Conflict Of Interest, Data and Funding Hiding, Flimsy and Misleading Claims…

…is that nowadays de rigueur for the average alarmist paper or what?

(the following is a version of a letter recently sent by Walter Starck to PNAS; republished with the author’s consent)

19 March 2010

To:
Randy Schekman
Editor-in-Chief
PNAS
pnas@nas.edu

From:
Walter Starck
Australia

Re:
McCook, L.J., et al. 2010. Marine Reserves Special Feature: Adaptive management of the Great Barrier Reef. PNAS 2010 : 0909335107v1-200909335.

The above referenced study presents a number of concerns:

• The most serious concern is a major conflict of interest involving all of the 21 authors.

It should be noted that the lead author is employed by the Great Barrier Reef (GBR) Marine Park Authority (GBRMPA) and all of the 20 additional authors are either employed by them or are recipients of substantial funding from them. It is incongruous in the extreme that all these employees and repeated recipients of generous GBRMPA funding, could, “…declare no conflict of interest” when they are in fact assessing the benefits of their own work and that of the organisation which supports them.

Combined with the rather unrestrained positive spin on the benefits and cost effectiveness achieved by GBRMPA management, the appearance of this report is that of a promotion piece which the most productive and respected beneficiaries of their research funding have been invited to endorse. In such case, it would have been very difficult for them to decline or to offer much objection to the claims made. At the same time, their names and status would provide credibility and deterrence of criticism while greatly increasing the prospect of acceptance for publication in a prestigious journal such as PNAS.

• In addition, on PNAS “Authors must acknowledge all funding sources supporting the work”. There appears to be no such disclosure in this study.

• PNAS papers must also, “…make materials, data, and associated protocols available to readers

McCook et al. state that, “Another important observation emerging from this review is the extent of relevant data that are not published or readily accessible. A full picture of the effects and effectiveness of zoning on the GBR has required extensive use of gray literature, previously unpublished data, and collation of separate data sources”. GBRMPA has been the sponsor of most of the research cited and, through the permit system, they exercise control over the terms of all other research conducted there. They are also a major publisher of GBR literature, both scientific and non-technical. The extent to which relevant data is not published or readily accessible is their direct responsibility.

As the data referred to for this review has obviously been assembled, why has it not been made available?

• The major claim of a doubling of fish on protected reefs appears to rest on a single example.

This is inconsistent with abundant other evidence including that which is presented in the report itself. Only one reef area of the 8 featured in the report showed a 2-fold increase and that area had the lowest level to begin and lowest difference between fished and unfished reefs. In 5 of the 8 areas featured in the report the protected reefs actually showed a decline in coral trout numbers. On fished reefs, three areas showed increases in biomass while 5 showed declines. This is hardly the “extraordinary” 2-fold increase in protected areas being bannered.

• McCook et al. state, “The economic value of a healthy GBR to Australia is enormous, currently estimated to be about A$5.5 billion annually….Relative to the revenue generated by reef tourism, current expenditure on protection is minor…Tourism accounts for the vast majority of reef-based income and employment. …income from tourism is estimated to be about 36 times greater than commercial fishing“. These claims are highly misleading.

The economic value cited includes the total value for all tourism in the region when half of all tourists do not even visit the reef. For those who do, the reef component of the large majority is a one day, one time participation in a reef tour and the value of reef tours is similar to the value of commercial fishing. If one also considers the economic value of recreational fishing, retail fish sales and seafood meals in restaurants, the total value of fishing is closer to twice that of reef tours.

In addition, the reef tour industry regularly uses only about 2 dozen out of the 2500 reefs of the GBR and, on those which are used, the actual area visited would only be about 1% of the area of even those reefs. Unfished reefs to optimize scenic value for tourism could easily coexist with an order of magnitude greater fishing effort, and no detriment at all to tourism. The attribution of total tourism value to the reef is no more justifiable than attributing it to the similar numbers who visit the rainforest or who eat seafood meals while visiting the region.

Such claims have been repeatedly made by GBRMPA and would, if used by a business, constitute violations of advertising and corporate law. To see it done repeatedly and included in a report in a leading scientific journal is a sad indictment of GBRMPA sponsored science as well as basic honesty.

• Babcock et al., 2010 (in another study published in PNAS on the same day as McCook et al.) also examined the ecological effects of marine protected areas. However, this report is much more widely based geographically and longer term.

Although the observed effects were generally positive, they were decidedly less large, rapid, extensive, and uniformly positive than those reported for the GBR. All of them also involved areas subject to much greater fishing pressure than the GBR. One might reasonably expect that increased protection for the least impacted areas would result in a less marked beneficial effect rather than the much more widespread rapid and dramatic benefits claimed by McCook et al.

For example, Babcock et al., “…found that the time to initial detection of direct effects on target species … was 5.13 ± 1.9 years…”. Note that this was the time to initial detection, not the even longer time required to reach a doubling of population. When compared to the much greater effects claimed for the GBR over two years, the latter do indeed appear to be “extraordinary”.

• Various key claims are contradicted by other more extensive work by the same researchers with no acknowledgment or discussion of this.

In reading over McCook et al., some 40 such discrepancies were noted and more detailed examination would surely reveal more. However, without going further it should be clear that PNAS has been badly used. The serious and obvious conflict of interest alone can neither be ignored nor credibly explained away. If not addressed, it makes a farce of the declaration of no conflict. It alone must surely be more than sufficient grounds to retract this study. Although doing this may be unpleasant it would be far less damaging than to try to examine and defend all of the sad and disreputable details.

Coming at a time when public credibility in science is being seriously eroded by ongoing revelations of malpractice in what the public was assured was irrefutable fact and settled science regarding climate change, these “extraordinary” (their own description) claims regarding the GBR are well positioned to become a “Reefgate”. This is especially so in that a key claim in this report and widely made elsewhere, is that a major benefit of protected areas on reefs is the increased resilience they provide against climate change.

Although controversy regarding the management of the GBR may appear of minor public interest from a U.S. perspective, it will be national news here in Australia and PNAS could find itself very much involved in a most difficult to defend position should prompt and decisive action not be taken.

A public release on all this will be made here in the near future. Whatever the decision of PNAS, it would be better made sooner than later.

Sincerely,

Walter Starck