Click for next page ( 146


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 145
6 Genetics and Conservation Managing salmon requires an understanding of the biological dynamics of the populations in which they occur and reproduce. In particular, knowledge of the structure of the genetic variation in salmon is needed to make decisions about how to identify and protect the local reproductive units, which are the fundamen- tal biological units. That knowledge is needed to guide conservation strategies. Without it, the safest conservation strategy requires conserving virtually every- thing. Even the first people who thought about conservation of western North American salmon were aware of differences between local populations and of the importance of the differences in the success of the fish. R.D. Hume packed canned salmon with a booklet that he wrote on conserving the valuable salmon resource (Hume 18931: I firmly believe that like conditions must be had in order to bring about like results, and that to transplant salmon successfully they must be placed in rivers where the natural conditions are similar to that from which they have been taken. Moulton (1939) and Thompson (1959, 1965) provided the first modern descrip- tion of the importance of local populations and genetic diversity for the manage- ment of Pacific salmon. They pointed out that each local population was geneti- cally adapted to its own environment ("home-stream colony") and had a characteristic level of abundance around which it varied. In any river system, some colonies are capable of supporting exploitation, and others are not; com 145

OCR for page 145
46 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST mercial exploitation is likely to lead to the disappearance of some colonies (i.e., the extinction of local populations). This chapter considers the fundamental role of genetics in the long-term sustainability of salmon. It examines the organization of genetic diversity within species and stresses the importance of recognizing the local reproductive group as the primary demographic and genetic unit. The objective of increasing run sizes (NPPC 1987) and the biological need for preserving existing genetic diversity have led to the characterization of a choice of "meat versus museums" or productivity versus genes (Beckman and Berg 19921. That is, we are faced with a choice between increasing production and protecting wild and natural runs. But that simple view is incorrect and dangerous. Genetic diversity is part of the fabric of a biological resource. The resource (production) cannot be separated from its genetic basis. Sustained salmon and steelhead productivity can be maintained only if the genetic resources that are the basis of such productivity are maintained. Those interested in pro- tecting genetic diversity are not interested in maintaining genes (i.e., DNA se- quences) for their own sake; rather, they are interested in protecting genetic diversity because it is necessary for the long-term persistence of the fish them- selves. Examination of the 1987 Columbia River Basin Fish and Wildlife Program reveals that genetics has been part of the development of the program (NPPC 19871. However, a closer reading indicates that genetics has not been incorpo- rated into planned activities. There is little sign of recognition that all manage- ment activities (e.g., activities related to fishing, passage, habitat, and produc- tion) affect genetic resources and that genetic diversity is part of the fabric of the resource. It is important that potential genetic effects of all management activi- ties be considered in making decisions. STRUCTURE OF GENETIC VARIATION Genetic diversity within a species occurs at two primary levels: genetic differences between individuals within local breeding populations called denies in population genetics, and genetic differences between breeding populations. The total genetic diversity within a species is often viewed as a hierarchy of levels (see Figure 6-1 for a simplified structure). The two levels of genetic diversity listed above are of primary importance in the genetics and evolution of a species. Genetic differences between individuals within a local breeding popu- lation are the basis for natural selection and adaptive evolution, and genetic differences between breeding populations reflect local adaptation to past environ- ments and random events. The genetic differences between breeding populations represent the broadest pool of genetic variation and are a valuable component of diversity. Adaptation to future environments is dependent on variation within the local breeding population as well as variations between populations.

OCR for page 145
GENETICS AND CONSERVaTION o Species \ Subspecies, ecotypes = : ~2 i,. ., 2,: ,,:: :' '' 2." ~ ~ ji:'i ,' ~ ' '''I j i ~ j j j: j j i , ~ ~ , ?. . ~ ~..j~2.:2.'i',.~j.:,' it' ~ , ~:,:.,.~:.; . Am 4,~,, : jet Jo. [;;~? USA A{ ~;~18 (.: ~:, ~,. ~:'.:j:: : ',, , ~ v;.i ~d.e ~'> ~ ail:, it,, ~ ,~ I: I: I,,, i:,,: .:.::": '::" j '" '"j>"' "':"'"'' '':'iti~ W ~. I , : rat V~ 2' ~ a.,:....,:, jj,,\li. ,...j..,..:' ::. I,:, \'~ ..2s'.''\~2) ('.j.'' ""i "2\. . ;2""'\, ~,.,i.~.~. i2...~..~.;lir W.',2~'l' 147 7 7 FIGURE 6-1 Schematic representation of structure of genetic diversity in Pacific salmon. Inverted triangle emphasizes that locally adapted and largely reproductively isolated, local breeding populations are basic unit of diversity. Varied operational definitions of stock would place this term in the range identified by dashed lines. Source: modified from Riddell 1993b. Animal and plant breeders over the centuries have used artificial selection among individuals within breeding populations to improve their stocks. One simply breeds individuals that have some favorable variations in the characters wanted. These variations are due almost universally to both environmental and genetic influences. A breeder looks at successive generations to see whether the selection process has worked. Even though the scientific basis of selective ad- vance was not understood until this century, artificial selection has produced extraordinary results. Humans have profoundly changed the inherited character- istics of dogs, cattle, grains, ornamental plants, and many other species. Darwin was the first to emphasize that natural ecological processes select for hereditary changes in a manner very similar to that of human selection. In nature, it is the harshness and challenge of a continually changing environment that determines which individuals successfully reproduce under natural conditions. Natural selection results in improved adaptation to environmental conditions.

