Columbia Basin salmon production has declined dramatically in recent years. Thirty-eight populations throughout the basin are less than half their former abundance and are showing statistically significant declining trends over a fifteen-year period (listed as BDD in Table 3.1). Of these, 66% are found in the Snake River, 18% in the mid-Columbia area, and 16% in the lower Columbia above Bonneville Dam. Eighty-seven percent of the declining populations are spring chinook and 13% are fall chinook. Many other populations also exhibit recent declining trends in abundance, but the information base was too brief to allow the same analysis.
Because the problem is greatest for chinook populations in the Snake River, and because the information base is most complete for those populations, we have focused our analysis of the impacts of this plan on Snake River spring and fall chinook. Other salmon populations in other portions of the basin should respond in a qualitatively similar manner, although the magnitude of the response would probably vary.
The following analyses focus first on the effects of the those recommendations affecting all salmon populations. These effects are characterized in terms of naturally reproducing populations. Supplementation strategies are a key part of the tribal approach to restore anadromous fish populations. However, supplementation projects act on single populations and their cumulative effects are less dramatic. Their effects are described separately.
A total of seven scenarios were examined for Snake River spring and fall chinook:
- Prior to European development (Historical);
- A recent base period (1990-1993) at about the time Snake River populations were listed (Recent);
- Two short term (7-year) scenarios:
- a. NMFS proposed recovery plan for 1995;
- b. a tribal short term proposal (for the next 7 years);
- Three medium-term (25-year) scenarios which differed only in the amount of river draw down:
- a. John Day and Ice Harbor projects to natural river level;
- b. John Day project to spillway crest, Ice Harbor and Lower Monumental to natural river levels;
- c. John Day and four lower Snake River projects to natural river levels. This is the preferred tribal drawdown option.
Survival rates appropriate for each life stage under the various scenarios were evaluated and the results are presented in Table 5C.1 for scenarios 1 through 3b and for scenario 4c.
These scenarios were compared based upon 1) the probability of population persistence and recovery within 50 years, 2) the rate at which spawning populations would rebuild, and 3) the fisheries they might support. The effects of various hydropower system operation scenarios were estimated from technical analyses conducted as part of the IDFG v NMFS negotiations (Schaller et al. 1992; Schaller and Cooney 1992; BRWG 1994) and from output of the FLUSH salmon passage model. Salmon population characteristics were obtained from the Coordinated Information System Stock Summary Report for Idaho (Kiefer et al. 1992). The effects of changes in harvest regulations were estimated by CRITFC staff using techniques developed by the U.S. v Oregon Technical Advisory Committee and Pacific Salmon Commission (PSC) Chinook Technical Committee. The effects of habitat protection and restoration was estimated based upon the outcome of the Coarse Screening Process (Rhodes et al. 1995) and the professional judgement of its authors.
Most of the expected results of the preceding hypotheses and associated actions were estimated using a simplified spreadsheet model. Spring chinook survival rates were drawn from survival profiles of five modeled populations from the Snake River system (Schaller et al. 1992; Schaller and Cooney 1992; BRWG 1994) and represent a composite survival profile for Snake River spring chinook. Fall chinook survival rates were estimated from model results for Snake River fall chinook.
The analysis was organized by life stage and environment, similar to the hypotheses and our proposed monitoring program. We did not include a production function in the analysis, but rather looked at the fate of a unit of production over its lifetime under steady state conditions. This approach allowed rapid estimation of results while retaining consistency with assumptions and data of earlier analyses. The key assumptions used in these analyses were:
- Egg-to-parr survival for Snake River spring chinook would decline 10% under NMFS plan because it is an action-by-action approach and would be unlikely to halt present declining trends in habitat quality, which result from cumulative non-point sources;
- Egg-to-smolt survival would double for Snake River spring chinook by year 25 due primarily to survival increases associated with restoration of degraded habitat if the tribes’ habitat protection and restoration measures were implemented immediately;
- Ocean natural survival rates will fluctuate around a mean value over time. We recognize there are decade-long cyclic patterns in ocean survival, but have not attempted to simulate them in this analysis;
- The average population structure of Snake River spring chinook is 50% female with an average fecundity of 4300 eggs per female. For fall chinook it is 40% females with an average fecundity of 5000 eggs per female.
While some level of detail was sacrificed by this approach, the results are appropriate for comparing the relative outcomes of various management scenarios.
Probability of Persistence and Recovery During the IDFG v. NMFS negotiations, state and tribal fishery scientists estimated the probabilities that fish populations would a) persist over the next 50 years and b) rebuild to ESA delisting levels (STFA 1995a; 1995b). We have used that analysis as one measure of the management scenarios described in this plan. Readers are referred to the source documents for a full description of the analysis.
Rate of Rebuilding The time for populations to rebuild to delisting levels was estimated assuming steady-state conditions and using the rate of spawner increase (Spawner/spawner Ratio in Table 5C.2) per generation.
Fishery Impacts We examined each scenario for its ability to support various levels of in-river fisheries consistent with court adjudicated tribal treaty rights. We assumed total runs above Bonneville Dam would be 50,000-128,000 spring chinook and 400,000 fall chinook within seven years and 200,000 spring chinook and at least 400,000 fall chinook after 25 years. Harvest rates described in the CRFMP consistent with these run sizes were used.
