PPR 5: IBIA proposes way ahead on sulphur testing and compliance verification

Two papers submitted by IBIA will be taken into consideration as the International Maritime Organization (IMO) works on a framework to ensure consistent implementation of the 0.50% sulphur limit taking effect on 1 January, 2020. The two papers, submitted to the 5th session of the Sub-Committee on Pollution Prevention and Response (PPR 5), propose a way ahead on sulphur testing and compliance verification. Both papers will be forwarded to a PPR intersessional meeting in July for consideration.

IBIA introduced the papers to PPR 5 as follows:

QUOTE
PPR 5/13/9:
We have been tasked with ensuring consistent application of Regulation 14.1.3 of MARPOL Annex VI. As we know, regulation 14 defines sulphur limits for fuel oils as % by mass with two decimal points.

What is lacking, however, is a clear definition of how sulphur content in fuel oil should be measured. This is absolutely fundamental to ensuring a consistent approach to determining whether applicable limits have been met.

We observe that the approach to testing and reporting of test analysis results and determination of non-compliance is not consistent with regards to in-use samples from ships, with examples heard of test results being reported with five decimals causing a ship to be deemed non-compliant even if that result, when rounded off to 2 decimals, would meet the regulatory limit.

Even for the MARPOL sample, although we have a sulphur verification procedure, there is no clear guidance with regards to which test method should apply and how test results should be reported.

This leaves room for ambiguity and uncertainty for stakeholders as to how analysis and reporting of sulphur test results will be applied and carries a risk that what is considered compliant by one Administration will be considered as non-compliant by another.

We believe the final and binding analysis should be carried out by using the same reference test method for all types of samples being tested for MARPOL compliance. This would ensure that analysis and test result reporting protocol is uniform and applied consistently in all jurisdictions. This does not preclude the use of other test methods for initial inspections, but where those results indicate noncompliance, it should be verified in an accredited laboratory using the reference test method.

Our document, PPR 5/13/9, therefore proposes adding a definition of sulphur content in regulation 2 of MARPOL Annex VI, using the most suitable test method.

We believe this should be ISO 8754, which is already referenced in Appendix V of MARPOL Annex VI regarding which test method to use to determine the sulphur content of fuel oil delivered to ships.

Adopting this solution would bring the clarity needed to ensure consistent implementation of Regulation 14.1.3 of MARPOL Annex VI.

PPR 5/12/1:
Distinguished delegates, our paper, PPR 5/12/1, proposes establishing appropriate guidelines for verifying the sulphur content in fuel oil samples taken from ships fuel systems (in-use samples). There is no doubt that having clear guidelines for this is important already today, as we know the majority of testing of fuel oil samples to verify MARPOL compliance in established emission control areas is done on such samples. As of 2020, a robust implementation of the 0.50% sulphur limit globally will likely lead to even more testing of in-use samples.

At MEPC 71, the Committee discussed a proposal to apply the same sulphur verification procedure to in-use samples as the one found in Appendix VI to MARPOL Annex VI to verify the sulphur content in the representative fuel oil sample taken at the time of delivery, known as the MARPOL sample.

Ladies and gentlemen, this may seem like a convenient approach, but our paper demonstrates why this can be flawed, and carries a risk of unfairly penalising ships on the basis of unsound evidence, also bearing in mind that the method of sampling is different between in-use samples and  those drawn at the bunker manifold.

Let me summarise some key points:

1. Every test method has an inherent uncertainty, which has been statistically and scientifically defined, but recognises that the actual “true value” can never be established. In layman’s terms, if a specific sample is tested in 100 different laboratories using the same test method, 95 of the test results would fall within a certain range that would be evenly distributed around an average value – this range is the 95% confidence level. For a single test result against a specific limit, if it falls within this range, you can say with 95% confidence that it has met that limit. The method prescribed in appendix VI of MARPOL Annex VI has not fully encompassed this principle. By extension, this method therefore risks an unfair indictment of the ship for non-compliance based on the average of just four test results when another set of test results, or a further four, six, ten or even 100 test results may have given a compliant final average result.
2. It should be noted that 95% confidence is a recognised principle in other IMO instruments, such as testing ballast water for compliance with applicable standards. It is also applied when testing against sulphur limits in road fuels throughout the European Union and elsewhere, and for marine bunkers on a commercial basis.
3. Accepting that a fuel oil meets the sulphur limit based on a single test result which is marginally above the limit, but within 95% confidence, does not weaken the regulation. It merely recognises the scientifically proven limitations of the test method. In fact, the sulphur verification method in Appendix VI for the MARPOL sample effectively shifts the actual limit to slightly below the statutory limit by not accepting the 95% confidence level.

With this in mind, and also keeping in mind that we are likely to see even more testing of in-use fuel oil samples from 2020, we propose that the guidelines for verifying sulphur content of in-use samples should be different from the method prescribed for the MARPOL sample, in order to reduce the risk of ships being subject to statistics randomly working against them, and also reduce cost for administrations to perform testing of in-use samples to ensure consistent implementation of sulphur limits. We have provided concrete proposals in paragraph 14 and 15 of our paper for the sub-committees consideration.
UNQUOTE

The proposal in paragraph 14 and 15 of PPR 5/12/1 reads as follows:

14. Based on the observations above, IBIA proposes the establishment of appropriate guidelines for verifying the sulphur content in fuel oil samples taken from ships (in-use samples). The guidelines should specify the use of the ISO 8754 test method and recognize the 95% confidence principle.

15. This could be achieved by applying the first stage of the verification procedure described in appendix VI for in-use samples, accepting the average of the first two test results as compliant as long as it does not exceed the limit plus 0.59 x R. Alternatively, there could be a separate guideline for sulphur verification of in-use samples based on the principles applying to ship samples in ISO 4259 for a single test result.

Report by Unni Einemo
unni@ibia.net