size distribution analysis of phytoplankton within the 10-50 µm size category. Cytometric forward light scatter was used as the proxy for cell size using certified standard beads for size calibration throughout the 10-50 µm range.

Figure 57: Example results for shipboard ballast water tests with treatment using UV treatment. The bulk FDA responses of size fractionated samples (10-50 µm) for uptake (control) and treatment water (upper right plot) were converted to equivalent numeric concentrations (lower left plot) using the calibrated equation given in the figure. The bulk FDA numeric equivalent concentrations are compared to FDA-tagged flow cytometric numeric counts given in the lower right plot.

67/90

0014-00007 R0

Proof of Concept: Calculation of numeric estimates of live cells from bulk FDA analysis An illustration of the conversion of bulk FDA results from shipboard ballast water testing is shown in figure 57. The formula given in figure 57 makes use of the calculated cell-specific fluorescein production rate for 15 µm cells derived from the regression equation in figure 54. The ballast water treatment test clearly showed drastically reduced fluorescein production in the treatment sample relative to the untreated uptake water. After application of the numeric conversion equation the FDA results yielded estimates of numeric live concentrations that were remarkably similar to those derives from high precision FDA-based flow cytometry. The observed reduction of FDA response resulting from UV treatment was determined after a 5 day hold in the ballast tank, as recommended in IMO shipboard trials. We note that recently there have been numerous observations of ‘false-positive’ live detections when using microscopic cell-specific tagging procedures with FDA as recommended in EPA ETV ballast water protocol (First and Drake, 2014). We have noticed that the reduction in bulk FDA response measured immediately after UV exposure is measurable but not as substantial as shown in Fig. 51. It is likely that the UV-based treatment must assessed several days (>5 days) after UV inactivation to promote full reduction of bulk FDA activity as shown in Fig. 51; more experiments are warranted to understand this problem. In the case of ballast treatment by electrochlorination we find immediate reduction of bulk FDA response, to levels approaching the blank.

8.3 BALLAST WATER TRAINING SEMINARS

Within the frame of this project SGS Institut Fresenius GmbH executed three training seminars for on-board ballast water sampling and analysis:

• The international training seminar for SGS employees from USA and Singapore at the INMT Flensburg, Germany, April 14th to 19th, 2014

• The national training seminar for SGS employees in Singapore at the laboratories of the

SGS Testing & Control Services Pte. Ltd. Singapore, July 28th to August 23rd, 2014

• The national training seminar focused on FDA at the INMT Flensburg, November 4th, 2014

Training materials were developed that provide additional background on the equipment and methods used, these training documents are intended to be used in combination with the Standard Operating Procedures for the methods.

68/90

0014-00007 R0

8.3.1 International ballast water training seminar Flensburg

8.3.1.1 Participants

The following SGS employees attended the training seminar at the INMT in Flensburg:

Table 11: Participating trainees in the Flensburg training seminar

Name Professional Background From Mr. Richard R. Reed Laboratory manager SGS USA

Mr. Wong Sze Beng Chemist SGS Singapore

Mr. Chris Mimms General Environmental Services SGS USA

Mr. Daniel Lodato Administrative Assistant SGS USA

8.3.1.2 Program

The program was split into theoretical lessons and practical exercises in the laboratory of the INMT in Flensburg. The theoretical lessons covered the following aspects: (i) the IMO Ballast Water Management Convention, (ii) Planktonology, (iii) On-Board Working Conditions, (iv) the SGS sampling system, (v) analytical methods. The practical lessons covered all detailed and indicative analysis methods, as well as the operation of the SGS ballast water sampling system 02 together with the sampling and backflush installations presented in figures 32 and 33. The complete program is displayed in para 12.3 of this report, also reference is made to the training documents accompanying this report as listed in para 12.2 of this report. The experiences from this ballast water sampling and analysis seminar are described in para 9 of this report.

8.3.2 National ballast water training seminar Singapore


The following SGS employees attended the training seminar in Singapore:

Table 12: Participating trainees in the Singapore training seminar

Name Professional Background From Mr. Sze Beng Wong Chemist

SGS Singapore

Mr. Don Chin Wee Kuan Engineer

Mr. Chee Choy Yen Chemist

Mr. Omar bin Osman Ship inspector

Mr. Abdul Rashid Bin Ahmad Ship inspector

Mr. Muhammed Osman bin Shainin

Ship inspector

69/90

0014-00007 R0

8.3.2.2 Program

The training seminar for SGS employees in Singapore was executed in a very different frame, although the basic program of the Flensburg seminar was used. The training seminar in Singapore could be enlarged by ballast water sampling and analysis on-board six ships in Singapore harbor. Details of these on-board activities as part of the training seminar are given in para 8.4 of this report. The experiences from this ballast water sampling and analysis seminar are described in para 9 of this report.

