Limiting Factors Analysis of the Nooksack River

Robert Bilby, NMFS; Pete Bisson, USFS; Timothy Hyatt, NWIFC; Robert Naiman, UW Fisheries; Thomas O'Keefe, UW Fisheries; Jeffrey Richey, UW PRISM program; Lisa Steubing, UW Computer Science & Engineering


With continued concern over the future of anadromous salmonid stocks in Puget Sound and throughout the Pacific Northwest, a critical need exists for the development of an effective modeling approach and useful data resource to characterize and quantify their habitat. The need for a method of using habitat data to prioritize protection and restoration efforts, and institute beneficial riparian management strategies has become more acute with Endangered Species Act (ESA) listings both in Puget Sound and throughout the region.

Scientists, resource managers, politicians, industry representatives, and citizens recognize an immediate need for a framework that clarifies tradeoffs and allows us to most effectively manage environmental and human needs for watershed resources. Of particular concern, is the status and quality of habitat in aquatic and riparian environments of the Puget Sound basin, and their role in the decline of wild salmonids. However, without knowing precisely which factors are limiting the production of salmonids, it is impossible to priortize habitat restoration and protection projects. Efforts to preserve and enhance habitat are hampered in many basins by a dearth of information on habitat quantity and quality. What is needed is a consistent framework for cataloging and comparing stream basins, and a database of stream inventory information at both the sub-watershed and multiple-watershed scales. The Northwest Indian Fisheries Commission (NWIFC) has taken the first steps necessary to develop these tools.

Currently, NWIFC is working on the Salmon and Steelhead Habitat Inventory and Analysis Project (SSHIAP). SSHIAP is a database of reach-level habitat information. The SSHIAP model segments all the river and tributary reaches in western Washington by slope, confinement, and hydrology, and adds attributes for stock distribution, anadromous barriers, and anthropogenic modifications to the channel (levees, riprap, culverts, etc). Ultimately the SSHIAP data will be integrated into a GIS to optimize the usefulness and potential distribution of the habitat and distribution information. The primary goal of our proposed research will be to develop procedures to enter these data into a GIS and to develop a modeling approach based on habitat condition and limiting factors analysis that allows us to make immediate and effective use of these data for prioritizing salmon restoration and protection efforts. We will utilize the GIS resources available through the Puget Sound Regional Synthesis Model (PRISM) program at the University of Washington to develop the modeling approach in the Nooksack River basin, selected for the range of habitats and land-use practices it contains.

We have selected the Nooksack River because it provides an effective microcosm for examining the decline of salmonids in the Puget Sound region. The lower mainstem is lined with levees, dairy farms, and the remains of enormous sloughs, side channels, and beaver marshes. The headwaters consist of steep ephemeral channels in commercial forest land, frequently crossed by logging roads. The middle reaches provide, or could provide, some of the very best chinook spawning grounds in the Pacific Northwest. The Kendall Creek hatchery was constructed on the North Fork in 1899, and the basin has been managed largely for hatchery production ever since. Despite a long history of intense fisheries management for Nooksack fish, aspects of the physical habitat affecting wild salmonids are still largely unknown. On-the-ground habitat surveys of the mainstem and tributaries are relatively rare, and detailed knowledge of the factors limiting wild salmonid production is simply not available. In addition, temperature differences between the glacial-fed North Fork and the lethally-warm South Fork will provide an incomparable opportunity to investigate the effects of different temperature regimes on salmonid production.

Problems and Opportunities

The data resource and this modeling effort would clearly be useful for the development of strategies relevant to ESA listings and a greater understanding of salmon habitat throughout Puget Sound. Lunetta et al. (1997) have shown that knowing the distribution and quality of habitat and having it in a GIS enables regional resource managers to prioritize habitat protection and restoration efforts. An analysis of the anthropogenic factors that limit salmonid production in a given watershed is a necessary precursor to prioritizing the habitat restoration strategies that will remove population bottlenecks caused by habitat modification (Beechie et al. 1994). The SSHIAP database has been designed to facilitate such a limiting factors analysis, either within basins for individual stocks, or among all basins in a watershed. Although our initial efforts supported through this proposal will be in the Nooksack basin, limiting factors analysis for the entire Puget Sound region will be the ultimate goal and can be accomplished given the combined resources and mandates of the PRISM and SSHIAP programs.

