Pierce County

Watershed Analysis for the Development of Salmonid Conservation and Recovery Plans Within Pierce County

Completion Report June 2001

Prepared for

Pierce County

Submitted by

MOBRAND BIOMETRICS, INC Vashon Island, Washington

ACKNOWLEDGEMENTS

Information used in the Pierce County analysis was contributed by a large number of individuals, many of who are listed in Appendix C tables. We extend our appreciation to each of them for their assistance.

Special acknowledgements are due to Debby Hyde of Pierce County, for recognizing the importance of the analysis and making it possible to accomplish it. Chris Schutz, working with Debby Hyde, did much of the coordination with the many participants. Key individuals who participated in a number of long workshops included the following: Dave Adams (Audubon Society), Chuck Baranski (WDFW), Randy Brake (Pierce County), Paul Bucich (Federal Way), Steve Chapel (Lakewood), Dick Gilmur (Port of Tacoma), Bill Graeber (WDNR), Glenn Grette (Pacific Engineering International), Jon Houghton (Pentec Environmental), Hans Hunger (Pierce County), Russ Ladley (Puyallup Tribe), Don Nauer (WDFW), John O'Loughlin (Tacoma), Tyler Patterson (USFS), Dave Renstrom (Pierce County), Dave Risvold (Pierce County), Joanne Schutt-Hames (WDOE), Blake Smith (Puyallup Tribe), Gene Stagner (USFWS), Jeanne Stypula (King County), and Bill Sullivan (Puyallup Tribe). We extend our thanks to each.

TABLE OF CONTENTS

LIST OF TABLES ...... iii LIST OF FIGURES...... iv

1.0 INTRODUCTION ...... 1-1 1.1 Project Objectives ...... 1-1 1.2 Project Overview...... 1-2 1.3 Use of the EDT Method...... 1-2 1.4 Document Organization...... 1-4

2.0 THE EDT METHOD AS APPLIED TO PIERCE COUNTY...... 2-1 2.1 Conceptual and Information Framework...... 2-1 2.1.1 The Framework Concept ...... 2-1 2.1.2 Ecological Information Structure ...... 2-2 2.2 Analytical Model...... 2-5 2.3 Step-by-Step Procedure...... 2-7 2.3.1 Identification of Goals and Values ...... 2-8 2.3.2 Resource Assessment...... 2-8 2.3.3 Analysis of Actions...... 2-10 2.3.4 Considerations for Monitoring and Adaptive Implementation...... 2-11

3.0 BACKGROUND...... 3-1 3.1 Goals and Values...... 3-1 3.2 The Environment...... 3-2 3.2.1 General Characteristics...... 3-2 3.2.2 WRIA 10 – Hylebos and Puyallup Systems...... 3-10 3.2.3 WRIA 12 – Chambers-Clover System...... 3-17 3.2.4 WRIA 15 – Kitsap Peninsula Systems ...... 3-21

4.0 RESOURCE ASSESSMENT ...... 4-1 4.1 Hylebos Basin ...... 4-2 4.1.1 Chinook Salmon...... 4-2 4.1.2 Coho Salmon...... 4-8 4.1.3 Inferences to Bull Trout for Hylebos Basin...... 4-12 4.1.4 Data/Information Uncertainties for Hylebos Basin...... 4-12 4.1.5 Hylebos Basin Conclusions...... 4-13 4.2 Puyallup Basin (excluding White) ...... 4-14 4.2.1 Chinook Salmon...... 4-14 4.2.2 Coho Salmon...... 4-21 4.2.3 Inferences to Bull Trout for Puyallup Basin ...... 4-28 4.2.4 Data/Information Uncertainties for Puyallup Basin ...... 4-28 4.2.5 Puyallup Basin Conclusions...... 4-30 4.3 White Basin ...... 4-30 4.3.1 Chinook Salmon...... 4-30 4.3.2 Coho Salmon...... 4-37 4.3.3 Inferences to Bull Trout for White Basin...... 4-45

i 4.3.4 Data/Information Uncertainties for White Basin...... 4-45 4.3.5 White Basin Conclusions...... 4-47 4.4 Chambers-Clover Basin...... 4-49 4.4.1 Chinook Salmon...... 4-49 4.4.2 Coho Salmon...... 4-52 4.4.3 Inferences to Bull Trout for Chambers-Clover Basin ...... 4-56 4.4.4 Data/Information Uncertainties for Chambers-Clover Basin ...... 4-58 4.4.5 Chambers-Clover Basin Conclusions...... 4-59 4.5 Kitsap Watersheds...... 4-59 4.5.1 Chinook Salmon...... 4-60 4.5.2 Coho Salmon...... 4-75 4.5.3 Inference for Bull Trout for Kitsap Watersheds ...... 4-94 4.5.4 Data/Information Uncertainties for Kitsap Watersheds ...... 4-94 4.5.5 Kitsap Watersheds Conclusions...... 4-96

5.0 ANALYSIS OF ACTIONS...... 5-1 5.1 Hylebos Basin ...... 5-2 5.1.1 Identification and Ordering of Actions for Hylebos Basin...... 5-2 5.1.2 Projections of Benefits for Hylebos Basin...... 5-4 5.1.3 Summary for Hylebos Basin ...... 5-7 5.2 Puyallup-White Basin...... 5-8 5.2.1 Identification and Ordering of Actions for Puyallup-White Basins ...... 5-8 5.2.2 Projections of Benefits for Puyallup-White Basins ...... 5-8 5.2.3 Summary for Puyallup-White Basins...... 5-8 5.3 Chambers-Clover Basin...... 5-17 5.3.1 Identification and Ordering of Actions for Chambers-Clover Basin...... 5-17 5.3.2 Projections of Benefits for Chambers-Clover Basins ...... 5-17 5.3.3 Summary for Chambers-Clover Basin...... 5-22 5.4 Kitsap Basins...... 5-22

6.0 ADAPTIVE IMPLEMENTATION PROCESS ...... 6-1 6.1 Uncertainty ...... 6-1 6.2 Adaptive Management...... 6-3 6.3 Monitoring and Evaluation...... 6-5

LITERATURE CITED...... L-1

APPENDIX VOLUME ONE A The EDT Method B Ecological Attributes and Related Survival Factors C Application Steps for the Pierce County Project

