9514 - Laserfiche WebLink

Home Browse Search

FISH CULTURE-PERSPECTIVE Corollary 5: Divergence of the hatchery population can be avoided if random sampling of broodstock is practiced. Regarding this myth, it is important to recognize that even a random sample is just that-a sample. Taking a random sample will, on average, avoid directional bias, but in any particular case there is no guarantee that the sampled individuals will be representative of the pop- ulation as a whole (Hard et al. 1992). By chance, some types will generally be overrepresented while others are underrepresented. The effect can be pronounced in small samples, and diminishing as the sample size increases. Of course, effects of brood stock sampling can be avoided entirely by taking the whole population into captivity, but this extreme strategy substantially increases other risks. A related approach uses a form of stratified random sampling to collect adults from throughout the duration of the spawning season in proportion to their occurrence in the natural population. This strategy has some attractive features. The difficulty is that although it may be possible to collect a sample of brood stock that is more or less repre- sentative of the population as a whole for one trait (e.g., run timing), it generally will not be possible to collect a sample that is simultaneously representative of the popu- lation for all traits of interest (sex ratio, age, size, etc.). Therefore, it should be recognized that broodstock sam- pling will result in a captive population whose genetic composition differs from that of the natural population for one or more traits of interest. At best, one can hope to min- imize the extent of these genetic differences. Risk Tradeoffs Inescapable risk tradeoffs associated with alternative management strategies also make it impossible to avoid all undesirable effects of fish hatcheries. For example, al- though opportunities for genetic change in the hatchery are reduced if fish are released early in their life cycle (e.g., as fry or parr rather than as smolts), doing so increases opportunities for competition with natural fish and also reduces the survival benefit provided by the hatchery. Similarly, taking a large fraction of the population for broodstock minimizes founder effect in the hatchery popu- lation but exposes a larger proportion of the population to the risks of fish culture and could affect the remaining nat- ural population demographically or genetically. These inherent tradeoffs in risk are a major reason why it is difficult to develop comprehensive guidelines for broodstock collection, fish culture, and release strategies. There are no simple, universal answers to questions such as, "Is it better to mark all hatchery fish in a supplementa- tion program to facilitate monitoring and enhance man- agers' ability to meet program guidelines (such as control- ling hatchery-wild spawning ratios), or is it better to minimize marking to reduce harm to the fish?" In this and many other cases, no strategies exist that will simul- taneously minimize every type of risk. Guidelines can, and should, outline the various risks and tradeoffs involved, but which strategy should be preferred in any 16 . Fisheries given situation must be evaluated on a case-by-case basis within the context of the program goals. Summary Some level of genetic change relative to the natural population cannot be avoided in a cultured population. The difference in selective environments experienced by hatchery and natural fish also means that changes that do occur as a result of fish culture are unlikely to aid survival and reproduction in the wild. It is important that fish cul- turists, fisheries managers, and fishery biologists consider these realities in evaluating the appropriate nature and scale for hatchery programs. What the above analysis does not tell us is how exten- sive the genetic changes in cultured populations will be and how strongly they will affect natural populations. This is a complex topic that needs further research, and developing the information that will help us address this key information gap should be a major focus of effort in the near future. Myth 3: Hatcheries will always have unintended and deleterious effects on natural populations. Whether genetic change in cultured populations will affect natural populations will depend on the nature of the program. In supplementation programs, which involve the intentional integration of natural and hatchery production, some genetic change to the natural population also is inevitable. Therefore, the net value of a supplementation program must be determined by weighing these genetic changes (and potential ecological effects such as competi- tion, predation, etc.) against benefits (such as alleviating severe demographic and genetic risks of small popula- tions) that a successful program may be able to provide to natural populations. In contrast, harvest augmentation hatchery programs can (at least in theory) avoid deleterious effects on natural populations if strong enough isolation of hatchery and natural fish can be maintained. Whether such isolation is feasible will vary widely from program to program and species to species. Here, I identify two incidental risks to natural fish populations posed by hatcheries that can occur but sometimes have been overstated in the popular press. These are examples of a blurring of fact and specu- lation about the effects of hatcheries mentioned by Camp- ton (1995). Corollary 1: Hatchery fish stray more than wild fish. This assertion is not necessarily true. What is known about straying in salmon can be summarized as follows (see Quinn 1993, 1997 for reviews and discussion): · The proportion of cultured fish that return to sites other than the hatchery or release site varies greatly among programs; some hatcheries consistently produce a relatively high proportion of strays, while others produce few. Limited data for natural fish suggest that they also exhibit considerable variation among populations in rates of straying. Vol. 24, No.2