  |
|
|
The purpose of this document is to present a first draft of Ecology’s river-basin characterization work for internal and Snohomish Basin peer review. In the next six months, we plan to refine these methods with the help of technical experts from the Snohomish Basin and state and federal agencies. We also will begin initial tests to evaluate sub-basin scale analysis tools. We ask that this document not be quoted or referenced until peer review is complete. This summary document is intended for non-technical users, while the detailed technical methods will be available as appendices in the final draft.
The Opportunity
Snohomish Basin (Map 1) has long been known for its enviable quality of life characterized by attractive job opportunities, fertile agricultural lands and extensive timber resources, diverse outdoor recreation, vast areas of public land, and abundant natural resources extending from Puget Sound to the Cascade crest. These characteristics have drawn hundreds of thousands of people to the Snohomish River Basin and have made its cities and towns grow and prosper.
With this growth come impacts to natural resources. To minimize these effects and to compensate for unavoidable environmental damage, extensive regulatory initiatives (e.g., local zoning, Shoreline Management Act, State and National Environmental Protection Acts, Growth Management Act, Clean Water Act, and the Endangered Species Act) have emerged.
While daunting to some, existing environmental regulations are a powerful tool that can provide a significant opportunity for resource protection and recovery. Further, a new collaborative effort has begun, between local jurisdictions, state and federal agencies, tribes, conservation organizations, and basin residents, that stresses cooperation with landowners to protect and restore the icon of the Pacific Northwest, salmon. A unique opportunity exists for significant environmental recovery if regulatory and non-regulatory tools can be employed in a focused, coordinated manner.
The Problem
While many of the necessary regulatory and non-regulatory programs are in place, problems persist and often have worsened. The continued decline in the health of aquatic species and ecosystems in Puget Sound and the Snohomish River Basin indicates that something is dramatically wrong with our approach to resource management (Karr 1995). That something, appears to be a lack of watershed-based tools that provide a conceptual framework for organizing and coordinating recovery actions (Adler 1995, Angermeier and Schlosser 1995, Frissell 1996).
The Need For A River Basin Scale Approach
The causes of this lack of success appear to be two-fold. First, because existing regulations and programs focus on a site or reach scale, no over-riding framework for ecosystem recovery is possible. Second, recurring attempts to focus on individual species (e.g., chinook salmon) or species guides (e.g., migratory ducks and geese) force resource managers to focus on the site-specific dynamics of species distribution and abundance and not on the landscape-scale processes that create and maintain the habitat structure (Angermeier and Schlosser 1995).
River-basin-scale tools are needed because they fit the basic nature of aquatic ecosystems, including the interaction between land and water resources, the links between water quantity and quality, the connections between groundwater and surface water, and the heterogeneity of aquatic ecosystems (Adler 1995).
River-basin-scale tools are needed because much of our lack of success can be attributed to the complexity of natural systems and our resulting inability to identify core problems. But when we can understand cause-and-effect relationships, measurable success is possible. For example, when a perched culvert (the cause) results in a fish passage barrier (the effect), we fix the culvert, the true cause of the problem. But rarely are cause-and-effect relationships that straight forward. A more likely scenario occurs when chinook biologists believe that riverbed scour is a potential limiting factor for chinook on a reach of the Snoqualmie River. The task of identifying limiting factors is monumental in its own right, but when the only available choice for resource managers is to correct alterations on the reach where scour is occurring, success becomes unlikely. Why? Because scour is not the core problem, but the effect of one or more natural process changes that occurred upstream of the site. Scour may be the limiting factor, or key problem, for chinook production in this particular reach of river, but it really is the symptom of a human-induced change in how the watershed delivers and routes water, sediment, and wood. Unless we begin to focus on the core problems, recovery efforts will not be successful.
Finally, river-basin-scale tools are needed because a focus on conventional site-specific planning has failed to stem the ongoing decline in water quality, baseflow, and aquatic species like salmon and ecosystems, despite millions of dollars
The Approach
The delivery and routing of water, sediment, large wood, nutrients-toxicants-bacteria, and heat are the key natural processes that create and maintain the physical structure that supports good water quality, fish habitat, and adequate summer low flow. Our approach is to first understand and characterize watershed processes at a river-basin-scale, then use this new understanding to direct more focused assessment at the sub-basin scale, and finally, use sub-basin assessment to direct restoration and preservation at a site or reach scale. At each finer scale, more specific recommendations can be made.