APPENDIX VOLUME TWO D Reach Analysis

APPENDIX VOLUME THREE E Catalog of Actions

ii LIST OF TABLES

2-1. Hierarchical organization of Ecological Attributes (Level 2) by categories of major stream corridor features. Corresponding salmonid Survival Factors (Level 3) are shown associated with groups of Level 2 attributes (other associations may also be used in conversion rules). Associations can differ by species and life stage. See Appendix B for association matrices...... 2-4 2-2. Numbers of stream and marine reaches analyzed...... 2-9 3-1. Themes of the goals and values pertaining to salmon recovery in WRIAs 10, 12, and 15...... 3-2 4-1. Geographic areas applied in identifying strategic priorities in Hylebos Basin ...... 4-6 4-2. Geographic areas applied in identifying strategic priorities in Puyallup Basin...... 4-19 4-3. Geographic areas applied in identifying strategic priorities in Puyallup Basin...... 4-36 4-4. Geographic areas applied in identifying strategic priorities in Chambers-Clover Basin...... 4-52 4-5. Geographic areas applied in identifying strategic priorities in Crescent Basin ...... 4-68 4-6. Geographic areas applied in identifying strategic priorities in Burley Basin ...... 4-70 4-7. Geographic areas applied in identifying strategic priorities in Minter Basin...... 4-72 4-8. Geographic areas applied in identifying strategic priorities in Rocky Basin...... 4-75 5-1. Actions grouped by Action Scenario analyzed for Hylebos salmon performance. Action Scenarios are Status Quo Future (SQ) and four scenarios (1-4) with new actions. Actions are organized based on application of strategic priorities described in the text. See Appendix E for action descriptions. The Action Code box is shaded for actions applied to the estuary or Commencement Bay ...... 5-3 5-2. Actions grouped by Action Scenario analyzed for Puyallup and White salmon performance. Action Scenarios are Status Quo Future (SQ) and four scenarios (1-4) with new actions. Actions are organized based on application of strategic priorities described in the text. See Appendix E for action descriptions. The Action Code box is shaded for actions applied to the estuary or Commencement Bay ...... 5-9 5-3. Actions grouped by Action Scenario analyzed for Chambers-Clover salmon performance. Action Scenarios are Status Quo Future (SQ) and four scenarios (1-4) with new actions. Actions are organized based on application of strategic priorities described in the text. See Appendix E for action descriptions ...... 5-18

iii LIST OF FIGURES

2-1. The EDT conceptual framework ...... 2-1 2-2. Data/information pyramid—information derived from supporting levels ...... 2-2 2-3. Ecological information structure...... 2-3 2-4. Measures of biological performance ...... 2-6 3-1. WRIA basins 10, 12, and 15...... 3-3 3-2. Topography in WRIA basins 10, 12, and 15...... 3-6 3-3. Land Cover Classification in WRIA basins 10, 12, and 15...... 3-7 3-4. Hylebos Basin—WRIA 10 ...... 3-11 3-5. Puyallup River Basin—WRIA 10 ...... 3-12 3-6. Tacoma Basin—WRIA 12 (Chambers-Clover Creek) ...... 3-18 3-7. Kitsap Basin—WRIA 15 ...... 3-22 4-1. Hylebos chinook (naturally produced) performance measures based on modeling results...... 4-3 4-2. Relative importance of geographic areas for restoration and protection measures for Hylebos chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contributions of performance measures to rankings are graphed ...... 4-5 4-3. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Hylebos watershed for chinook salmon...... 4-7 4-4. Hylebos coho (naturally produced) performance measures based on modeling results...... 4-9 4-5. Relative importance of geographic areas for restoration and protection measures for Hylebos coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-10 4-6. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Hylebos watershed for coho salmon...... 4-11 4-7. Overview of strategic priorities for restoration and protection measures by geographic area within the Hylebos watershed...... 4-14 4-8. Puyallup chinook (naturally produced) performance measures based on modeling results ...... 4-15 4-9. Estimated spawning distributions of Puyallup chinook based on modeling. Lower Puyallup consists of all areas downstream of the Carbon River; Mid Puyallup includes all areas upstream of the Carbon River but downstream of Electron Dam; Upper

iv Puyallup encompasses areas upstream of Electron Dam; and South Prairie includes all areas within its drainage ...... 4-17 4-10. Productivity estimates for Puyallup chinook subpopulations based on modeling...... 4-17 4-11. Relative importance of geographic areas for restoration measures for Puyallup chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-18 4-12. Relative importance of geographic areas for protection measures for Puyallup chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-20 4-13. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Puyallup watershed for chinook...... 4-22 4-14. Puyallup coho (naturally produced) performance measures based on modeling results...... 4-23 4-15. Relative importance of geographic areas for restoration measures for Puyallup coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-25 4-16. Relative importance of geographic areas for protection measures for Puyallup coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-26 4-17. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Puyallup watershed for coho...... 4-27 4-18. Overview of strategic priorities for restoration and protection measures by geographic area within the Puyallup basin (excluding White basin) ...... 4-31 4-19. White chinook (naturally produced) performance measures based on modeling results...... 4-32 4-20. Estimated spawning distributions of White chinook upstream and downstream of PSE Diversion Dam based on modeling...... 4-33 4-21. Relative importance of geographic areas for restoration measures for White chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-35 4-22. Relative importance of geographic areas for protection measures for White chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-37

v 4-23. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the White watershed for chinook...... 4-38 4-24. White coho (naturally produced) performance measures based on modeling results...... 4-39 4-25. Estimated spawning distributions of White River coho in areas upstream and downstream of PSE Diversion Dam based on modeling...... 4-40 4-26. Numbers of adult coho trapped and hauled at PSE Diversion Dam. Dashed line represents the estimated number arriving to the trap site based on modeling ...... 4-41 4-27. Relative importance of geographic areas for restoration measures for White coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-42 4-28. Relative importance of geographic areas for protection measures for White coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed...... 4-43 4-29. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the White watershed for coho...... 4-44 4-30. Overview of strategic priorities for restoration and protection measures by geographic area within the White basin ...... 4-48 4-31. Chambers-Clover chinook (naturally produced) performance measures based on modeling results...... 4-50 4-32. Relative importance of geographic areas for restoration and protection measures for Chambers-Clover chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-51 4-33. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the White watershed for chinook...... 4-53 4-34. Chambers-Clover coho (naturally produced) performance measures based on modeling results...... 4-54 4-35. Relative importance of geographic areas for restoration and protection measures for Chambers-Clover coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-56 4-36. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Chambers-Clover watershed for coho ...... 4-57 4-37. Overview of strategic priorities for restoration and protection measures by geographic area within the Chambers-Clover basin...... 4-60 4-38. Crescent chinook (naturally produced) performance measures based on modeling results ...... 4-61

vi 4-39. Burley chinook (naturally produced) performance measures based on modeling results...... 4-62 4-40. Minter chinook (naturally produced) performance measures based on modeling results...... 4-63 4-41. Rocky chinook (naturally produced) performance measures based on modeling results...... 4-64 4-42. Relative importance of geographic areas for restoration and protection measures for Crescent chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-67 4-43. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Crescent watershed for chinook salmon ...... 4-69 4-44. Relative importance of geographic areas for restoration and protection measures for Burley chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-70 4-45. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Burley watershed for chinook salmon ...... 4-71 4-46. Relative importance of geographic areas for restoration and protection measures for Minter chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-72 4-47. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Minter watershed for chinook salmon ...... 4-73 4-48. Relative importance of geographic areas for restoration and protection measures for Rocky chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-74 4-49. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Rocky watershed for chinook salmon ...... 4-76 4-50. Crescent coho (naturally produced) performance measures based on modeling results...... 4-77 4-51. Donkey coho (naturally produced) performance measures based on modeling results...... 4-78 4-52. Burley coho (naturally produced) performance measures based on modeling results...... 4-79

vii 4-53. Minter coho (naturally produced) performance measures based on modeling results...... 4-80 4-54. Rocky coho (naturally produced) performance measures based on modeling results...... 4-81 4-55. Flow diversion in Donkey Creek – Harborview Drive culvert...... 4-82 4-56. Relative importance of geographic areas for restoration and protection measures for Crescent coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-84 4-57. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Crescent watershed for coho salmon...... 4-85 4-58. Relative importance of geographic areas for restoration and protection measures for Donkey coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-86 4-59. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Crescent watershed for coho salmon...... 4-87 4-60. Relative importance of geographic areas for restoration and protection measures for Burley coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-88 4-61. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Burley watershed for coho salmon...... 4-89 4-62. Relative importance of geographic areas for restoration and protection measures for Minter coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-90 4-63. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Minter watershed for coho salmon...... 4-91 4-64. Relative importance of geographic areas for restoration and protection measures for Rocky coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed ...... 4-92 4-65. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Rocky watershed for coho salmon...... 4-93 4-66. Overview of strategic priorities for restoration and protection measures by geographic area within Kitsap watersheds...... 4-97