8.3.3 National ballast water training seminar Flensburg


The following participants attended the training seminar at the INMT in Flensburg:

Table 13: Participating trainees, third ballast water training, INMT Flensburg

Name Professional Background From

Mrs. Hannelore Nissen Head of laboratory, Chemistry INMT, Flensburg

Dipl. Eng. Wolf Eggert Ship Operation Engineering INMT, Flensburg

Dipl. Eng. Michael Pohl Mechanical Engineering Boll Filter, Kerpen

Florian Kalsow Student, Naval Engineering University of Flensburg

8.3.3.2 Program

After the initial ballast water seminar held in the Institute for Nautics and Maritime Technologies, University of Flensburg and the second ballast water seminar in Singapore, a third ballast water training seminar for the FDA analysis was held in the Institute for Nautics and Maritime Technologies, University of Flensburg. The third ballast water seminar was split into a theoretical and practical part and concentrated on these aspects :

• Introduction to the IMO Ballast Water Management Convention

• Introduction to planktonology

• Regulatory framework for on-board safety

• Regulatory framework for execution of chemical analysis

• Ballast water sampling and sample processing

• Analysis of samples with the Fluorescein-diacetat method

8.4 ON-BOARD TRIALS

In connection with the training seminar in Singapore there was the opportunity to perform sampling and analysis of treated ballast water on-board six ships in the Singapore harbor. This opportunity allowed for the assessing of:

70/90

0014-00007 R0

- The applicability of the sampling and analysis methods under on-board conditions. - How well the training prepares the participants for performing sampling and analysis on-

board. The scope of work on-board these six target ships comprised the sampling of treated ballast water from isokinetic sampling ports already installed in the ballast water main pipe system of these ships using the SGS ballast water sampling system 02. Since no backflush port was installed on any of the six target ships, the SGS ballast water sampling system 02 was operated as an open circuit sampling system. The analysis of the ballast water samples comprised the methods as presented in the table below.

Table 14: Applied analytical methods for ballast water analysis on-board six ships in Singapore harbor

Organism group Method applied Execution

Plankton ≥50µm ATP Microscopic count PAM

On-board

Plankton >10µm<50µm ATP Microscopic count PAM

On-board

Bacteria quantitative ATP Plate counts

On-board Landbased SGS laboratory

Bacteria specific Escherichia coli Enterococci Vibrio cholerae

FISH Landbased SGS laboratory

The details on the assessment of sampling and analysis methods as well as the effectiveness of training methods are given in paragraph 9.3.2 and paragraph 9.4.

9. EXPERIENCES, LESSONS LEARNT AND DISCUSSION

9.1 THE DEVELOPMENT OF THE SAMPLING SYSTEM

The SGS ballast water sampling system 02 developed within the frame of this project proves that the representative sampling directly from ballast water main pipes is feasible. In addition the dimensions of the sampling system allow for an easy transport and use on-board ships. However, the time needed until a full functional sampling system could be presented, i.e. approximately three years seems to be a long development period.

71/90

0014-00007 R0

Although during these years of development SGS contacted quite a number of ship owners to convince them to execute tests on-board their ships, most of these ship owners withdrew their initial willingness to cooperate for on-board tests. In this respect the compliance tests executed on the six ships in Singapore harbor were the first tests, for which the SGS ballast water sampling system 02 could be used for sampling. Although the SGS ballast water sampling system 02 proved its functionality, the fact that on-board all ships only one sampling port was present prevented that the sampling system could be as a closed-loop system. A full listing of practical considerations for use of the SGS ballast water sampling system 02 on-board is given in paragraph 10.

9.2 THE DEVELOPMENT OF ANALYTICAL METHODS

The analytical methods developed within the frame of this project proved to be adequate for rapid on-board compliance testing, although the time from sample to result differs considerably between the developed methods. The ATP method as well as the FDA method are indicative methods and can be executed on-board ships. The PAM method, too, proved to be an adequate method for indicative on-board compliance testing. Both the ATP method and the PAM method generate reliable data within only a few minutes whereas the FDA method requires up to 60 minutes until the results from the analysis are available. The FISH method is a detailed method for the detection of the IMO requested bacteria species and should be executed in a land-based laboratory. The FISH method approximately needs 15 hours to generate analytical results; this is much faster than the classical plate count methods with species specific agars which require 24/48 hours of incubation time.

9.2.1 Execution of the FDA method

Since the FDA method was developed by MLML, a separate test was set up so SGS personnel could get experience with the FDA method. This paragraph details those experiences. The FDA method uses, among other substances, Dimethylsuflfoxid-DMSO. According to the most actual material safety data sheets (Merck, April 29th, 2014, cf. documents accompanying this report) several safety measures have to be taken, when working with DMSO. The personal safety equipment comprises, laboratory glasses, special gloves, impenetrable laboratory coat and others. Wastes generated during the analysis using DMSO have to be discharged under special conditions. The FDA method for the analysis of ballast water uses the FDA-DMSO solution in very small quantities (2 drops per analysis). In this respect it was agreed with the head of the laboratory at the INMT to execute the FDA analysis with laboratory glasses and special gloves. The waste generated during the analysis was collected in adequate containers for special discharge. The FDA method for the detection of viable plankton organisms >10µm<50µm is executed in simple analytical steps and the sample processing to be conducted prior to the analysis is kept to a minimum.