OCR for page 145
48 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST Simply put, both artificial and natural selection result in descent with genetic change; this is biological evolution. For the evolution and continued existence of species, genetic differences between populations are as important as genetic differences between individuals within a population. Consider the extreme where no differences exist between local populations. In that case, a species consists of many copies of the same genetic population and is extremely vulnerable to environmental change. For example, a new disease might be introduced to which most individuals are geneti- cally susceptible; the disease would jeopardize all populations and therefore the entire species. However, in the usual case, where genetic differences do exist between local populations, it is likely that some populations would have a higher frequency of genetically resistant individuals and thus would be relatively unaf- fected by the disease. LOCAL REPRODUCTIVE UNITS Because of homing, the fundamental unit of replacement or recruitment for anadromous salmon is the local population (Rich 1939, Ricker 19721. That is, an adequate number of individuals for each local reproductive population is needed to ensure persistence of the many reproductive units that make up a fished stock of salmon. The homing of salmon to their natal streams produces a branching system of local reproductive populations that are largely demographically and genetically isolated. The demographic dynamics of a fish population are deter- mined by the balance between reproductive potential (i.e., biological and physi- cal limits to production) and losses due to natural death and fishing. "Population persistence requires replacement in numbers by the recruitment process" (Sissenwine 1984), so fishery scientists have focused on setting fishing intensity so that adequate numbers of individuals "escape" fishing to provide sufficient recruitment to replace losses. The distinction between a local breeding population and a fished stock is critical (Beverton et al. 1984~. Whereas a local breeding population has a spe- cific meaning a local population in which mating occurs- stock- is essentially arbitrary and can refer to any recognizable group of population units that are fished (Larkin 1972, Chapter 41. The literature has often been unclear on this distinction (e.g., Helle (1981) defined a stock as a local breeding population, citing Ricker (1972) as his authority). In practice, it is extremely difficult to regulate losses to fishing on the basis of individual local breeding populations. Thousands of local breeding populations make up the West Coast salmon fishery, and many of these are likely to be intermingled in any particular catch. Neverthe- less, the result of regulating fishing on a stock basis and ignoring the reproductive units that together constitute a stock is the disappearance or extirpation of some of the local breeding populations (Clark 1984~. The homing behavior of anadromous salmon results in a complex pattern of

