In our latest feature on Gulf snapper, we discuss the misguided efforts by lobbyists and congress instructing the National Marine Fisheries Service (NMFS) to count the number of Gulf snapper. Counting fish (stock assessment) is an important part of fisheries management, but the approach taken by congress is misinformed. Perhaps the most famous and important work on stock assessment, Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty, was written by Ray Hilborn and Carl Walters in 1992 and has become the international standard for approaching a fish counting task – it is the 9th most cited fisheries-related work ever published according to Trevor Branch’s running list (worth a browse). Here, Carl Walters gives his thoughts on the Gulf snapper counting issue.

Comment by Carl Walters, University of British Columbia.

“Can red snapper fishermen expect increases in allowable catch if an absolute abundance estimate is obtained?”

A key general point about managing fisheries sustainably is that one of the most critical needs is not the stock size, but rather the fishing mortality rate: which is the ratio of catch to stock size, i.e the relative impact of fishing. We can compare this rate to estimates of the rate expected to give highest yield, and estimation of that “optimum” rate in turn depends, not on stock size, but on estimates of natural mortality rate, growth, recruitment, and size-age vulnerability.

For most stock assessments like red snapper, we do not obtain estimates of the fishing mortality rate by just dividing catch by measured stock size. Instead, we work backward by using size-age composition and abundance trend data to estimate trends in the fishing rate, then basically dividing the catches by these fishing rate estimates to get historical stock size estimates.

I suspect that fishermen hope that the assessment calculations have resulted on overestimates of fishing mortality rate, perhaps due in part to lack of survey trend data on oil platform aggregations, and that the rate will be corrected downward if/when a direct stock estimate shows the stock size to be larger. There is in fact reason to distrust the mortality rate estimates from size-age composition data, because we are uncertain about the natural mortality rate component of decline in catches at age, and because snappers have what we call “dome-shaped vulnerability” meaning that older fish that have moved into deeper water and were at least historically less vulnerable to fishing since only longlining caught them.

If I were a fisherman, I would not get my hopes up much about a total biomass survey resulting in any substantial increase in allowable catch, and I certainly would not be willing to invest much in paying for such a survey if the costs were to be borne by my fishing community. There are two reasons for this. First there is a good chance that there will not be substantial upward revision in stock estimates, because catches from places like oil platforms (and presumably size-age data too) are already being included in the assessment estimates, and will change the estimates only if there has been a really big difference in abundance trends between surveyed areas and areas that have not been surveyed. Second, if abundance estimates are revised upward, these corrections will also be applied to historical data in order to re-estimate historical recruitments and productivity. Those corrections will very likely result in downward corrections in the productivity estimates (if the stock was historically larger, it must have been less productive in order to have exhibited historical patterns of decline).

Where I would spend a large funding windfall would be on three other things: (1) getting better information on how snappers move, particularly the onshore-offshore movement with age and movement into and away from aggregation sites, both natural and oil platform ones; (2) getting better estimates of natural mortality rates; and (3) getting better estimates of the net effect on how much survival to recruitment really has been affected by shrimp trawl bycatch (this is controversial because there may be higher natural mortality rates after bycatch reduction). One really neat way of getting at all of these things would be to use the latest neat idea in fish tagging, namely “gene tags” where fish are tagged in-situ (without bringing them onboard) by collecting a bit of flesh and doing a unique DNA fingerprint for each fish; there then needs to be a large (and relatively expensive) program to resample fish at sea and the catch for these fingerprinted individuals. I understand from colleagues in the region that just such a program will be proposed, and I strongly urge fishermen to support it.

Carl Walters is a Professor Emeritus at the Institute for the Oceans and Fisheries at the University of British Columbia. His area of expertise includes fisheries assessment and sustainable management and has used that expertise to advise public agencies and industrial groups on fisheries assessment and management. He is a member of the Royal Society of Canada and received the Volvo Environmental Prize in 2005. He has been a member of a number of NSERC grant committees since 1970, and received the AIFRB Award for Outstanding Individual Achievement in 2011. Walters is considered the ‘father’ of adaptive management.