72/90

0014-00007 R0

The incubation time for each analysis is given with 30 minutes, a fact, which prolongs the time from sample to result considerably, especially, if several samples have to be analyzed simultaneously. The method uses two different syringes, fixed on both sides to the same, central syringe filter, on which the viable organisms are collected, before being further processed with the relevant chemicals of the method. This technical arrangement still needs a further optimization step, since the female connectors of one of the syringe cannot be tightly fixed. As a consequence, for many samples the liquid contained in the syringes seeped out subsequently impacting on the final results. It has to be explicitly noted, that the liquid in the syringes already contains DMSO and thus there is a strong demand to solve this minor technical problem. The fluorometer, which finally measures the content of Fluorescein in the sample, is easy to handle and the results are available only a few seconds after activation of the fluorometer.

9.3 THE TRAINING SEMINARS

9.3.1 The Flensburg seminar

The ballast water sampling and training seminar in Flensburg revealed that the theory and the practical exercises with the SGS ballast water sampling system connected to a land-based ballast water system can easily be learnt also by trainees with limited or no technical and scientific background. The same applies for the developed analytical methods ATP and FISH as the necessary material and supplies can easily be operated or used. However, the practical exercises with the analytical methods were executed in the laboratory at the INMT which does not reflect on-board conditions. Although the developed analytical methods can be easily operated, the scientific background of these methods can be difficult to understand for trainees with a limited or no scientific background. The same applies for theoretical lectures of basic physiology, e.g. the importance of osmosis for the processing of marine plankton samples. In contrast to these specific experiences stands the practical exercise for microscopic counts of live plankton organisms. All of the trainees easily and quickly learnt how to operate the microscope and the different plankton count chambers. In addition the differentiation into “alive”, i.e. “viable” individuals and “dead”, i.e. ”non-viable” individuals, by simply observing the organisms for their species specific movements or by agitating apparent lifeless individuals using adequate dissection tools was not a major obstacle for the participants However, the experiences gained during the Singapore training seminar following the Flensburg seminar revealed an important gap of the Flensburg seminar: the practical exercises under “real life” conditions on-board a ship. The Flensburg seminar proved that a seminar duration of one week for theory and practical exercises with a land-based ballast water system is sufficient for the trainees to learn the basics of ballast water sampling and analysis, even if the trainees, as it was the case for the Flensburg participants, do not have a technical and/or scientific background.

73/90

0014-00007 R0

However the Singapore training seminar covering on-board compliance tests on several ships with the entire spectrum of the at present available indicative and detailed methods for ballast water analysis execution of on-board compliance testing, revealed that a much wider range of general experience is necessary to ensure smooth operations during on-board compliance testing. This general experience comprising e.g. the conduction of interviews with the ship’s crew to receive the necessary data or the quick solution of suddenly occurring technical problems can only be acquired during the execution of the on-board compliance tests.

9.3.2 The Singapore seminar

The Singapore training seminar used the same program which was developed for the Flensburg seminar. The participating trainees in contrast had different professional backgrounds: two chemists, one engineer and three ship inspectors. Already the first days of the training made clear that the success of the seminar and the time needed for theory and practical exercises largely depend on the professional background of the trainees. While the theoretical part of the Singapore seminar required almost the same time as during the Flensburg seminar, the practical exercises, i.e. the training on the indicative analytical methods in the laboratory, could be finished in a much shorter period of time. As it was planned to execute on-board compliance tests after the first week of the training seminar an additional lecture for on-board safety regulations was held prior to the first on-board compliance test. Again it was obvious that the professional background of the trainees trigger the success of the seminar: three of the trainees for the Singapore seminar are ship inspectors, already worked on many ships and were perfectly familiar with all relevant safety regulations. The experiences gained during the compliance tests on-board the six ships in Singapore harbor are given special attention in paragraph 9.4 of this report.

9.3.3 The second Flensburg seminar

The ballast water training for FDA analysis started with introductory lessons regarding the Ballast Water Management Convention, planktonology, on-board safety issues and the regulatory frame of chemical analysis, i.e. safety measures to be taken when using chemicals for analysis as presented in material safety data sheets. The practical part of the training started with an introduction into the FDA method, the production and use of iso-osmotic, sterile filtered seawater and the processing of samples with artificial seawater. The brief introductions were followed by demonstrations and practical exercises for the production of artificial seawater, processing of samples with artificial seawater and the execution of the FDA method. Each of the participants was asked to execute the FDA analysis. During the FDA analysis the participants were asked to wear laboratory glasses and laboratory gloves. The training started at 09:00 a.m. and was finished around 04:00 p.m. Similar to the ballast water training seminars already executed, the participants for this third ballast water FDA training seminar reacted differently according to their educational background (cf. table 13) to the content of teaching.

74/90

0014-00007 R0

While the head of laboratory could easily understand the execution of the FDA method, it took more time for the engineers to understand the theory of the FDA method. In contrary the engineers were familiar with the safety regulations on board of ships, while the head of the laboratory needed more time to learn about these regulations. For all of the participants it was not easy to understand the necessity, why an iso-osmotic solution of artificial seawater is needed. Again the principle of osmosis and its implications for the processing of ballast water samples was difficult to learn for this group of participants. All participants agreed, that the FDA method is easy to learn and easy to execute also, with the view to execute it on-board a ship. All participants confirm that wearing special protective laboratory gloves makes it more difficult to achieve a smooth handling of the material, especially seen to the female syringe fixed to the syringe filter. During all analysis the participants complained about solution seeping out between the female syringe and the syringe filter. The time needed to execute the FDA analysis by untrained persons is split into 4 steps:

• Processing of sample (filtration, recovery of organisms with artificial seawater): 6 min.

