Fish & wildlife Independent science ISAB documents

Ecological impacts of the flow provisions of the Biological Opinion for endangered Snake River salmon on resident fishes in the Hungry Horse, and Libby systems in Montana, Idaho, and British Columbia

March 4, 1997  |  document ISAB 97-3

read full document > (1.7m PDF)

[The HTML version below doesn't include appendices or figures.]

I. Introduction

Background of Assignment

Within the community of biologists in the basin, contention exists with respect to the scientific rationale for late summer flow augmentation in the mainstem Columbia River intended to assist outmigration of endangered Snake River fall chinook, as called for in the Biological Opinion (BiOp). Contention primarily exists with respect to three concerns:

1. Flow augmentation in August and September is provided by deep drafting of Hungry Horse and Libby reservoirs in the headwater reaches of the Columbia River in Montana, leading to negative effects borne by residents there (Figures 1, 2 and 3). Such drafting can also impact refill schedules, leading to negative effects with basin- wide implications. 2. Summer drafting, following on the heels of deeper flood control drafting in early spring, might cause significant negative impacts on the ecology of both reservoirs, as well as lakes and river segments downstream to the target reaches in the mainstem , and; 3. Flow augmentation in the mainstem during summer and fall (normally low flow periods) may or may not significantly benefit endangered fall chinook.

Litigation has been threatened or pursued owing to a lack of resolution of these concerns by policy makers. People in the headwater areas believe that the impacts of drawdowns on resident fisheries are substantial and not warranted on the basis of a presumed weak or non-existent flow-survival relationship for endangered fall chinook using the mainstem. On the other side, the Biological Opinion (BiOp) concluded the flows were needed because slow water movement in the lower Columbia River, and high water temperatures at that time of year negatively impact the endangered fish. Additionally, downstream constituents want the BiOp implemented as called for under the authority of the Endangered Species Act.

Statement of Assignment

In a March 27, 1996 letter from Donna Darm (National Marine Fisheries Service) to Chip McConnaha, ISAB Coordinator, the Independent Scientific Advisory Board (ISAB) was asked to address two questions.

1. Does available evidence/analysis demonstrate that resident fish populations are at risk of extinction in Libby Reservoir? In Hungry Horse Reservoir? If yes (or maybe): 2. Does available evidence/analysis demonstrate that the Biological Opinion operations increase the risk of extinction of resident fish in Libby Reservoir? In Hungry Horse Reservoir?

Rather than providing narrowly focused answers to these questions, we approached our review from an ecosystem perspective. In our proposal of September 27, 1996, we stated our objective to "Evaluate the type and extent of ecological impacts of the flow provisions of the Biological Opinion on resident fishes in the Hungry Horse/Libby systems, including the potential of the provisions to increase the probability of their extinction". Specifics of the ISAB review process are presented in Appendix 1. The ISAB noted that the two questions are couched in the language of the Endangered Species Act and reflect the region's ongoing policy struggles in this area. They are not forthright scientific questions, when considered in an ecosystem context. The questions focus on extinction of reservoir fishes, which undoubtedly have persisted, and in some cases likely even proliferated, in these reservoirs since they were built, rather than focusing on the negative impacts of headwater drafting on the ecology of all upstream waters affected by the drafting. Finally, these questions explicitly do not ask for an evaluation of any flow-survival relationship for fall chinook in the mainstem, the lack of which could compromise the rationale for drafting headwater reservoirs. To be most useful to policy makers, the questions should ask whether there are negative effects of the August flow augmentation that would require balancing against benefits to endangered Snake River salmon. Unfortunately, we were not able to develop information for this report on benefits to Snake River salmon, because of the limits of time and the complexity of the problem. We are continuing to pursue the question in other contexts, such as our participation in design of studies to address the NWPPCs mainstem flow hypothesis and our participation in the PATH process. Our analysis in this report therefore focuses on the effects of reservoir drafting on resident fishes in the upstream areas. We are sensitive to the fact that the result of approaching the problem from this perspective is that it may lead to an impression of a one-sided point of view in our response, which is not intended.

II. ISAB CONCLUSIONS

Overall Conclusions of the ISAB Review

(A) The biological effects of the summer drafting are not likely to drive resident fish populations to extirpation in Hungry Horse and Libby reservoirs. However, the two questions posed are overly simplistic. Certain resident fishes (Kootenai River White Sturgeon and Bull Trout) in the Hungry Horse and Libby systems are at some risk of extinction.

