Investigation of losses of juvenile salmonids to non-indigenous predators in the FCSRPS is called for by the 2008 BiOp, the NPCC Fish and Wildlife Program, and the Adaptive Management Implementation Plan (AMIP). The significance of this work to the BiOp objectives and regional plans was well supported. This project has provided important new information on the early life history of Snake River fall Chinook salmon. Over-wintering of juvenile fall Chinook salmon in the hydrosystem reservoirs was shown to be a viable life history strategy. This realization has influenced decisions regarding operation of the juvenile bypass systems at Snake River dams and also the transportation program for juvenile fish. Much remains to be learned about the implications of the over-wintering life history strategy for management operations, including bypass, transportation, spill, and summer flow augmentation. Results to date have shown that subyearlings migrating from spawning areas in the Clearwater River migrate rapidly in the free-flowing portion, but then slow and delay in the area above the confluence with the Snake River (transition zone). Survival through the confluence area is low, but the survivors grow to relatively large sizes and apparently survive well during subsequent migration and marine life. The net benefit for survival through the entire life cycle has yet to be determined. They also hypothesize that predation is higher now than 15 years ago, especially in summer. This project would produce needed information on current losses of juvenile fall Chinook to predators, updating and expanding on studies done over 20 years ago. Understanding of how predation, gas-bubble disease, and temperature interact to affect survival of reservoir-type fall Chinook salmon would be useful for management of the ESU. This project is a collaborative effort between the USGS, USFWS, NOAA, PNNL, and the University of Idaho. The activities funded by this proposal would not duplicate other efforts. The relationship of this project to other research efforts in the upper Snake River mainstem, particularly the other predation study (199007700), was clearly described. Project proponents made a good effort to differentiate the activities outlined here from other studies. This was a well-prepared proposal that meets scientific criteria in most respects. The project has contributed to our knowledge of life history diversity in upper Snake River fall Chinook salmon, and its record of publication in peer-reviewed journals is exemplary. The questions being asked (role of predation in limiting survival of overwintering reservoir-type Chinook parr, importance of gas bubble disease, and the influence of water temperature) are difficult to answer in such a large aquatic ecosystem, and the project has shown that it can successfully carry out large-scale studies and reach scientifically supported conclusions. For the most part the deliverables, work elements, metrics, and methods were adequately described. The technical background was adequately summarized. The objectives, in general, were clear although a little more background on how the results of this project could alter transportation strategies for fall Chinook would have been helpful. Project proponents have done a good job of differentiating this project from the other large Snake River fall Chinook project (199102900). The personnel have experience in the work elements and are well-qualified to conduct the work. A response is requested on a few points. A more thorough discussion is needed of why smallmouth bass and channel catfish were selected as the predators of interest out of the suite of potential animals feeding on fall Chinook parr in the lower tributaries and reservoirs of the Snake River. It was not clear how predation from other fish species or birds were to be accounted for in the study. Second, a more complete description of the predation-trial chamber is needed, as well as the methods used to prepare the fish for the feeding trials. How well does the chamber simulate the lower river-reservoir transitional area habitat (e.g., will there be any cover in the chamber for Chinook parr that might emulate what they could use along the shoreline)? According to the proposal, predators will be habituated to juvenile salmonids as food before each trial. Could this cause them to form an unnatural search image for parr, even if alternative prey were present? A hungry predator used to eating small salmonids will likely concentrate on them in an enclosed environment, so what steps will be taken to ensure that the importance of predation will not be overestimated in this experimental setup? Third, some clarification of the expected resolution of the otolith microchemistry work is needed. The otolith microchemistry component would provide a tool to differentiate between natural-origin and hatchery-origin adults returning to the Snake River. This is a necessity in order to determine if ESU recovery goals are met. This tool would potentially also allow determination of the overall life-cycle success of hatchery and natural fish and of different life-history types. In theory, this approach would provide more complete data than available for a smaller number of fish using radio telemetry and would be especially useful for fish too small to radio tag. It is not clear, however, that the methods will produce the desired level of resolution. Hatchery/wild differences may be detectable, but will reservoir versus river or tributary rearing be detectable? Studies in other places have shown that resolution can be an issue, and there is the danger that this technique is being oversold at its present level of resolution and sophistication. Some evidence based on other studies that the desired resolution is obtainable is requested. If the proposed work is highly developmental and evidence of adequate resolution is not available, they should perhaps get a baseline of chemistry in the various habitats before launching into a full-blown investigation. Perhaps a one-year feasibility study should be considered, that if shown to work could be continued and expanded. A limitation that should be further addressed is the assumption that returning adults that have no fin clips or hatchery-implanted tags are of natural origin. What percentages of hatchery-produced fry in various years have had no fin clips or tags? Finally, a summary of how the results of the work could specifically affect transportation strategies will make the project’s relevance and importance more apparent. A little more explanation of how the results of the studies will influence transportation strategies would be useful. The importance of understanding migration timing and the apparent effect of a delayed reservoir-type life history on the SAR parameter was clear, but how will this information be used to adjust barging activities?