• Analytical procedure until incubation step: 5 min.

• Incubation: 30 min.

• Terminating analytical procedure, including measurement: 2 min.

It turned out that FDA samples can be processed in an interlaced way : for a sequence of three samples the time until the result of the third sample is available a time span of 93 min. is necessary.

9.4 THE ON-BOARD COMPLIANCE TESTS IN SINGAPORE HARBOR

General Apart from the necessary and obligatory preparations, which should precede any on-board compliance test and which are also described in the Guidelines on Port State Control under the 2004 BWM Convention (IMO MEPC67/20) an initial meeting with the ship master and chief engineer on duty is indispensible prior to the execution of any other activity within the frame of on-board compliance testing. During this meeting the ship master has to be informed about the testing program, about the support from the ships´ crew necessary to perform the compliance test and the team supposed to execute the compliance test needs the complete hydraulic data of the ship. These data will finally enable the team to secure the adequate sampling of the ballast water with the SGS ballast water sampling system 02. The initial, informal meeting is followed by the establishment of the two working places on-board the ship: (i) the sampling location and (ii) the work place for the execution of the analysis. As soon as the two work places are functional, the sampling procedure can be started, which should be accompanied by a crew member for the entire duration. The analysis procedures can start as soon as the first ballast water sample is available. Installation and operation of the SGS ballast water sampling system 02

75/90

0014-00007 R0

The modular concept of the SGS ballast water sampling system 02 simplifies the transport of the material down to the technical decks, on which the sampling has to be executed. Once the adequate sampling port of the ballast water main pipe has been identified, the SGS ballast water sampling system 02 is connected to the sampling port and the modular parts of the sampling system are fixed. Although the technical details of the SGS ballast water sampling system 02 allow for an easy installation it turned out on almost every ship that the installation of the sampling system requires a considerable long time span. This is mainly due to the fact that almost all isokinetic sampling ports on-board the tested ships are difficult to access. In addition these isokinetic sampling ports do not provide uniform dimensions. As a consequence the SGS ballast water sampling system 02 has to be adapted to the specific conditions on each ship and the installation has to be done by the aid of several crew members. The photographic documentation in the annex to this report presents some of the isokinetic sampling ports encountered on the tested ships (cf. para 10.4). The assembly of the SGS ballast water sampling system 02 is done in a very short period of time, due to the simple connectors used for the different, modular parts of the sampling system. As none of the tested ships provided a backflush port for the SGS ballast water sampling system 02, the sampling system had to be operated as an open circuit sampling system. The ballast water, which had run through the sampling system was collected in a recipient, from which a pump transferred it into the bilge of the ships. On all ships this sampling condition limited the sampling procedures. As the ships masters are obliged to secure the adequate treatment and disposal of the bilge waters, no ship master will allow the bilge of his ship to be filled up with several cubic meter of ballast water. As a consequence the maximum allowable volume of ballast water to be used during the sampling procedures was limited to only 2.0 m³. On-board analysis of ballast water The on-board analysis revealed the advantages of the indicative methods. The ATP method can be used for the detection of all IMO target organism size classes. Subsequently only one set of material and supplies is needed for the detection of plankton ≥50µm, plankton >10µm<50µm and bacteria (quantitative). The time needed from sample to result ranges within 4 to 8 minutes. The detection of active chlorophyll as indicator for live microalgae, too, is a fast method, which generates results in less than 3 minutes. The microscopic counts for the detailed analysis executed on-board the ships demand a considerable long period of time, however, on none of the ships this detailed analytical method generated any delay of the compliance tests. In order to reduce the time needed for the on-board compliance tests the sampling and analytical procedures should be arranged in an interlaced structure, which allows for the simultaneous execution of sampling and analysis: while the filtration of 1.0m³ of ballast water to retain the live plankton organisms ≥50µm might require a filtration time of 25 minutes, the samples for the analysis of live plankton organisms ≥10µm<50µm can be generated via the small isokinetic

76/90

0014-00007 R0

bypass sampling port of the SGS ballast water sampling system, as separate, single samples or as a continuous drip sample. At present, the term “gross exceedance” or similar “gross non-compliance” are not clearly defined and this gap complicates the decision to formulate a non-compliance statement for indicative analyses: which value correspond to the term “gross exceedance”? One of the lessons learnt is that to capture a representative and reliable glimpse or snapshot of the ballast water with its complex characteristics of an artificial water body within minutes, the ballast water has to be analysed as complete as possible under the given conditions on-board ships. Another lesson learnt is that providing additional and uniform sampling port installation on-board the ships would optimize the generation of representative samples with the SGS ballast water sampling system 02 to a large extent.

10. BALLAST WATER SAMPLING AND ACTUAL IMO REGULATIONS

At present the basic technical conditions for sampling of ballast water on-board ships faces the following challenges, even when using the SGS ballast water sampling system 02.

