Let us eat (other people’s) fish

Should Americans really eat more fish?

In a recent op-ed in the NYT titled “Let Us Eat Fish” Dr. Ray Hilborn, a fisheries scientist at the University of Washington, argued we should. Ray thinks that because some of the hundreds of fish populations harvested in U.S. waters appear to be recovering and approaching levels that will maximize their productivity, then we should increase our fish consumption. He went on to argue this would be an environmentally positive choice since it would reduce terrestrial habitat loss and pollution. Ray is known in marine ecology, my academic field, for pushing his colleagues to stick to the facts and avoid exaggerating environmental degradation – a message I support. Unfortunately, in his op-ed, Ray left out a few key facts himself and makes an argument driven more by personal bias than the scientific record.

Let’s examine the factual accuracy of six statements from Ray’s op-ed and then decide how well the evidence supports his overall argument.  

1) “Over the last decade the public has been bombarded by apocalyptic predictions about the future of fish stocks”

I would categorize the recent science on fish and fishing a little differently. Rather than focusing on future fish stocks, this enormous literature has documented the past collapse of fish populations around the world and their current highly depleted status.

My research is focused on Caribbean coral reefs and there is no question that these systems are extremely overfished. I rarely see top predators on Caribbean reefs and only in no-take marine reserves do we ever see large numbers of fish or any fish larger than a dinner plate. This graph from Paddack et al. 2009 illustrates a broadly documented pattern: continued decline in fish abundance over the last several decades across the Caribbean basin:

Change in reef fish density across the Greater Caribbean. Bars are 95% confidence intervals. From Paddack et al. 2009. Based on a meta-analysis of fish surveys performed with SCUBA on 318 reefs, ie, fisheries independent data. Sample sizes are given in parentheses and represent the number of individual fish density estimates included in the analysis for each temporal grouping.

This Jamaican fish trap illustrates a phenomena that is pretty-much Caribbean-wide. The fish inside are not bycatch; they are the target fishery. These reefs have very few other fish – everything else was fished to local extinction decades ago:

Here is another example of fisheries-independent data collected and analyzed by my UNC colleagues Dr. Pete Peterson and Dr. Frank Schwartz (the figure is from Myers et al 2007):

Frank has been assessing the abundances of sharks off of North Carolina for over 40 years and like many similar studies, his work documents the rapid decline in top predators like tiger and bull sharks due primarily to sport fishing. This freed rays like the cownose ray from predation and caused an increase in their abundances that in turn decimated our scallop fishery, ie, large sharks eat rays and rays eat scallops. So one key function of sharks was to “protect” scallops from overconsumption by their own predators. Ecologists call this indirect facilitative relationship a “trophic cascade”. Such cascading effects are ubiquitous in nature and are just one of countless ways species are bound to each other in a complex web of interactions.

There are countless papers and graphs documenting the same pattern over and over again all around the world. But I don’t think any graph is convincing as a picture (and a time capsule) like the ones Dr. Loren McClenachan dug up in the Monroe County Library in Key West, Florida. The images document the striking decline in fish size on a single dock over time:
[cincopa AcJAtlazc91R]

2) “Subsequent research, including a paper I co-wrote in Science in 2009 with Boris Worm, the lead author of the 2006 paper, has shown that such warnings were exaggerated”

Really? Well… no. Unfortunately, Worm et al 2009 did not conclude “such warnings were exaggerated”.  Instead, the Abstract and Conclusion sections of the paper include statements like:

“63% of assessed fish stocks worldwide still require rebuilding, and even lower exploitation rates are needed to reverse the collapse of vulnerable species”

“Unfortunately, effective controls on exploitation rates are still lacking in vast areas of the ocean, including those beyond national jurisdiction”

“Ecosystems examined in this paper account for less than a quarter of world fisheries area and catch, and lightly to moderately fished and rebuilding ecosystems comprise less than half of those. They may best be interpreted as large-scale restoration experiments that demonstrate opportunities for successfully rebuilding marine resources elsewhere.”

3) “The overall record of American fisheries management since the mid-1990s is one of improvement, not of decline…Few if any fish species in the United States are now being harvested at too high a rate, and only 24 percent remain below their desired abundance

I agree with the first statement: there has been progress in the management of some fisheries in U.S. waters, however, why Ray concludes that “Few if any fish species in the United States are now being harvested at too high a rate” is unclear. The NOAA report Ray linked to includes this map of species currently “overfished” is the U.S.:

That is quite a few ecologically and commercially important species that NMFS categorizes as overfished.

