Confusing Nemo (3): Enjoy your fish in 2049

Next in my little fact-checking exercise of The Black Fish's movie Losing Nemo:

If the fishing industry keeps fishing at its current rate, science predicts that all fish will be gone by 2048

This statement is based on a paper in Science (see a copy of the article here) by marine scientist Boris Worm, published in November 2006. Intriguingly, the objective of the article is not to predict anything like stock collapse: rather, it investigates the importance of biodiversity for the provision of a host of marine ecosystem services, such as fisheries, nursing juveniles of marine species, and filtering of waste from human sources. The article finds a crude but nevertheless convincing positive correlation between species richness and such traits as productivity and speed of recovery from overfishing: in other words, species-rich ecosystems produce more biomass than species-poor ecosystems, and also recover more easily from overexploitation. So is that where the 2048 prediction comes from, a projection of biodiversity decline that should lead to worldwide stock collapse around 2048?

Actually not. One of the driving forces considered in the article is fishing pressure, so it describes how more and more fish stocks have collapsed (defined as catches dropping below 10% of the highest catch ever recorded) since 1950, according to data from the Sea Around Us project. Based on these data the authors estimate a mathematical formula to describe this trend:

y = 0.0168*x^1.8992

where y denotes the percentage of fish species currently collapsed and x denotes the number of years after 1950. Later in the article the authors make the following remark:

This trend is of serious concern because it projects the global collapse of all taxa currently fished by the mid–21st century (based on the extrapolation of regression in Fig. 3A to 100% in the year 2048).

2048 is 98 years after 1950: indeed, 0.0168*98^1.8992 is about 101.

The perils of extrapolation
So that's where the 2048 comes from. A trend, based on data over about 50 years, extrapolated another 50 years until we reach 100 percent. This claim, and how the authors arrived at it, has attracted a lot of criticism, which I won't discuss in detail here, but in my view the most fundamental objection is that extrapolating any trend, especially an exponential trend describing a number with a natural maximum (like a percentage), by as far as twice the observed range is bound to give extreme and unrealistic outcomes. In this particular case, as more and more stocks 'collapse', you are bound to have stocks left that are actually quite well-managed. Most Atlantic pelagic stocks, such as herring and mackerel, are healthy and well-managed. Not only are the stocks large enough to ensure plenty of replenishment that compensates fishing mortality, the institutions to manage the fishery, like Exclusive Economic Zones, scientists doing stock assessments, and stringent government policies, are also in place. North Sea herring is MSC certified. Iceland has practically its own national cod stock which it would be crazy to deplete. A lot of overfishing is due to poor exclusivity of stocks and the ensuing Prisoner's Dilemma, but these institutional flaws are largely absent for most North Atlantic stocks. And this is just one example.

How a side remark came to define a paper

If the 2048 estimate is so shaky, why do the authors make this claim? Remember that the objective of the article has never been to predict fisheries collapse: the authors wanted to assess the importance of species richness for marine ecosystem services. The 2048 claim was a side remark that ended up as a red flag in the associated press release. Interestingly, in the same month the 2048 article was published, the American fisheries scientist Ray Hilborn complained in the American magazine Fisheries (see a copy of the article here) that more and more articles in marine science, especially in leading journals such as Nature and Science, were published not for their scientific merit but for the public impact of their press release:

These four examples [not including the 2048 article, RG] illustrate a failure of the peer review system and lack of the basic skepticism needed in science, and are unfortunately but a few of the many papers now appearing with similar sensational but unsubstantiated headlines. [...] Critical peer review has been replaced by faith-based support for ideas and too many scientists have become advocates.

Indeed, Worm's paper was accompanied by a press release running the secondary headline "Current trends project collapse of currently fished seafoods by 2050". Apparently, the headline was meant as a wake-up call:

In a note to colleagues that was mistakenly sent to The Seattle Times, Worm wrote that the projection could act as a "news hook to get people's attention." "One reason why nobody cares about marine biodiversity is that there seemed no clear end in sight," he continued. "... Well, it's time to wake up — IF the current trend continues we will see drastic consequences in our own lifetime." (Excerpt from The Seattle Times)

Trevor Branch recently published a very insightful analysis of how the article was received in the scientific community. He found that among the articles citing the Worm et al. paper, the ones referring to the 2048 claim "had characteristics that suggest unfamiliarity with the controversy surrounding this projection, namely papers with few authors, published in journals with low impact factors, in fields far removed from ecology and fisheries, and sharing no coauthors with the Worm et al. paper." In other words: it's the rebels without a clue who cite the article for its doomsyear 2048.

The usual disclaimer
I'm no environmental Pangloss. There is plenty to be worried about, like the criminally exploited bluefin tuna stocks, or the continuous decline of European eel. Institutional problems abound in many regions of the world, making it exceedingly difficult to implement a sustainable system of stock management. And then there are pressures like ocean acidification, proliferation of invasive species, accumulation of plastics, and so on.

But I also believe that activist science will eventually eat itself. If a scientist bases his claims on wild extrapolations such as these, he or she is making it the pseudo-skeptics just too easy.