Installed isokinetic sampling ports are very difficult to access

⇒ Loss of time through difficult installation of any sampling system

Installed isokinetic sampling ports are not uniform

⇒ Requires additional time to adapt the sampling system to the sampling port

No backflush ports available

⇒ Connected sampling systems can only be operated as open circuit systems

⇒ The flow velocity in open circuit sampling systems can hardly be controlled

⇒ The adequate flow velocity directly at the filtermaterial to safely sample live plankton organisms is difficult to control

⇒ Open circuit sampling systems generate large volumes of waste water

⇒ Due to IMO regulations under MarPol the discharge of waste ballast water volumes into the ships bilge is critical, since it increases the bilge water volume to be treated considerably

77/90

0014-00007 R0

Under these conditions the generation of several ballast water samples each with a large volume of > 1.0 m³ and the generation of continuous drip samples over a long period of time seems to be almost excluded.

The installation of uniform, easy accessible ports for ballast water sampling and especially the installation of back flush ports would secure that future on-board sampling of ballast water can be performed easily.

11. CONCLUSIONS AND FUTURE PERSPECTIVES

Methods for on-board sampling and (indicative) analysis were developed and successfully tested on-board ships. All analysis methods mentioned here are available on the open market and SGS offers sampling and analysis of ballast water as one of its services.

Sampling: the ballast water sampling system 02 has no electrical components and the robust design allows for sampling of large volumes of ballast water according to the IMO guidelines.

Adenosine triphosphate methods: ATP is an indicative method that can be used for ≥50 and ≥10 - <50 organisms, as well as total heterotropic bacteria. The time from sample to result is 10-15 minutes and the equipment can be used on-board ships.

Fluorescein diacetate: FDA is an indicative method that can be used for ≥10 - <50 organisms. The time from sample to result is 40-45 minutes and the equipment can be used on-board ships. Some issues were identified getting accurate results using the FDA method on samples treated with UV-based ballast water management systems.

Pulse-amplitude modulated fluorometry: PAM is an indicative method that can be used on photosynthetic organisms. The time from sample to result is 5-10 minutes and the equipment can be used on-board ships.

Fluorescent in-situ hybridization: FISH is a detailed method that can be used for the D-2 indicator microbes Escherichia coli, Enterococci and Vibrio cholerae. The time from sample to result is 13-14 hours and the equipment is too bulky to be used on-board ships. However, it presents an improvement over classical bacterial incubation methods which require 24-48 hours.

Both the sampling system and the analysis methods are fully functional, however SGS is continuing the further development of all methods based on practical experiences.

78/90

0014-00007 R0

12. ANNEX

12.1 LIST OF PROJECT COOPERATION PARTNERS

Co-operation Partner Field of Expertise

Moss Landings Marine Laboratories, USA

Official project partner

Institute for Nautics and Maritime Technologies, Germany

Project workbase, provision of workshop, BW laboratory and personnel

BlueBioTech, Germany Provision of live marine microalgae, laboratory workspace and personnel

Microbi Maris Biotech, Germany Marine microbiological services

ANKON, Germany Steel construction works at project workbase

Hydro Bios, Germany Production of marine scientific apparatus

Institute for Marine Science-GEOMAR, Germany

Exchange of information on marine planktonology

Institute for Hydrobiology and Fisheries Science, Germany

Exchange of information on marine planktonology

Bollfilter Protectionsystems, Germany

Production of ballast water filters

Vermicon, Germany Development of FISH microscopy for on board analysis

AQUA TOOLS, France Production of ATP test kits

Luminultra, Canada Production of ATP test kits

Royal Netherlands Institute for Sea Research, Netherlands

Consultations on BW sampling and analysis

Marine Eco Analytics, Netherlands Consultations on BW sampling and analysis

Norwegian Institute for water research, Norway

Consultations on BW sampling and analysis, validation services

TAKIN Extrusion Technology, ItalyItalian engineering company, optimization of ballast water sampler prototoype 01

79/90

0014-00007 R0

12.2 LIST OF SEPARATE DOCUMENTS APPENDED TO THIS REPORT

The following list presents the separate documents, which accompany this report

Document Group Document

Standard Operating Procedures (SOP)

Artificial Seawater ATP Bacteria ATP Plankton 10-50µm ATP Plankton >50µm FISH Entero Ecoli FISH Vcholerae FDA Plankton 10-50µm PAM

Protocols Artificial Seawater ATP Bacteria ATP Plankton 10-50µm ATP Plankton >50µm FISH Entero Ecoli FISH Vcholerae FDA Plankton 10-50µm PAM

Material Safety Data Sheets (MSDS)

ATP Luminase UltraCheck UltraLute UltraLyse

FISH Finisher Solution B2 Solutions D1, D7 VitPCNCSolc

FDA DMSO FDA MesHy NaOH Reagent A1 Reagent A3

Training Documents ATP method Artificial seawater FISH method Flow velocities Indicative ./. in depth analysis Interview with ship’s crew Orientation of isokinetic pipe Microscopic plankton counts Identification of plankton Processing of plankton samples Closed sampling system Open sampling system Questionnaire for ships crew

80/90

0014-00007 R0

List of separate documents appended to this report (cont.d)

Document Group Document

Validation reports Norwegian Institute for Water Research – NIVA 2013: Preliminary validation of an ATP analytical procedure for ballast water NIVA Journal no. 1894/13 v1.4

Norwegian Institute for Water Research – NIVA 2013 Preliminary validation of a sampling procedure developed by SGS Germany NIVAJournal no. 1912/13 v2.1

Microbi Maris Biotec, Germany Report on the validation of a method for the dtermination of bacteria in marine water samples based on the measurement of cellular Adenosin-triphosphate (cATP)

Microbi Maris Biotec, Germany Report on the validation of a method for the determination of bacteria (Escherichia coli, Enterococci and Vibrio cholera) in marine water samples based on the VIT® gene probe technology Fluorescence-In-Situ-Hybridisation (FISH)

81/90

0014-00007 R0

12.3 BALLAST WATER SAMPLING AND ANALYSIS TRAINING SEMINAR: PROGRAM

82/90

0014-00007 R0

12.4 PAMAS DATA

Data produced by the PAMAS device.