OCR for page 145
GENETICS AND CONSERVATION 149 genetic differentiation among local breeding populations because individual fish that home to different streams cannot breed with each other. The importance of the demic structure of a group of populations (or metapopulat~on as described below) for managing a fishery is related to how often fish stray or spawn in a nonnatal stream, although it is often not clear how much small differences in where or when salmon home are due to environmentally caused differences, random events, or genetics (Quinn 19931. It is difficult to measure empirically the amount of straying among natural local breeding populations of salmon (Quinn 1990~. However, analysis of genetic variation in a network of local breeding populations can provide insight into the pattern and amount of straying among local breeding populations. The chinook of the Klamath River drainage show a complex pattern of genetic differentiation at 36 variable protein loci detected electrophoretically; Bartley et al. (1992a) described genetic variation in chinook from a larger geo- graphical area, but we have analyzed their data only for the Klamath River. Samples were collected from 10 locations throughout the drainage (Figure 6-2~. To simplify the presentation, we analyzed the eight loci in which the mean frequency of the most common allele (or a form of gene) was less than 0.95 (Table 6- 1). We pooled all other alleles at each locus to standardize the statistical analysis among loci. There are highly significant differences in allele frequen- cies among the 10 samples at all eight loci individually (p < 0.001; contingency chi-square test, and except for populations 8, 9, and 10, which show no evi- dence of genetic differentiation between each other-the distributions of allele frequencies are significantly different between all other pairs of populations (con- tingency chi-square test summed over all eight loci). To display the pattern of genetic differentiation between the 10 populations, we used principal-component analysis of the covariance matrix of allele frequen- cies at the eight loci. About 75% of the total variation is explained by the first two principal components (Figure 6-31. In general, the samples tend to cluster according to geography (Figure 6-3J; the importance of geographical distance is especially apparent for the first principal component, which explains about 61~o of the total variation. Thus, straying appears to be more common among popula- tions that are closer to each another, consistent with patterns documented by Quinn et al. (1991) and Pascual and Quinn (19941. Those data indicate substantial genetic differentiation among chinook popu- lations from the Klamath River (except for population samples 8,9, and 10, from the extreme upstream portion of the river basin). The genetic differentiation results from the homing tendency of chinook and the resulting reproductive iso- lation among the breeding populations. Bartley et al. (1992a) estimated that the amount of genetic differentiation in the 10 samples is consistent with an average of about four migrants per generation among local populations. The average age of sexual maturity (i.e., generation time) in chinook from the Klamath River is

OCR for page 145
50 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST , ~,'_ \ ~OREGON ~ J~ it' CALIFORNIA ~N ~ _ FIGURE 6-2 Collection sites of 10 samples of chinook in Klamath River drainage. Source: Bartley et al. 1992a. about 4 years (Healey 19913. Therefore, the estimated rate of straying among these local breeding populations is about one stray per year within each reproduc- tive population. LOCAL ADAPTATION One important reason to protect local populations is that they are locally adapted to the streams that support them. In other words, evolution has made a local breeding population better able to survive and reproduce in its home stream than in other streams. These adaptations are of great importance. Re-establish- ing new populations through introductions once the local populations have been lost has proved to be extremely difficult. And even if a newly introduced popu

OCR for page 145
151 o o o o o ~ o oo o o o o ~ o oo o ~ C~ _ _ - o o V Cq o 2 o C) _ ,L, ~_ Ct ~ . _ _ 2 au ._ Ct - _I _ a~ o E~ V 4 4 O C~ C: 4 C: 4 4 4 C: V: ~ o ~ - Ct o o o o ~ ~ o o o ~ ~ ~ ~ o ~ ~ o o oo ~ G~ ~ O oO ~ O O . . . . . . . . . . o o o o - o o - - o ~q ~ U~ ~ C~ oo o o o _ o ~ _ U~ ~ o o ~ oO G~ oo ~ oo ~ O O ~ o o o o o o o ~ ~ o o V~ ~ ~ ~ oo V~ C~ ~ ~ o ~ ~ C~ ~ ~ oo vo U~ C~ ~ ~ ~ ~ ~ . . . . . . . . . . o o o o o o o o o o o o o o o oo oo ~ C~ oo oo ~ oo ~C~)-,-,~_____ . . . . . . . . . . o o o o o o o o o o 0 r~ o~ ~ ~ c~ ~ oo oo O C~ ~ ~ K0 ~ ~ ~ G~ O o0 oo ~ ~ ~D ~ ~ ~ - o o o o o o o o ~ o ~ o o ~ ~ cM - C7N =1 - ) ') - ) ~ - CA ~ ~ ~ ~ - ~ ~ ~ ~ ~ . . . . . . . . . . o o o o o o o o o o v~ ~ v~ ~ oo =, ~ ~ G~ oo ~ ~ oo ~ ~ C~ . . . . . . . . . . o o o o o o o o o o ~ ~ v~ O ~ ~ ~ v~ oo ~ o oo ~ ~ c~ u~ ~ ~ c~ r~ G~ ~ ~ ~ G~ . . . . . . . . . . o o o o o o o o o o _ ~ ~ ~ ~r, ~ ~ 00 C~ O ;^ c~ c~ . - o s~ a: 11 c~ .= .- 3 V: C~ C~ o C ~C) C ~- o _ _ Ct C~ C C~ C) a; C) _ o CQ

