ERRORS IN THE FISHING EFFORT SURVEY: MAINTAINING A RATIONAL RESPONSE

 

When news broke that the Fishing Effort Survey, a crucial component of the Marine Recreational Information Program’s efforts to gauge anglers’ impacts on marine fish stocks, was overestimating recreational effort, and thus also recreational catch and landings, it understandably sent a small wave of panic through folks involved in the fishery management process.

I know that I felt it myself.  If managers couldn’t rely on MRIP estimates, I thought, how were they going to credibly constrain recreational landings, which comprised the greatest source of removals in many important coastal fisheries?

It wasn’t hard to predict that the “anglers’ rights” crowd was going to use the error as an excuse to attack the fishery management process, and call for relaxed regulations that might well put the health of fish stocks at risk.  The other side of that coin was that, if managers lost MRIP, their only tool for estimating recreational landings and release mortality on a coastwide basis, the resultant management uncertainty might force them to adopt very restrictive management measures as a precautionary response to the lack of reliable data.

The good news, so far at least, is that managers seem to be finding a way between the two extremes, recognizing that, although a pilot study suggested that angler effort might be overestimated by 30 to 40 percent, we don’t know whether that initial finding reflects the true magnitude of the error.  We won’t know that until a far more extensive study, intended to run through all of 2024, reveals its findings.

Probably the first test for fishery managers came at the August 2023 meeting of the Mid-Atlantic Fishery Management Council and the Atlantic States Marine Fisheries Commission’s Summer Flounder, Scup, and Black Sea Bass Management Board, when 2024 specifications for the summer flounder, scup, and black sea bass fisheries were on the table.  A management track update of the summer flounder stock assessment had just found that the stock had experienced overfishing in 2022, and it seemed likely that representatives of the recreational fishing industry would challenge those findings, arguing that the error in the Fishing Effort Survey was enough to cast them into serious doubt.

Fortunately, before summer flounder discussions began, Michael Luisi, the Maryland fisheries manager who was chairing the meeting on behalf of the Council, managed to head off any knee-jerk responses and speculative challenges, thus helping to assure the rationality of the discussions that followed.  Noting that NMFS had just announced the Survey error the day before, Mr. Luisi advised that

“I knew that there would be people around the table, members of the public, and others who would be expressing their concerns regarding how we take the next steps based on our agenda for establishing specifications for species when we’re being told that there’s some error and some bias in the information that we are using for those purposes.  I don’t know that now is the time to get into the details about this, about the report that came out yesterday.  What I’d like to do is move forward with our agenda, get the specifications on the books and then there will be opportunity in the future for more detailed question and answer with [NMFS], and that’s going to be warranted in the future.”

While Mr. Luisi acknowledged that the 30 to 40 percent overestimate in effort suggested by the pilot study was substantial, and could impact the findings of stock assessments, he also made what might have been the most significant observation of the meeting:

“It’s not like this bias is understood.”

For those seven words hold the key to effectively navigating the Fishing Effort Survey issue.

The suggestion that angling effort is being overestimated by 30 to 40 percent came from a pilot study.  While it seems that a problem exists, the pilot study does not provide definitive information on either the extent or the magnitude of the error.  The overestimation might be restricted to just a few states, or may be a coastwide issue; the coastwide error rate, assuming that error exists, could be somewhere between 30 and 40 percent, but it could also be well over or under those values.

That being the case, those calling for immediate action of any kind are acting prematurely.  If you don’t yet understand the nature and extent of the bias in the survey, changes in current procedure, however well-intentioned, could just end up making things worse.

Thus, well-meaning calls for NMFS to do something right now to address Survey error, in order to prevent harm to fish stocks, are misguided.  When you think about it, the problem we’re facing is NMFS overestimating recreational catch, landings, and effort.  The error is actually adding a precautionary buffer to management measures—recreational removals are likely to be somewhat lower than what the MRIP estimates suggest.  So basing management measures on current MRIP estimates is unlikely to cause excessive fishing mortality accruing from the angling sector.

Those who try to use the supposed overestimates of recreational catch, landings, and effort to justify relaxing recreational regulations are equally mistaken.

After looking at the recent stock assessment update for summer flounder, which incorporated MRIP estimates (and so the presumed overestimates of recreational catch, landings, and effort), the Council and Summer Flounder, Scup, and Black Sea Bass Management Board voted to reduce 2024 commercial quotas and recreational harvest limits by about 40 percent.  In response, a recent piece in The Fisherman magazine, an outlet that has long opposed conservative fisheries management in favor of increased exploitation, commented that

“the MAFMC and ASMFC voted jointly on August 8 to reduce the recreational harvest limit on fluke by roughly 40% in 2024, which coincidentally is the same high-end disparity in fishing effort surveys coordinated by NOAA fisheries,”

and in doing so implied that such cut might not have been needed.  

