Thursday, February 9, 2017

MANAGING MANAGEMENT UNCERTAINTY

Fisheries management is not an exact science.
Although biologists are constantly refining their stock assessments and population models, there are still many sources of uncertainty.
Some of them fall into the category of “scientific uncertainty.” Such scientific uncertainty applies to all of the important parameters, such as stock size and the number of young fish recruited into the stock each year. Biologists deal with such uncertainty by specifying a “confidence interval,” stated as a probability that the actual stock size, recruitment figure or other parameter is included in a particular range of values.
The summer flounder stock assessment provides a good example. Biologists have a lot of good data available to them, and the assessment is updated each year, as new information becomes available. Even so, the confidence intervals used in summer flounder management are fairly broad.

Thus, while the Summer Flounder Stock Assessment Update for 2016 estimated the spawning stock biomass to be 36,240 metric tons (mt), it also recognized the uncertainty in that estimate, stating that “The 90% confidence interval for [spawning stock biomass] in 2015 was 32,605 to 44,425 mt,” meaning that there is a 90% probability that the actual biomass fell somewhere between those two numbers.

Obviously, management measures that are appropriate for a stock of 36,000 mt are a little more restrictive than they would need to be if the stock size is closer to 44,000 mt, but not restrictive enough if the stock size be closer to 32,000 mt. However, such measures would represent a statistically defensible way to strike a balance between those two extremes.
Because some degree of scientific uncertainty is inevitable, fisheries managers aren’t permitted to set harvest quotas at the overfishing limit (OFL), the threshold where overfishing begins. Instead, the Science and Statistics Committee (SSC) of each regional fishery management council must address scientific uncertainty by establishing an allowable biological catch (ABC), an upper limit on landings which represents the OFL, reduced by a buffer that allows for such uncertainty.

In the case of 2017 summer flounder harvest, such buffer reduced a 7,600 mt OFL to a 5,125 mt ABC, making it far less likely that overfishing will occur.
Scientific uncertainty is only half of the problem that managers face when setting annual catch limits (ACLs); they must account for management uncertainty as well. Unlike scientific uncertainty, which must be considered when setting ABCs, management uncertainty need not be considered at all.

Management uncertainty can arise in a number of ways. Harvest might not be accurately calculated; that’s a particular problem in recreational fisheries, where harvest is estimated based on a survey, and less of a problem in commercial fisheries, where landings must be reported. Managers can never be completely certain that the regulations that they adopt in any given year will prevent overfishing. Even if the regulations opposed are adequate, there is no guarantee that all fishermen will obey them.
Thus, management uncertainty can have a real impact on the success of management measures. However, because accounting for uncertainty will inevitably reduce ACLs, regional fishery management councils are often reluctant to do so.
That became a particular problem for the Gulf of Mexico Fishery Management Council (Gulf Council) a few years ago, which couldn’t keep recreational fishermen from exceeding their ACL for red snapper.
Anglers caught too many red snapper in every year between 2008 and 2013, with the exception of 2010, when the catastrophic oil spill that occurred after BP’s Deepwater Horizon oil well blew up kept fishermen off the water.
The Gulf Council admitted that “management uncertainty was high for the recreational sector,” but refused to adopt the 20% buffer recommended by its SSC. It argued that a new benchmark stock assessment and better methods of gathering recreational data rendered such a large buffer unnecessary.

A group of commercial fishermen, concerned about the damage that constant recreational overharvest was doing to the stock, sued NMFS in federal court, seeking a decision that would compel the agency to hold anglers accountable for their repeated overages.
The commercial fishermen won the lawsuit.
The court that decided the case, Guindon v. Pritzker, said that accountability measures (AMs) “can and should be used to address management uncertainty. NMFS must disapprove and return for revision any Council proposal that does not contain adequate AMs.”