(B) There is a complex interaction of the requirements for flow augmentation in the BiOp for endangered Snake River salmon with requirements for flood control, power production, irrigation, provision for resident fishes, including endangered Kootenai River White Sturgeon, as well as recreational pursuits and aesthetics, and associated effects on reservoir elevations and fluctuations in river flow in the Hungry Horse/Libby systems. This interaction may be described as a collision of objectives (Ripley, 1971).

(C) Drawdown of the reservoirs adversely affects resident fishes, including adfluvial populations. Flow augmentation in August leads to increased flows in the streams and lakes below the two reservoirs and adversely affects resident fish populations to the points where the streams join the Columbia River. Increased flows also make possible wider fluctuations in daily and weekly flows as the power system follows the load, and these fluctuations in turn adversely affect resident fishes.

(D) Although the primary effects of drawdown in August are different from those in late winter and spring, they can not be viewed as separate actions, because there are interactions .

(E) These adverse effects on resident fishes are among the trade-offs that must be considered by policy makers for water management in the region, as noted by Wright, (1996)

(F) Integrated Rule Curves (IRCs) offer policy makers a tool to evaluate and optimize the trade-offs associated with decisions on river reregulation that can take into account varying objectives in water management and uncertainties in supply. Model evaluations of the IRCs have been used to explore the interactions between effects of flows called for by NMFS in the BiOp for endangered Snake River salmon on resident fishes upstream, as well as on flood control, power production and other factors (Marotz, et al.., 1996). While various flow scenarios have been modeled in this way, the quantification of benefits to Snake River salmon would require use of another model.

(G) The question of benefits of August flow augmentation to endangered Snake River salmon is a complex one with information that is subject to more than one interpretation. The ISG concluded that a flow-survival relationship remains to be demonstrated (ISG 96-6). The question is being pursued by the ISAB in other contexts, such as our participation in design of studies to address the NWPPC's mainstem flow hypothesis and our participation in the PATH process.

III. FINDINGS

1. Description of the Systems and Their Operations

Hungry Horse and Libby dams are located in Montana, Figure 1. Their reservoirs are drafted in late winter or early spring to elevations usually 85 feet and 95-110 feet below full pool respectively to provide space for spring runoff as a flood control measure, as well as storage space for later power production, irrigation, and many other water uses, Figures 2 and 3. The flow augmentation provision of NMFS BiOp for endangered Snake River salmon leads to drawdown in August to elevations 20 feet below full pool, as a maximum. A more complete description of the Hungry Horse and Libby systems and their operations is found in Appendix 2.

Operations at Hungry Horse and Libby dams interact with operations of projects downstream. There are five other dams downstream of Hungry Horse Reservoir in what might be referred to as the Hungry Horse system, before it enters the Columbia River. In the Libby system, there is one dam, Corra Linn Dam, below Libby Dam. Duncan Dam impounds the Duncan River which is a side stream entering Corra Linn Reservoir (Lake Koocanusa). Further information on downstream operations is given in Appendix 3.

2. Risk of Extinction of Resident Fishes in the Reservoirs and Systems

Studies relating to the risk of extinction of resident fishes that occur in the Hungry Horse and Libby systems include one of Kootenai River white sturgeon which resulted in listing under the ESA in 1994 (USFWS, 1996), and several studies of bull trout (Thomas, 1992; Montana Bull Trout Scientific Group, 1996A, 1996B, etc.). Bull trout are thought to be at high risk of extinction throughout much of their range (Rieman and McIntyre 1993, 1995, 1996). The studies by Thomas and the Montana Bull Trout Scientific Group are specific to Montana, and include information on factors thought to affect the risk of extinction of bull trout in the waters of the Hungry Horse and Libby systems. The study by Thomas (1992) was conducted prior to implementation of the BiOp for Snake River salmon and therefore does not include information directly bearing on whether implementation might be expected to lead to an increased likelihood of extinction. The series of bull trout studies by the Montana Bull Trout Scientific Group generally point to the lack of quantitative information on historic abundance of bull trout.

The Kootenai River study identified a need for further study of the effects of dam operation on bull trout. Neither the bull trout studies nor the Kootenai River White Sturgeon study can be used as they stand to assess whether or to what extent the NMFS Biological Opinion operations might lead to an increased probability of extinction of these resident fishes. The Kootenai River White Sturgeon Draft Recovery Plan includes provisions for maintaining minimum and stable flows during the spawning season in July, as well as other measures. The IRCs are cited in the draft Recovery Plan as the best available guidelines for operation of Libby Dam to balance requirements for white sturgeon with the needs of other species. The plan notes that the IRCs could be affected by NMFS Section 7 requirements relative to flows for listed Snake River salmon (USFWS, 1996, p.46).

