NOTE:
The following is a study proposal not a peer reviewed journal article. However, it is included here because it proposes a study designed to evaluate the reproductive success of hatchery-reared and wild coho salmon and to test the theory of hatchery supplementation to increase wild salmonid populations.
CHIP Project Proposal Narrative
I. Project Title: Nonpariel Dam Adult Trap and Coho Genetic Pedigree
II. Contact: Kathryn Kostow, Genetics Program Leader; Dave Loomis, Umpqua District Fish Biologist, Dr. Michael Blouin, Oregon State University.
III. Project Abstract:
This proposal would investigate several areas of uncertainty about the use of hatcheries to increase the abundance of wild populations. There is a considerable interest in using hatcheries to speed the recovery of wild populations. However the value of such programs is untested. Substantial literature exists that indicates hatchery programs may pose high risks to wild populations, rather than aid them (see the following reviews: Hindar et al 1991, Waples 1991, Waples 1999, and Lichatowich 1999 and literature cited therein). If the risks are real, hatcheries may interfere with recovery, rather than speed it. Until recently, analytical methods to explore the critical questions and risks associated with hatchery programs were unavailable because we were not able to track lineages in streams once hatchery and wild fish were allowed to spawn together. New molecular genetics methods now allow us to use DNA fingerprints to pedigree entire populations under some circumstances and develop lineages that continue for multiple generations under natural spawning conditions. We can finally produce direct evidence of the success or failure of hatchery supplementation programs and provide direct measurements of some of the risks predicted by genetics theory. We propose to utilize these methods on an experimental supplementation program for coho salmon on the Calapooya River, a tributary of the Umpqua River on the Oregon Coast.
IV. Proposal:
A. Project Need:
1. Intent:
The effective use of hatchery fish to increase the size of an extant wild population has not been demonstrated. The concept is to take part of a small wild population into captivity, disproportionately increase the number of offspring produced by them, release those offspring into the wild, and then allow them to spawn naturally as adults thereby significantly increasing the total number of natural spawners. If this larger spawning population reproduces successfully in the stream it should produce a much larger naturally-produced (wild) population in a small number of generations. The benefit of this larger population size may out-weigh the impact of genetic risks caused by the action (Figure 1).
However the success of this approach has not been evaluated or demonstrated. We know we are able to substantially increase the number of natural spawning fish by adding hatchery adults to a stream. But to date we have not been able to demonstrate that this action increases the number of naturally-produced (wild) adults in the stream. We also expect, based on genetics theory, that substantial genetic risks to the wild population may occur as a result of this action, but we have not been able to directly measure the risks. Our biggest handicap to evaluating these efforts has been our inability to determine the parentage of naturally-produced offspring in a natural stream setting. New developments in molecular genetics now allow us to pedigree entire populations, provided we are able to handle the adults. These methods let us exactly match offspring to parents. The results are straightforward and unambiguous. We are able to follow lineages from parents to offspring to grand-offspring. We finally will have a clear answer as to whether hatchery fish breed as successfully in streams as wild fish do, which will measure the success of the hatchery program. We will also be able to directly measure several genetic risk factors.
Reproductive success by hatchery fish spawning in a stream is expected to be lower than that of wild fish. The lower fitness of hatchery-born adults manifests itself in two ways: First, hatchery-born adults do not compete for mates or build nests as successfully as wild fish (Fleming and Petersson, 2001, Chebanov and Riddell 1998). Second, the survival of their offspring is reduced owing to relaxed natural selection and to domestication selection that occurs during the egg-to-smolt stage in the hatchery (Lynch and O’Hely 2001, Reisenbichler and McIntyre 1977, Reisenbichler and Rubin 1999). Successful reproduction by the hatchery fish spawning in the stream ö specifically production of adult offspring is required if the benefit of an increased wild population size is to occur. We will be able to directly measure the reproductive success of the hatchery fish relative to wild fish by knowing exactly how many adult offspring are produced by each natural spawning individual.