Methods presented here focus on three distinct landscape scales that function as a series of filters: course-sieve characterization at river-basin scale, moderate-sieve analysis at sub-basin scale, and fine-sieve assessment at site-specific scale.
Value of This Approach
A three-step approach of characterization, assessment, and analysis activities at the river basin, watershed, and site scales can enhance our understanding of the Snohomish Basin and provide a framework for better decision-making at the site-specific scale. Specifically, river-basin characterization can:
River-basin characterization seeks to merge traditionally disparate scientific disciplines and data sets into a more integrated problem-solving framework. To do this, an interdisciplinary team of scientists compiles and analyzes technical information and then collaborates with technical experts in the river basin to share information and interpretations.
Limiting Factors vs Characterization
Limiting factors analysis is a reach-specific, technically-based tool that assesses the critical structural elements of habitat and identifies one or more elements that are restricting habitat use or population grow of a target species. Limiting factors is an essential component of any fish or wildlife recovery effort. River-basin characterization neither duplicates nor competes with limiting factors analysis. Rather, it provides the process-based, landscape scale understanding needed to increase the capability of limiting factors analysis to identify core process-based problems at the reach scale. Limiting factors and basin characterization are supportive.
Change Is Needed NOW
Despite dramatic increases in effort, strong mandates, and massive expenditures for environmental protection over the past 20 years, the overall condition of natural ecosystems continues to decline (Karr 1995, Montgomery et al. 1995). A growing body of work indicates that declines in ecosystem integrity are perpetuated by existing policies and traditional techniques that treat local symptoms of habitat damage and fail to address the root biological and physical causes of ecosystem degradation and population decline (Frissell 1993, Angermeier and Schlosser 1995, Montgomery et al. 1995, Reeves et al. 1995, Ebersole el al. 1997).
CHARACTERIZATION METHODS
Background
In April 1998, an internal technical team recommended that Ecology establish an interdisciplinary technical team to develop and evaluate tools that provide local watershed councils with river-basin-scale technical support (Roberts et al. 1998). By September of that year, a team was formed consisting of a stream geomorphologist, hydrogeologist, fisheries biologist, water quality specialist, ecologist, GIS analyst, and a GIS technician. Funding for this team came from Ecology, Washington Department of Transportation and the US Environmental Protection Agency. The Snohomish and Lewis River Basins were identified as pilot river basins to develop and evaluate river-basin characterization methods.
The following questions guided team actions:
Technical Framework
We suggest that an over-arching framework be established to coordinate salmon recovery and water quality and baseflow improvement for the Snohomish River Basin. This framework should be based on a foundational understanding of key watershed processes and a more detailed understanding of structural elements required to support good water quality, thriving salmon runs, and baseflow. This means that for the recovery of chinook salmon, reach-specific limiting factors analysis (i.e., the assessment of critical structural elements of habitat) must be developed from an understanding of the natural processes that create and maintain structural elements of habitat. Recovery plans developed with only an understanding of the structural limiting factors at a site or reach scale have increased risk of focusing on symptoms rather than the real cause of problems.
Our goal is to develop tools that will assist local watershed efforts in finding solutions to declining salmon stocks, degrading water quality, increasing flood peaks, and declining baseflow. To do this, the team developed landcover estimates for pre-disturbance (circa 1870), current, and future (GMA buildout) conditions and then subdivided the Snohomish Basin into 60 sub-basins. Team members then developed river-basin-scale tools (Appendix A-G) that characterize where human landuse has changed, and will change, the delivery and routing of water, sediments, nutrients, toxicants, large wood debris (LWD), and heat.
Characterization Tools
A general summary of how each process was characterized follows. Detailed technical write-up of each characterization tool can be found in the Appendices.
Characterizing Change in Landuse/Land Cover
Geographic Information System (GIS) coverages for pre-disturbance, current, and future landuse/land cover allowed team members to assess process change both from pre-disturbance to current conditions and from current to future conditions. A pre-disturbance coverage was developed using vegetation data compiled from 1869 to 1873 Government Land Office surveys for Snohomish County (microfiche at the Washington State Library, Olympia, WA). Our intent was to sample tree species and seral stage within generalized vegetation regions throughout the Snohomish Basin. Current conditions were developed using the most recent landuse/land cover developed for Puget Sound from satellite imagery. And finally, a future landuse/land cover coverage was developed by synthesizing Growth Management Act planning documents from King and Snohomish Counties and some cities in the Snohomish Basin. Detailed land cover information is presented Appendix A.