ISRP “A more thorough discussion is needed of why smallmouth bass and channel catfish were selected as the predators of interest out of the suite of potential animals feeding on fall Chinook parr in the lower tributaries and reservoirs of the Snake River. It was not clear how predation from other fish species or birds were to be accounted for in the study.”

Response:  Our study would ideally provide information to manage native predators (e.g., northern pikeminnow and birds) and target removals of smallmouth bass and channel catfish, while considering background management practices and the food web (e.g., Harvey and Karieva 2005).  Though we suspect that an ongoing effort by the ISAB will bring forth the importance of the food web in ongoing research and management, this effort is still in progress and was not available as a basis for proposed research in the present categorical review.  Thus, we selected the species of study for our project based on existing regional programs, the amount (or lack thereof) of baseline research, and our desire to reduce our project’s expenditures but still produce useful information.

We selected to study smallmouth bass and channel catfish because the 2008 Biological Opinion, the FCRPS Adaptive Management Implementation Plan (AMIP), and the Council’s 2009 Amendments to the Fish and Wildlife Program, specifically call for predation research on smallmouth bass and channel catfish.

We chose to study smallmouth bass because our experience in the field suggests that predation by this species has likely become a factor for mortality of fall Chinook salmon subyearlings in recent years and there is baseline research we can build on to test this theory.  Naughton et al. (1994) studied smallmouth bass in 1996 and 1997 and estimated that: (1) abundance of smallmouth bass in the Snake and Clearwater river arms of Lower Granite Reservoir in 1996 was 11,877 (95% C.I.; 8,818–16,002) and 3,820 (95% C.I.; 2,328–6,283), respectively in 1996; (2) abundance of smallmouth bass in these two areas in 1997 was similar to 1996; (3) subyearling Chinook salmon made up less than 10% of the diet of smallmouth bass in these two locations during 1996 and 1997; and (4) monthly loss of subyearling Chinook salmon in these two locations during 1996 and 1997 ranged from only 552 to 2,672.  It is important to note that these baseline results were obtained during the late 1990s when relative abundance and apparent mortality of fall Chinook salmon subyearlings passing downstream in Lower Granite Reservoir were low.  During 1995–1999 for example, CPUE during seining of fish destined to enter Lower Granite Reservoir and the apparent mortality of these fish after passage through the reservoir averaged 4.4 ± 1.3 (SE) fish per seine haul and 30.0 ± 1.06%, respectively (Connor and Tiffan 2010).  Mean CPUE (48.1 ± 10.7 SE) and apparent mortality (66.6 ± 4.4% SE) increased during 2000–2008 and the relation between these two variables across 1995–2008 took the form of a Ricker curve (Connor and Tiffan 2010).  Collecting additional abundance and predation data for comparison to those of Naughton et al. (1994) will help to determine if overcompensation by smallmouth bass is a factor for density dependent mortality of fall Chinook salmon subyearlings. 