Despite the notable prior efforts at limiting factors analysis, estimates of coho and chinook smolt production in Washington are based on data of dubious quality. The Washington State Department of Fish and Wildlife bases their smolt production and fishery escapement goals on a limited number of studies, most of which were conducted in Oregon and British Columbia, and very few of which have been published in peer-reviewed literature (Zillges 1977, Baranski 1989). It is our intent to improve and enhance the predictive capability of coho and chinook production models with improved data and techniques crafted toward local conditions.


Success of this project ultimately depends on close cooperation among local and regional agencies , tribes, non-governmental organizations, and individual scientists who are actively working on the link between riparian habitat and anadromous salmonid production. Individuals directly collaborating on this project will include:

The individuals representing these different agencies and organizations will provide the necessary oversight and/or conduct the research. Our proposed research will contribute towards efforts already in place to address salmon habitat issues in Puget Sound watersheds. These efforts have been initiated by groups and agencies that include local governments (e.g. King County and Snohomish County) and non-governmental orgaizations (e.g. Washington Trout and American Rivers).

The Role of PRISM

Although the project will include input and collaboration with individuals from different entities it will be centrally coordinated through the PRISM program at the University of Washington. PRISM represents an innovative new program at the University that takes advantage of the fact that the university is not bound by traditional jurisdictional boundaries and has resources across a broad range of areas. Their expertise in GIS and considerable modeling resources will be key to the success of our project. PRISM uses a modeling approach to synthesize information and bring together people working with similar issues in Puget Sound's watersheds. The program provides the opportunity to integrate key pieces of information on the Puget Sound ecosystem. Riparian areas and their effect on salmonid habitat are key components of this endeavor.

PRISM's modeling effort has started primarily with the inclusion of geomorphic and hydrologic variables. The research project described here will provide a meaningful link between the evolving fish/river layer and land use layer allowing us to better understand different management scenarios in riparian areas. This project like others coordinated through PRISM takes an applied approach and seeks to concentrate on research that is immediately beneficial to people and institutions outside the university. The research described here will provide us with valuable information about the limits to salmonid production that will in turn help us weigh management choices we face now and in the future.

Limiting Factors Analysis

The technique of limiting factors analysis was first described for coho salmon in Reeves et al. (1989). It is designed to identify physical limitations to fish production that may be addressed by habitat restoration or enhancement. This approach assumes that when habitat required by a species during a particular season is in short supply, a bottleneck results and this habitat becomes limiting (Reeves et al. 1989). Other factors limit production, but this method and our efforts will be primarily focused on habitat.

The basic methodology is to first quantify the different habitat types in a watershed (pools, riffles, cascades, side-channels, sloughs, etc.) and then estimate the smolt production capacity for each habitat type, by season. Simple calculations of production capacity can be refined and expanded based on more comprehensive and higher quality data, which this project is designed to elucidate. The comparisons of which habitat types are lacking, under different conditions of temperature, life-cycle phase, and habitat complexity, are the basis for identifying population bottlenecks. Currently the estimates of parr density and smolt production in various habitat types are based on limited studies from other regions of the Northwest (Chapman 1965, Zillges 1977, Everest et al 1987). Updated estimates, based on local conditions, are sorely needed.

Limiting factors analysis was implemented by Beechie et al. (1994) as a means of estimating smolt production for coho salmon in the Skagit River and quantifying the factors that limit their production. In the case of the Skagit River, Beechie et al. (1994) found that hydromodification was the primary factor contributing to habitat loss (73% of summer habitat and 91% of winter habitat), followed by culverts (13% of summer habitat and 6% of winter habitat) and forestry (9% of summer habitat and 3% of winter habitat).

A major advantage of using a limiting factors approach is that it allows one to determine smolt production by habitat type and determine the cause of habitat alterations where present. It considers land use activity collectively in the context of the whole watershed rather than trying to identify habitat problems on a case-by-case basis. This approach allows us to prioritize restoration efforts. As an example, it might be more cost effective to remove a culvert blocking good habitat or open access to slough channel than to initiate an expensive restoration project for a stream where riparian habitat has been degraded 10-20% (Beechie et al. 1994).

Proposed work plan

Our immediate objectives under the current proposal will be to map and characterize the habitat types based on standard procedures (Schuett-Hames et al. 1994, Bisson et al. 1982). Much of this first phase has been carried out by SSHIAP over the past three years. Under this proposal the SSHIAP data will be field-verified and refined to identify to what degree pool-riffle ratios and habitat quality can be gathered from maps and aerial photos. Part of the habitat mapping and characterization will involve advanced GIS support in an attempt to model habitat character based on streambed slope, confinement, hydrology, riparian condition, and other physical features (Lunetta et al. 1997). Phase II will involve field studies to refine the estimates of parr density in various habitat types. These densities will be necessary for the basic modeling exercises in Phase III. Phase III will combine the field-verified SSHIAP database with improved smolt production estimates to calculate smolt production for the entire Nooksack River basin.

We will develop a locally-based coho (and possibly chinook) production model based on a limiting factors analysis of the entire Nooksack basin, using both remote and field-verified examinations of habitat quality and quality. The SSHIAP model has already grouped river and tributary reaches into similar habitat types, the field verification would establish habitat character and quality for each SSHIAP segment. Intensive field studies would use a combination of electroshocking and snorkel surveys (Hankin and Reeves 1988) to estimate fish populations and densities in various habitat types. Field studies would be conducted in the summer of 2000, with modeling efforts to commence immediately afterward.

Longer term objectives are to demonstrate methods by which riparian characterization in conjuction with a GIS modeling approach can be used as an effective tool for habitat preservation, enhancement, and restoration. Ultimately these tools will be used throughout Puget Sound as part of the SSHIAP evaluation and monitoring programs and be available for those interested in the status of salmon habitat.



Baranski, C. 1989. Coho smolt production in ten Puget Sound streams. Washington Department of Fisheries Technical Report #99. 29p

Beechie, T, E Beamer, and L. Wasserman. 1994. Estimating salmon rearing habitat and smolt production losses in a large river basin, and implications for habitat restoration. N. Amer. J. Fish. Mngt. 14:797-811.

Bisson, P.A., J.L. Nielsen, R.A. Palmason, and L.E. Grove. 1982. A system of naming habitat types in small streams, with examples of habitat utilization by salmonids during low streamflow. In N.B. Armantrout [ed.] Acquisition and Utilization of Aquatic Habitat Inventory Information: Proceedings of the Symposium. [Portland, OR, October 1981].

Hankin, D.G., and G.H. Reeves. 1988. Estimating total fish abundance and total habitat area in small streams based on visual estimation methods. Can. J. Fish. Aquat. Sci. 45: 834-844.

Lunetta, R.S., B.L. Consentino, D.R. Montgomery, E.M. Beamer, T.J. Beechie. 1997. GIS-based evaluation of salmon habitat in the Pacific Northwest. Photogrammetric Engineering and Remote Sensing 63(10):1219-1229

Reeves, G.H., F.H. Everest, T.E. Nickelson. 1989. Identification of physical habitats limiting the production of coho salmon in western Oregon and Washington. USDA Forest Service General Technical Report. PNW-GTR-245.

Schuett-Hames, D. A. Pleus, L. Bullchild, and S. Hall. 1994. Timber-Fish-Wildlife Ambient Monitoring Program Manual. TFW-AM9-94-001 Northwest Indian Fisheries Commission. Olympia WA.

Zillges, G. 1977. Methodology for determining Puget Sound coho escapement goals, escapement estimates, 1977 pre-season run size prediction and in-season run assessment. Washington Department of Fisheries Technical Report #28. 65p.