As David Newman points out “Hilborn is also wrong when he claims that “few if any fish species in the United States are now being harvested at too high a rate.”  In truth, 20 percent of species tracked as part of the government’s Fish Stock Sustainability Index – the same source cited by Hilborn – are still being fished at too high a rate, and thus subject to overfishing.”

Talking Fish agrees; “[Hilborn’s] piece correctly states that populations of haddock and redfish have increased dramatically in New England, it neglects to mention that many other stocks managed as part of the Northeast multispecies groundfish complex are still subject to overfishing and/or are overfished, including Atlantic cod, Atlantic halibut, white hake, windowpane flounder, winter flounder, witch flounder and yellowtail flounder. The rebound of haddock and redfish is excellent news, but it does not speak for the state of the marine ecosystem as a whole.

Also see Eric Heupel’s excellent post explaining the Fishery Stock Sustainability Index (FSSI) from which these assessments were derived. He argues that the most recent FSSI fisheries grade  is a D; not exactly something to gloat about.

It is also important to consider that when a fisheries scientist tells you a population is overfished, they mean fishing pressure is too high to maximize productivity, i.e., reducing fishing would increase production and catch, the optimal fisheries density generally being roughly half the populations natural density. This is probably not the definition of “overfished” that came to many readers minds and not one ecologists or conservation biologists would use. I’d consider a population “overfished” when its numbers are reduced to a degree that its function in an ecosystem is significantly altered. This nearly always occurs long before a population is commercially “overfished”.  Furthermore, many fisheries scientists recognize the need for more conservative catch limits, as Ray and his co-authors argued in their Science paper (Worm et al 2009): “in fisheries science, there is a growing consensus that the exploitation rate that achieves Maximum Sustainable Yield (uMSY) should be reinterpreted as an upper limit rather than a management target. This requires overall reductions in exploitation rates”

I understand that people need to eat and have an income, but Ray is an exception among scientists in continuing to argue that the ecological function and ecosystem impacts of fishing should not be taken into account when determining optimal population sizes and catch limits.

Ray’s view of wildlife as a mere resource waiting for consumption is on display throughout his essay with quotes like this; “Rebuilding, however, has come at a cost: to prevent overharvesting and protect weak species, about 30 percent of the potential sustainable harvest from productive species (those that can be harvested at higher rates) goes untapped

Finally, the status of most fish harvested in U.S. waters is not even assessed by NOAA and don’t show up in these statistics. For example, none of the six clearly overfished (by any definition) sharks in the Myers et al 2007 study (described above) do. In Puerto Rico, there are dozens of fish species that are so depleted they are functionally extinct from nearshore environments, yet only grouper and conch appear to be on NMFS’s radar. There are also many species, like the North Atlantic Right Whale that are no longer fished but have not recovered from earlier over-explotation that are not considered in Ray’s analysis. Other species, like the Loggerhead Sea Turtle, that are the indirect victims of fishing (adult turtles get caught up in fishing gear and drown) don’t make the list either. Neither do invertebrates, many of which have been fished to ecological extinction (and are far below MSY), including abalone in California, sea urchins in New England, and blue crabs and horseshoe crabs along Atlantic shores, lobsters in the southeastern U.S., etc.

4) “On average, fish stocks worldwide appear to be stable…”

This is unlikely to be true, although it is difficult to assess since the data are dodgy or nonexistent for most of the world’s harvested populations. Recall, that in their analysis of the state of the world’s fisheries, Ray and his coauthors were only able to assess 25% of world fisheries area and catch, with almost no data for the tropics and most of the developing world. The fisheries-independent data that do exist (such as the meta-analysis by Paddack et al. 2009, also see Newman et al 2006 and Stallings 2009) certainly don’t jibe with this statement. As Talking Fish points out, “it is difficult to understand how a scientist of Hilborn’s stature could come to a conclusion about the overall health of the world’s fish populations given his concession in the op-ed that little is known about the stock status of fish populations in ‘much of Africa and Asia.’ “.