Further information on both the bull trout and white sturgeon is provided in Appendix 4.

3. Adverse Effects of Drawdown.

On the basis of experience elsewhere, as reported in the scientific literature, adverse effects on resident fish populations are to be expected from reservoir drawdown. A brief review of the literature is provided in Appendix 5. Drawdown is a recognized management technique for reduction of some fish populations considered to be undesirable (e.g. Bennett, 1954; Shields, 1957; EPA, 1996).

While we could find no evidence that deep summer drafting of Hungry Horse and Libby reservoirs as called for in the BiOp will lead to extirpation of resident fishes in the reservoirs, studies in the reservoirs and rivers have demonstrated that deep drawdowns adversely affect the ecological dynamics of the reservoirs and their food webs, which in turn is likely to adversely affect the resident fish populations (see Appendix 4 for detailed discussion).

Hungry Horse Reservoir contains fish populations that are of particular significance, because the full complement of species native to the headwater reaches of the upper Columbia River are present. There have been few introductions of non-native fishes upstream of Hungry Horse Dam, and the management objective is to reduce or eliminate them. (Brian Marotz, personal communication) The construction of the dam has prevented non-native fish stocked elsewhere in the basin from invading the South Fork of the Flathead River drainage. Consequently, Hungry Horse Reservoir and the South Fork of the Flathead River above it, may serve as an important reserve for both westslope cutthroat trout and bull trout. Populations of both of these species appear to be stable in the Hungry Horse and Libby systems, based on annual net surveys for cutthroat in the Hungry Horse system (Brian Marotz, pers. comm.) and redd counts of bull trout in spawning tributaries (Montana Bull Trout Scientific Group, 1996A, 1996B). Elsewhere in the northwest, both of these species are believed at risk of local and regional extinction (Rieman and McIntyre, 1996). Hungry Horse Reservoir and its main tributary, the South Fork of the Flathead River would be a likely candidate for special protection as a native salmonid reserve owing to the presence of healthy core populations of bull trout and westslope cutthroat trout. This concept is discussed in greater detail in the ISG's recent report (96-6), Return to the River, which discusses the role of and need for salmonid reserves in the preservation and recovery of Columbia River salmonid stocks. The fish populations in Libby Reservoir have a lower conservation profile than those in Hungry Horse, but support a valuable sport fishery, especially the kokanee and cutthroat trout.

Limnological studies were conducted in the Hungry Horse and Libby Reservoirs in the late 1970s and 1980s with Bonneville Power Administration funding in response to the NWPPC Fish and Wildlife Program. The research described seasonal productivity of the food web in relation to drawdown and refill of the reservoirs and population dynamics and growth of the fish taken in the sport fisheries. The studies at Hungry Horse Reservoir focused on factors affecting abundance of westslope cutthroat trout, bull trout, mountain whitefish, northern squawfish, largescale suckers and longnose suckers, the most abundant fishes in the reservoir (May et al., 1988). In Libby Reservoir (Lake Koocanusa), those factors were considered in their effects on peamouth (most abundant), suckers, northern squawfish, yellow perch, kokanee, mountain whitefish, bull trout, and other salmonids, in descending order of abundance (Chisholm, 1989).

Summer and fall growth periods for these fishes were driven mainly by abundance of zooplankton and benthic midges, although terrestrial insects were of considerable importance to cutthroat in Hungry Horse. Availability of all of these forage sources were found to be influenced by temperature seasonality and interannual drawdown schedules. In summary, the reservoir environments were more productive and fish grew faster when the reservoirs filled early and were not deeply drafted in the summer (Fraley 1986). While the impetus for these studies arose from the drawdowns in late winter and early spring, their results can be applied as principles to the August drawdown. August drawdown affects a somewhat different complex of invertebrates, leading to reductions in the food supply for resident fishes at a critical time for growth. Because those invertebrates have a life cycle extending for more than one year, their reduction in the fall carries over into the spring, exacerbating the changes brought about by spring drawdown (Marotz et al, 1996). Further details are given in Appendix 5.

