Statement: Marine Institute’s rebuttal of research criticisms fails to offer answers or insights
12 March 2014
Dr Jackson of the Marine Institute has issued a deceptive rebuttal of the critique of his work published in the latest issue of the Journal of Fish Disease.1, 2
An immediate and emotive public statement was released by Dr Peter Heffernan, CEO, Marine Institute, noting ‘I welcome the acknowledgement by the ‘Journal of Fish Diseases’ for their error in publishing as a ‘Short Communication’ the piece by Krkosek et al. claiming ‘fundamental errors’ in Marine Institute science.’ 3
Both the rebuttal and statement from Dr Heffernan deserve a closer examination. For at no stage did the Journal apologise for anything more than an administrative error. Nor does the rebuttal actually answer the key issue at stake – do sea lice from salmon farms have an impact on salmon conservation.
A question of statistics
To start, in Jackson’s rebuttal1 there is detailed discussion of the validity of the statistical methodologies used by his team and those used by Krkosek’s team who reanalysed his data and criticised his work.2
Both approaches have been previously used in published peer reviewed papers examining the impact of sea lice on salmon populations, and both can be deemed valid statistical methods. But, what this ongoing argument fails to address is did the statistical analysis answer the relevant question?
Jackson et al.’s original study focuses on the impacts of sea lice as a percentage of marine mortality rates.4 Given natural marine mortality is in the region of 90-95% (as salmon are subject to predators, disease, pollution, starvation and so on) examining additional impact of sea lice offers little insight into the burden of these parasites on numbers of adult salmon returning to sustain future generations.
For example, if, in the absence of sea lice 6% of migrating smolts return as adult salmon, then the effect of sea lice mortality reduces that return rate to 4% – it would be fair to say sea lice are reducing return rates of adult salmon by 33%. However, if Jackson’s approach is followed, the impact of sea lice is only 2% of overall mortality rates, as sea lice increase the mortality rate from 94% to 96%. This misses the point. Such a figure offers no insight into whether or not sea lice impact the salmon being relied upon to spawn the next generation. It was this issue Krkošek, et al. wished to clarify in his critique.2
In their rebuttal, Dr Jackson’s team seem to deliberately avoid this fundamental point. Instead they bamboozle the reader with statistics.1
As a result, it could be concluded that the Marine Institute is only concerned with maintaining a reputation for using recognised mathematical methods. And, that they do not wish to offer Ministers, who must make decisions based on their data, a greater understanding of the sea lice problem.
Norway’s similar results but differing conclusions
The Marine Institute go on, both in their rebuttal and statement, to claim a Norwegian study led by Dr Skilbrei, found “results are similar to those of Jackson et al.’ 1,3,5
While, it is true there are similarities in the results, it isn’t true they drew the same conclusions.
Jackson’s study reported an odds ratio 1.14:1 between return rates of treated and untreated smolts; while Skilbrei and his Norwegian team reported an odds ratio of 1:1.17. What these figures equate to is a decrease in returns in untreated smolts, due to sea lice of 14% and 17% respectively.
However, rather than drawing conclusions based on these figures Jackson instead focussed on his findings that ‘the level of sea lice-induced mortality is small as a proportion of the overall marine mortality rate, which is in the region of 90%, and in absolute terms represents 1%’. His papers summarises ‘…while sea lice-induced mortality on outwardly migrating smolts can be significant, it is a minor and irregular component of marine mortality in the stocks studied and is unlikely to be a significant factor influencing conservation status of salmon stocks.’ 4
In contrast Skilbrei’s team drew the conclusion that ‘…salmon lice appeared to impose an average additional marine mortality of ~ 17% (odds ratio of 1.17 for recapture of treated/control fish). According to the considerations by Norwegian expert groups aiming to quantify the impact of salmon lice, this level of influence would be expected to represent a moderate regulatory effect on a salmon population.’5
So while Jackson suggested sea lice had little, if any, impact on salmon conservation, the Norwegians did quite the opposite stating not just a mild, but a moderate effect. Hardly a ‘similar result’.
When you examine the year on year impact of a 17% decrease in salmon returns, it becomes obvious why such a conclusion was drawn, for salmon populations could be effectively wiped out in less than two decades. The same can be said of a 14% decrease in returns, as reported by Jackson: 4
In another bid to dupe readers, Dr Heffernan states that he ‘welcomes the acknowledgement by the ‘Journal of Fish Diseases’ for their error in publishing as a ‘Short Communication’ the piece by Krkosek et al. claiming ‘fundamental errors’ in Marine Institute science.’ 3
Such a statement suggests that the Journal is apologising for publishing Krkosek’s work altogether. However, the Journal’s actual apology reads:
‘Due to a procedural error, Dr Jackson et al. were not given the opportunity to reply to this Comment before it was published on Early View. Wiley apologises to Dr Jackson et al. for this oversight. The Editors wish to clarify that Comments are not subject to the same level of peer review as Original Articles and Review Papers. Dr Jackson et al.’s response will be published alongside this Comment in the next available issue of Journal of Fish Diseases.’