Yet Ray’s points gets us to the real heart of the matter: most of the seafood Americans eat is imported, therefore, the state of US fisheries shouldn’t influence our seafood consumption. By arguing we should eat more fish, what Ray is really saying is that we should be eating more of other people’s fish. In 2009, NOAA reported that 84% of the seafood consumed in the US was imported and this figure is growing.

In 2009 the U.S. imported 5.2 billion tons and exported 2.5 billion tons of seafood; a huge trade deficit that has global ecological and socioeconomic impacts. Like our spending deficit, China meets nearly a quarter of our seafood deficit. The rest of Asia and South America make up the rest:

I suppose a more reasonable (and factually supported) argument would have been that since the management of US fish stocks seems to be improving, Americans can begin to eat more American fish. The problem is, this would be impossible to do in practice for the simple reason that nobody would have any idea where the fish they were buying and eating were coming from. You don’t even know what species you are really consuming most of the time due to fraud in the sea food industry.

5) “if fish are off the menu, we will likely eat more beef, chicken and pork. And the environmental costs of producing more livestock are much higher than accepting fewer fish in the ocean: lost habitat, the need for ever more water, pesticides, fertilizer and antibiotics, chemical runoff and “dead zones” in the world’s seas”

First, nobody is arguing that fish should be “off the menu”. Second, I doubt there is a direct relationship between the consumption of fish and beef, chicken and pork. Third, these environmental costs, eg, dead zones, are due primarily to the farming of plants like corn and oil palm and not to livestock agriculture (although there are some important exceptions like deforestation for beef production in Brazil). Fourth, there are enormous environmental costs of fishing that Ray apparently is not aware of. These indirect impacts of harvesting and farming (via aquaculture) fish leads to precisely the impacts Ray mentions, i.e., habitat loss, chemical pollution, etc.

For example, as Eric points out over at DSN, bottom trawling – the method of choice for catching bottom fish in New England – is highly destructive. The method essentially entails dragging a weighted net across the sea floor; it is effective, but catches and kills indiscriminately and also destroys essential fish habitat (imagine cutting down a forest to catch a squirrel).[youtube]http://www.youtube.com/watch?v=bUHcD_jTgVA[/youtube]

Bottom trawling and other forms of net fishing also have lots of “bycatch“, a polite term for species, i.e., animals, that are accidentally caught and then thrown overboard usually after they sufficate on deck. In some fisheries, the ratio of bycatch to catch (of targeted species that are kept) can be 10:1 or higher, particular in the wild shrimp fishery where bycatch ratios can reach 20:1 and consist of endangered species like sea turtles. The photo below is the catch from a shrimp trawl; hard to see any shrimp in that pile of flesh isn’t it. 

Speaking of shrimp, the top three seafood imports in 2009 as recorded by NOAA were shrimp (1.2 billion tons), tuna (718 million pounds) and salmon (501 million pounds). Nearly all of that shrimp comes from man-made ponds (shrimp farms) in Asia and Ecuador created by cutting down coastal mangroves. This is the primary reason roughly half the world’s mangroves have been lost over the last several decades.

The image above is of Muisne, Ecuador, where roughly two thirds of mangroves forests have been replaced by shrimp farms.

A large majority of our third most popular import, salmon, are also farm-raised. Salmon farming causes serious environmental problems as well and negatively effects wild salmon populations we are trying rebuild and prevent from going extinct.

Finally, as in the shark-ray-scallop case I described above, substantial reductions in key species leads to community shifts and the loss of ecosystem functioning that hurt us economically and wreak ecological havoc in the ocean ecosystems.

Again, people clearly need to eat and fish protein is an important component of the diet of hundreds of millions of us. Yet to argue that fishing is any less destructive than terrestrial agriculture is misleading (to put it lightly).

6) “Suddenly, that tasty, healthful and environmentally friendly fish on the plate looks a lot more appetizing”

Tasty? For the most part (although I am not a big fan of “monkfish“, maybe unfairly judging the taste by the appearance). Healthful? Sure, some fish, such as salmon, are very healthy to eat (in moderation). But not farmed-raised shrimp from China and lots of other fishes and seafood species.  Environmentally friendly? Not exactly. There are a few species that are wild-caught (not farmed) and harvested sustainably with “minimal” ecological impacts (yup, there’s an app for that), but those are the rare exception.