viii 5-1. Hylebos chinook (naturally produced) performance measures under historic, current, and five action scenarios ...... 5-5 5-2. Hylebos coho (naturally produced) performance measures under historic, current, and five action scenarios...... 5-6 5-3. Comparison of action components for the Hylebos Basin ...... 5-7 5-4. Puyallup chinook (naturally produced) performance measures under historic, current, and five action scenarios ...... 5-12 5-5. White chinook (naturally produced) performance measures under historic, current, and five action scenarios...... 5-13 5-6. Puyallup coho (naturally produced) performance measures under historic, current, and five action scenarios...... 5-14 5-7. White coho (naturally produced) performance measures under historic, current, and five action scenarios...... 5-15 5-8. Comparison of action components for the Puyallup-White basins ...... 5-16 5-9. Chambers-Clover chinook (naturally produced) performance measures under historic, current, and five action scenarios ...... 5-20 5-10. Chambers-Clover coho (naturally produced) performance measures under historic, current, and five action scenario...... 5-21 5-11. Comparison of action components for the Chambers-Clover Basin...... 5-23 6-1. Hypothetical salmon production time series. The working hypothesis explains cause and effect based on all available information. It is updated as better information becomes available through monitoring of key habitat attributes, hypothesis testing (e.g., point-to-point survival of treatment groups), and the literature...... 6-2 6-2. Uncertainty classification ...... 6-3 6-3. Mind map of the adaptive management feedback loops. The outermost loops require more policy involvement and are less frequent than the inner loops...... 6-4 6-4. The elements of a simple adaptive management process. The diagram displays the annual cycle of activities that support informed policy decisions. The process revolves around two annually updated reports indicated by the shaded boxes in the diagram ...... 6-5 6-5. Monitoring questions. Implementation monitoring addresses question 1, effectiveness monitoring addresses question 2, and validation monitoring addresses questions 3, 4, and 5...... 6-6

Pierce County Watershed Analysis Executive Summary

EXECUTIVE SUMMARY

Purpose and Scope This document presents a set of strategic priorities for planning and implementing salmon habitat restoration and protection actions in watersheds within Pierce County, Washington. The priorities are based on an assessment of habitat conditions and their effects on the performance of naturally produced salmon. This information is meant to be a valuable aid to the many entities and agencies working within the watersheds to develop and coordinate action plans.

Salmon survival depends on the condition of diverse habitats along the aquatic landscape. The quality and quantity of these habitats, from gravel beds in headwater streams to eel grass beds in nearshore marine areas, can all affect the performance of salmon populations. Protecting or restoring these habitats in a strategic manner will require locally based solutions, suited to the needs of each watershed. This can only be achieved through coordinated multi-jurisdictional efforts, based on a rational process for identifying and prioritizing actions aimed at the factors most affecting salmon survival.

Pierce County has initiated such a process. As a regional agency, it seeks to develop a rational basis for guiding and coordinating salmon conservation and recovery actions within the county. To help do this, Pierce County is applying an analytical approach called Ecosystem Diagnosis and Treatment (EDT), a habitat-based procedure for relating environmental conditions to the performance of salmon populations. EDT captures a wide range of information and makes it accessible to planners, decision-makers and scientists as a working hypothesis of the ecosystem. A strategic assessment was performed and action priorities were derived using this methodology. This document presents the results.

Four basins were included in the analysis: Puyallup-White (WRIA 10), Hylebos (WRIA 10), Chambers-Clover (WRIA 12), and Kitsap (WRIA 15). The analysis focused on chinook and coho salmon and provided inferences about bull trout.

The project had two primary objectives: 1. To complete watershed assessments in the four basins for the focus species, assessing current and historic measures of population performance relative to habitat conditions, and derive strategic priorities for protection and restoration actions. 2. To develop sets of candidate actions for the four basins—each action identified with respect to its strategic priority—and analyze potential benefits to the focus species.

Approach The project was completed in two phases, corresponding to the two objectives: 1) watershed assessment and 2) analysis of action alternatives, providing an overall set of strategic priorities for salmon recovery and protection planning in Pierce County.

June 2001 Mobrand Biometrics, Inc. Page x Pierce County Watershed Analysis Executive Summary

In the assessment phase, we characterized two baseline reference scenarios in terms of environmental conditions and population performance measures: (1) predevelopment or historic conditions and (2) current conditions. We structured the assessment to draw conclusions at basin, subbasin, and stream reach scales. The comparison of these scenarios formed the basis of the diagnostic conclusions about the ways in which the environment and associated salmon performance have been altered by human development. The historic reference scenario also serves to define the natural limits to potential recovery actions.

In the final step of the assessment phase, we derived hypothesis-driven strategic priorities for conservation and recovery actions. These priorities identify the relative importance of geographic areas and their associated environmental factors for protection and/or restoration of salmon. This information is needed for both near and long range action planning, as WRIA steering committees and various stakeholders seek to identify, prioritize, implement, and monitor conservation and recovery actions.

In the second phase of the project, action alternatives were posed as experimental hypotheses to be tested through an adaptive management program. We identified and helped basin stakeholders prioritize near-term conservation actions to protect and restore the ecosystem processes and functions that create and maintain habitat for salmonid species. Each action was described by its approach, feasibility, community support, and cost. Criteria for prioritizing actions focused only on potential benefits to salmon populations.

The EDT method was used to analyze the various past, present, and future habitat scenarios. The principal outputs of the EDT analysis are the parameter estimates of biological performance for the fish populations of interest. We define biological performance in terms of three elements: productivity, capacity, and life history diversity, which are among the performance measures used by NMFS as part of its viable population concept (see McElhany et al. 2000).

Assessment Results The assessment is presented separately for each species by basin, arranged into the following topics: 1. Population performance summaries 2. Strategic priorities for restoration and protection measures 3. Data uncertainties 4. Assessment conclusions

When interpreting and using the information presented in the following sections, the reader should be aware of the following:

Results from the EDT analysis of three Scenarios are presented: (1) the historic environmental conditions as an approximation of maximum potential, (2) current environmental conditions with harvest and genetic fitness loss as an approximation of current potential, and (3) current environmental conditions with no harvest and no loss in fitness. Scenario (2) is the EDT assessment of the current situation—if this assessment is

June 2001 Mobrand Biometrics, Inc. Page xi Pierce County Watershed Analysis Executive Summary accurate it should be consistent with recently observed population numbers. The difference between Scenarios (1) and (3) is an approximation of the loss in potential due to environmental factors alone. To the extent that environmental losses can be mitigated, it also approximates the maximum benefits of habitat restoration.

Examples of analytical results are provided here to illustrate the types of displays and information that are produced by the analysis. The examples focus on the Chambers-Clover Basin, using coho as the diagnostic species.