4. Effects Downstream of the Projects

Storage of water in the two reservoirs has resulted in reduction of spring flows and increase in flows the rest of the year, accompanied by the ability to follow power loads in the region on a daily basis, leading at times to rapid changes in flow over short time periods (Stanford and Hauer, 1992). For example in the Kootenai River, August flows from Libby Dam since 1982 have created late summer flows outside of the previously observed ranges of both maximum and minimum flows recorded in the 71 years prior to closure of Libby Dam (Figures 4 and 5). Increased flows under the BiOp might heighten this problem.

Loss of seasonality of flow (i.e., no scouring spring flows to maintain quality salmonid habitats... see Return to the River; ISG Report 96-6) occurs in both river systems, but is less pronounced in the Flathead system because of the confluence of the unregulated flows from the North and Middle Forks of the Flathead River with the regulated flows in the South Fork only a few miles downstream from Hungry Horse Reservoir.

Stanford and Hauer (1992) reviewed effects of altered hydrographs caused by Hungry Horse and Kerr Dams on the ecology of the Flathead River system. The main problem is that an unusually wide varial zone (i.e., that portion of the river bottom that is alternately flooded and dewatered by flow fluctuations associated with drafting and with daily variations due to load following by the power system) exists in the regulated segments of this river. Many important organisms cannot survive in this wide varial zone.

In the 1980s the effects of regulation on river ecology in the Kootenai were studied extensively (Perry 1984; Perry and Perry 1986; Perry et al., 1986; Perry and Perry 1991). As a result of the changes in river conditions, benthic communities survive to some extent in the Kootenai River only in the deepest parts of the channel (thalweg). Recent studies show that the area of the varial zone is increasing and the diversity and productivity of the benthic food web has declined in relation to the studies in the mid-1980s (F.R. Hauer, Flathead Lake Biological Station, personal communication). Loss of habitat and food web integrity extends all the way to Kootenay Lake and is correlated with the decline of the Kootenai River white sturgeon. No recruitment of white sturgeon has been observed in the river since Libby Dam was closed, leading to the listing of the species under the Endangered Species Act (USFWS, 1996). Species that need resting and feeding habitats in the near shore area, like young-of-the-year fishes, either are left high and dry or washed downstream, as a result of the rapid daily fluctuations in flow.

For many years, summer drafting from the bottom of Hungry Horse Reservoir also caused rapid and extreme (> 10oC in a few seconds of time) temperature changes. That problem was solved by the retrofitting of the dam in 1996 with a depth-selective release structure. Temperature in the tailwaters in 1996 was designed to follow seasonal norms that occurred in the unregulated Flathead River. While this thermal regulation can be expected to produce positive results with respect to the thermal energetics of river biota, the fluctuating flows remain a problem.

The recent summer drafting associated with the BiOp flows has produced summer high water conditions in the rivers downstream that previously were not experienced by resident fishes. Summer drafting of the reservoirs causes high and variable flows in the tailwaters and river reaches downstream through Flathead and Kootenay lakes at a time of year when this is not normal. Adverse effects are likely to occur. These river reaches contain important resident fishes, including the endangered white sturgeon in the Kootenai River and rapidly declining Flathead Lake bull trout. Several detailed studies have demonstrated the adverse effects that hydropower operations have had on the food webs of these rivers (Perry and Perry 1986; Spencer et al.. 1991; Stanford and Hauer 1992).

When the flows pass through Flathead Lake, they affect the operations of Kerr Dam on the outlet of the lake. The operators of Kerr Dam are required by the terms of their FERC license to hold Flathead Lake at full pool from mid June until October. Hence, summer drafting at Hungry Horse has to be accommodated by high discharges from Kerr Dam. This introduces a problem, because the operators of Kerr Dam are facing implementation of a recent FERC order to baseload the powerhouse, rather than follow power loads. The problem is that if the operators at Hungry Horse Dam follow the load during the August flow augmentation period, operators at Kerr Dam may be forced to do likewise.

Flathead Lake is also of concern with respect to summer drafting of Hungry Horse. Long term limnological studies on Flathead Lake have related declining water quality to anthropogenic loading of nitrogen and phosphorus. Because mass flux of plant growth nutrients through Flathead Lake is controlled by the rate of exchange of water, water quality in the lake is directly influenced by both Hungry Horse and Kerr Dam operations (Stanford and Hauer 1992; Stanford et al.. 1994). The State of Montana has initiated an aggressive nutrient control strategy for Flathead Lake (Flathead Basin Commission 1994) and the US Environmental Protection Agency currently is in the process of implementing a Total Maximum Daily Load allocation under authority of the Federal Clean Water Act. These control strategies are affected by summer drafting of Hungry Horse Dam for BiOp flows. The specific effects of those operations are unknown at this time.

