Skip to content

Advocacy Considered: Mind the Data Gap

August 17, 2011

The ocean conservation community is big on numbers, and that is because hard numbers grab people’s attention far better than any abstract argument that something is “hurting the environment.”

How bad is overfishing?  Put it in numbers.  Tell me how many more tons of fish we should have each year.  Or even better, tell me how much money we’ve lost and compare it to something else for perspective.  (Great example here)

But I think sometimes we need to take a step back and talk about the limits to our quantifications.  And maybe even our occasional abuse of those numbers, such as the abuse of shark kill statistics recently highlighted by Shelly Clarke at SeaWeb. As Clarke explains:

I almost never see any reference to the 38 million, which was after all, my best estimate.  Frequently I see “73 million” without any reference to this being my highest estimate, and almost as often I see “100 million,” an estimate that was published in Time magazine in 1997 but for which I can find no scientific basis.

As for me, I get a good view of the limits to our knowledge in my work with fisheries statistics.

For example, an oft-quoted set of numbers are the FAO’s biennial status estimate of world fisheries.  The most recent 2010 State of World Fisheries and Aquaculture noted that 32 percent of world marine fish stocks are (as of 2008) either over-exploited, depleted, or recovering, and another 53 percent of stocks are fully exploited. The statistic is frequently used to explain the need to better manage stocks, while at the same time not expecting a whole lot more from them in the future.

I generally agree with this assessment, but think the data deficiencies matter quite a bit in figuring out how bad things are and what we might expect.

The FAO tells us that there is a much we don’t know about world fish stocks:

The statistics presented in The State of World Fisheries and Aquaculture 2010 on the status of marine fisheries are frequently referred to in international policy documents and in the media to draw attention to the issue of the sustainability of the world’s fisheries.

While this information represents a unique effort to provide a global overview on the state of fishery resources, it should be noted that the stocks included in this analysis, and for which assessments are available, represent only a fraction of the total number of exploited stocks around the world…[a]lthough these assessed stocks include the largest single-species stocks and account for almost 80 percent of the total declared landings… [bold added]

This points out two big issues.  First, the evaluation of world marine fisheries is not weighted according to the size of the fishery.  So…what if the largest stocks are doing really well?  You won’t see that, and it’d mean that we might not be doing so poorly from a global biomass perspective.

Second, what about the remaining 20 percent?  If we assume an even distribution of stock statuses, and assume that the unknown stocks are doing just fine, our numbers drop to 26 percent over-exploited, depleted, or recovering, and 42 percent full-exploited.  Not great, but not as bad.  (Though, yes, if they were doing poorly, we’d have an even stronger argument for fisheries reform.)

Another problem with FAO catch data it is only as good as those doing the reporting.  As Pauly et al point out, a great deal of fish go unreported due to discarding at sea and IUU fishing.  They suggest that world catches actually look more like this:

Notice the big bump up when discards and IUU are included?  And another estimate suggests that as many as 25% of the world catch is illegal and unreported.

Meanwhile, attempts to get a better handle on world fisheries using other data sets are also challenged by data deficiencies.  Branch et al rightly argue that reported catches alone are a poor proxy for the state of fisheries.  But at the same time, their proposed better metric – biomass assessments – is even more data poor than catch statistics.  They share:

We obtained stock assessments from the RAM Legacy database. The extracted data contained information on 234 stocks of 124 species; catches from these stocks summed to 17-25% of global catches in each year from 1970 through 2006.  We obtained estimates of Bmsy either directly from stock assessments (n=126) or by fitting a Schaefer model (logistic population growth) to time series of catch and biomass data (n=108), as described elsewhere. [bold added]

I think the big takeaway here is that we have to acknowledge that uncertainty is still the dominant condition.  If we start thinking we have the full picture, then we risk losing our ability to think outside the box and maybe even verge on the ideological.

Do I see this uncertainty as a problem in advocacy?  Not really.  After all, it cuts both ways.  The proponents of maximum/optimum sustainable yield face the same problems in getting the numbers right for their exploitation.  This just further enforces that we need a system based not on optimization, but risk minimization.  As the story of cod tells us, optimizing with imperfect information can have terrible results.

Branch TA, Jensen OP, Ricard D, Ye Y, & Hilborn R (2011). Contrasting Global Trends in Marine Fishery Status Obtained from Catches and from Stock Assessments. Conservation biology : the journal of the Society for Conservation Biology, 25 (4), 777-786 PMID: 21535149

Pauly D, Christensen V, Guénette S, Pitcher TJ, Sumaila UR, Walters CJ, Watson R, & Zeller D (2002). Towards sustainability in world fisheries. Nature, 418 (6898), 689-95 PMID: 12167876

2 Comments leave one →
  1. anthony permalink
    August 17, 2011 12:03 pm

    if fish depletion is anything like the american healthcare system, public interest advocates would be smart to publish only their most conservative estimates. it is counterproductive to embellish, when even the best-case numbers can almost always help make your point.

    “when scientists disagree, everyone else stops listening.”

    • August 17, 2011 12:29 pm

      “it is counterproductive to embellish, when even the best-case numbers can almost always help make your point.” Great point.

Leave a Reply to Mark Gibson Cancel reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: