Research Project Description
Chinook salmon stocks in the Snake River have been listed under the Endangered Species Act (ESA) since 1992. In response to declining populations and ESA requirements, agencies have adopted policies that attempt to conserve and restore remaining Chinook salmon populations. These efforts have included measures to maintain genetic integrity of remaining wild stocks, reduce passage mortality by improving conditions in the migration corridor, reduce the effects of exotics, restrict sport and commercial harvest, and conservation or restoration of remaining critical habitat (Lee et al. 1997; NWPPC 2000; IDFG 2001). The conventional approach to managing critical habitat has been focused on conserving or restoring the quality of remaining habitats; i.e. conserving and restoring those habitats considered necessary for Chinook salmon to complete their complex life cycle from an incubating egg to a mature fish depositing eggs in natal spawning areas. While conservation of the quality of critical habitats is essential, there is growing concern that the size and spacing of habitats also needs to be considered (Krohn 1992). Simberloff (1988) suggested that effective conservation may require maintaining or restoring a critical amount or mosaic of habitat, as well as habitat of certain quality.

Further, emerging conservation theory suggests that recolonization and persistence of widely ranging species may be strongly influenced by the spatial geometry of remaining habitats. The relevance of these concepts to the persistence of declining stocks of Chinook salmon Oncorhynchus tshawytscha is unknown. If patterns in the distribution and spatial structure of Chinook salmon populations are important to their persistence in stochastic environments, effective conservation may imply maintaining or restoring a critical amount or mosaic of habitat as well as smaller scale habitat characteristics. This research will describe factors influencing the spatial distribution and persistence of wild Chinook salmon. Results will advance current understanding of the relationship between landscape characteristics and the distribution, pattern, and persistence of Chinook salmon. Such information could be key for development of conservation and restoration strategies. While this research will focus on larger scale spatial questions about persistence, it will simultaneously provide information useful for intensively monitoring an ESA listed Chinook salmon stock. Our annual estimates of wild Chinook salmon redds will enable managers to estimate total annual redds in order to monitor stock status and evaluate the influences of various mitigation and restoration efforts.

This research addresses at least three critical needs identified in Regional Program documents. 1) the need for long-term information to assess trends in wild Chinook salmon populations; 2) the need for evaluation of broad scale population sampling and inventory methods; and 3) the need for analysis of the spatial structure of wild Chinook salmon populations.

Project history and accomplishments
In 1995, we developed a study plan and coordinated with State and U.S. Forest Service biologists. We developed a Memorandum of Understanding- No.INT-95121-MOU to coordinate our research with State of Idaho biologists and fisheries managers. We developed a cost-share agreement with the Payette National Forest to assist funding of a portion of the field work. From 1995 to 2002 we have annually completed surveys of the mainstem Middle Fork Salmon River and 12 tributaries and used a GPS unit to map the location of potential spawning areas and redds. In 2001 we began a cooperative agreement with the University of Idaho (UI) to examine geomorphic controls on spawning gravel availability. Although the Project has been in progress since 1995, FY 1999 was the initial year of BPA funding and FY 2005 is the seventh year of BPA funding. Previous year costs were paid by RMRS with some assistance from cooperators. Ten years of data have been gathered. The analysis of spatial structure will be strengthened by collecting additional years of data that monitor the progeny of prior years spawners (spawner-to-spawner) returns.  

     FY 1995-1998: Annually in September 1995-1998, we flew reaches of the mainstem MFSR and 12 tributaries and used a GPS unit to map the location of potential spawning areas and redds. We completed ground-based counts in four streams that were not visible from the air. GPS files have been corrected and transferred into GIS for spatial analysis. Summaries of redd surveys were submitted to collaborators and other interested parties. From 1995 to 1998, annual redd counts ranged from 20 to 424.