Chambers-Clover coho show a sharp reduction in population performance measures from historic to current conditions (Figure 4-34). Much of this loss can be attributed to watershed development. The relative importance of geographic areas within the basin to coho for restoration or protection benefits reflects the many alterations that have occurred in the basin (Figure 4-35). Areas of highest priority for restoration are located upstream of Steilacoom Lake and include all of the Clover mainstem, North Fork Clover Creek, and Spanaway Creek. Restoration of others areas would contribute substantial benefits to coho as well. (See Table 4-3 within the main report for a description of geographic areas.). The differences in the relative importance of the geographic areas for coho and chinook are shown in Figure 4.37.

Reach specific strategic priorities for Chambers-Clover coho are provided in Appendix D— Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in watershed planning related to salmon conservation and recovery.

Analysis of Actions The purpose of this step in the analysis was to identify candidate actions and evaluate their potential benefit to the fish populations of interest. These candidate actions should be considered as a starting point for an adaptive planning process where routine updates of the diagnosis and associated watershed goals may lead to additions and/or modifications of actions. As new information regarding action effectiveness becomes available, attention can be refocused on the most successful methods for achieving watershed goals.

June 2001 Mobrand Biometrics, Inc. Page xii Pierce County Watershed Analysis Executive Summary

Chambers-Clover Summary Of Projected Performance Measures Under Three

Scenario Abundance Productivity Diversity index Historic 5,000 27.5 100% Current with harvest and fitness loss 100 6.5 30% Current without harvest and fitness loss 200 9.4 36%

Note: Abundance is the equilibrium spawning population size. Productivity is number of spawners produced per parent spawner at low population density. Diversity index is the percent of life history trajectories modeled that are sustainable (productivit

Coho spawner abundance 1,200 5,000

900

of fish 600 er b m u 300 N

0 Historic Current-with harv Current-no harv Scenario

Coho productivity

30 r 25 20 spawne 15 s per

n 10 tur

e 5 R 0 Historic Current-with harv Current-no harv Scenario

Coho life history diversity 100%

75%

cent 50% r e P 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-34. Chambers-Clover coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page xiii Pierce County Watershed Analysis Executive Summary

Chambers-Clover Coho Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 15 D Puget Sound 14 D Chambers Bay 12 D Chambers mainstem 6 B Leach 8 B Flett 5 B Steilacoom Lake 13 D Ponce de Leon 11 C Lower Clover (to Spanaway) 2 A Morey 6 B Lower Spanaway 1 A Spanaway Lake 10 C Upper Spanaway 8 B NF Clover 2 A Upper Clover 4 A

Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA Chambers Bay NA Chambers mainstem 2 A Leach 9 C Flett 7 B Steilacoom Lake 5 B Ponce de Leon 5 B Lower Clover (to Spanaway) 4 B Morey 11 C Lower Spanaway 1 A Spanaway Lake 7 B Upper Spanaway 11 C NF Clover 10 C Upper Clover 2 A

June 2001 Mobrand Biometrics, Inc. Page xiv Pierce County Watershed Analysis Executive Summary

Chambers-Clover Basin Overview Strategic Assessment of Geographic Areas

Chinook Coho strategic strategic priority priority

Geographic Area Restoration Protection Restoration Protection Chambers Bay Chambers mainstem Leach Flett Steilacoom Lake Ponce de Leon Lower Clover (to Spanaway) Morey Lower Spanaway Spanaway Lake Upper Spanaway NF Clover Upper Clover

Key to Strategic Priority (Benefit Category letter shown)

D & E C B A Indirect or General Low Medium High

Figure 4-37. Overview of strategic priorities for restoration and protection measures by geographic area within the Chambers-Clover basin.

June 2001 Mobrand Biometrics, Inc. Page xv Pierce County Watershed Analysis Executive Summary

Figure 4-36. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Chambers-Clover watershed for coho.

June 2001 Mobrand Biometrics, Inc. Page xvi Pierce County Watershed Analysis Executive Summary

We identified candidate actions in two ways. First, we solicited candidate actions from entities and agencies working in the basins in order to incorporate actions already in progress or under consideration by planners and resource professionals. The procedure used to solicit this information is presented in Appendix E. Information on technical feasibility, community support, outcome certainty, and cost was requested as part of this process. The second means of identifying candidate actions was through the diagnosis: we identified strategies that would address specific problems revealed in the EDT diagnosis step.

Action were organized into four groups or Action Scenarios based on the following conservation and restoration principles: 1. Maintain (or protect) habitat quality and quantity that support the existing core (or most productive) life history patterns of the population; 2. Improve (or restore—even partly) habitat quality and quantity that support the existing core (or most productive) life history patterns of the population; 3. Maintain or restore habitat quality and quantity that support secondary life history patterns (less productive population components) of the population; and 4. Improve habitat quality, or reconnect habitat segments as needed, to recover lost life history patterns that existed in the historic population.

In addition, a Future Status Quo Scenario was constructed that included all actions already implemented or in progress, as well as projected changes in land use as a result of population growth and development over the next 20-25 years.

The evaluation of scenarios for Chambers-Clover Creek are shown here as examples of the way in which results are presented in the report. As shown in Figure 5-10, the analysis suggests that coho salmon would be highly responsive to the Action Scenarios. Notably, we project substantial improvements in performance under the Status Quo Scenario due to the effect of the 28 actions already in progress.

A graphic comparison of all actions related to the Chambers-Clover Basin is seen in Figure 5-11. The display compares actions with respect to effects on attributes, extent of dispersal of effect, technical feasibility, likelihood of outcome, community support, and cost. Details associated with the latter four items are provided in Appendix E.

Uncertainty, Adaptive Management, and Monitoring Our knowledge about the biology of the salmonid species we seek to manage, the present and future condition of their environment, and the effects of human interventions is imperfect. Research progressively increases our understanding of the salmonid species and their habitat requirements and sensitivities, but some uncertainty always remains. Annual returns of chinook salmon, for example, vary greatly from year to year. We can explain some of the variation in terms of parent spawning escapements, ocean and fresh water survival conditions, and so on, but much of the variation remains unexplained. The solution to “how to act in the face of uncertainty” is adaptive management.

June 2001 Mobrand Biometrics, Inc. Page xvii Pierce County Watershed Analysis Executive Summary

Chambers-Clover Coho Summary Of Projected Performance Measures Under Status Quo and Expanded Action Scenarios Scenario Abundance Productivity Diversity index Historic 5,000 27.5 100% Current with harvest and fitness loss 100 6.5 30% Future with status quo mgmt (SQ) 500 9.5 46% Action Scenario-1 (S-1) 700 9.9 60% Action Scenario-2 (S-2) 1,000 12.5 67% Action Scenario-3 (S-3) 1,000 12.3 71% Action Scenario-4 (S-4) 1,000 12.5 71%

Coho spawner abundance 2,500 5,000 2,000 fish f 1,500 o er

b 1,000 m u

N 500 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho productivity

30 r e 25 awn 20 sp r

e 15 s p

n 10 r

tu 5 e R 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho life history diversity 100%

75%

cent 50% r e P 25%

0% Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Figure 5-10. Chambers-Clover coho (naturally produced) performance measures under historic, current, and five action scenarios.

June 2001 Mobrand Biometrics, Inc. Page xviii Pierce County Watershed Analysis Executive Summary

Figure 5-11. Comparison of action components for the Chambers-Clover Basin.