Characterizing Change In Ground-water Discharge To Streams and Baseflow
For this study, baseflow in streams and rivers is defined as ground-water discharge to seeps and springs, as well as snow and glacier melt. These are the primary natural sources of streamflow between storms and during the dry season of summer through early fall. Because there is so little data on streamflow for the Snohomish River Basin, we used surrogate information to estimate the reduction in baseflow due to human activities. The annual amounts of ground-water rights and claims were compared to estimated annual ground-water recharge (natural replenishment). Annual recharge was estimated using published equations relating the factors of annual precipitation and surficial geology (Woodward, et al., 1995). Sub-basins with appropriations exceeding 10% of recharge were highlighted as potentially having significant reductions in baseflow due to ground-water withdrawals. The annual amounts of surface-water rights and claims were compared to estimated annual runoff (stormwater plus baseflow). Annual runoff for sub-basins was interpolated from a published map (Gebert, et al. 1987). Sub-basins with appropriations exceeding 5% of runoff were highlighted as potentially having significant reductions in baseflow. The reduction in annual recharge was estimated using landuse and road information. Recharge was reduced by 30% in urban areas and by 100% for roads (effective width 100 feet) in areas of other landuses. Estimates were calculated for current and future conditions (Map 2). Sub-basins with the highest (4th quartile) estimated reductions of baseflow for current conditions were highlighted as potentially having significant reductions in baseflow. Ground-water characterization methods are presented in detail in Appendix B.
Characterizing Change In Surface Water and Sediments
Four basic steps were required to characterize change in the delivery and routing of surface water and sediments. Sub-basins were first grouped by natural drainage features. Second, a Potential Sediment Transport Coefficient was developed for each sub-basin. This coefficient characterizes a stream’s ability to transport and store sediment. Next, an equation for the Cumulative Sediment Source Component was developed. This coefficient is composed of two basic inputs: 1) natural, or background, surface erosion and mass wasting; and 2) mass wasting and surface erosion as the result of management activities. Lastly, Sediment Yield Hazard was calculated by multiplying the Potential Sediment Transport Coefficient and the Cumulative Source Coefficient. Each sub-basin was defined in terms of its geomorphic and hydrologic properties. After calculating a set of quantitative, dimensionless indices, each sub-basin was ranked relative to its pre-disturbance, current, and future sediment production, yield, and delivery. A list was then developed of sub-basins having the greatest potential for human alteration in the delivery of water and sediment (Map 3 & Map 4). Methods are based on Fitzgerald et al (1998). Surface water and sediment methods are presented in Appendix C.
Characterizing Change In Nutrient Loading
Within each sub-basin, nutrient loading rates were determined using literature values based on landuse and land cover. Of all the compiled literature values of loading rates for total nitrogen and total phosphorous, the lower and upper quartile loading rate values were determined. In addition, a "most likely" loading rate was established by matching each land cover with a specific study that most closely reflect the geologic and climatic conditions observed in the Snohomish River Basin. The estimated nutrient loading rates for each land cover were summed within each sub-basin. Resulting loading rates are presented for each sub-basin in pre-disturbance, current, and future land cover conditions (Map 5). Methods follow the approach by Reckhow et al. 1980. Results are presented in Appendix D. Additional water quality work compiled existing and historic (1970s) ambient monitoring data that was converted to GIS coverages to assist in the interpretation of nutrient loading results.
Characterizing Change In Heavy Metal Loading
Estimates for mean annual loadings of copper, lead, and zinc were developed for each sub-basin using the nationwide regression models developed from the National Urban Pollution Project (Tasker and Driver 1988). These regression models estimate loadings of pollutants for a single storm event whose rainfall is greater than 0.05 inches in an urban area. Loadings are reported on a per area basis (kg pollutant per storm event, per square kilometer). The average number of storms per year was developed from precipitation data and multiplied by the average loading rate per area to estimate annual loading. It should be noted that estimated loading rates only represent non-point sources associated with urban development and do not include natural metal loading rates for the weathering of Cascade rock. Results are summarized in Map 6 and Appendix E.
Characterizing Change In Large Wood
Few tools are available to characterize change in the delivery and routing of LWD. Using pre-disturbance and current land cover coverages, sub-basins having greater than 50% of their riparian stream length in mid- and late-seral stage trees were considered to have the best potential for delivering large wood to streams. Sub-basins having the least proportion of stream length in mid- and late-seral stage trees (upper quartile) were considered to have the greatest potential for alteration in the delivery of wood. The potential for change in the routing of wood was measured by calculating average stream crossings per mile of stream for each sub-basin. Sub-basins with the highest stream crossings per mile of stream (upper quartile) were assumed to have the greatest potential for a change in the routing of LWD (Map 7).
