Counting Fish


Bruce Barcott's essay last week in the online magazine Yale e360 reveals an interesting environmental paradox, one in which Pacific salmon might be both endangered and, simultaneously, too populous.  As Barcott writes, "How can numerous Pacific salmon runs be on the verge of extinction while total salmon numbers are straining the limits of the ocean's capacity to support them?"  The fulcrum for this is a recent Marine and Coastal Fisheries article [PDF].  Noting that salmon populations have doubled since 1960 because of ever escalating production of pink and chum hatcheries in Alaska, Japan, and Russia, the authors worry, as Barcott explains, that "the north Pacific Ocean may be nearing the limit of its salmon-carrying capacity."

This is a concern, but our temporal and spatial frames of reference require refinement.  Fisheries scientists tend to treat the start of their records as the beginning of time, especially when assessing hatcheries, but this makes for bad history.  That production doubled in the last fifty years is interesting and may be troubling, yet the amount of juvenile salmon entering the Pacific had so shrunk since 1800 that the recent doubling might mean little.  Merely counting fish is insufficient.  Context is everything, so we must expand our gaze in time and space.  There is a long history to ocean capacity, and increased productivity is also attributable to factors such as harvests, habitat protection, and fish passage.  This is about more than the last half century and a few superhatcheries.

The same applies to the ocean.  Barcott notes that concerns about hatcheries have been heightened because of the North Pacific's recent decline in productivity.  Carrying capacity--what scientists call density dependence--is contingent on ecological context.  Alaska's desire to pump huge numbers of juvenile salmon into this fluctuating pool of nutrients might exacerbate already stressed conditions for non-hatchery fish, but the ocean's falling productivity, though unsettling, is unsurprising, having been brought on by a periodic hydrological shift called the Pacific Decadal Oscillation (PDO).  Every several decades the Alaska and California Currents oscillate in productivity.  Depending on the relative strength of the Aleutian Low, the Subarctic Current delivers nutrients primarily southward or northward, and this latest switch favors the California Current.  Thus when Barcott says the "North Pacific" is declining in productivity, he actually means the Alaska Gyre.

The Pacific Decadal Oscilattion - Relevant large-scale upper level physical oceanography of the Subarctic North Pacific and the Bering Sea. Photo courtesy John Wiley and Sons. Click to view larger image.

A convergence of policy and ecology might be imperiling non-hatchery salmon and, perhaps, other parts of the ecosystem, but this is literally only half of the story.  These events primarily affect runs from northern British Columbia, Alaska, and Russia.  Most ESA-listed runs are not affected because they inhabit the California Current, which just improved.  Nor is it clear how hatchery-reared salmon affect naturally reproducing runs.  Although one study suggests an inimical relationship between pinks and sockeyes, two more recent studies find hatchery salmon are generally less fit [PDF] and reproduce poorly [PDF].  After 130 years salmon hatcheries still mostly create expensive fodder.  Historically, their best documented impact is not outcompeting wild fish--which might be true--but inflating harvests that inordinately depress wild stocks.  Pinks also cause concerns because they enter the ocean first in spring and thus get preferential access to declining nutrients, and because studies suggest that early migration boosts hatchery stock fitness.  This is why Barcott regards pinks as "built for 21st century reproductive success."  Early migration is only contingently advantageous, however.  Broughton Sound pink runs imploded in 2001 when juveniles swam through salmon-farm-induced irruptions of sea lice (Morton), and Sacramento River chinooks crashed after encountering a seasonally-induced dead zone [PDF] off San Francisco Bay in 2005.

Because interactions between salmon and sea are so contingent, often differing by race, species, and watershed, we must ask, "Which salmon and which ocean?"  Similarly, when we see reports on single studies, we should know the conversations in which they are embedded.  On his own blog, Barcott offers access to several documents he consulted.  Included is a 2001 reply to a comment on a study about carrying capacity and Alaska hatcheries.  Missing are both the study and the original  comment.  This is like watching only one hole of a golf tournament, and I say this as someone who agrees with Barcott and with the comment he cites.  Scientific debates are rarely short.  We need all parts of the discussion to fully capture the intellectual push and pull within a community making sense of confusing, contradictory data.  The debate about hatcheries and carrying capacity began not in the 2000s but in the 1970s or even earlier.  From then to now the question of what happens to salmon during their ocean phase has been suffused with uncertainty, so we create false clarity by ignoring the partial and contested nature of the Marine and Coastal Fisheries study, which I find intriguing and proudly claim one of the authors as a colleague.

There is a growing tendency for science journals and news media to highlight research in isolation to its intellectual context.  In the last few years I have seen ever more press releases on new studies about, say, salmon migrations, interactions between farmed and wild salmon, and organic pollutants in salmon.  Rarely do these live up to the billing, which is not to say they are unimportant, but that each is only one iteration in a longer conversation. 

Science is a process that rarely proves anything, and in the last 130 years researchers have saved exactly zero runs.  Over the same period the PDO has switched at least five times, each time inspiring someone to pour more money into hatcheries and allowing others to rationalize habitat changes that ultimately did far more permanent damage to salmon than the Pacific, which in any case we can do squat about.  Nor we can know what all affects any run, even who captures the fish.  The ocean is a mess but more because of the social structure of the modern fishery than anything it naturally does outside the subject of warming and acidification. But that's a topic for a separate post.

Joseph Taylor teaches in the history and geography departments at Simon Fraser University, in Vancouver. He is the author of Pilgrims of the Vertical: Yosemite Rock Climbers and Nature at Risk and Making Salmon: An Environmental History of the Northwest Fisheries Crisis, which won the American Society of Environmental History’s best book award. He lives in Oregon.