OCR for page 145
152 -0.10 LL -0.1 5 Ad o o C ) -0.30 it G -0.45 UPSTREA SALMON AND SOCIETY IN THE PACIFIC NORTHWEST 5 _ W' (\ 1; ,' 4 . . 3 1 1 1 1 1 1 -1 .65 -1 .50 -1 .35 -1 .20 -1 .05 -0.90 PRINCIPAL COMPONENT FIGURE 6-3 Plot of first two principal-component scores derived from allele frequencies at eight loci in Table 6-1. Samples enclosed by solid line are not statistically different. Samples enclosed by dashed line have statistically different allele frequencies (0.05 > p > 0.001); all other pairs of samples have highly significant differences in allele frequencies (p < 0.0011. ration is initially successful, it might not be adapted to the range environmental conditions that have happened in the past and can be expected to occur again in the future. That is, important local adaptations might be of importance only in extreme environments that are unlikely to occur within the time frame that we are concerned with but that are extremely likely to occur on an evolutionary time scale. Adaptation to local conditions is demonstrated by the timing of spawning among populations of sockeye in the Fraser River system (Brannon 19873. A1- though the timing of spawning varies little from year to year in a given spawning location, there are great and consistent differences among spawning areas be- cause of adaptations to the most favorable local conditions for incubation, timing of emergence, and juvenile feeding (Burgner 19911. The timing of spawning for local populations in the Fraser River is influenced primarily by the temperature regime of the spawning site (Figure 6-41. Spawning is later in the warmer incubation environments. That pattern of differences among populations results in similar emergence timing of the progeny during the next spring because of the greater rate of development at higher temperatures. Smoker et al. (in press)

OCR for page 145
GENETICS AND CONSERVATION 91 84 7 6- o - ~- - as 4- 3- 2- Forfar Scums Harrison Weaver Adams .. Chilko ~ Stellako Scotch Nadina 1 AUG SEP OCT NOV Month 153 FIGURE 6-4 Spawning times and mean incubation temperatures of nine local breeding populations of sockeye from the Fraser River. Source: Brannon 1987. showed similar genetic variation in the timing of migration in pink salmon in Alaska. Stearns (1976) has defined a life-history "strategy" as a set of coadapted reproductive traits resulting from selection in a particular environment. Some anadromous salmon have spectacularly complex life histories. For example, Snake River sockeye emerge in freshwater at an elevation of 2,000 meters, mi- grate to a nursery lake, and generally spend 2 years growing in the lake. They then smoltify and migrate some 1,500 km downstream to the ocean, where they spend 2 or 3 years. In the ocean, they undergo long feeding migrations. On approaching sexual maturity, they return to the mouth of the Columbia River and then retrace their journey of 1,500 km upstream to their nursery stream, where they spawn. The life history of those fish includes three distinct habitats: the nursery

OCR for page 145
154 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST lake, the ocean, and the spawning stream. The fish undergo complex behavioral, physiological, and morphological transformations in transition from one habitat to another. In addition, they must undergo four major migrations: from stream to lake, from lake to ocean, a feeding migration in the ocean, and from ocean to natal stream. The timing of each of those events must be precise. Migration us? and down the freshwater system to the ocean must be timed to correspond with appropriate water bows and the availability of food. In some cases, arrival in the freshwater habitat can precede actual spawning by many months. A number of studies have demonstrated that nearly all those aspects of life history are influ- enced to some degree by genetic differences among individuals and populations (reviewed in Taylor 1991 and Levings 1993~. A well-studied aspect of local adaptation in salmon is the migratory behavior of newly emerged fry. Studies with sockeye, rainbow trout, cutthroat trout, and arctic grayling (a freshwater member of the salmon family, Thymallus arcticus) have demonstrated innate differences in migratory behavior that correspond to specializations in movement from the spawning and incubation habitat in streams to lakes favorable for feeding and growth (Raleigh 1971, Brannon 1972, Kelso et al. 1981, Kaya 1991~. Fry emerging from lal~e-outlet streams typically migrate upstream on emergence, and fry from inlet streams typically migrate down- stream. Quinn (1985) showed that differences in compass-orientation behavior of newly emerged sockeye correspond to movements in feeding areas (Figure 6- 51. . 24~_ Cedar River Art 1 it. -a x.;. . ~ Chilko River Weaver Creek At' FIGURE 6-S Mean compass orientations of newly emerged sockeye fry from three populations: Cedar River in Washington and Chilko River and Weaver Creek in Fraser River system in British Columbia. Compass orientation of experimental fry corresponds to the direction that will facilitate migration in the nursery lakes. Source: Quinn 1985.