June 2001 Mobrand Biometrics, Inc. Page xix Pierce County Watershed Analysis Executive Summary

Adaptive management is an action-oriented approach. It is a carefully audited process of selecting, implementing, monitoring, and continually reevaluating and adapting strategies to accomplish long-term ecosystem goals. It is important to keep in mind that decisions should be made in an integrated, multi-species context, since salmonid populations and species within a watershed are interdependent and components of a common ecosystem. The adaptive management process requires economic, philosophic, institutional, and policy commitments, as well as the support of scientific expertise and tools.

Adaptive management also requires a strong commitment to monitoring and evaluation. We propose, in the report, an adaptive management program as a mechanism for implementing salmon recovery in the face of uncertainty. The program incorporates monitoring and evaluation and community involvement.

June 2001 Mobrand Biometrics, Inc. Page xx

Pierce County Watershed Analysis Section 1

1.0 INTRODUCTION

This document presents a set of strategic priorities for planning restoration and protection of salmon habitat in watersheds within Pierce County, Washington. The priorities are based on an assessment of the relative contributions of geographic areas, and the environmental factors operative within each, to the performance of naturally produced salmon. This information is meant to be a valuable aid to the many entities and agencies working within the watersheds to develop and coordinate action plans.

Salmon survival depends on the condition of diverse habitats along the aquatic landscape. The quality and quantity of habitats from gravel beds in headwater streams to eel grass beds in nearshore marine areas can all affect the performance of salmon populations. Protecting or restoring these habitats in a strategic manner will require locally based solutions, suited to the needs of each watershed. This can only be achieved through coordinated multi- jurisdictional efforts based on a rational process for identifying and prioritizing actions aimed at those factors that most affect salmon survival.

Pierce County has initiated such a process. As a regional agency, it seeks to develop a rational basis for guiding and coordinating salmon conservation and recovery actions within the county. To this end, Pierce County is applying an analytical approach called Ecosystem Diagnosis and Treatment (EDT)—a habitat-based procedure for relating environmental conditions to the performance of salmon populations. EDT captures a wide range of information and makes it accessible to planners, decision-makers, and scientists as a working hypothesis of the ecosystem. Pierce County contracted with Mobrand Biometrics, Inc., (MBI) to apply the method in analyzing conditions in its watersheds with respect to salmon performance and to derive a strategic assessment of action priorities. This document presents the results of these analyses.

Four basins within the county are addressed: Puyallup-White (WRIA 10), Hylebos (WRIA 10), Chambers-Clover (WRIA 12), and Kitsap (WRIA 15). The analyses focus on two species, chinook and coho salmon, and draw inferences about a third, bull trout. Chinook salmon produced in this geographic area were listed as threatened under the federal Endangered Species Act (ESA) in March 1999 as part of the larger Puget Sound Evolutionary Significant Unit (ESU). At approximately the same time, all bull trout populations within the contiguous United States were listed as threatened, thereby including those in Pierce County. Coho salmon produced in the area are currently a candidate species for ESA listing, as part of the Puget Sound/Georgia Straits ESU.

1.1 Project Objectives The project, as contracted to MBI, had two primary objectives: 5. To complete watershed assessments in the four basins for the focus species, assessing current and historic measures of population performance relative to habitat conditions, and to derive strategic priorities for protection and restoration actions.

June 2001 Mobrand Biometrics, Inc. Page 1-1 Pierce County Watershed Analysis Section 1

6. To develop sets of candidate actions for the four basins—each action identified with respect to its strategic priority—and to analyze possible benefits to the focus species, if implemented.

1.2 Project Overview The project consisted of two phases—corresponding to the two objectives: 1) watershed assessment and 2) analysis of action alternatives. Combined, both phases provide an overall set of strategic priorities for recovery and protection planning within Pierce County.

In the assessment phase, we characterized baseline reference conditions with regard to both environmental conditions and population performance measures. We structured the assessment to draw conclusions at basin, subbasin, and stream reach scales. We characterized two baseline reference scenarios: predevelopment or historic conditions and current conditions. The comparison of these scenarios forms the basis of the diagnostic conclusions about how the basins and associated salmon performance have been altered by human development. The historic reference scenario also serves to define the natural limits to potential recovery actions within the basins.

To perform the assessment, we assembled baseline information on habitat and human-use factors and fish life history patterns. We characterized stream, estuarine, and marine water reaches by a set of defined attributes, drawing on available data and local and regional experts. We analyzed the data sets from species-specific life history perspectives in order to describe population performance in relation to habitat and human-use factors. These characterizations of the environment and resultant species performance are the operating hypotheses about the ecosystem—guiding the strategic assessment and the near-term and long-range salmon recovery planning.

In the final step of the assessment phase, we derived hypothesis-driven strategic priorities for conservation and recovery actions. These priorities identify the relative importance of geographic areas for protection or restoration (or both) and the associated environmental factors. This information is needed for both near-term and long-range action planning as WRIA steering committees and various stakeholders seek to identify, prioritize, implement, and monitor conservation and recovery actions.

In the action analysis phase, action alternatives were posed as experimental hypotheses to be tested through an adaptive management program. We identified and helped Pierce County and other basin stakeholders prioritize near-term conservation actions to protect and restore the ecosystem processes and functions that create and maintain habitat for salmonid species. Criteria for identifying and prioritizing action recommendations included (but were not necessarily limited to) the following: benefits for salmon habitat and salmon recovery, cost- effectiveness, and technical feasibility.

1.3 Use of the EDT Method Ecosystem Diagnosis and Treatment (EDT) is an analytical method relating habitat features and biological performance to support conservation and recovery planning (Lichatowich et al. 1995; Lestelle et al. 1996; Mobrand et al. 1997; Mobrand et al. 1998a ; Mobrand et al.

June 2001 Mobrand Biometrics, Inc. Page 1-2 Pierce County Watershed Analysis Section 1

1998b). It acts as an analytical framework that brings together information from empirical observation, local experts, and other models and analyses.

EDT emphasizes the importance of a science-based approach to recovery planning. Fundamental to the scientific method is the use of an explicit conceptual framework within which information about the natural system is gathered, organized, and analyzed. A logical linkage between actions and events within the watershed and their effect on values and objectives must be presumed and explicitly addressed—a requirement of EDT.

EDT differs from models often used in fish and wildlife management and offers important features that can augment conventional methods. EDT is best described as a scientific model (see Hilborn and Mangel 1997). A scientific model attempts to explain the mechanisms behind phenomenon to form an overall hypothesis. This contrasts with conventional statistical models that provide correlation-based predictions of events without necessarily explaining the underlying mechanism. As a scientific model, EDT constructs a working hypothesis of a watershed as a basis for planning and for comparison of alternative futures. This hypothesis provides metrics to gauge progress and testable hypotheses to refine knowledge. EDT helps us understand and describe the inevitable complexity of ecological systems in order to plan effective recovery strategies. A statistical model, on the other hand, seeks to reduce complexity to a small number of predictive or correlated variables. A scientific model like EDT provides the hypothesis while a statistical model can provide the test. The hypothesis is the rationale that links actions and expected outcome.

Validation of a scientific model as a planning tool means establishing its applicability and utility to the problem at hand. We suggest three criteria or questions for judging the usefulness of such a model: 1) Does it produce results that are consistent with what we observe; 2) How well does it explain what we observe; and 3) Is it useful for guiding future actions?