Characterizing Change in Heat
Substantial effort was invested in evaluating river-basin-scale tools for the delivery and routing of heat. Efforts focused on quantifying stream canopy closure, stream depth, and groundwater discharge by sub-basin. After numerous attempts, landscape scale values could only be compiled for stream canopy closure. Additional work will be required to develop usable river-basin-scale tools.
Characterization of Salmon Runs and Habitat
To assist team members in beginning to understand relationships between anadromous fish habitat and process alterations, a summary document was prepared that presents information on the condition of anadromous fish runs and habitat within the Snohomish River Basin. This information is presented in Appendix F.
The Technical Framework For River Basin Characterization
River-basin characterization is relatively new to Western Washington. To assist technical experts in evaluating the conceptual framework and the technical literature it is based on, an in-depth rationale is presented in Appendix G.
Finally, all characterization products were combined into a single map (Map 8) to provide a general picture of the extent of processes altered, by sub-basin. This information established the foundation for all recommendations presented below.
Recommendations
There are no silver bullets when it comes to addressing environmental problems. However, river-basin characterization provides new insight that can be used to develop recommendations that improve our chances to see measurable resource recovery.
Appendix H presents our process-based, river-basin-scale recommendations for water quality, chinook recovery, peak flow, and baseflow and the methods used to develop them. We present these recommendations strictly as examples of how local technical experts can use this information in their recovery efforts. Process-based information blended with local technical expertise may produce the best recommendations.
Because of the length of the recommendations document and the limited space available in this summary report, we will not present all recommendations here. Rather, we have chosen to summarize important points that show some potential differences between this approach and current recovery strategies.
Water Quality Recommendations
Significant differences among characterization methods exist in the spatial and temporal scales of evaluation. Recovery efforts tend to focus solely at a site scale and look almost exclusively at current conditions. Examples of exceptions to this are the multi-faceted work by George Pess and others (1999) and companion documents by Michael Pollock (1998) and Brian Collins (1997) on historic conditions in the Stillaguamish River Basin, as well as Andy Haas’ work on pre-disturbance historic estuarine conditions in the Snohomish Basin (Andy Haas unpublished data). This type of work provides restorationists with a perspective of change that is necessary for assessing recovery potential. With this concept in mind, river-basin characterization looked at three temporal scales: pre-disturbance, current, and future. You will see that water quality recommendations clearly show the value of understanding past and future landuse change.
Of particular interest is the value in understanding the effects of future planned development. Using water quality recommendations as an example, there are 13 sub-basins that have the greatest potential for current nutrient problems. Long-term ambient monitoring work has identified most of these sub-basins as existing problems. However, water quality characterization also identifies 15 additional sub-basins having high potential to become major non-point nutrient sources in the future, if development follows current GMA planning.
Our recommendations are simple. We suggest continued focus on the 13 degraded sub-basins with assessment, retrofits, and restoration to correct these problems. But, we also suggest that a significant portion of water quality efforts go toward the 15 sub-basins that are currently in good condition but clearly at risk from future development. Experience has shown that it is easier and more cost effective to maintain good water quality conditions than it is to recover degraded ones.
Recommendations for a Conceptual Framework for Salmon Recovery
Most exciting however, may be the potential that a process-based, river-basin-scale characterization effort has at providing a conceptual framework for salmon recovery. The framework that we propose was adapted from concepts presented by the Pacific Rivers Council in the early 1990s. What is different is the process-based context in which the framework is placed. This framework has potential to provide the landscape-scale organization needed to maximize the extensive work that has gone into limiting factors analysis.
We recommend that natural resource planners consider using a four-step process-based approach to structuring a salmon recovery plan. With this approach, we are not implying rank or priority in the order that each is presented, although some logical order is apparent. We do suggest that some level of assessment work proceed within each of the four key areas. Recommendations for a conceptual framework for chinook recovery in the Snohomish Basin are based on the following four steps:
River-basin characterization indicates that three sub-basins meet the process-based criteria for existing refugia. A rationale and recommendations for early action are provided for existing refugia sub-basins in Appendix H.
The second part of this approach calls for the identification of a well-dispersed network of future refugia sub-basins. Refugia sub-basins must be well distributed throughout all major drainages that support chinook in the Snohomish River Basin. When existing refugia do not meet this criteria, addition sub-basins must be identified and targeted for restoration and preservation actions that elevates them to refugia status in the future. Using process-based criteria, 14 sub-basins are proposed for future refugia status. Rationale for the selection of future refugia sub-basins and additional recommendations for early action are also provided.