OCR for page 145
GENETICS AND CONSERVATION 155 Strong evidence exists that spawning populations of anadromous salmon exhibit highly specific local adaptations for a number of traits such as the compli- cated homing behavior, temperature adjustments, unique local mating behavior, and adjustment of smelts to local feeding conditions. These adaptations are most likely to be quantitative characters that are dependent on the effects of many genes, each of which has only a small effect individually (polygenes). These principles are basic to genetics of populations but are known mostly from the study of species better suited than salmon to this kind of genetic analysis. For the above reasons, we would expect it to be difficult to "replace" a local population with transplants from nonlocal populations. The more complex the life cycle, the more difficult it would be to replace a local population. Complex- ity would be influenced by the number of migrations and major habitat shifts. A break in continuity between any one of these stages would cause failure of repro- ductive success. For a newly introduced population to be successful, at least a few individuals must be able to complete the entire life cycle and reproduce. The experience with attempted introductions of salmon populations supports these conclusions. Attempts to establish self-sustaining anadromous populations generally have failed (Steward and Bjornn 1990, Burgner 1991, Heard 1991, Ridell 1993b). For example, a logging dam was built in 1908 on the Adams River, a tributary of the Fraser River in British Columbia (Riddell 1993b). The dam blocked access of sockeye to the upper Adams River from 1908 to 1921; this run had been among the largest sockeye runs in the Fraser River system. The upper Adams River area has 1.2 km2 of spawning area, which should be suffi- cient to support 6 million adult sockeye per year on the basis of the productivity of other sockeye populations in the region. Sixteen attempts between 1949 and 1975 to reintroduce sockeye to these spawning areas from other areas have not been successful in reestablishing this run. Today, very few fish return to spawn in the upper Adams River. In contrast with the lack of success with anadromous sockeye, introductions of kokanee (nonanadromous sockeyes have been successful. The greater success of introduced kokanee is probably related to their having a simpler life history than anadromous sockeye. Similarly, introductions of trout into lakes and streams throughout the intermountain west of North America have been generally suc- cessful (Allendorf and Leary 19881. METAPOPULATION STRUCTURE The individual local breeding populations within a drainage basin or other geographical area are usually connected in a higher level of organization by exchange of individuals through "straying." The set of local breeding popula- tions connected by exchange of individuals is a metapopulation or a "population of populations" (Hanski and Gilpin 19911; an example of a metapopulation is the fall-run chinook of the lower Columbia River (Pascual and Quinn 19941. The

OCR for page 145
56 UPSTREAM: SALMON AND SOCIETYIN THE PACIFIC NORTHWEST local breeding population is the primary demographic and genetic unit, as dis- cussed in the previous sections. However, the metapopulation organization be- comes very important when time scales longer than several generations are con- sidered. From a demographic perspective, metapopulation dynamics is the balance between extinction and recolonization of local breeding populations. That is, individual local breeding populations are expected to have a limited time of persistence on an evolutionary time scale. New populations are expected to be established by strays from other local breeding populations within the metapopu- lation. The metapopulation concept is thus closely linked with the process of local extinctions and the establishment of new local breeding populations. The geographical arrangement of salmon into discrete spawning populations where environmental conditions are appropriate for successful reproduction makes the metapopulation model appropriate for salmon. Local breeding populations of salmon are small enough and exist in such variable environments that they are likely to have relatively short persistence times. From a genetic perspective, the concept of the metapopulation is similar to Wright's view of species subdivided into many small local breeding populations that breed largely within themselves but are connected with other local breeding populations by migration and genetic exchange (Wright 1951, 19699. Wright (1940) also considered the potential evolutionary importance of local extinction and recolonization of vacant habitat patches in his analysis of genetic population structure. Partial isolation of local breeding populations allows the evolution of adaptations to local environmental conditions, and the infrequent exchange of individuals ensures that all the alleles within a metapopulation will be present in each local breeding population and can be acted on by natural selection (Allendorf 1983~. Local genetic adaptations play an important role in the demographic dynam- ics of metapopulations. As discussed previously, efforts to re-establish local breeding populations by transferring fish from other locales has generally not been successful. That lack of success has at least a partially genetic basis in that the transferred fish do not have the appropriate genetic adaptations to complete their life cycle in the habitat to which they are transferred. Thus, straying itself might not be adequate for recolonization; the straying individuals must be from similar-enough environmental conditions for their progeny to be able to complete their life cycle and successfully reproduce in the new environment. The total genetic diversity in a species is the sum of the variability over many hierarchical levels from the smallest local breeding population to the metapopu- lation to larger geographical areas that contain many metapopulations. It is impossible to draw sharp boundaries between different levels of spatially struc- tured populations, but it is useful to identify the mechanisms that operate in the population dynamics on different spatial scales (Hanski and Gilpin 19911. For example, Figure 6-6 shows the distribution of chinook on different geographical

