Salmon Recovery and Local Breeding Populations

Editor,

In a recent Holiday Market E-Mail Update (Oct. 15, 2001) there was a partial reprint of an article by Jim Buchal entitled Environmentalists, Terrorists, Patriotism and the War Against America: A Speech to the Kitsap Alliance of Property Owners. In it, Buchal argues that salmon are not endangered, and that the concepts of distinct population segments and evolutionarily significant units (ESU’s) are just lies that the “buffoons that the clueless majority sends to Washington, D.C.” have perpetrated on the public. He also argues that fish raised in hatcheries are not meaningfully different from ones bred in the wild, that streamside habitat has little to do with salmon survivability, and that anyone who says otherwise is simply a “powermonger” who wants nothing more than to get people’s private property for some undisclosed end. He then attempts at various points to somehow relate all of this to “crazy environmentalists”, “morons”, “eco-nazis” and even terrorists.

The lack of scientific understanding in these arguments is considerable (not to mention the paranoia). Mr. Buchal appears to have little or no formal understanding of the science of salmon ecology, or for that matter, even how to do basic research and think critically. Furthermore, the hysteria, cheap shots and general immaturity displayed in his comments make this lack of understanding insufferable. His book “The Great Salmon Hoax” also contains many errors, which likewise are the result of poor research, improper or severely out of context citations, and conclusions that do not follow from either his premises or his sources (See Footnote 1). These errors and his confrontational tone both require comment. I apologize in advance for the fact that the references used in what follows are relatively few in comparison with the much larger body of literature on the subject (though they do represent the general consensus of research to date). The sheer volume and severity of the errors in “The Great Salmon Hoax” and this article forbids a detailed treatment of them all in this space (though I hope to provide this in another forum soon). For now, let me just concentrate on Mr. Buchal’s statements in the article referenced above.

First of all, let’s look at his “science”. The concept of the Local Breeding Population he disparages is based on the fact that the health of any breeding population of any species in the wild, including salmon, is determined by the genetic diversity of the population in question. This diversity is driven by genetic mutation, population dynamics, and the action of natural selection on the population as determined by a wide range of environmental variables and symbiotic relationships with the surrounding habitat. Salmon, as we all know, return at maturity to their parent streams to spawn. They return to specific locations in specific streams, and thus breed within specific subpopulations (for instance, in the Skagit River, pink salmon which breed in Illabot Creek are a distinct population from those that would breed in, say Gilligan Creek). Because of homing instinct, these subpopulations (called “Demes” from the same root word demographic is derived from) are the critical determinant of recruitment and biodiversity (Rich, 1939; Ricker, 1972). The persistence of such demes is dependent on recruitment, biological and physical constraints on reproductive potential, and losses due to natural death and fishing. If the recruitment process does not replace these losses, demes can collapse (Sissenwine, 1984). In this process, the distinction between a deme (local population) and a larger stock is critical (Beverton et. al., 1984). This is because breeding only happens within demes that exist within certain habitat regions that have very specific characteristics and vary widely from location to location. It is these characteristics which drive the natural selection process, and thus the genetic diversity of salmon stocks. Within any given deme, genetic mutation will produce a certain amount of diversity, which will be larger in proportion to the population size. Over the course of evolutionarily significant times, a given larger stock of salmon will be separated into demes by a variety of factors. Each time a deme is separated from a larger metapopulation, only a random sampling of the parent metapopulation’s genome will be represented in the new habitat. Typically, through what scientists call the Founder Effect (Mayr, 1942), this “subgenome”, which characterizes the deme, will not be fully representative of that of the parent metapopulation. Given the distinct geographic and biological characteristics of the new habitat, evolutionary mechanisms will produce a distinct subpopulation unique to that habitat. This is known as Allopatric Speciation. In the case of salmon, it is actually somewhat more complex than this in that most subpopulations occasionally receive strays from nearby other subpopulations within the larger metapopulation and are thus not fully isolated (Hanski & Gilpin, 1991). The degree to which this happens is driven by geography proximity (Quinn et. al., 1991; Quinn & Pascual, 1994). This allows genetic information from the larger population to enter into demes from time to time, though on a small scale, thus helping to insure genetic diversity.

Within any given such subpopulation, a variety of forms of genetic mutation will increase diversity. In other words, mutation will increase the number of Alleles (separate forms of a gene at any particular locus within the gene) which code for any given trait, harmful or helpful. Most of these traits will be heterozygous within the population (not shared by both chromosomes, and thus not expressed in a trait). Against this, breeding within the population will tend to drive the genetic diversity toward homogeneity, and thus to homozygosity, for harmful alleles. This is Inbreeding Depression. Against this background, natural selection will take place, tuning the subgenome of the deme to maximal fitness for the particular geographic habitat where the population spawns and nurses, and for the spawning journey back to this habitat at adulthood. In addition, a variety of stochastic factors external to the habitat ecology itself (catastrophes, natural and man-made, random dispersion within the genome through breeding dynamics, etc.) act to occasionally disrupt this process in constructive and destructive ways. This adaptation is strongly coupled to local habitat conditions and geographical relationships to the parent metapopulation. An excellent example of this was provided by Bartley et. al. (1992) and the National Research Council (NRC, 1996). They studied the genetic differentiation at 8 loci in 10 subpopulations of Klamath River drainage chinook using electrophoric analysis. Analysis of their data shows statistically significant genetic variation in allele frequencies for all 8 loci occur for each subpopulation. These differences were then shown to be strongly related to geographic proximity (NRC, 1996). Thus, salmon have a complex, highly symbiotic relationship to their watersheds that involves hierarchical relationships between demes (subpopulations) and their larger metapopulations.