Supplementation The recommendations in this plan are intended to support healthy naturally reproducing salmon populations throughout the Columbia Basin for populations of average productivity or higher. We realize, however, that salmon populations with lower than average rates of production are also worthy of protection and long-term persistence, but may not be able to maintain themselves by natural reproduction alone. Also, it may not be possible to achieve all the objectives of this plan using only natural reproduction. We believe hatchery technology can play a significant role in the following cases:
- To reduce the risk to those populations most in danger of short term extirpation;
- To restore production to areas where the original population has been extirpated;
- To provide partial mitigation for populations where society is unable or unwilling to take necessary preservation actions in other parts of the salmon’s natural ecosystem;
- To maintain individual populations with much lower-than-average productivity over the long term; and
- To replace production lost because of permanent blockage or alteration of former habitat.
We estimated the potential effect of supplementation by examining the demographic impacts on a single hypothetical population of 1000 fish. We used CRFMP harvest rates as described in Table 5C.2, 70% egg-to-smolt survival in the hatchery, and equal survival of hatchery and naturally produced smolts. In addition, the model addresses only a single generation of salmon and does not account for effects over multiple generations.
Survival rates for all life history stages under all the management scenarios were calculated for spring chinook and fall chinook. Only the effects of option 4c for the 25-year scenarios are presented in further analyses for the sake of simplicity and because that is the preferred tribal drawdown option. The impacts of the other two 25-year options are similar but of somewhat lesser magnitude.
Spring Chinook under the NMFS Plan
Comparing historical and recent survival rates (Table 5C.2) shows that nearly all of the change has occurred in the tributary habitat (egg-to-smolt survival) and mainstem passage portions of the life cycle.
The proposed NMFS recovery plan fails all three of the comparison measures described above (persistence, recovery, and fishing). First, it does not meet the proposed interim standards for short-term persistence and long-term recovery described by the State and Tribal Fishery Agencies Analytical Team (STFA 1995a; 1995b). Eighty percent of the modelled populations failed to meet the persistence standard and none of the populations achieved the recovery level over the next 48 years.
Second, the NMFS recovery plan produces a relatively small increase in the rate of rebuilding (27% greater than recently). It could be difficult to verify that a benefit of this magnitude actually resulted, given the amount of natural variation in the system. In most cases and over short periods, tests to measure changes in overall survival have low statistical power (DeLibero 1986; RASP 1992b). It has been estimated (Lichatowich and Cramer 1979) that it could take 20-30 years of monitoring to have an 80% chance of detecting even a 50% change in overall salmon survival, given the level of natural variation. At best, rebuilding would be slowest under the NMFS plan.
Finally, the NMFS plan does not provide for reasonable fishing levels consistent with recognized treaty rights. Snake River runs are one component of the spring chinook run above Bonneville Dam. Regardless of their actual abundance in a particular year, it is desirable that all population components of the run be able to sustain a fishery consistent with the cultural and treaty requirements of the tribes. The NMFS plan would produce a S/S ratio of 0.78 using the tribes’ 25-year harvest rates, for instance.
Spring Chinook under the Tribal Plan
The proposed tribal restoration actions, in contrast, meet all three of the evaluation measures. First, eighty percent of the modelled populations meet both persistence and recovery standards proposed by the state and tribal fishery managers (STFA 1995a). Second, the spawning index is sixty percent greater than recently in the short term, and is over six times greater after 25 years. This projected level of change would be easier to identify with a modest monitoring program and to incorporate results into an adaptive management decision process. Finally, these populations could support increased harvests consistent with reserved treaty rights as Columbia Basin salmon runs are restored.
Current survival rates (egg-to-adult) for Snake River spring chinook are so low that the populations are not replacing themselves. Implementing the actions under the tribal plan would increase the survival rate for naturally reproducing Snake River spring chinook 60% in seven years and over 6-fold in 25 years. Figure 5C.1 compares the present and projected survival rates and the necessary survival rate for replacement.
The rate of growth of the population will determine the time necessary for recovery. Figure 5C.2 compares the estimated population rebuilding rates for naturally reproducing Snake River spring chinook under existing conditions, the NMFS Proposed Recovery Plan, and the tribal plan. Under existing conditions the populations will continue to decline. The estimated effects of the NMFS Plan are for a slow rate of increase, while the estimated growth rate under the tribal plan is about five times greater. It must be noted that for the purposes of this example that all proposed actions are assumed to be in place immediately and density dependence is handled through a simple logistic function based on a target escapement goal. In reality, actions would be phased in over time under all plans and population growth will depend on a number of factors including environmental variation.
The analysis thus far has dealt with actions which affect all Snake River populations above Lower Granite Dam. The benefits are described in terms of an average population. Small populations in poor habitat are at greater risk of extirpation and may require additional help if they are to survive. Active habitat restoration and short-term supplementation projects are two methods for addressing the needs of these individual populations.