OCR for page 145
GENETICS AND CONSERVATION , , \` OREGON `' ~ ~: 70 . 60 50 40 _ 157 / ~:~ Katzeb,e it; f ~ Sound - ~ CoDoerm'=~ < , ~ ~ BERING NAUSEA OF ~ ~ SEA by, OKHOTSK ,~ ~ n ~ Ka'mchatka R. ~ ~ "~\~ : ~ Islands Her fly Vancouver~_ ~ _. Islander, ~ . Columbia R. ~ Sacramento B.~ $ Sara Joaquin ft.N 1 40W 1 20W ~> ,,,:'' ~ ; Ado 1 20E 1 40E 1 60E : . ,,; ~ - ' ~\~ NORTH PAClPlC OCf~ , . . . . . 180 1 60W 70N 60N 50N 4noN FIGURE 6-6 Three scales of geographical population structure of chinook. A, metapop- ulation of chinook in Klamath River drainage (Source: Utter et al. 1992). B. U.S. West Coast (Source: Utter et al. 19921. C, North Pacific Ocean and Bering Sea, showing distribution of chinook salmon through their natural range (Source: Healey 19911.

OCR for page 145
58 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST scales beginning with the metapopulation of chinook in the Klamath River drain- age (A). The chinook in western North America (B' consist of many such clusters of local breeding populations or metapopulations. Finally, the entire range of chinook (C) contains all the genetic diversity present in the species. The different geographical subdivisions of chinook have adaptive differ- ences that are partially genetically based. For example, chinook returning to a coastal river in California are adapted to survive and reproduce in an environment very different from a river of comparable size in Alaska. On a smaller scale, it is extremely likely that the enormous chinook that used to inhabit the Elwha River on Washington's Olympic Peninsula were genetically different from the smaller chinook of other rivers in the region. The importance of geography In the distri- bution of genetic variation in chinook was examined by Utter et al. (1992), who found that 43% of the variability in allele frequencies at 24 polymorphic loci in nine local breeding populations of chinook from the West Coast of the United States is explained by the correlation between geographic distance and genetic similarity (Figure 6-79. Studies of biochemical variation do not permit any interpretation of the adaptive significance of genetic differences among local breeding populations, because we do not completely understand the relationship between biochemically detected genetic variation and phenotypic vanation. It is likely that the pattern of o.o~o O.~ o.oos UJ ~ 0.007 cn it 0.006 o.oos 0.004 0.003 0 002 _ O.W1 0.000 o - 100 200 300 400 500 600 7= 800 900 1000 GEOGRAPHIC DISTANCE (km) FIGURE 6-7 Correlation between geographic and genetic distance for nine local breed- ing populations of chinook from coastal drainages of California, Oregon, and Washington (r: = 0.43; P < 0.001). Line is principal axis of correlation. Source: (Jtter et al. 1993.

OCR for page 145
GENETICS AND CONSERVATION 1.4 t.3~ l 0 1.2 1.t al 0-9 W o 0.8-\ ~ 0.14 CY 0.6 0.5 0.4 0.3 0.2 0.1 O-, ,o 0 200 400 600 Distance Transferred (km) . 159 - - o - - - 800 FIGURE 6-8 Relative recovery rate (in comparison to local fish) for transferred hatchery coho versus distance transferred. Source: Reisenbichler 1988. genetic divergence at protein loci among chinook populations largely reflects the operation of genetic exchange and genetic drift on selectively neutral, or nearly neutral, genetic variation. However, as gene flow decreases and genetic diver- gence increases, the probability that populations will acquire adaptations to their local environment increases. For example, Reisenbichler (1988) showed that the distance from the home stream is inversely related to the success of hatchery- reared coho (Figure 6-8~. LEVEL OF GENETIC ORGANIZATION TO BE CONSERVED The study of salmon genetics does not by itself lead to an answer to the question, What should we conserve? It does make clear that genetic variation- evolutionary potential is present at all population levels, and it does make clear how important natural population-genetic structure and dynamics are. The En- dangered Species Act (ESA) leads to attempts to decide what level of genetic organization should be conserved by specifying that it applies to "distinct popu- lation segments" of vertebrate species. For that reason, the concept of "evolu- tionarily significant unit," or ESU, has been introduced (see, e.g., Waples 1991, Waples in press). This concept has formed the basis for the policy on distinct population segments for salmon adopted by the National Marine Fisheries Ser