The bottom line

Rays’s argument is not supported by the facts. He thinks about seafood consumption with a myopic national view and appears to be unaware of where most of our seafood comes from and what the social and ecological impacts of importing fish from developing nations is in those regions.  (Ironically, Ray and his colleagues argued for a “global perspective on rebuilding marine resources” in Worm et al (2009)).

I am not arguing people shouldn’t eat fish. I eat wild salmon roughly once a month and I love it. But I don’t think we should be gobbling up others people’s fish just because we have the economic leverage to do so. In addition to depriving them of protein in their own diets, we are indirectly degrading their marine environments and increasing our own carbon footprints by transporting seafood internationally.

[Editorial postscript: This post generated a lot of email responses and conversation, which have been collated together and can be read at SeaMonster’s “Forum on fish, food, and people”]


Posted

in

by

Tags:

Comments

15 responses to “Let us eat (other people’s) fish”

  1. RCompton

    An absolutely disgraceful misinterpretation of Hilborn’s paper. I encourage people to read Hilborn’s paper themselves. The comparison with climate change sceptics is inappropriate and in fact contrary to the real situation. The apocalyptic predictions of Pew sponsored environmental advocates such as Daniel Pauly are in fact as rare an opinion amongst quantitative fisheries scientists as climate sceptics in the world of informed climatologists. Scientists have become sponsored by environmental groups to produce sensational headlines. In reality Professor Hilborn should not be seen ‘at the other end of the spectrum’ rather smack bang in the middle representing the most defensible view point. Decide for yourself and treat articles such as this with caution.

    1. NOTE FROM THE EDITORS: We originally didn’t plan to publish this, primarily because we don’t want SM to turn into one of those sites where people argue about whether the earth is round, if Elvis is still alive, etc. We haven’t formulated a comment policy yet, but this is not the kind of thing we want to spend a lot of time responding to. That said, we decided to let this through, in large part because it illustrates our point about the eerie similarity between overfishing deniers and climate change deniers. RCompton – like so many climate change deniers – thinks it is all a conspiracy driven by corrupt scientists bought out by left wing think tanks. Sound familiar?

      As for the comment; (1) We don’t know what paper RCompton is referring to (we were only responding to the Hilborn NYT op-ed, nothing else. In fact, we agree with and admire much of Ray’s other work). (2) We did not misrepresent Ray; in fact we quoted him directly and extensively (nearly 25% of his op-ed) and only fact-checked his published statements. (3) Ray misrepresented himself, eg, claiming falsely that Worm et al 2009 showed “such warnings were exaggerated”. (4) We don’t believe there are polling data out there for how many scientists are on Team Ray and Team Daniel (or is it Team Edward and Team Jacob) but based on the email we have been getting today, there isn’t any question that most marine ecologists don’t agree with Ray’s op-ed. (5) We don’t want to get into who’s viewpoint is most defensible; why don’t we instead take a look at the facts and start the conversation by agreeing that liberties were taken with said facts in the op-ed in question.

      John and Emmett

  2. Great critique John (especially your point about the amount of imported fish eaten in the US – and across here in the UK too). I say bring on the lentils…

  3. Billo Heinzpeter Studer

    Great answer to Hilburn, and very helpful with good arguments. Thanks for this!

  4. Statements such as this one by Ray Hilborn and published in the NYT are widely spread to millions of Americans. Unfortunately, the general and simplistic message is captured fairly quickly by the general population. It would be great to be able to reply to his essay in the same OP-ED section of the NYT.

    1. Thanks Abel. A friend who heard about the op-ed actually said to me “hey, I heard it is OK to eat fish now”. The NYT did publish a few very short responses here; as you can see they are all pretty flaccid. But I agree, it is hard to fight words and anti-science with fewer words and no ability to display and link to evidence. I did write the public editor at the NYT to see if they will consider a more detailed debunking.

  5. Trevor Branch

    Such a lengthy post requires a lengthy response, which there is no time to make.
    1) You say that scientists are not predicting future collapse. Worm et al. (2006) projected collapse of all fisheries by 2048. Pauly & Watson (2003) “…if we don’t manage this resource, we will be left with a diet of jellyfish and plankton stew”. Jackson (2008) “transforming complex food webs topped by big animals into simplified, microbially dominated ecosystems with boom and bust cycles of toxic dinoflagellate blooms, jellyfish, and disease”.