83/90

0014-00007 R0

12.5 PHOTOGRAPHIC DOCUMENTATION

Photo 1, 2: Ballast water sampling port on-board a small container ship (left) and the connection of the SGS ballast water sampling system 02 (right)

Photo 3, 4: The SGS ballast water sampling system 02 as an open circuit system installed on a VLCC ship (left and on a small container ship (right), not availability of space to execute the sampling

84/90

0014-00007 R0

Photo 5, 6, 7 and 8: Sampling ports on different ships

85/90

0014-00007 R0

Photo 7: The work place for ballast water analysis in the engine control room on-board a ship

86/90

0014-00007 R0

12.6 LITERATURE LIST

12.6.1 IMO documents

Guidelines for Ballast Water Sampling (G2), October 10th, 2008; MEPC 58/23, Annex 3, Resolution MEPC.173/58) )

Guidelines for Approval of Ballast Water Management Systems (G8), October 10th, 2008; MEPC 58/23, Annex 4, Resolution MEPC.174(58) )

Guidelines for Port State Control under the International Convention for the Control and Management of Ship’s Ballast Water and Sediments April 10th, 2014; III 1/8 IMO Sub-Committee on Implementation of IMO Instruments, 1st Session, Agenda Item 8, Annex 1, document page 7ff

International Convention for the control and management of ship’s ballast water and sediments; BWM/CONF/36, February 16th, 2004

12.6.2 Plankton sampling and separation

Boll & Kirch Filterbau GmbH 2009: Muschellarvenabtötung in Spaltfiltern; not published, pp 5

Chavez, F.P., Bidigare, R.R., Karl, D.M., Hebel, D., Latasa, M., Campbell, L. 1995: On the chlorophyll a retention properties of glass-fiber GF/F filters; Limnolo.Oceanogr. 40(2), 1995, 428-433

Evans, M., Sell, D.W. 1985: Mesh size and distroibution characteristics of 50cm diameter conical plankton nets; Hydrobiologia 122, 97-104

Harris, R.P., Wiebe, P.H., Lenz, J., Skjoldal, Huntley, M. (eds.) 2000: ICES Zooplankton Methodology Manual; Academic Press, London; ISBN 0-12-327645-4; p684

Hwang, J.-S., Kumar, R., Dahmes, H.-U., Tseng, L.-C., Chen, Q.-C. 2007: Mesh size affects abundance estimates of Oithona spp. (Copepoda, Cyclopoida); Crustaceana 80(7): 827-837

Kelso, W.E., Keller, M.D., Rutherford, D.A. 2013: Collecting, processing and identification of fish eggs and larvae and zooplankton; from: Fisheries techniques, third edition, ISBN: 978-1-934874-29-5.

Logan, B.E. 1993: Theoretical analysis of size distributions determined with screens and filters; Limnol.Oceanogr. 38(2), 1993, 372-381

Postel, L., Fock, H., Hagen, W. 2000: Biomass and abundance; in: Harris et al eds. 2000: ICES Zooplankton Methodology Manual; Academic Press, London; pp 83-164

Seda, J., Dostalkova, I. 1996: Live sieving of freshwater zooplankton: a technique for monitoring community size structure; J.Plankt.Res. 18(4) 1996, pp513-520

Sommer, U. 2005: Biologische Meereskunde (Marine Biology) 2nd Edition; Springer Berlin, ISBN3-540-23057-2, pp 404

Taguchi, S., Laws, E.A. 1988: On the microparticles which pass through glass fiber filter type GF/F in coastal and open waters

U.S. EPA 2003: Standard operating procedure for zooplankton analysis; U.S. EPA LG403

U.S.EPA GLNPO 2005: Standard operating procedure for zooplankton sample collection and preservation and Secchi depth measurement field procedures; U.S. EPA LG402

Wu, C.-J., Shin, C.-M., Chiang, K.-P. 2011: Does the mesh size of the plankton net affect the result of statistica analysis of the relationship between the copepod community and water masses; Crustaceana 84(9): 1069-1083

87/90

0014-00007 R0

12.6.3 PAM – Variable Chlorophyll a fluorescence

Baker, N.R. 2008. Chlorophyll fluorescence: A probe of photosynthesis in vivo. Annu. Rev. Plant Biol. 59:89-113.

Genty, B.; Briantais, J.M.; Baker, N.R. 1989. The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochim. Biophys. Acta 990: 87–92.