OCR for page 145
160 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST vice. The policy has been used in ESA listing determinations for Pacific salmon (summarized by Waples, in press). The ESU concept provides a consistent, scientifically based framework for interpreting the meaning of distinct populations of salmon under the ESA. The unifying theme of the ESU concept is conservation of the evolutionary legacy and potential of the biological species that is, the genetic variability that is a product of past evolutionary events and that represents the reservoir on which future evolutionary potential depends. The goal is thus to ensure viability of the biological species by conserving enough of its basic components to allow the dynamic processes of evolution to proceed. Waples (1991) advocated a holistic approach to identifying ESUs and pro- vided specific guidance on how to integrate genetic, phenotypic, life-history, ecological, and geographic information. The ESU concept is inherently hierar- chical and applies equally to local breeding populations and to groups of popula- tions (metapopulations). The focus is on units that are largely independent over evolutionarily important periods, and the "bottom-line" test for an ESU is whether its loss would represent a significant loss of ecological and genetic diversity to the species as a whole. It is important to recognize (as did Waples, in press) that decisions about what constitutes "significance" and about the resource tradeoffs implicit in recovery plans are largely societal decisions that cannot be based on scientific grounds alone. A recent National Research Council report (NRC l995b) developed the con- cept of the evolutionary unit (EU). The EU concept is similar, but not identical, to the ESU; differences between them are largely a matter of emphasis. The EU concept does not stress reproductive isolation as a criterion because reproductive isolation is often more difficult to assess directly than it is in anadromous salmon. The EU was meant to be applicable to all animals and plants, while the ESU was developed for anadromous salmon. The NRC report concluded that the applica- tion of either the EU or ESU concept would lead to similar results most of the time and also recommended that identifying an EU be separate from deciding whether it is in need of protection (NRC 1995b). One aspect of the problem was discussed by Allendorf and Waples (in press) in analyzing the importance of local adaptation to conservation programs. They pointed out a paradox: small, locally adapted populations (demes) are clearly essential units for conservation efforts, but the kinds of local adaptations that they contain have evolved many times in salmon in a relatively short time and hence are likely to evolve again. Such recurrent adaptations may actually be different genetic solutions to an environmental challenge and different demes adapted to the same challenges may have different genetic complements. For some future challenge, an appropriate solution may require components of several of them, even if they appear to be adapted to the same environment. If that is true, why should we expend a lot of time and money on conserving wild runs? One answer to the question concerns the time scale of concern. Although

OCR for page 145
GENETICS AND CONSERVATION 161 salmon have adapted relatively quickly and frequently in evolutionary time- perhaps over periods of hundreds to a few thousand years and even faster where they have been introduced into new environments, such as New Zealand (Quinn and Unwin 1993) the depletion of salmon and their habitats in the Pacific Northwest is happening over periods of decades. It is not appropriate to imagine that a process that takes hundreds or thousands of years can compensate for one occurring at least 10 times as fast. In addition, human impacts on the environ- ment have probably affected the types of environmental variations and their variability, and that change is also likely to decrease the ability of populations to adapt to local conditions. Another answer concerns the number and sizes of natural populations. The local adaptations that produced the diversity of salmon life histories in the Pacific Northwest took place in a large and undepleted metapopulation structure. Today, parts of the metapopulation structure are missing, other parts are reduced in size, some local breeding populations have been extirpated, and many areas are popu- lated largely or only by hatchery fish. It is therefore likely that even given hundreds or a few thousand years, local adaptation would not occur as quickly as it did in the past. So, although the evolutionary plasticity of salmon gives us hope that rehabilitation is possible, it is not a reason to diminish efforts to conserve diverse wild runs if long-term sustainability of salmon in the Pacific Northwest is a goal. Finally, it is likely that some of the responses to local conditions is pheno- typic, as well as genetic, at least in its early stages. If the genetic variability in salmon populations is reduced by local extinctions, increases in the proportions of hatchery fish, and reductions in population sizes, it is likely that phenotypic (or environmentally caused) plasticity will be reduced as well. Indeed, the relation- ship between phenotypic plasticity and genotypic variability in fishes has been a research topic for many years (e.g., Alm 1959, Policansky 1983, Stearns 19893. A better understanding of the relationship in salmon would clearly be helpful in any rehabilitation and conservation program. EFFECTS OF HUMAN ACTIVITIES ON GENETIC DIVERSITY The genetic structure of salmon populations in the Pacific Northwest has been modified drastically over the last 100 years. Many locally adapted popula- tions have been lost because of dams and loss of habitat. The remaining wild spawning populations have been modified to an unknown degree by fishing, interbreeding with fish released from hatcheries, and habitat modifications. Relatively few of the local breeding populations of salmon that existed 150 years ago in the Pacific Northwest exist today. It is difficult to estimate accu- rately what has been lost. Pacific salmon have disappeared from about 40% of their historical range in Washington, Oregon, Idaho, and California (Chapter 4~. Many local breeding populations in the remaining 60C%c of the historical range