While the tribal plan proposes supplementation as a means of rescuing imperiled populations or reintroducing populations to barren habitat, supplementation can also compensate for lower than desired natural survival rates for individual populations.
Figure 5C.3 shows that, in combination with all other actions in the NMFS recovery plan, populations would have to be supplemented at a rate of 35% or 50% to equal the benefits of the tribal short- and medium-term strategies, respectively.
Supplementation should not be viewed as a panacea for maintaining salmon in the face of actions that create dysfunctional ecosystems–systems with a mortality schedule that prevents long-term maintenance of any natural production. That approach would necessarily evolve into hatchery-only production without meaningful natural production.
Supplementation was analyzed as potential compensation for the lack of increased survival at certain life-cycle stages and its use yields significant results. Table 5C.3 and Figure 5C.2 describe alternative supplementation scenarios and the effects of these scenarios in a framework that compares the tribal options against existing conditions and the NMFS plan. For instance, assuming that recent conditions were to continue in the future, to achieve the restoration levels resulting from implementation of tribal options on or three, affected Snake River populations would need to be supplemented by 50% and 35% respectively. In other words, to achieve the restoration level that would be provided by option three, 5o% if the returning adult Snake River spring chinook would be captured, reared and released in accordance with the supplementation guidelines of Cuenco et al. 1993.
Because fall chinook spawn primarily in the mainstem Columbia and Snake rivers and move into the ocean during their first year of life, it is more difficult to observe and measure their freshwater production and behavior. Consequently, we have assumed the primary change has been not in the rate of survival from egg-to-smolt, but in the amount of spawning area available, which has been drastically reduced by the construction of mainstem dams. Although it is likely that presmolt survival has decreased due to the cumulative effects of development, we did not have information to estimate the magnitude of that change.
Comparing historical and recent survival rates (Table 5C.2) shows that the major changes have occurred in the mainstem passage and harvest portions of the life cycle. Harvest rates shown for recent years are reduced from previously higher levels.
As with spring chinook, NMFS’ proposed scenario does not meet the survival and recovery standards (STFA 1995b). Unlike spring chinook, however, there are large increases in the spawner-to-spawner survival, even with short-term harvest rates as described above. Tribal Option 3 meets the survival and recovery standards, incorporates restoration of viable fisheries, and provides significant increases in spawner indices (Table 5C.2). Supplementation of Snake River fall chinook may be appropriate in the short term to provide additional buffering against environmental variation, but does not appear to be needed in the longer term under tribal scenarios.
Since so much of the fall chinook’s former mainstem habitat has been lost, permanent hatchery programs will be required to replace this lost production.
Other Salmon Populations
The above analyses are directly applicable to Snake River spring/summer chinook and fall chinook populations above Lower Granite Dam. We can say qualitatively, however, that many other salmon populations in other locations will also benefit to varying degrees from actions called for in this plan. All populations migrate through the lower Columbia River above Bonneville Dam and will benefit from improved passage conditions in that area. Additionally, tribal habitat protection measures will apply basin-wide and will increase egg-smolt survival in many damaged areas. The need for active habitat restoration projects and/or supplementation should be judged on a case-by-case basis.
Lamprey and Sturgeon
We have not attempted to quantify the benefits of this plan for lamprey and sturgeon populations above Bonneville Dam. Qualitatively, both populations will benefit from restoration of conditions under which they evolved. Improved mainstem passage conditions will also improve lamprey migration. If natural river levels are reestablished in the Snake River, this will restore important sturgeon habitat. Improved tributary habitat, especially in the lower gradient portions of mainstem tributaries, will increase the amount and quality of habitat needed by juvenile lamprey.
Salmon populations throughout the Columbia Basin are declining. All areas of the basin and all species are affected to some degree, although the problems are most critical in the Snake River portion of the basin.
This situation is the result of decades of fragmented use, development, and management of the natural resources in the Columbia Basin. Decisions have been made without proper thought for their impacts on anadromous fish, whose ecosystem encompasses tributary watersheds, mainstem portions of the Columbia and Snake rivers, and the North Pacific ocean. Consequently, the natural buffering and compensatory mechanisms, which allowed salmon to persist in times of environmental stress and thrive in times of average and good conditions, are now des-troyed, and salmon populations are being extirpated at an alarming rate.
Future management of these culturally and economically valuable anadromous fish must take a more holistic view, integrating decisions across all stages of their life cycles. The bottom line in the tribal approach toward restoring anadromous fish populations is to increase their total survival rate to the point where we can maintain healthy naturally reproducing populations which can survive natural cycles of environmental variation and allow full exercise of fishing rights reserved by our tribes in treaties with the United States government.
Toward this end, we have identified survival standards for each life-history stage of spring and fall chinook (Table 5C.2) and actions which we feel will achieve these standards. We will work toward achieving these standards in an adaptive manner, realizing that present knowledge is imperfect and changes in this plan will be necessary over time, and that trade-offs can be made in the path we take, so long as the long-term objectives are met. We will hold ourselves and the people of the Columbia Basin accountable to the same rigorous standard:
We must increase total survivals to the point that anadromous fish can maintain healthy naturally reproducing populations which allow full exercise of treaty reserved fishing rights.