The EDT method has been widely applied throughout the Pacific Northwest in a variety of rivers. Most noteworthy for Pierce County, it is being used in the Nisqually watershed by the Nisqually Tribe, thereby providing nearly full coverage of waters within the county using this method. Only a portion of WRIA 12 (Sequalitchew watershed) has not been analyzed. Also, the Co-Managers1 are using EDT to assess conditions for chinook salmon in most other rivers within the Puget Sound ESU. That project provided an important information source for the Puyallup estuary and Commencement Bay for this analysis.

1 The Co-Managers are the State of Washington (represented through the Washington Department of Fish and Wildlife) and treaty tribes within the Puget Sound area.

June 2001 Mobrand Biometrics, Inc. Page 1-3 Pierce County Watershed Analysis Section 1

1.4 Document Organization This document is organized into six sections: 1.0 Introduction 2.0 The EDT Method as Applied to Pierce County —a description of the principal parts of the EDT method as it as has been applied in this analysis 3.0 Background —information relevant to the project on goals and values within the county related to salmonid resources and an overview of the relevant watersheds 4.0 The Assessment —the assessment of the focus watersheds with respect to the performance of fish species of interest here 5.0 Analysis of Actions —identification of candidate actions and an analysis of their relative benefits to the species of interest 6.0 Considerations for Monitoring and Implementation —recommendations for application of the results of the analysis

Five appendices accompany this report: A. The EDT Method B. Ecological Attributes and Related Survival Factors C. Application Steps for the Pierce County Project D. Stream Reach Analysis for Species Performance E. Catalog of Actions

June 2001 Mobrand Biometrics, Inc. Page 1-4 Pierce County Watershed Analysis Section 2

2.0 THE EDT METHOD AS APPLIED TO PIERCE COUNTY

This chapter describes the basic components of the EDT method as it was applied in the Pierce County analysis. A more complete description of the conceptual design and application of EDT can be found at http://www.edthome.org. Additional information is also provided in Appendix A of this document.

The EDT method consists of three components: Conceptual and Information Framework—a way of organizing information to describe a watershed ecosystem for analyzing biological performance Analytical Model—a tool used to analyze environmental information and draw conclusions about the ecosystem Step-by-Step Procedure—the steps followed in applying EDT; these are described as applied in the Pierce County analysis

2.1 Conceptual and Information Framework 2.1.1 The Framework Concept The conceptual framework consists of three major elements: the vision, the set of biological objectives, and the strategies for moving the watershed toward the vision (Figure 2-1). The vision describes a set of desired future conditions with regard to biological, economic, and social values. In an ESA context, these desired conditions address recovery objectives for salmon species. The biological objectives describe the vision with respect to the characteristics of the environment and associated biological performance of species under those conditions. The strategies are those actions intended to achieve the biological objectives. This simple framework forms the core of the EDT method—it is the framework that has been adopted by the Northwest Power Planning Council for planning recovery actions in the Columbia Basin (EWG 1998).

Biological objectives

Ecological Biological Rationale Rationale Rationale Actions attributes performance Vision

Figure 2-1. The EDT conceptual framework.

This framework is the pathway for linking various potential watershed actions to desired outcomes. It provides the rationale for identifying how actions are transferred through the ecosystem into resource outcomes. The framework explains possible consequences of actions in a manner consistent with existing knowledge and information, and it requires that assumptions necessary to watershed planning be identified—thus it becomes a vehicle for learning and communicating.

June 2001 Mobrand Biometrics, Inc. Page 2-1 Pierce County Watershed Analysis Section 2

2.1.2 Ecological Information Structure The Information Structure and associated data categories are defined at three levels of organization. Together, these can be thought of as an information pyramid in which each level builds on information from the lower level (Figure 2-2). As we move up the through the three levels, we take an increasingly organism-centered view of the ecosystem. Levels 1 and 2 together characterize the environment, or ecosystem, as it can be described by different types of data (Figure 2-3). This provides the characterization of the environment needed to analyze biological performance for a species. The Level 3 category is a characterization of that same environment from a different perspective: “through the eyes of the focal species" (Mobrand et al. 1997). This category describes biological performance in relation to the state of the ecosystem described by the Level 2 ecological attributes.

Umbrella attributes (classes of Level 3- Biometrics attributes) - "through the eyes of species" - short list

Level 2-Ecological attributes

Level 1- wide range of data types

Figure 2-2. Data/information pyramid—information derived from supporting levels.

The organization and flow of information begins with a wide range of environmental data (Level 1 data) that describe a watershed, including all of the various types of empirically based data available. These data include reports and unpublished data. Level 1 data exist in a variety of forms and pedigrees. The Level 1 information is then summarized or synthesized into a standardized set of attributes (Level 2 ecological attributes, see Table 2-1) that refine the basic description of the watershed. The Level 2 attributes are descriptors that specify physical and biological characteristics about the environment relevant to the derivation of the survival and habitat capacity factors for the specific species in Level 3. Definitions for Level 2 and Level 3 attributes are given in Appendix B, together with a matrix showing associations between the two levels.

The Level 2 attributes represent conclusions that characterize conditions in the watershed at specific locations, during a particular time of year (season or month), and for an associated management scenario. Hence an attribute value is an assumed conclusion by site, time of year, and scenario. These assumptions become operating hypotheses for these attributes under specific scenarios. Where Level 1 data are sufficient, these Level 2 conclusions can be derived through simple rules. However, in many cases, experts are needed to provide knowledge about geographic areas and attributes where Level 1 data are incomplete. Regardless of the means whereby Level 2 information is derived, the characterization it provides can be ground-truthed and monitored over time through an adaptive process.

June 2001 Mobrand Biometrics, Inc. Page 2-2

Biological objectives

Ecological Biological Rationale Rationale Rationale Actions attributes performance Vision

Ecosystem Characterization Biological Performance

Population Level 2: Ecological Level 3: Survival Level 1: Input Data Performance (by Attributes Factors species) Expert Environmental Digital Species - Capacity ( or Rating or abundance) description data: characterization of Rules performance EDT Model - Location Rules ecosystem: responses: - Standard list of - Expected response - Productivity - Dimensions attributes (survival parameters) - Life history diversity - Land use - Attributes rated - Defined through using guidelines "umbrella attributes" - Land cover - All geographic units - Species specific - Environmental data (biotic & abiotic) - Across calendar year

Figure 2-3. Ecological information structure.

Pierce County Watershed Analysis Section 2

Table 2-1. Hierarchical organization of Ecological Attributes (Level 2) by categories of major stream corridor features. Corresponding salmonid Survival Factors (Level 3) are shown associated with groups of Level 2 attributes (other associations may also be used in conversion rules). Associations can differ by species and life stage. See Appendix B for association matrices.