    2) Fish stocks worldwide are indeed roughly stable since about 1990 if we look at biomass, see Worm et al. (2009, Science), Branch et al. (2011, Cons. Biol.), Hutchings et al. (2010, Can J. Fish Aq. Sci.). It is, of course, an interesting argument that stocks that we know little about must therefore be in a bad state. Or could they be in a good state? Until someone gathers data, it is hard to say.

    3) Clearcutting rainforests, prairies etc, with 100% loss of biodiversity, to grow crops is clearly a bigger impact than fishing. What percent of land cover remains pristine or lightly impacted compared to oceans? How many birds, mammals, reptiles, and freshwater fish are extinct vs. how many marine fish species are extinct? Clearly eating food from land has a bigger impact than eating fish.

    1. Thank you Trevor for your comments and sorry for the length of my post.

      1) I actually did not say that. But I do see and don’t dispute your broader point.

      2) Beginning with the caveat that we have no temporal biomass data for at least 75% of the world’s oceans and probably 99% of harvested species, I suppose I agree that quite a lot of studies (based on the very limited data that is available) are suggestive of recent biomass stability. However, isn’t it also true that in nearly all cases, this stability only occurs after a rapid and drastic decline in biomass, eg look at the curves in Myers and Worm 2003 and Myers et al 2007. I know you and others have pointed out serious problems with using fisheries catch data to infer biomass, yet so many studies with fisheries independent data, eg. Myers et al 2007, show precisely the same patterns, I have to wonder whether studies based on catch data are actually fairly representative of biomass trends. One interpretation is that current biomass stability is simply the result of overfishing, ie, reducing biomass to a point where extraction is no longer possible or profitable. I am curious how likely you think that is. I considered bringing this up in the original post, but it feels quite speculative (and as you said, it was getting long!)

      3) I honestly don’t see it like this. For one, a far greater proportion of the oceans is fished (99%) than the land (~40% is used for agriculture). Second, is the percentage of biodiversity loss from terrestrial clear cutting any higher than the impacts of bottom trawling (which covers a far greater amount of area than forestry) or clearcutting mangroves for aquaculture? Third, increasing our seafood consumption will in no way reduce terrestrial agriculture. How exactly would eating more farmed salmon and shrimp and Pacific Tuna lead to Iowa corn fields being restored to native prairie? It won’t. It will only increase the indirect effects of fishing and the habitat loss and degradation associated with aquaculture. Beyond that, I can’t see how a diet based on carnivorous animals could generally have less environmental impact than a plant-based diet.

      “What percent of land cover remains pristine or lightly impacted compared to oceans?” Good question. Strictly speakly, none of either. But I’d guess a far greater percentage of the land, primarily northern Europe, Asia and North American as well as northcentral South America (Amazonia) and central Australia is much closer to “pristine” than all but tiny remnants of the oceans. Think about it; there are at least dozens of uncontacted tribes of people in Brazil and many thousands of square miles of northern forest and tundra that no human has even hunted or farmed. How much of the oceans has never been harvested? As for extinctions: you are surely right, but this is due more to the life history of marine critters than lack of effort on our part!

  6. Trevor Branch

    John, I believe some of your responses are factually wrong.

    “Beginning with the caveat that we have no temporal biomass data for at least 75% of the world’s oceans and probably 99% of harvested species”
    This is factually wrong. We have temporal assessment data for 20-25% of world catch, and temporal FAO biomass status reports for 80% of world catch. So any species not represented are likely (1) of no commercial value, (2) therefore lightly harvested.

    “However, isn’t it also true that in nearly all cases, this stability only occurs after a rapid and drastic decline in biomass, eg look at the curves in Myers and Worm 2003 and Myers et al 2007.”
    First, these are for sharks, pelagic tunas and billfishes not for “nearly all cases”, even in Myers et al. (2007) half the [smaller] species are increasing not decreasing as you show above. Second, *after* the period of apparent 80% decline in CPUE (Myers and Worm 2003), tuna and billfish catches went up four-fold and are still rising. So either the oceans are now 20 times more productive (4X more catch from 0.2 of the population), or the real declines are much smaller than found by Myers and Worm (2003). Assessments support the latter.