Hatchery programs, where substantial numbers of hatchery fish spawn naturally in a wild population, theoretically cause five major genetic risks to wild populations. The risks are demonstrated in Figure 1 and include the following:
Risk 1.
Population Bottleneck (Ryman and Laikre 1991): This risk occurs when a small number of parents (those taken into the hatchery) contribute more offspring per parent to the supplemented population than the rest of the population (those left in the wild). This difference in family size causes a decrease in the effective population size of the total population.
Risk 2:
Increased Inbreeding (Ryman et al 1995): This risk occurs when only a small number of parents (those taken into the hatchery) produce a substantial proportion of the fish in the supplemented population. Since they share so few parents, the hatchery fish in the supplemented population are more likely to be related to each other, thus increasing the incidence of inbreeding.
Risk 3:
Increased Genetic Load (Lynch and OâHely 2001): This risk results from the increased reproductive success and survival that occurs while fish are in the captive environment. Increased reproductive success and survival in captivity occurs because natural selection pressures are intensely relaxed which leads to an increase in the level of genetic load.
All of these risks are inevitable in any hatchery supplementation program. However, if the hatchery fish breed successfully, and the program succeeds in increasing the size of the wild population, and it stabilizes at the larger size, and the hatchery program stops removing further risk, a net benefit to the wild population may occur. If, on the other hand there is reproductive failure by the supplemented population, further genetic risks will occur:
Risk 4:
Genetic Variation is Lost (Nei et al, 1975): When an offspring population is smaller than itâs parent population genetic variation is lost. This is due to reproductive failure by some parents and the loss of the genetic material they carry. Additional random loss of genetic variation may occur when populations are very small.
And finally, if the hatchery program continues over multiple generations the impacts of these risks will accumulate in the wild population due to the nature of the genetic mechanisms involved (Risk 5).
Direct measurements of effective population size, inbreeding coefficient, and reproductive success or failure can be made using pedigrees. Occurrence of increased genetic load and loss of genetic variation can be inferred from the measures of individual reproductive success.
Additional questions exist about the best protocols to use in implementing a supplementation program. For example, using single-generation hatchery broodstock (parents taken from the wild each generation) rather than old hatchery stocks should minimize the genetic effects, but there has never been a test of this hypothesis. Similarly, releasing unfed fry should reduce the extent to which selection is relaxed in the hatchery to only that experienced during the egg-to-fry stage, and to selection on any parental behaviors such as maternal nest building ability. Therefore, although survival from egg to adult of fish released as unfed fry is much lower than that of fish released as smolt, the hatchery adults that return from the unfed fry releases may be nearly as successful at natural reproduction as completely wild fish. This hypothesis has also never been tested. It is not possible to test all possible protocols in a single experiment. This study proposes to investigate the following strategies:
a. Is a first-generation wild-type broodstock a better choice than an older, multi-generation broodstock? Theoretically, the first-generation broodstock should have less genetic load and domestication build-up than an older one and should succeed better. The existing Rock Creek Hatchery coho broodstock is an older and also partly mixed-origin broodstock. The success of these will be compared to wild fish collected at Winchester Dam in 2001 and at Nonpariel Dam in 2002-03 to form a first generation broodstock.
b. Is a less invasive hatchery program better than a more invasive one? In a less invasive program, fish are held captive through a lesser portion of their life cycle, which should decrease genetic load build-up. The down-side of holding fish captive for a shorter period is that the survival benefits, and therefore the rapid increase in number of fish, are compromised. In our experiment we compare two options:
i. Captivity during reproduction and rearing through hatching (release of unfed fry); and
ii. Captivity during reproduction and rearing through smoltification (release of smolts).
c. The reproductive success of adults returning from all of the hatchery treatments will be compared to that of wild fish returning at the same time (in years 2004 through 2007, including both jacks and adults, with their offspring returning in 2007 through 2010, including both jacks and adults).
The potential benefits of a supplementation program also depend on the carrying capacity of the basin. The naturally-produced population can increase in size only if the basin is capable of producing more fish than are currently present. It is therefore important to evaluate the apparent carrying capacity of the supplemented basin at the beginning of the program.