Related Survival Factors Ecological Attributes (Level 2) (Level 3) 1 Hydrologic Characteristics 1.1 Flow variation Flow - change in interannual variability in high flows Flow Flow - changes in interannual variability in low flows Withdrawals (entrainment) Flow - Intra daily (diel) variation

Flow - intra-annual flow pattern

Water withdrawals 1.2 Hydrologic regime Hydrologic regime - natural Hydrologic regime - regulated 2 Stream Corridor Structure 2.1 Channel Channel length morphometry Channel length Channel stability Channel width - month maximum width Channel width Channel width - month minimum width Habitat diversity Gradient Key habitat Obstructions 2.2 Confinement Confinement - hydromodifications Sediment load Confinement - natural 2.3 Habitat type Habitat type - backwater pools Habitat type - beaver ponds Habitat type - glides

Habitat type - large cobble/boulder riffles Habitat type - off-channel habitat factor Habitat type - pool tailouts Habitat type - primary pools Habitat type - small cobble/gravel riffles

2.4 Obstruction Obstructions to fish migration 2.5 Riparian and channel integrity Bed scour Icing Riparian function Wood 2.6 Sediment type Embeddedness Fine sediment (intragravel) Turbidity 3 Water Quality 3.1 Chemistry Alkalinity Chemicals (toxic substances) Dissolved oxygen Oxygen Metals - in water column Temperature Metals/Pollutants - in sediments/soils Miscellaneous toxic pollutants - water column Nutrient enrichment 3.2 Temperature variation Temperature - daily maximum (by month)

Temperature - daily minimum (by month) Temperature - spatial variation

June 2001 Mobrand Biometrics, Inc. Page 2-4 Pierce County Watershed Analysis Section 2

Table 2-1 continued. Related Survival Factors Ecological Attributes (Level 2) (Level 3) 4 Biological Community 4.1 Community effects Fish community richness Competition with hatchery fish Fish pathogens Competition with other fish Fish species introductions Food Harassment Harassment Hatchery fish outplants Pathogens Predation risk Predation Salmon carcasses 4.2 Macroinvertebrates Benthos diversity and production

In the Pierce County process, conclusions regarding Level 2 attribute conditions were derived by a group of natural resource-related professionals with knowledge of the watersheds of interest. These individuals had expertise in such disciplines as fish habitat, hydrology, geomorphology, water quality, and civil engineering.

The link between Level 2 attributes and Level 3 factors is made through sets of rules. The rules translate the Level 2 characterization of the environment into biological performance by life stage for a focus species. Biological performance describes how a species reacts to characteristics of its environment in terms of survival (productivity) and capacity. The rules are defined through the Level 3 Survival Factors (Table 2-1), which act as "umbrella attributes" grouping Level 2 attributes together.

A separate set of biological rules for doing the conversion from Level 2 to Level 3 has been derived for chinook and coho salmon. The rules are provisional—they are currently being reviewed through a formal process in the region. They are a characterization of our understanding of the relation between the environment and salmon survival at the current time. We expect that the rules will be refined through the review process. Additional information on the rules and the review process can be found at http://www.edthome.org.

The Level 3 Survival Factors serve as the input to the EDT model for estimating population response measures. These measures are the currency for formulating and comparing strategic priorities and sets of conservation and recovery actions.

The remaining component that is incorporated into the Information Structure is the set of candidate actions to be considered for implementation. Actions—defined through assumptions about effectiveness, dispersal of effect, and time lag to achieve full effect—are evaluated by examining how they result in changes to Level 2 attributes, which in turn affect Level 3 factors and population performance measures. In the Pierce County process, assumptions about action effectiveness were made with the aid of a working group of civil engineers and biologists. These assumptions represent objectives for the actions that can, if implemented, be monitored for effectiveness.

2.2 Analytical Model The tools essential for applying the EDT method have been assembled into the EDT model: a repository of data, information, and knowledge, as well as a collection of analytical

June 2001 Mobrand Biometrics, Inc. Page 2-5 Pierce County Watershed Analysis Section 2

procedures. It includes a database that stores and documents information about the geography and physical characteristics of the watersheds of interest. Also included are databases that describe and document the biology, life history characteristics, and environmental sensitivities of a set of indicator species. The EDT model includes a module for developing alternative future scenarios by defining action strategies and targeted environmental attributes.

The EDT model makes it possible to manage the complexity and quantity of detailed information needed to use the EDT method. The model allows us to address tractable issues and problems in the context of a broad framework, which integrates a wide range of scientific disciplines. The model is a tool for achieving accountability: it expands the ability of scientists to keep track of complex relationships and opens broader horizons for creativity.

The analytical tools included in the model compute the various diagnostic indicators described and displayed elsewhere in this document. The principal output are the parameter estimates of biological performance for the fish populations of interest. These parameters are then used by the model in deriving other diagnostics of interest, such as strategic priorities for conservation and recovery actions.

We define biological performance in terms of three elements: productivity, capacity,2 and life history diversity (Figure 2-4). These measures are characteristics of the ecosystem that describe persistence, abundance, and distribution potential of a population. They are the core performance measures used by the National Marine Fisheries Service (NMFS) as part of its viable population concept (McElhany et al. 2000). Each measure is defined briefly below.

Life history diversity

Productivity Capacity

Figure 2-4. Measures of biological performance.

Productivity. This element represents the relative success of the species to complete its life cycle within the environment it experiences.3 It determines resilience to mortality pressures,

2 We use the terms productivity and capacity as defined by Hilborn and Walters (1992). Capacity is the maximum population size for one or more life history segments. Capacity and productivity are not independent. 3 The productivity rate is the reproductive rate measured over a full generation that would occur at low population density, i.e., when competition for resources among the population is minimal.

June 2001 Mobrand Biometrics, Inc. Page 2-6 Pierce County Watershed Analysis Section 2

such as from fishing, dams, and further habitat degradation. Habitat quality (including water quality) is a major determinant of a population’s productivity. This performance element is especially important when efforts are being made to reverse long-term downward trends in population abundance. The model estimates productivity for the population of interest under specific management scenarios, expressed as the average number of adult progeny produced per parent spawner (at low population density). A life cycle productivity less than 1 for any part of the population is, by definition, unsustainable. As population productivity approaches 1 (e.g., values less than 2),4 the population is clearly at risk.

Capacity. This element defines how large a population can grow within the environment it experiences, as a result of finite space and food resources. It determines the effect of this upper limit on abundance to survival and distribution. Habitat quantity is a major determinant of the environmental capacity to support population abundance. In the analysis presented here, we frequently refer to "abundance" rather than capacity. Here we are describing the equilibrium run size abundance (or average abundance under steady state conditions), which highly correlates with capacity. The model estimates both capacity and equilibrium abundance for the population of interest corresponding to specific management scenarios.

Life History Diversity. This element represents the multitude of pathways through space and time available to, and used by, a species in completing its life cycle. Populations that can sustain a wide variety of life history patterns are likely to be more resilient to the influences of environmental change. Thus a loss of life history diversity is an indication of declining health of a population (Lichatowich and Mobrand 1995) and perhaps its environment. The model computes an index of life history diversity as the percentage of possible life cycle pathways (i.e., life trajectories in space and time that members of a population might follow across the aquatic landscape) having a productivity greater than 1.

The algorithms used to calculate population parameters are based on the Beverton-Holt survival function (Beverton and Holt 1957). All of the estimates are made for steady state conditions. The derivation of some of the key relationships used in the EDT analysis are presented in Appendix A.

2.3 Step-by-Step Procedure The EDT method consists of a series of steps adapted for the Pierce County analysis (Lestelle et al. 1996). The steps are outlined below: 7. Identification of goals and values 8. Resource assessment (or diagnosis) 9. Analysis of actions Considerations for monitoring and implementation

Each step is described below, and further details are provided in Appendix C.

4 The life cycle productivity needed to sustain a population in the face of environmental uncertainty has not been defined.

June 2001 Mobrand Biometrics, Inc. Page 2-7 Pierce County Watershed Analysis Section 2

2.3.1 Identification of Goals and Values Watershed goals for fish resources are derived from social, cultural, political, and legal considerations in a policy environment. The EDT process does not presume agreement between the various values and goals, but it emphasizes the importance of identifying all of them. Goals and values provide the currency whereby projected outcomes of actions can be evaluated.