The Integrated Rule Curves (IRCs) for Hungry Horse Reservoir, described below, provide rational guidance for minimizing this problem (Marotz et al., 1996).

5. Models of Effects of Dam Operation on Resident Fishes

From the limnological studies in Hungry Horse and Libby reservoirs and their immediate environs, mathematical models were developed for use in describing the biological systems in the reservoirs (Fraley et al., 1989; Marotz et al., 1996). The models can be used to illustrate the probable extent to which drawdowns, failure to refill, and other water management options are harmful or beneficial to aquatic life in the reservoirs and downstream within the boundaries covered. Resident fishes incorporated in the models were cutthroat trout in Hungry Horse Reservoir and kokanee in Libby Reservoir. Further information on the models is provided in Appendix 5. The models have been reviewed in their particulars by Andersen, (1991), and Swartzman, (1995). The ISAB itself has not reviewed the mathematical basis of the models. While the models may not provide precise quantitative results, we believe they produce reasonable results in agreement with studies on effects of drawdown elsewhere.

6. Balancing Downstream Flow Needs With Resident Fish Needs: Development of the Integrated Rule Curves

The biological models were used as a basis for developing Integrated Rule Curves (IRCs) (Fraley et al., 1989; Marotz et al., 1996; Marotz in prep., and briefing to ISAB). The IRCs are a family of operational rules for dam operation that incorporate incremental adjustments to allow for uncertainties in water availability. Their use is similar to the use of flood control and power rule curves. The IRCs are themselves FORTRAN models that simulate the physical operation of the dams including the water budget and downstream flood concerns, and predict the resulting thermal structure of the reservoir and tailwater operation, which in turn predict biological responses (Marotz et al., 1996). The IRCs can be used to specify flows and to examine their effects on the interrelated requirements for water uses, such as those called for by NMFS in the Biological Opinion for endangered Snake River salmon and resident fish upstream, as well as the requirements for flood control, power production and irrigation. The IRCs have the flexibility to incorporate new information as it becomes available. Consequently, they offer a tool for use by policy makers in balancing the needs of salmon downstream against the need to maintain high pool elevations to enhance reservoir productivity for the benefit of resident fishes, as well as the needs of other water users. A recent study conducted for NMFS included a set of findings of a Steering Committee which reported that "There are legitimate biological trade-offs relating to flow augmentation for migrating salmon vs. protection for resident fish populations in reservoirs" (Wright, 1996).

Iterations of the IRCs are currently being exchanged between affected parties including the State of Montana, the Confederated Salish and Kootenai Tribes of Montana, and Bonneville Power Administration. Communications are taking place with the Corps of Engineers with respect to flood control rule curves in the two systems.

Similar analyses of water release schedules are occurring elsewhere in the basin. Geist et al.. (1996) developed a spreadsheet simulation model to examine drawdown alternatives and their respective tradeoffs in benefits and costs among river uses/resource groups, among species, and among reservoirs..

The Northwest Power Planning Council adopted the IRCs in the 1994 Fish and Wildlife Program, but the concept was not implemented in 1995 or 1996 because operators began to implement the NMFS BiOp for endangered Snake River salmon (Marotz, in prep.). The differences in operations would be substantial during the summer after July. Under both alternatives refill is normally achieved by August 1. The BiOp requires drafting the reservoirs, usually 20 feet, during August, while the IRCs would maintain reservoir elevations near full pool through September (Marotz et al., 1996).

7. Recommendation

Identification and assessment of the effects of water allocation actions associated with the Biological Opinion are among the region's central management and research questions. NMFS and BPA employed a facilitator in 1995 to attempt to arrive at an agreement among federal, state and tribal parties on modifications to the operations called for in the BiOp (Wright, 1996). We agree with the Steering Committee Findings in that report that, "The need for and level of August flow requirements should be one of the region's top salmon monitoring and evaluation priorities." We recommend that NMFS undertake an assessment of the tradeoffs between the benefits of spring and summer flow augmentation for Snake River and other salmon, and the accompanying effects on other water uses, including resident fishes. In the meantime, decisions by policy makers will have to be based on the best information currently available. The ISAB can assist in identifying that information, and in identifying studies needed to improve the information.

^ top

Use Adobe Reader to view PDF documents