We chose to examine predation by channel catfish because little is known about the abundance of this species in Lower Granite Reservoir and virtually nothing is known about whether this species consumes juvenile anadromous salmonids in the reservoir.  Available information does suggest that predation by channel catfish is possibly a cause for concern.  Zimmerman and Parker (1995) used CPUE as an index of relative abundance and found that channel catfish were slightly less abundant in Lower Granite Reservoir than in lower Columbia River reservoirs including John Day and McNary reservoirs.  Juvenile fall Chinook salmon are relatively abundant these two reservoirs during July (Tiffan et al. 2000).  In John Day Reservoir during July, Poe et al. (1991) found that juvenile anadromous salmonids made up about 30% of the diet of channel catfish.  Vigg et al. (1991) found channel catfish consumed up to 1.85 juvenile salmonids per day in the vicinity of McNary Dam during July.  Harvey and Karieva (2005) modeled the food web in John Day Reservoir and estimated that removing channel catfish would reduce annual predation on juvenile Chinook salmon by approximately 5%.

We did not choose to study northern pikeminnow because there is already a sport reward program in place to reduce the numbers of this predator and we assumed that this program is being monitored and evaluated.  We chose not to study bird predation because the U.S. Army Corps of Engineers is currently examining avian predation in the Snake River under their Anadromous Fish Evaluation Program and we assumed that avian predation will be monitored and evaluated.  Moreover, our study area only includes the Snake and Clearwater arms of Lower Granite Reservoir and a several kilometer stretch downstream.  To our knowledge, avian predation is limited to transient seagulls, a small population of mergansers, and one colony of cormorants.

We take peer-review seriously and ask the ISRP members to share their ideas about the direction of our proposed research after reading this response.  We still have the option of modifying our proposal according to ISRP review and funding any additional work with money we intended to turn back to BPA.  At the very least, we could freeze the stomachs of northern pikeminnow collected during bass and catfish sampling for later analyses.  We could also visit the cormorant nesting site, scan it for PIT tags, and fecal samples for later analyses.   

 ISRP:  “…a more complete description of the predation-trial chamber is needed, as well as the methods used to prepare the fish for the feeding trials. How well does the chamber simulate the lower river-reservoir transitional area habitat (e.g., will there be any cover in the chamber for Chinook parr that might emulate what they could use along the shoreline)”?

Response:  Predation trials will occur in two replicated chambers measuring 5.07-m long, 1.22-m wide, and 0.76-m deep.  Each chamber will be uncovered, will have one-way windows for observers and video cameras to record predation events, and fresh river or well water will be circulated through the chamber at a rate of about 2.5 l/s.  Each chamber will contain a set of same-species predators found at the Clearwater-Snake confluence (e.g., smallmouth bass, channel catfish, or northern pike minnow), the “consumption potential” of the predator groups between tanks will be equal, and the individual predators within these chambers will remain the same throughout all experimental trials.  The number and size of predators will be determined based on the expected cumulative consumption rate of the group during the experiment, so that we are able to make statistical comparisons between time-to-death and number-of-juveniles-consumed during each trial.  Physical structures that simulate the environment at the Snake-Clearwater confluence (e.g., large cobble) will be provided as substrate refuge for juveniles able to seek it.  Enclosed non-natural structures (e.g., PVC pipe; Mesa et al. 1998, Mesa et al. 2002) may also be used to provide prey refuge.  These physical refuges will also be available in acclimation chambers so that salmon can acclimate to their presence before being placed in predation chambers.  Refuges will allow “healthy” salmon to seek refuge during the predation trials and reduce their probability of predation, similar to how they may act in the wild whereas impaired fish will be more subject to predation if they are unable to seek refuge. 