     FY 1999-2004: Annual aerial redd counts were continued in September 1999-2004 within reaches of the mainstem MFSR and 12 tributaries. A GPS unit was used to map the location of potential spawning areas and redds. We completed ground-based counts in four streams that were not visible from the air. GPS files have been corrected and transferred into GIS for spatial analysis. Summaries of redd surveys were submitted to collaborators and other interested parties. From 1999 to 2004, annual redd counts ranged from 110 to >2,200. In 1999 the principal investigator attended a technical conference, presented a paper, and published a companion paper in the conference proceedings (see Thurow 2000). In FY 2001 we submitted and were granted a request for a within year increase in funds to evaluate the bias and precision of redd counts. In 2002, we submitted a proposal (#28001) that was funded to expand the redd count analysis research and is currently Project #2002-049-00. In June 2001, we initiated a cooperative research program with the University of Idaho to examine geomorphic controls on the spatial distribution of spawning gravels in the MFSR. We constructed a digital elevation model of the basin, and predicted substrate size as a function of channel hydraulics (channel type and bankfull discharge). Pilot field studies were conducted to quantify reach-average channel type, substrate size, and channel dimensions (bankfull depth, width, and channel gradient). Some of these data were used to drive the hydraulic model, while others were used to validate our predictions of substrate size. Initial results show a strong correspondence between predicted locations of suitable spawning gravels and observed locations of redds, indicating that much of the spatial pattern of Chinook spawning may be explained by the effects of channel hydraulics on substrate size. These findings were presented at the American Fisheries Society (AFS) Transboundary conference (Buffington et al. 2002). In June 2002, we continued and expanded the above research to sample a broader range of locations and underlying geologies in the basin. To date we have sampled over 120 stream reaches. We have also initiated a pilot study to examine the effects of sediment supply on substrate size and quality. The 2002 field work was conducted with USFS and UI funding external to BPA. Additional results were presented as invited talks at both the National AFS meeting and the Fall AGU meeting (Buffington et al. 2003a,b).

As part of the UI cooperative agreement we have also begun statistical analyses of annual patterns of spawning locations and density. Results have been presented at the AFS Transboundary conference (Isaak and Thurow 2002), Idaho chapter AFS meetings (Thurow and Isaak 2003, Isaak and Thurow 2003, Isaak et al. 2005), AFS National meetings (Isaak et al. 2003, Thurow and Isaak 2004, Isaak and Thurow 2004), Western Division AFS meetings (Thurow and Isaak 2004, Isaak and Thurow 2004), the North American Benthological Society meeting (Isaak and Thurow 2004), and the Biobío River Scientific Forum, University of Concepción, Chile (Buffington et al. 2004). One manuscript has been published in the Canadian Journal of Fisheries and Aquatic Sciences (Isaak et al. 2003) and three manuscripts are in review. Quarterly and Annual reports have been submitted to BPA and additional manuscripts are in progress.

Since 1995, redds were observed at elevations between 1100m and 2100m with a majority were constructed between 1500m and 2000m. Chinook salmon spawned in both mainstem reaches of the Middle Fork Salmon River and tributaries with about 98% of the redds to date observed in tributaries.

Location of Project:
We selected the Middle Fork Salmon River (MFSR) drainage as the study area (see Thurow 2000 and Servheen et al. 2001 for detailed descriptions). The study area was selected for several reasons: 1) Remaining Chinook salmon stocks are wild and indigenous, unaltered by hatchery supplementation. Consequently, the ability of the salmon population to respond to the quality and quantity of the available habitat has not been altered. Wild, indigenous, Chinook salmon populations like those in the MFSR are rare; Thurow et al. (2000) reported their presence in 4% of the potential historical range and 15% of the current range in the Columbia River basin and portions of the Klamath River basin. 2) Most of the drainage has been lightly disturbed by anthropogenic activities so habitat quality has not been substantially altered in most areas. Widespread degradation of habitat would be expected to confound a spatial analysis of freshwater habitat by influencing fish distribution and abundance. 3) The large area provides an opportunity for a large sample size. About 650 km of tributaries and 170 km of the mainstem are accessible to Chinook salmon (Mallet 1974; Thurow 1985). This increases the likelihood of a sample size large enough to complete a robust spatial analysis. 4) Opportunities exist for extensive collaboration with other agencies and tribes who are already conducting Chinook salmon redd counts in the drainage; and 5) the principal investigator has more than 25 years of experience working in this drainage and has an intimate knowledge of the MFSR and the spawning ecology of its Chinook salmon.