OCR for page 145
62 UPSTREAM: SALMON AND SOCIETY IN THE PACIFIC NORTHWEST have been extirpated. Williams et al. (1992) listed 106 stocks of salmon and steelhead that have been extirpated in this region. Most of these stocks probably comprised two or more local breeding populations (Williams et al. 19921. In addition, many local breeding populations have been swamped by introgression from hatchery salmon, the original local breeding population having been re- placed by genetic material from straying hatchery fish. For example, coho still spawn and reproduce in the lower Columbia River. However, it appears that the native local breeding populations of these fish have been replaced by feral hatch- ery fish (Johnson et al. 19914. In addition, a substantial proportion of the remaining demes of salmon are threatened by imminent extirpation (Williams et al. 19923. Nehlsen et al. (1991) clearly summarized the current status of salmon and steelhead in their review: With the loss of so many populations prior to our knowledge of stock structure, the historic richness of the salmon and steelhead resource of the West Coast will never be known. However, it is clear that what has survived is a small proportion of what once existed, and what remains is substantially at risk. The continual erosion of the locally adapted groups that are the basis of salmon reproduction constitutes the pivotal threat to salmon conservation today. Extant salmon populations have also been genetically modified by human actions over the last 100 years. Most management practices affect some aspects of the genetic makeup of salmon resources. Perhaps the largest effect has been produced by hatcheries, described briefly by Allendorf and Ryman (1987) and discussed in detail in Chapter 12. But other management practices also have effects. The potential and actual effects of fishing have been reviewed recently by Nelson and Soule (1987), Thorpe (1993), and Policansky (1993~. For pink salmon in particular, fishing appears to have changed the genetic makeup of many stocks and resulted in smaller adults. It has likely affected other species as well and might have changed the frequency of early-maturing small males (jacks) of several salmon species. Changing the habitat of salmon also has the potential for indirectly causing genetic changes in their populations, although this has received little attention. There is evidence of selective effects of hatchery environments (see Chapter 12), and there is every reason to believe that changes in water flows, temperatures, and chemistry and other changes in the physical and biological environments of salmon in streams caused by human activities have resulted in genetic changes. In summary, a large array of human activities have affected salmon genetics in such a way as to reduce genetic diversity at all levels of population structure. CONCLUSIONS Sustained productivity of anadromous salmon in the Pacific Northwest is possible only if the genetic resources that are the basis of such productivity are

OCR for page 145
GENETICS AND CONSERVATION 163 maintained. We have already lost a substantial portion of the genetic diversity that existed in these salmon species 150 years ago. The possible genetic effects of any actions must be considered when any management decisions are made. The local reproductive population, or deme, is the fundamental biological unit of salmon demography and genetics. An adequate number of returning adults for every local breeding population is needed to ensure persistence of all the repro- ductive units. The result of regulating fishing on a metapopulation basis and ignoring the reproductive units that make up a metapopulation is the disappear- ance or extirpation of some of the local breeding populations and the eventual collapse of the ~netapopulation's production. The metapopulation model of geographical structure is important for salmon because of the geographical arrangement of salmon into discrete spawning popu- lations adapted to the environmental conditions in which they reproduce. Local demes of salmon are small enough and exist in variable-enough environments for it to be likely that they will have relatively short persistence times on an evolu- tionary scale. Although the deme is the functional unit of salmon genetics and demography, the cluster of local populations (the metapopulation) connected by genetic exchange via natural straying is the fundamental unit on an evolutionary time scale. This conclusion is crucial because it leads to many other conclusions and recommendations about salmon management. For example, most of this report's conclusions and recommendations about hatcheries, fishing, and habitat rehabilitation are founded on the importance of maintaining appropriate diversity in salmon gene pools and population structure, which has not been adequately recognized.