It quoted a section of the administrative record which stated that “Considerable uncertainty exists in projecting season estimates given variability in average weights, catch per day, and implementation of incompatible state regulations,” and ultimately found that “NMFS had a statutory duty to…require whatever accountability measures are necessary to constrain catch to the quota.”
NMFS responded by implementing the 20% buffer that the Gulf Council had previously rejected, in order to avoid further recreational overfishing.
Now, management uncertainty may be undermining the effort to manage summer flounder.
Everyone admits that there is a serious management uncertainty issue in the recreational summer flounder fishery. The issue was well-described in a letter sent by James Gilmore, Director of the New York State Department of Environmental Conservation’s Marine Division, to the Chair of the Atlantic States Marine Fisheries Commission’s Summer Flounder, Scup and Black Sea Bass Management Board (Management Board):

“There seems to be a poor relationship between the recreational measures (derived from calculations based on the [Marine Recreational Information Program]) and the performance (as estimated by [such program]). Regional summer flounder recreational management has been in place for the last 3 years (2014-2016). During this period regulations adopted by each region have not changed with the exception of NJ’s portion of Delaware Bay…Under consistent measures there are numerous factors that may influence recreational harvest in a state, with weather and fish availability to anglers among the most important. Harvest estimates are in turn influenced by the actual magnitude of the harvest and the variability inherent in a survey…Under 3 years of consistent regulations from 2014-2016, coastwide harvest estimates in numbers of fish have ranged from 1.6-2.5 million fish, varying as much as 50% between years. When we consider a smaller geographic scale, this variability increases to 66% between years in the CT-NJ region, and an average of 139% at the individual state level. It is difficult to say how much of this variability is due to estimation vs. actual harvest magnitude.”
Given such a high degree of management uncertainty, and considering the court’s decision in Guindon v. Pritzker, one might expect NMFS to establish a recreational ACL somewhat lower than the ABC, in order to create a buffer against such uncertainty.
However, in materials provided to the Mid-Atlantic Fishery Management Council (MAFMC) and Management Board, the MAFMC’s Summer Flounder, Scup and Black Sea Bass Monitoring Committee advised that “The Monitoring Committee recommends no reduction from the commercial ACLs to the [Annual Catch Targets] in 2017-2018 to address management uncertainty…The Committee also recommends no reduction from the recreational ACLs to the recreational [Annual Catch Targets] for 2017-2018…”

The lack of a buffer for management uncertainty helped anglers in 2016. Because angling effort was probably underestimated in 2015,harvest estimates for that year were surprisingly low, which allowed managers to keep regulations unchanged in 2016, even though the ACL was reduced by 29%.

However, when effort returned to normal levels in 2016, such regulations did not adequately constrain recreational landings. As a result,anglers overfished, and so will be forced to reduce landings by 40% in 2017, even though the ACL will only decline by 30%.

Such a large one-year reduction has caused a lot of anger in the recreational fishing community, hostility that might have been avoided had managers been willing to admit that the 2015 harvest estimates were probably too low, and incorporated a buffer for management uncertainty into the 2016 regulations.

On the other hand, had such a buffer been adopted for 2016, many members of the recreational fishing community probably would have protested that…
That’s why it’s difficult to properly account for management uncertainty. Fishery managers know that it exists, at some level, in every recreational fishery; they also know that, eventually, it will probably result in anglers overfishing their ACL.
However, because accounting for management uncertainty leads to smaller recreational ACLs, which in turn leads to unhappy recreational fishermen, managers are reluctant to put such constraints on landings even when, as was true with red snapper, anglers regularly overfish their ACLs.
A court decision, such as Guindon v. Pritzker, can force managers’ hands, but managers shouldn’t have to be forced to take needed actions. By incorporating a buffer for management uncertainty into each year’s ACLs, fishery managers better assure that regulations will be consistent from year to year.
Consistent regulations, even if they result in somewhat smaller ACLs, tend to strengthen anglers’ faith in fisheries management, while regulations that change every year, without seeming cause or direction, lead anglers to question managers’ decisions.
Since fishermen are far more likely to obey regulations that they believe in, properly accounting for management uncertainty is an important, yet too often ignored, part of the management process.
-----


This essay first appeared in “From the Waterfront,” the blog of the Marine Fish Conservation Network, which can be found at http://conservefish.org/blog/

No comments:

Post a Comment