During the 4-week acclimation, predators will be fed both pelletized food (e.g., Nelson and Sons, Inc. Silver Cup floating fish food) and live food (i.e., juvenile fall Chinook salmon) to minimize the number of juvenile salmon needed for the experiment.  Juvenile fall Chinook salmon will be periodically fed (e.g., 1-2 times per week) to predators to maintain their habituation to eating live juvenile salmon (since they will be captured from the wild).  This is common practice among peer-reviewed predation studies using smallmouth bass (Mesa et al. 1998, Mesa et al. 2002, Anglea et al. 2004) and ensures piscine predators do not avoid a food item because of its novelty (Thomas et al. 2010).  During the feeding habituation, predators will be confined to one end of the experimental chamber while live salmon are released into the enclosure, and predators will then be released to consume the live prey.  This will ensure that predators are comfortable eating a known food item in a novel habitat.  During the actual experimental trials, the predators will also be confined to one end of the experimental chamber while juvenile salmon become accustomed to the chamber.  This will give prey a chance to realize the presence of predators and seek out shelter, if they are able, before the predators are released.

ISRP:  “According to the proposal, predators will be habituated to juvenile salmonids as food before each trial. Could this cause them to form an unnatural search image for parr, even if alternative prey were present?  A hungry predator used to eating small salmonids will likely concentrate on them in an enclosed environment, so what steps will be taken to ensure that the importance of predation will not be overestimated in this experimental setup”?

Response:  Because the purpose of the predation-trial experiment is not to estimate predation rates on juvenile Chinook salmon in the wild, but is to evaluate the difference in predation potential between salmon that are exposed to sudden changes in temperature versus controls, we believe that habituating predators to a specific prey is necessary for success of this comparison.  If predators are not habituated to the desired prey item in our experiment, we will not be able to adequately assess the predation probability differences between salmon that undergo temperature extremes at the Clearwater-Snake confluence versus salmon that migrate through a more natural thermal regime.  As mentioned in the previous response, other experiments published in the peer-reviewed literature have similarly fed live Chinook salmon to predators to habituate them for similar experiments. 

ISRP: “Some clarification of the expected resolution of the otolith microchemistry work is needed. The otolith microchemistry component would provide a tool to differentiate between natural-origin and hatchery-origin adults returning to the Snake River. This is a necessity in order to determine if ESU recovery goals are met. This tool would potentially also allow determination of the overall life-cycle success of hatchery and natural fish and of different life-history types.  In theory, this approach would provide more complete data than available for a smaller number of fish using radio telemetry and would be especially useful for fish too small to radio tag. It is not clear, however, that the methods will produce the desired level of resolution. Hatchery/wild differences may be detectable, but will reservoir versus river or tributary rearing be detectable? Studies in other places have shown that resolution can be an issue, and there is the danger that this technique is being oversold at its present level of resolution and sophistication. Some evidence based on other studies that the desired resolution is obtainable is requested. If the proposed work is highly developmental and evidence of adequate resolution is not available, they should perhaps get a baseline of chemistry in the various habitats before launching into a full-blown investigation. Perhaps a one-year feasibility study should be considered, that if shown to work could be continued and expanded.”

Response:  Below we present results from preliminary analyses concerning the resolution of otolith microchemistry data. We highlight two areas: water samples throughout fall Chinook rearing areas and analyses to place fall Chinook (known and unknown origin) to their rearing areas.  We believe these results indicate we will have the resolution to perform the proposed research, and we also believe we will be able to improve the resolution as the research progresses.

Over the past several years, we have collected water samples from a variety of locations (see Figure 1).  Based on these samples, we found that ⁸⁷Sr/⁸⁶Sr varied across the major reaches within the Snake River Basin (ANOVA, F=99.4, p=<0.001). Samples have also been analyzed for a suite of elements to determine elemental ratio discrimination, but those data are still being analyzed and are not presented. Post hoc comparisons of Sr isotope data alone (Tukey’s HSD) distinguished four groupings: 1) upper Snake; 2) lower Snake River; 3) Clearwater and Salmon rivers; and 4) Tucannon, Grande Ronde, and Imnaha rivers (Table 1, Figure 2).