The third component of the framework calls for the restoration and preservation of watershed processes in sub-basins adjacent to, or immediately downstream from, existing and future refugia areas. The concept here is to build on intact processes that create and maintain high quality habitat for chinook. Again using process-based criteria, eight sub-basins were identified as primary sub-basins because of their potential for process recovery and landscape position immediately adjacent to, or downstream from, refugia areas. An additional, eight sub-basins located immediately adjacent to, or downstream of, primary sub-basins received secondary priority status for sub-basin assessment and restoration actions.
And finally, the "grubstake" areas, those large wetland/riparian/floodplain complexes low in the drainage system that, regardless of upstream process alteration, are capable of influencing processes on and downstream of the site. Recommendations provide examples such as the 6000 acre Marshland complex, the Snohomish estuary, and the large wetland/riparian complexes on the Snoqualmie River that meet critical habitat requirements of chinook in the lower, more highly altered, floodplain system.
The concept for this approach is also a simple one (Map 9). Protect intact sub-basins (existing refugia), restore a well-dispersed network of sub-basins that can be returned to an intact condition (future refugia), build on intact sub-basins by restoring processes in sub-basins downstream from refugia (restoration sub-basins), and finally restore large wetland/riparian/floodplain systems low in the drainage (grubstake sites). The key to this approach is focusing on the condition of landscape-scale processes like the delivery and routing of water, sediment, and wood, rather than the reach-specific structural elements of habitat.
Appendix H also provides a process-based approach to looking at habitat degradation factors identified by the Snohomish Basin Salmonid Recovery technical Committee. These recommendations have been developed with an understanding of process alteration and a spatial context that promotes an understanding of downstream effects. Characterization facilitates better understanding of the river basin as a continuum rather than a collection of disparate parts. At the river-basin scale, alterations in sediment inputs in a headwater sub-basin can be followed down the drainage system and linked to symptoms such as riverbed scour further downstream. It is nearly impossible to develop this type of insight at strictly a site-specific scale.
Recommendations for Baseflow and Peak Flow Alteration
Recommendations for baseflow and peak flow were the most straight-forward due to their direct relation to the delivery and routing of water. River-basin characterization identified 20 sub-basins that have the greatest risk of baseflow alteration and another 20 sub-basins that are most likely to experience increased peak flows. Recommendations presented in Appendix H suggest that monitoring and assessment are needed to more clearly define the human landuses that are responsible for changes in the delivery and routing of water at a sub-basin scale.
Conclusions
In 1998, Washington State Department of Ecology formed an interagency a team consisting of a stream geomorphologist, hydrogeologist, fisheries biologist, water quality specialist, ecologist, GIS analyst, and a GIS technician. The goal of this team was to develop landscape-scale tools that assist local watershed efforts in finding solutions to declining salmon stocks, degrading water quality, increasing flood peaks, and declining baseflow.
To do this, the team first identified five key natural processes; the delivery and routing of water, sediment, large wood, nutrients/toxicants/bacteria, and heat; that create and maintain fish habitat structure, water quality, peak flow, and baseflow. Then the Snohomish Basin was subdivided in 60 sub-basins, and GIS coverages were developed for pre-disturbance, current, and future landuse/landcover conditions. Team members then developed and evaluated river-basin-scale tools that characterize where human landuse has changed, and will change, the delivery and routing of key processes.
This information then was used to develop recommendations for using characterization products to assist local technical experts in resource recovery efforts. We suggest a single framework be established to coordinate salmon, water quality, peak flow and baseflow recovery for the Snohomish River Basin. This framework should be based on a foundational understanding of key watershed processes and a more detailed understanding of structural elements required to support good water quality, thriving salmon runs and aquatic resources.
There are no silver bullets when it comes to addressing environmental problems. However, river-basin characterization is a tool that Ecology is developing to help local watershed councils begin to identify and address core natural resource problems of importance to residents.
Acknowledgements
Many individuals and agencies were instrumental in supporting the watershed characterization team. The authors wish to acknowledge the Washington State Department of Transportation and the US Environmental Protection Agency for providing financial support. Clearly, without the staff funding provided by these two agencies, this project could not have moved forward. We also wish to acknowledge the long-term technical support and encouragement provided by Dr. Jennifer Brown. Her knowledge and insight were instrumental in shaping the conceptual framework that you see in this document. Special thanks must also go to Dr. Tom Hruby for his work to compile the heavy metals data and Andy Haas for sharing his unpublished pre-disturbance vegetation data compiled for the Snohomish River and its estuary. Finally, we wish to thank all of the local jurisdictions, state and federal agencies, and the Tulalip Tribes for their willingness to share data and GIS coverages.