As can be seen the Journal apologises only for the administrative error of not giving Dr Jackson’s team the opportunity to respond in the same ‘Early View’ issue as the comment was first published. Nothing more. Indeed, the apology actually notes Krkosek’s comment will be published once more alongside Jackson’s response.
Dr Heffernan’s statement goes on to suggest that the critique was also down-graded to a ‘comment’. 3 A curious claim given its title upon publication in August 2013 was ‘Comment on Jackson et al…’ and has not changed since. The Journal’s doesn’t suggest any reclassification either. It simply confirms ‘comments’ are not subject to the same level of peer review, which is standard practice for opinion style articles that are not original research.
The apology for a ‘procedural error’ from Wiley publishers appears to have been exaggerated by the Marine Institute at many levels. This begs the question of why they feel it is necessary to do this? Do they not believe their science can stand up for itself?
A good reason to be concerned
The fact is, the Marine Institute have good reason to be concerned. Today, the European Commission is determining whether or not to reopen investigations into the impact of sea lice. The decision to previously close the case was based on this very research data.
Now they are having to reconsider that decision, and will be closely scrutinising Dr Jackson and the Marine Institute’s work. When doing so, the European Commission will examine all the data now available including Krkosek’s criticism and re-analysis of Jackson’s data, and studies published by Dr Gargan’s team at Inland Fisheries Ireland.6
Interestingly these did report similar findings with Krkosek’s reanalysis of the Jackson data concluding salmon returns would drop by 34%,2 and Gargan reporting a 39% in the Inland Fisheries Ireland study.6 Again, if the cumulative impact of Gargan’s study is examined, it can be seen the risk is salmon are wiped out in less than a decade:
All in all, it seems the scientists are actually in agreement – sea lice do indeed have an impact the return rates of salmon. The only questions remaining are the scale of impact and the speed at which they destroy local salmon populations. Will it be a decade or two?
Dr Jackson’s changing position
Adding a final twist to this tale is Dr Jackson split opinions on this subject. While the Marine Institute are now standing firmly behind his science claiming sea lice have little impact on nearby salmon populations, less than two years ago Dr Jackson co-authored a paper that drew quite a different conclusion.7
The summary of the paper stated ‘‘Salmon lice originating from farms negatively impact wild stocks of salmonids, although the extent of the impact is a matter of debate… This is considered to have a moderate population regulatory effect.’
It went on to conclude ‘For the foreseeable future, salmon lice will continue to be a serious problem for the salmon farming industry and a threat to their environmental credibility.’ 7
A point on which few disagree.
Secretary, Save Bantry Bay, Alec O’Donovan, 087 7949227 (mobile) or 027 50508
Chair, Save Bantry Bay, Kieran O’Shea, 086 1280303 (mobile) or 027 60121
1. Jackson, D., et al. “Response to M Krkosek, CW Revie, B Finstad and CD Todd’s comment on Jackson et al.‘Impact of Lepeophtheirus salmonis infestations on migrating Atlantic salmon, Salmo salar L., smolts at eight locations in Ireland with an analysis of lice‐induced marine mortality’.” Journal of Fish Diseases (2014).
2. Krkoˇsek, M., et al. “Comment on Jackson et al.‘Impact of Lepeophtheirus salmonis infestations on migrating Atlantic salmon, Salmo salar L., smolts at eight locations in Ireland with an analysis of lice-induced marine mortality’.” Journal of fish diseases (2013).
3. Marine Institute stands over quality and accuracy of its research as the Journal of Fish Diseases acknowledges errors, Statement by Dr. Peter Heffernan, CEO, Marine Institute. Feb 2014. http://www.marine.ie/home/aboutus/newsroom/pressreleases/MarineInstitutestandsoverqualityandaccuracyofitsresearchastheJournalofFishDiseasesacknowledgeserrors.htm
4. Jackson, D., et al. “Impact of Lepeophtheirus salmonis infestations on migrating Atlantic salmon, Salmo salar L., smolts at eight locations in Ireland with an analysis of lice‐induced marine mortality.” Journal of fish diseases 36.3 (2013): 273-281.
5. Skilbrei, O. T., et al. “Impact of early salmon louse, Lepeophtheirus salmonis, infestation and differences in survival and marine growth of sea‐ranched Atlantic salmon, Salmo salar L., smolts 1997–2009.” Journal of fish diseases 36.3 (2013): 249-260.
6. Gargan, P. G., et al. “Evidence for sea lice-induced marine mortality of Atlantic salmon (Salmo salar) in western Ireland from experimental releases of ranched smolts treated with emamectin benzoate.” Canadian Journal of Fisheries and Aquatic Sciences 69.2 (2012): 343-353.
7. Torrissen, O., et al. “Salmon lice–impact on wild salmonids and salmon aquaculture.” Journal of fish diseases (2012).