We then analyzed otolith microchemistry in the region corresponding to natal (data not shown) and juvenile rearing (Figure 3) for 120 natural reared adults of unknown origin that were collected at Lyons Ferry hatchery. The range of ⁸⁷Sr/⁸⁶Sr of these fish corresponded to that of the water samples, with near 100% correct classification of 14 known-origin PIT-tagged fish to natal and rearing locations. When we assigned returning adults to the 4 groups (Table 2 and Figure 3), the proportions were roughly consistent with expectations based on redd surveys.  As part of our proposed research, we will continue to undergo these types of validation exercises.

We also note that based on previous research, the chemical signature of the lower Columbia River is vastly different than that of the lower Snake River. Therefore, we are confident that we will be able to detect any protracted residence in the Columbia River. Furthermore, seawater also has a distinct ⁸⁷Sr/⁸⁶Sr and strontium concentrations are much greater in seawater than in freshwater, so we have successfully been able to determine when fish entered the ocean.

An area of ongoing research is the ability to further differentiate among the four natal and rearing groupings we have established. In particular, we want to better spatially define the boundary between the upper and lower Snake River groups. We will need to collect more water and juvenile samples to do this. In addition, we will examine the utility of other isotopic ratios and elemental concentrations, which should also help in distinguishing the Salmon and Clearwater rivers.

 Table 1.  Water sampling ^87/86Sr data for the Snake River basin 2008-2009.

 Table 2.  Natal origin representation for adult fall Chinook salmon captured at Lower Granite Dam (2006-2008). LDFA was used to group fish to source river group based upon ^87/86Sr ratio area of the otoliths corresponding to rearing. Juvenile validation samples consisted of juvenile fish of known origin (Wild=7, Hatchery=9) and were used to test the classification.  

ISRP: “A limitation that should be further addressed is the assumption that returning adults that have no fin clips or hatchery-implanted tags are of natural origin. What percentages of hatchery-produced fry in various years have had no fin clips or tags?”

Response:  We recognize that some hatchery fall Chinook subyearlings and yearlings released above Lower Granite Dam are not marked.  For example, in 2010, a total of 3,858,702 hatchery fish were released above Lower Granite Dam, of which 2,837,349 (73.5%) were marked with AD clips, CWTs, or PIT tags (FPC 2010).  This means that when we sample unmarked adults at Lower Granite Dam, about a quarter of them could be hatchery-origin fish.  In the past, we have used scale pattern analysis to differentiate between unmarked hatchery and natural-origin fish.  We will continue to refine this procedure as well as use the reference chemical signatures from the known juvenile populations (hatchery and natural) that we propose to sample to confirm the origin of all adults sampled at Lower Granite Dam.

ISRP: “Finally, a summary of how the results of the work could specifically affect transportation strategies will make the project’s relevance and importance more apparent.  A little more explanation of how the results of the studies will influence transportation strategies would be useful. The importance of understanding migration timing and the apparent effect of a delayed reservoir-type life history on the SAR parameter was clear, but how will this information be used to adjust barging activities?”

Response:  Two pieces of information are needed to fully answer the question about transportation strategies.  The first relies on the findings of the ongoing COE-funded Snake River fall Chinook transportation study.  That study aims to determine the efficacy of in-river migration to that of transportation by comparing SARs.  The results of that study are sure to be contentious, and in all likelihood the “spread-the-risk” approach to fall Chinook transportation and in-river migration will continue.  However, our proposed work will specifically show which subpopulations are contributing most to SARs that may influence decisions to alter transportation strategies.  For example, this project has shown that most juvenile fall Chinook salmon produced in the Clearwater River do not reach Lower Granite Dam to be transported during the summer (Tiffan et al. 2010).  Rather, these fish begin passing in late September and October.  In fact, subyearlings transported from Lower Granite Dam at this time of year can have SARs approaching 10% (Doug Marsh, NOAA Fisheries, personal communication).  Determining what other subpopulations are passing at this time, through our otolith analyses, may provide justification for extending the transportation season past the end of October, which is when transportation ends each year. 