Schreiber, U., W. Bilger, H. Hormann and C. Neubauer. 2000. Chlorophyll fluorescence as a diagnostic tool: basics and some aspects of practical relevance, pp 320-326, In: A.S. Raghavendra (Ed.), Photosynthesis: A Comprehensive Treatise, Cambridge University Press.

Vidondo, B., Y.T. Prairie, J.M. Blanco and C.M. Duarte. 1997. Some aspects of the analysis of size spectra in aquatic ecology. Limnol. Oceanogr. 42:184-192.

Welschmeyer, N. 2013. Rapid, indirect ballast water compliance methods and their relation to regulated numeric ballast water discharge standards: Calibration based on plankton size distribution models in aquatic ecology. (in preparation).

12.6.4 FDA - Fluorescein-Diacetate Fluorometry

Blanco, J.M., F. Echevarria and C.M. Garcia. 1994. Dealing with size-spectra: Some conceptual and mathematical problems. Sci. Mar. 58:17-29

Breeuwer, P., J. L. Drocourt, N. Bunschoten, M. H. Zwietering, F. M. Rombouts, T. Abee (1995). Characterization of uptake and hydrolysis of fluorescein diacetate and carboxyfluorescein diacetate by intracellular esterases in Saccharomyces cerevisiae, which result in accumulation of fluorescent product. Appl. Environ. Microbiol. 61: 1614-1619.

Brussaard, C.P.D., D. Marie, R. Thyrhaug and G. Bratbak. 2001. Flow cytometric analysis of phytoplankton viability following viral infection. Aquatic Microbial Ecology 26:157-166.

Cavender-Bares, K.K., A. Rinaldo and S.W. Chisholm. 2001. Microbial size spectra from natural and nutrient enriched ecosystems. Limnol. Oceanogr. 46:778-789

Clarke, J.M., M.R. Gillings, N. Altavilla and A.J. Beattie. 2001. Potential problems with fluorescein diacetate assays of cell viability when testing natural products for antimicrobial activity. J. Microbiological Methods 46:261-267.

Dobroski, N., C. Scianni, D. Gehringer and M. Falkner. 2009. Assessment of the efficacy, availability and environmental impacts of ballast water treatment systems for use in California waters. Prepared for the California State Legislature, California State Lands Commission, 173p.

Dorsey, J., C. M. Yentsch, S. Mayo, and C. McKenna. 1989. Rapid analytical technique for the assessment of cell metabolic activity in marine microalgae. Cytometry. 10: 622-628.

First, M.R. and Drake, L.A. 2014. Life after treatment: detecting living microorganisms following exposure to UV light and chlorine dioxide. Journal of Applied Phycology 26: 227-235

Garvey, M., B. Moriceau, and U. Passow. 2007. Applicability of the FDA assay to determine the viability of marine phytoplankton under different environmental conditions. Mar. Ecol. Prog. Ser. 352: 17-26.

Menden-Deuer, S. and E.J. Lessard. 2000. Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankon. Limnol. Oceanogr. 45:569-579.

Montagnes, D.J.A., J.A. Berges, P.J. Harrison and F.J.R. Taylor. 1994. Estimating carbon, nitrogen, protein, and chlorophyll a from cell volume in marine phytoplankton. Limnol. Oceanogr. 39:1044-1060.

Platt, T. 1985. Structure of the marine ecosystem: Its allometric basis. Can. J. Fish. Aquat. Sci. 213:55-64

Platt, T. and K.L. Denman. 1978. The structure of pelagic marine ecosystems. Rapp. P.-V. Reun. Cons. Int. Explor. Mer. 173:60-65.

88/90

0014-00007 R0

Rotman, B., and B. W. Papermaster. 1966. Membrane properties of living mammalian cells as studied by enzymatic hydrolysis of fluorogenic esters. Proc. Nat. Acad. Sci. 55: 134-141.

Sheldon, R.W., A. Prakash and W.H. Sutcliffe. 1972. The size distribution of particles in the ocean. Limnol. Oceanogr. 17:327-340.

Sheldon, R.W., W.H. Sutcliffe and M.A. Paranjae. 1977. Structure of pelagic food chain and relationship between plankton and fish production. J. Fish. Res. Board. Can. 34: 2344-2353.

USEPA. 2009. Statistical analysis of ground water monitoring data at RCRA facilities: Unified Guidance. 888p. US EPA Publication 530/R-09-007.

Vidondo, B., Y.T. Prairie, J.M. Blanco and C.M. Duarte. 1997. Some aspects of the analysis of size spectra in aquatic ecology. Limnol. Oceanogr. 42:184-192.