References
Adler, R.W. 1995. Addressing barriers to watershed protection. Environmental Law Vol. 25:973-1106.
Angermeier, P.L. and I. Schlosser. 1995. Conserving aquatic biodiversity: beyond species and populations. In: Evolution and the aquatic ecosystem: defining unique units in population conservation. Nielsen, J.L. editor. American Fisheries Society Symposium 17:402-414.
Collins, B. 1997. Effects of land use on the Stillaguamish River, Washington, ~1870 to ~1990: implications for salmonid habitat and water quality and their restoration. Report to The Tulalip Tribes Natural Resource Department and others. 23 pp.
Doppelt, B, J. Smith, R. Wissmar, and C. Williams. 1996. What is watershed analysis? In: Healing the Watershed: A Guide to the Restoration of Watersheds and Native Fish in the West. The Pacific Rivers Council, Inc. 212 pp.
Ebersole, J.L., W. Liss, and C. Frissell. 1997. Restoration of stream habitats in the Western United States: restoration as reexpression of habitat capacity. Environmental Management Vol. 21(1):1-14.
Fitgerald, J., Hardy, T. and T. Geier, 1998. Refinement of a geomorphic risk assessment and application to a nonpoint sediment source analysis in the Middle Fork Payette River Subbasin, Idaho. Unpublished draft.
Frissell, C.A. 1996. A new strategy for watershed restoration and recovery of Pacific salmon I the Pacific Northwest. In: Watershed and Salmon Habitat Restoration Projects: Guidelines for Managers of State Trust Lands. by Larry Dominguez. Washington Department of natural Resources. Olympia, Wa. 90 pp.
Gebert, W. A., Graczyk, D. J., and Krug, W. R. (1987) Average annual runoff in the United States, 1951-80. U. S. Geological Survey, Hydrologic Investigations Atlas HA-710.
Karr, J.R. 1995. Clean water is not enough. Illahee Vol. 11, No. 1 and 2:51-59.
Maxwell, J. 1993. Ecosystem management by watersheds. Nonpoint Source News-Notes No.32:6-8.
Montgomery, D.R., G. Grant, and K. Sullivan. 1995. Watershed analysis as a framework for implementing ecosystem management. Water Resources Bulletin Vol. 31, No. 3: 369-386.
Pess, G.R., B. Collins, M. Pollock, T. Beechie, A. Haas, and S. Grigsby. 1999. Final Draft - Historic and current factors that limit coho salmon (Oncorhynchus kisutch) production in the Stillaguamish River Basin Washington State: implications for salmonid habitat production and restoration. Report for Snohomish County Department of Public Works and others.
Pollock, M.M. 1998. Current and historic riparian conditions in the Stillaguamish River Basin. Report to Stillaguamish Tribe of Indians Natural Resources Department, Arlington, Wa.
Reckhow, K.H., M. Beaulac, M. and J. Simpson. (DATE). Modeling phosphorous loading and lake response under uncertainty: a manual and compilation of expert coefficients. US Environmental Protection Agency. EPA440/5-80-011.
Reeve, G.H., L. Benda, K. Burnett, P. Bission and J. Sedell. 1995. A disturbance-based ecosystem approach to maintaining and restoring freshwater habitats of evolutionarily significant units of anadromous salmonids in the Pacific Northwest. In: Evolution and the aquatic ecosystem: defining unique units in population conservation. Nielsen, J.L. editor. American Fisheries Society Symposium 17:334-349
Roberts, D. and others. 1998. An integrated characterization process to support local watershed planning. Washington State Department of Ecology Publ. No. 98-12. 19 pp.
Tasker, G.D. and N. Driver. 1988. Nationwide regression models for predicting urban runoff water quality at unmonitored sites. Water Resources Bulletin 24:1091-1101.
Woodward, D. G, Packard, F. A., Dion, N. P., and Sumioka, S. S. (1995) Occurrence and quality of ground water in southwestern King County, Washington. U. S. Geological Survey Water Resources Investigations Report 92-4098, 69 p., 4 plates.
Return to River Basin Characterization
This page was last edited on January 26, 2007.
Copyright © Washington State Department of Ecology. See http://www.ecy.wa.gov/copyright.html.