    “For one, a far greater proportion of the oceans is fished (99%) than the land (~40% is used for agriculture).”
    This is factually false. The Arctic, Antarctic are basically unfished; abyssal depths alone comprise the majority of the ocean and are not fished; even in heavily trawled areas like the North Sea there are unfishable areas that are a sizeable fraction of the total area; and agriculture, cities etc. result in 100% biodiversity loss, while even in heavily fished areas biodiversity loss is only partial.

    “is the percentage of biodiversity loss from terrestrial clear cutting any higher than the impacts of bottom trawling (which covers a far greater amount of area than forestry) or clearcutting mangroves for aquaculture?”
    Yes, 100% loss is by definition higher than any loss from trawling. Bear in mind the North Sea is heavily overexploited but still producing wild caught fish. On the other hand, corn fields don’t produce any natural lumber. Granted, mangrove clearcutting is more akin to agriculture. And what about industry, mining, urban sprawl, destruction of virgin forest (what proportion is left?)?

    “Third, increasing our seafood consumption will in no way reduce terrestrial agriculture. How exactly would eating more farmed salmon and shrimp and Pacific Tuna lead to Iowa corn fields being restored to native prairie?”
    Simple laws of supply and demand. If there is no seafood, people will eat 80 million tons of something else. More crop lands will be needed for vegetarians, and even more for meat-eaters, assuming that most seafood eaters will not turn vegetarian. That means more land area will need to be turned over to agriculture.

    “Beyond that, I can’t see how a diet based on carnivorous animals could generally have less environmental impact than a plant-based diet.”
    Yes, ideally, we’d all be vegetarian. Alternatively, eat tilapia, they are vegetarian.

    “What percent of land cover remains pristine or lightly impacted compared to oceans?” Good question. Strictly speakly, none of either. But I’d guess a far greater percentage of the land, primarily northern Europe, Asia and North American as well as northcentral South America (Amazonia) and central Australia is much closer to “pristine” than all but tiny remnants of the oceans. Think about it; there are at least dozens of uncontacted tribes of people in Brazil and many thousands of square miles of northern forest and tundra that no human has even hunted or farmed.”

    How can untouched lands be inhabited by uncontacted tribes? I assume that such tribes eat plants and kill animals. How many wild animals live in New York vs. in the waters surrounding New York? How many buffalo and passenger pigeons survived compared to cod and anchovy? Global bird extinctions (190), mammal extinctions (~100 terrestrial, 5 marine), marine fish extinctions (zero). Need I continue?

    In the U.S., 27% of land area is protected (and this is fully 10% of global protected area), while 67% of total marine area is protected. I hate to source Wikipedia for “facts”, but that is the first thing that came up: http://en.wikipedia.org/wiki/Protected_areas_of_the_United_States

    1. Billo Heinzpeter Studer

      Trevor, you are partly right in saying «Alternatively, eat tilapia, they are vegetarian». They are omnivorous non-predators and could therefore well do with fully plant based feed. Why then is fishmeal/oil used in tilapia feed? Why does even a scheme for «responsible» aquaculture like ASC allow this? To let the industry boost the growth of the animals and to serve the Omega-3 hype…
      Besides that, more people turning to a predominantly vegetarian or even vegan diet is quite probable since – as you say – land use for animal based food production to serve a growing world population is becoming a core issue pretty fast…

  7. […] April 2011, and  John Bruno of the University of North Carolina (and my Co-Editor at SeaMonster) replied here at SeaMonster. John’s post was emailed to a list of experts in fisheries and marine ecology, and quickly […]

  8. […] then John Bruno really got things going with a response to Hilborn, Let us eat (other people’s) fish, at The Sea […]

  9. […] letters, and then ‘thangs got crazy’ with Dr. John Bruno’smost excellent response at The SeaMonster.  In response to this response…we got more responses.   And they came from some of the biggest […]

  10. […] Ray Hilborn made a splash when he suggested that the decline of fisheries has been overstated.  A rebuttal posted on the popular blog SeaMonster by John Bruno sparked a fruitful, and primarily civil […]

  11. […] Dr. Hilborn’s opinion. One respondent, Dr. John Bruno, wrote a long rebuttal entitled “Let us eat (other people’s) fish,” arguing that Dr. Hilborn’s advice may be credible in domestic U.S. markets, but not […]

Leave a Reply

Your email address will not be published. Required fields are marked *