The same piece in The Fishermen also tried to cast doubt on the need for the emergency regulations adopted by the ASMFC’s Atlantic Striped Bass Management Board last May, and extended at the Board’s August meeting, which imposed a 31-inch maximum size on recreationally-caught striped bass.  Such piece observed that

“ASMFC’s bombshell vote in the spring was based on recreational harvest surveys being called into question by the federal fisheries agency.  A final vote in favor of the Massachusetts emergency action was approved by a 15-1 vote, with the sole opposition coming from the state of New Jersey.  Then on August 1, the emergency 28- to 31-inch slot measures were upheld into 2024 throby ugh another vote by the ASMFC Striped Bass Management Board that passed 14-2, with both New Jersey and the District of Columbia voting in opposition…

“In terms of the ASMFC decision and the fact that NOAA Fisheries cast yes votes at the commission to continue the emergency regulations for striped bass, despite knowing about this effort and harvest announcement to come, Howell [Director of NOAA Fisheries Office of Science and Technology] said, ‘We have this in hand now, it is probably going to be part of discussions, I don’t see it going into action any time soon.  We still have to use the information that we have.’”

Again, there is implied criticism of the striped bass management measures.

But such criticism ignores a key fact.  While the pilot study suggests that the current MRIP estimates overstate recreational catch, landings, and effort, the trends identified in such estimates remain the same.  Thus, 2022 recreational striped bass landings were still nearly twice the level of landings in 2021, and still indicate a sharp increase in fishing mortality.  As NMFS noted in a recent release,

“If the agency shifts to a revised design—based on the findings of the follow-up study—the magnitude of historical estimates may change, but critical catch and effort trend information are expected to remain similar.  It’s important to note that stock status determinations are relatively consistent when trend information hasn’t changed.

“We will work closely with our partners to make informed decisions on how to proceed in light of the pilot study findings.  Until we have the full-scale study results in hand, data from the Fishing Effort Survey remains the best—and sometimes the only—available science for tracking relative year-to-year and long-term effort trends.  [emphasis added]”

Thus, while the presumably inflated estimates provided by the Fishing Effort Survey will inevitably increase management uncertainty, such estimates should still provide reasonably good guidance to managers and provide some assurance that recreational management measures based on such estimates will not cause harm to fish stocks.

Although it might seem a little counterintuitive, when higher recreational landings are input to some fish population models, they result in a higher estimate of spawning stock biomass. The easiest way to explain that is while anglers might be removing more fish from a population than previously believed, the model assumes that the population must also be larger than managers had thought, in order to absorb such removals without suffering a significant decline in fish abundance. 

After the 2018 benchmark striped bass stock assessment considered MRIP data derived from the Fishing Effort Survey, it estimated spawning stock biomass to be about 68,476 metric tons, higher than the approximately 61,000 metric ton estimate from the 2013 benchmark assessment.  Yet, unlike the 2013 benchmark assessment, the 2018 assessment found the stock to be overfished, because the MRIP data also contributed to an increase in the estimate of the spawning stock biomass target, to 114,295 metric tons, compared to 72,032 metric tons in 2013.

We saw the same thing occur in the 2019 operational stock assessment for bluefish, where inclusion of Fishing Effort Survey-derived MRIP data contributed to an increase in the biomass target from 101,343 metric tons to 198,717 metric tons.

Thus, at least for some stocks, estimates of spawning stock biomass and SSB targets will vary in rough proportion to the estimates of recreational landings, significantly reducing the impact of overestimates of recreational catch, landings, and effort on stock status assessments.

However, commercial quotas might still cause some problems.

That’s because such quotas essentially allocate a fixed percentage of the spawning stock biomass for commercial harvest, and such commercial harvest is measured, in near real time, from vessel trip reports and weigh out slips that represent the actual poundage of fish caught, and not merely estimated.  If, because of inflated MRIP estimates, a stock’s spawning stock biomass is estimated to be 30 percent higher than it actually is, then the commercial quota will be about 30 percent higher than it should be, given the actual size of the stock. 

That may be the only Fishing Effort Survey-related issue that might need more immediate remediation to assure the health of fish stocks.

Otherwise, NMFS is doing the right thing when it takes things slowly.  As Dr. Howell notes,

“The findings from this limited pilot study should not be taken as a final answer, and the results cannot be generally applied to all fisheries and fishing areas.  We have to do our due diligence in conducting a full-scale study prior to assessing the need for FES design changes or making large-scale changes to assessments or management measures.”

NMFS hopes to incorporate changes indicated by the large-scale study into the Fishing Effort Study by 2026.  In the meantime anglers, and the angling industry, should emulate the agency by approaching the issue with reason, not with emotion, and by refraining from suggesting purported fixes until we have enough data to know what aspects of the Survey require repair.