12.6.5 ATP – Adenosin-Triphosphate Fluorometry

Abelho, M., 2005. Extraction and quantification of ATP as a measure of microbial biomass; in “M.A.S. Graca, F.Bärlocher, M.O.Gessner (eds.) Methods to study litter decomposition: a practical guide; 223-230; Springer, Germany

Holm-Hansen, O. 1970. ATP levels in algal cells as influenced by environmental conditions; Plant Cell Physiol. Vol. 11 689-700 (1970)

Hammes, F., et al. 2010. Measurement and interpretation of microbial adenosine tri-phosphate (ATP) in aquatic environments; Water Res. (2010), DOI: 10.1016/j.watres.2010.04015

NIVA (2010) Land based testing of the CleanBallast ballast water management system of RWO. http://www.bsh.de/de/Meeresdaten/Umweltschutz/Ballastwasser/RWO/RWO_NIVA.pdf

Nyberg, H. 1989. Groth and ATP levels in Porphyridium purpureum (Rhodophyceae, Bangiales) cultured in the presence of surfactants; Brit.Phycol.J. 24, 1 9-98 (1989)

Siebel, E., et al. 2008. Correlations between total cell concentration, total adenosine tri-phosphate concentration and heterotrophic plate counts during microbial monitoring of drinking water; Drink.Water Eng.Sci. 1, 1-6 (2008)

12.6.6 FISH – Fluorescence-In-Situ-Hybridization Microscopy

Bargellini, A., et al. 2010. Inter-laboratory validation of a rapid assay for the detection and quantification of Legionella spp. In water samples; Letters Appl.Microbiol. 51, 421-427 (2010)

Energy Institute 2009. An investigation of fluorescence-in-situ-hybridization (FISH) as a routine tool to monitor sulphate reducing bacteria in oil field systems; ISBN 978 0 85293 546 0

12.6.7 Ballast Water General

Alvarez, E., et al. 2011. How to effectively sample the plankton size spectrum – a case study using FlowCam; J.Plank.Res. vol.33, 1119-1133 (2011)

Anwar, N. 2011: Ballast Water Management; Witherby Publishing Group Ltd. London, ISBN 978-1-85609-522-8

Bierman, S.M., et al. 2012. The development of a full standard methodology for testing ballast water discharges for gross non-compliance of the IMO’s ballast water management convention; EMSA/NEG/12/2012

Bogorov, B.G. 1927: Zur Methodik der Bearbeitung des Planktons. (Eine neue Kammer zur Bearbeitung des Zooplanktons) Russ.gidrobiol. Zh. 6:193-198

89/90

0014-00007 R0

Burkholder, JA. M. Et al. 2007. Phytoplankton and bacterial assemblages in ballast water of U.S. military ships as a function of port of origin, voyage time and ocean exchange practices; Harmful Algae vol.6 486-518 (2007)

Cohen, A. 2010. Non-native bacterial and viral pathogens in ballast water; Report prepared for NOAA by Center for Research on Aquatic Bioinvasions, Richmond, CA

DHI 2010. Development of guidance on how to analyze a ballast water sample; EMSA/NEG/10/2010

EMSA 2011. Testing sample representativeness of a ballast water discharge and developing methods for indicative analysis – research study; EMSA/NEG/09/2010

EMSA 2010. EMSA workshop on ballast water sampling and the development of a joint European ballast water sampling strategy – workshop report

First, M., R., Drake, L.A. 2013. Approaches for determining the effects of UV radiation on microorganisms in ballast water; Managem.Biol.Inv. vol 4, 87-99 (2013)

First, M., R., Drake, L.A. 2013. Life after treatment: detecting living organisms following exposure to UV light and chlorine dioxide; J.Appl.Phycol. DOI 10.1007/s10811-013-0049-9

GLOBALLAST 2011. Establishing equivalency in the performance testing and compliance monitoring of emerging alternative ballast water management systems; GloBallast Monograph Series No. 20, 201

Gonzales, A.G., et al. 2007. A practical guide to analytical method validation, including measurement uncertainty and accuracy profiles; Trends Anal.Chem. vol.26, 227-238 (2007)

Great Ships Initiative 2011. A ballast discharge monitoring system for great lakes relevant ships: a guidebook for researchers, ship owners and Agency officials

Harris, R., Wiebe, P., Lenz, J., Skjoldal, H.R., Huntley, M. eds. 2000: ICES Zooplankton Methodology Manual; monogr. Academic Press 2000, pp684

ICES 2004. Biological monitoring – general guidelines for quality assurance; ICES Techniques in marine environmental sciences no. 32

Matei, D. 2013. Ballast water sampling for compliance monitoring; WWF Report

Swamy, P.M. 2008: Laboratory Manual on Biotechnology; monogr. Pastogy Publications 2008, pp617

Tamburri, M. 2010. Test plan for performance evaluations of ballast water filter systems; MERC 2010

UNESCO 2010. Microscopic and molecular methods for quantitative phytoplankton analysis; IOC manuals and guides, no.55, IOC/2010/MG/55

US-California State Lands Commission 2013. 2013 Assessment of the efficacy, availability and environmental impacts of ballast water treatment systems for use in California waters

USCG 2012. Standards for living organisms in ship’s ballast water discharged in U.S. waters, Final Rule; 46 CFR Part 162

USCG 2010. Standard operating procedure for phytoplankton analysis; LG401

USCG 2008. Analysis of ballast water sampling port design using computational fluid dynamics; USCG report CG-D-01-08

USEPA 2013. Vessel general permit for discharges incidental to the normal operation of vessels (VGP)

USEPA 2010. Generic protocol for the verification of ballast water treatment technology; EPA/600/R-10/146

von Baur, M. 2013: Ballast Water Treatment Systems: Quo vadis ?; Hansa 3/2013

Wright, D.A., 2012. Logistics of compliance assessment and enforcement of the 2004 ballast water convention; J.Mar.Engin.Tech. vol. 11, 17-23 (2012)

90/90