 References

Anglea, S.M., D.R. Geist, R.S. Brown, K.A. Deters, and R.D. McDonald.  2004.  Effects of acoustic transmitters on swimming performance and predator avoidance of juvenile Chinook salmon.  North American Journal of Fisheries Management 24:162-170.

 Connor, W.P., and K.F. Tiffan.  2010.  Preliminary evidence for density dependent changes in the migrant growth, migration rate, and mortality of natural Snake River fall Chinook Salmon subyearlings.  Pages 2-29 in W.P. Connor and K.F. Tiffan, editors.  Research, monitoring, and evaluation of emerging issues and measures to recover the Snake River fall Chinook salmon ESU.  2008 Annual Report to the Bonneville Power Administration, Project 199102900, Portland, Oregon.

 FPC (Fish Passage Center).  2010.  Data available at www.fpc.org.

 Harvey, C. J., and P. M. Karieva.   2005.   Community context and the influence of non-indigenous species on juvenile salmon survival in a Columbia River reservoir.  Biological Invasions 7: 651–663.

Mesa, M.G., T.P. Poe, A.G. Maule, and C.B. Schreck.  1998.  Vulnerability to predation and physiological stress responses in juvenile Chinook salmon (Oncorhynchus tshawytscha) experimentally infected with Renibacterium salmoninarum.  Canadian Journal of Fisheries and Aquatic Sciences 55:1599-1606.

Mesa, M.G., L.K. Weiland, and P. Wagner.  2002.  Effects of acute thermal stress on the survival, predator avoidance, and physiology of juvenile fall Chinook salmon.  Northwest Science 76(2):118-128.

Naughton, G.P., D.H. Bennett, and K.B. Newman.  2004.  Predation on juvenile salmonids by smallmouth bass in the Lower Granite Reservoir system, Snake River.  North American Journal of Fisheries Management 24:534-544.

Poe, T.P., Hansel, H.C., Vigg, S., Palmer, D.E., and Prendergast, L.A. 1991. Feeding of predaceous fishes on out-migrating juvenile salmonids in John Day Reservoir, Columbia River. Transactions of the American Fisheries Society 120: 405–420.

Thomas, R.J., T.A. King, H.E. Forshaw, N.M. Marples, M.P. Speed, and J. Cable.  2010.  The response of fish to novel prey: evidence that dietary conservatism is not restricted to birds.  Behavioral Ecology 21(4):669-675.

Tiffan, K.F., and nine coauthors.  2010.  Migration delay and survival of juvenile fall Chinook salmon in the vicinity of the confluence of the Snake and Clearwater rivers.  Pages 1-67 in K.F. Tiffan, W.P. Connor, B.J. Bellgraph, and R.A. Buchanan, editors.  Snake River Fall Chinook Salmon Life History Investigations.  2009 Annual Report to the Bonneville Power Administration, Project 200203200, Portland, Oregon.

Tiffan, K.F., D.W. Rondorf, and P.G. Wagner.  2000.  Physiological development and migratory behavior of subyearling fall Chinook salmon in the Columbia River.  North American Journal of Fisheries Management 20:28-40.

Vigg, S., T. P. Poe, L. A. Prendergast, and H.C. Hansel.   1991.   Rates of consumption of juvenile salmonids and alternative prey fish by northern squawfish, walleyes, smallmouth bass, and channel catfish in John Day Reservoir, Columbia River.  Transactions of the American Fisheries Society 120:421-438.

Zimmerman, M. P., and R. M. Parker.   1995.  Relative density and distribution of smallmouth bass, channel catfish, and Walleye in the Lower Columbia and Snake Rivers.  Northwest Science 69: 19-28.