Typically, in a population large enough to be healthy, these factors will largely cancel each other and the population will be fit, healthy and well adapted to handle any of the stochastic “catastrophes” described above. But when populations become small, harmful alleles are a much more significant part of the subgenome and genetic drift becomes much more important (the proportion of harmful alleles with respect to positive traits, as determined by natural selection, is called the Genetic Load of the population). Likewise, mutation rates, which are proportional to the population size, will be correspondingly depressed. Thus, as the population shrinks in size, genetic drift begins to win out over positive mutation and the population becomes less capable of adapting to selective pressures in its environment. It becomes “threatened”, or even “endangered”. Eventually, if the population falls to a critically low level, it will no longer have the genetic information or population size to negotiate any of these natural pressures and the population will be lost. Typically, when any species goes extinct, it is not because the last member was hunted or fished out – it goes extinct because the population grew too small to genetically resist a cascade of natural and anthropogenic events which kill off the remaining numbers.

This is a key point. Salmon populations do not go extinct because they are fished out. They go extinct because natural, and more commonly, anthropogenic pressures lower their numbers to a point below viable genetic diversity, and this genetic diversity has meaning only within specific habitats at specific geographic locations. This is the first point where Mr. Buchal shows his ignorance of life science. He argues that any given subpopulation could be lost without losing the entire “species”. Though this is technically true, it is completely irrelevant given actual salmon populations. Pacific salmon “species” are, in fact, defined entirely by the collective sum of their Local Breeding Populations, all of which exist in specific habitats to which they have adapted over many millennia. Very few of these Local Breeding Populations are not at risk to degree. They’re all in at least some danger, not just one or two. Many are in critical danger, and hatcheries have had little success in replacing what has already been lost (Steward & Bjornn, 1990; Burgner, 1991; Heard, 1991; Ridell, 1993). In fact, due largely to the factors described above, acting on LBP’s all over the Northwest, salmon have now disappeared from over 40% of their historical range in Washington, Oregon, California and Idaho in the last century. If the areas where they are now threatened or endangered are added, less than one third of their historic range has not experienced dangerous losses, and this situation is getting worse with each passing year (NRC, 1996).

Furthermore, Mr. Buchal’s shrill rants aside, once a fish representative of a given deme is removed from it’s respective habitat to be spawned artificially in a hatchery (such as the Samish hatchery in the notorious photo accompanying the article) it’s genetic diversity no longer has a context. That diversity was honed by mutation and natural selection for a completely different set of circumstances which may or may not be compatible with the environment it has been moved to. Fish well adapted to a tributary of the Upper Stillaguamish, for instance, may not be as viable for the lower Samish, or vice versa. Also, since timing of runs as a function of geographical location is often an important part of subpopulation adaptation (Brannon, 1987; Burgner, 1991), a chum salmon adapted to spawn in the upper Skagit near Marblemount won’t necessarily do well in a breeding population from the Lower Skagit (this is particularly true because homing plays such a crucial role in subpopulation recruitment). Issaquah Creek chinook, which depend on lake nursery areas like Lake Sammammish, might not do as well in either situation. In general, separating a fish from the environment where it's genotype has evolved will significantly reduce the advantages it provides. The fact that the fish being clubbed in the article’s notorious photo (see Footnote 2) was genetically part of some threatened population was meaningless, because the fish was no longer in the habitat where it’s genetic makeup was precious (which is why hatchery workers don’t generally worry about clubbing them to get eggs). Furthermore, since hatcheries tend to promote inbreeding depression, after a few generations, much of that genetic uniqueness will be lost anyway, and it will be lost at least 3 to 4 times faster than evolution can replace it. This is a large part of why hatcheries have generally been unsuccessful at restoring fish populations (Steward & Bjornn, 1990; Burgner, 1991; Heard, 1991; Ridell, 1993). Furthermore, decadal variations in North Pacific feeding grounds can mask the seriousness of these problems in the short term. Such variations, driven primarily by ENSO and PDO (El Nino Southern Oscillation and the Pacific Decadal Oscillation) – large scale low frequency temperature fluctuations which are in turn are thought to be driven by large scale climate change effects and the thermohaline cycle of the open ocean - lead to shifts in deep upwelling of nutrients that directly affect the fecundity of the open ocean food chain (UNESCO, 1992). These decadal shifts can have a profound effect on the survival and growth rates of salmon populations, and thus on returning adults (Francis & Sibley, 1991; Beamish & Bouillon, 1993; Francis & Hare, 1994). Such decadal variations can temporarily (repeat, temporarily) mask the effects of the problems described above (Lawson, 1993).

Mr. Buchal appears to be completely ignorant of all this. After addressing the subject of Local Breeding Population, he then takes up the subject of riparian habitat and property ownership. In a particularly hysterical rant, he states that,

“Nothing is going to change until you people wake up and start fighting. It doesn't take much. It just takes an angry mob to start showing up at these meetings and saying who the hell are you to tell me I can't remodel my house the way I want because of some fish? Are you people nuts? Do you think fish swim through my living room? Do you think a fish can feel the difference between plowing twenty-five feet away from the water and plowing 125 feet away? Do you think that a fish can tell the difference between whether ‘native vegetation’ or ordinary lawn grass is growing on your property twenty feet away from the river?”