Many statements regarding goals and values for salmonid resources within the boundaries of Pierce County have been issued by various entities and agencies. We reviewed documents published by these entities in order to identify major themes within the range of statements issued. These representative statements are presented in Section 3 of this report. We do not suggest that these are definitive statements of goals and values for fish resources within the county—merely that they reflect a basis for developing more specific and comprehensive goals with regard to salmon conservation and recovery actions.

2.3.2 Resource Assessment During the resource assessment step, we diagnosed the environmental impediments to achieving the goals and values associated with the salmon resources of Pierce County. This step was structured to produce conclusions drawn at basin, subbasin, and stream reach scales. The assessment, thus, provides a comprehensive, analytically-derived limiting factors analysis5 of each watershed, from which we formulated strategic priorities for conservation and restoration measures.

The resource assessment consisted of two tasks: 1) baseline information assembly and 2) analysis and diagnosis.

2.3.2.1 Baseline Information Assembly To perform the assessment, we assembled baseline information on habitat and human-use factors and fish life history patterns for the watersheds of interest and adjoining estuarine, nearshore, and deep water marine areas. We first structured the entirety of the relevant geographic areas, including marine waters, into distinct habitat reaches. We identified reaches on the basis of similarity of habitat features, drainage connectivity, and land use patterns (Table 2-2). This task required that all reaches be completely characterized by the relevant environmental attributes.

We formed technical work groups for each of the basins (see Appendix C) for the purpose of deriving the Level 2 attribute conclusions for the freshwater stream reaches. Expert knowledge about habitat identification, habitat processes, hydrology, water quality, and fish biology was incorporated into the process. The work groups drew upon published and unpublished data and information for the basins to complete the task.

5 The term "limiting factors analysis" is widely used in the Pacific Northwest to refer to various types of analyses of the importance of different environmental factors to salmon performance. Often these are not analytically derived. Notably, the EDT method does provide an analytically-derived analysis—one that examines the relative contributions of all factors to the loss in salmon performance.

June 2001 Mobrand Biometrics, Inc. Page 2-8 Pierce County Watershed Analysis Section 2

Table 2-2. Numbers of stream and marine reaches analyzed.

Watershed6 or marine area No. of reaches Hylebos (WRIA 10) 25 Puyallup-White (WRIA 10) 261 Chambers-Clover (WRIA 12) 31 Kitsap (WRIA 15) 61 Puget Sound 35

We employed a similar process for estuarine and marine areas on a project working with the Co-Managers assessing Puget Sound rivers using the EDT method. For that project, we formed a technical work team of experts to characterize the estuaries and nearshore areas with respect to the Level 3 survival factors (see Appendix C). A process that follows the entire ecological information structure depicted in Figure 2-3 is still being formulated. For the Co-Manager's assessment, and for the Pierce County project being reported here, the Level 3 factors act as umbrella attributes that served the same purpose that the Level 2 attributes served for freshwater reaches.

For the Pierce County project, the technical work team, with expertise in the Puyallup estuary and Commencement Bay, reviewed and refined the results of the work performed under the Co-Managers' project.

We characterized two baseline reference scenarios for each watershed and marine area: predevelopment (or historic) conditions and current conditions. The comparison of these scenarios formed the basis for diagnostic conclusions about how the basins and associated salmon performance have been altered by human development. The historic reference scenario also served to define the natural limits to potential recovery actions within the basins.

2.3.2.2 Analysis and Diagnosis We analyzed the data sets from species-specific, life history perspectives using the EDT model to estimate population performance measures in relation to the habitat and human- use factors associated with each scenario. The estimates provided an approximation of the extent that environmental change has affected performance of these salmon populations. The analysis also incorporated information on harvest and genetic fitness effects, enabling us to estimate the portion of lost performance that is due to environmental effects.

The objective of the diagnosis then became identifying the relative contributions of environmental factors to the losses in salmon performance. To accomplish this, we performed two types of analyses, each at a different scale of overall effect.

6 Includes estuarine portion of watershed.

June 2001 Mobrand Biometrics, Inc. Page 2-9 Pierce County Watershed Analysis Section 2

The first analysis was done across geographic areas relevant to populations, where each geographic area typcially encompasses many reaches. This analysis, called the Geographic Area Analysis, identified the relative importance of each area for either restoration or protection actions. In this case, we analyzed the effect of either restoring or further altering environmental conditions on population performance.

The second analysis considered conditions within individual stream reaches and identified the most important factors contributing to a loss in performance corresponding to each reach. This analysis, called the Stream Reach Analysis (Appendix D), identified the factors (classes of Level 2 attributes) that, if appropriately moderated or corrected, would produce the most significant improvements in overall fish population performance. It identified the factors that should be considered in planning habitat restoration projects.

Together, these two analyses formed the basis for identifying strategic priorities for conservation and restoration measures within each watershed.

2.3.3 Analysis of Actions The purpose of this step in the analysis was to identify candidate actions and analyze them for their potential benefit to the fish populations of interest. This identified set of candidate actions should be considered only for near-term planning. The adaptive planning process requires routine updates of the diagnosis (incorporating new information) and associated watershed goals, prompting possible realignments in strategic priorities. Also, as new information becomes available about action effectiveness, it will refocus attention to the most successful ways of achieving watershed goals.

We identified candidate actions in two ways. For the first method, we solicited candidate actions from entities and agencies working in the basins. By collecting ideas that were already being considered by planners and resource professionals as reasonable and relevant to salmon conservation and recovery, we hoped to avoid analyzing actions that had no reasonable merit within the socio-economic setting of Pierce County. Information on technical feasibility, community support, outcome certainty, and cost was requested as part of this process. Appendix E includes a description of the solicitation procedure as well as the actual information gathered (Catalog of Actions).

The second method for identifying candidate actions was through the diagnosis. We identified strategies that might be employed to address specific issues.

Once candidate actions were identified, they were organized and sequenced for analysis. Our task was not to arrange the actions into alternative basin plans—it was simply to analyze their potential benefits and provide the results back to community decision makers. Because the actions were numerous, the constraints on the analysis required that we organize them into groups, arranged according to strategic priority. We then performed the analysis on these strategically sequenced groupings.

In the process of identifying candidate actions, we learned that many actions were already in various stages of implementation. We incorporated these actions into the analysis to characterize a future condition represented by existing or status quo management. This characterization became a third baseline reference condition, together with reference

June 2001 Mobrand Biometrics, Inc. Page 2-10 Pierce County Watershed Analysis Section 2 conditions for current and historic scenarios, for analyzing the effects of new actions. We included in this new characterization projected patterns of human population growth in the watersheds and the associated effects on environmental conditions. A technical work team consisting of planners in the region informed us about likely growth patterns and changes in the percent of impervious surfaces in the watersheds.

Assumptions about action effectiveness were made with the aid of a working group of civil engineers and biologists. These assumptions represent objectives for the actions that can, if implemented, be monitored for effectiveness.

2.3.4 Considerations for Monitoring and Adaptive Implementation The final step in the EDT process consisted of summarizing the principal findings of the project into a set of recommendations to Pierce County for action monitoring and adaptive implementation.

June 2001 Mobrand Biometrics, Inc. Page 2-11

Pierce County Watershed Analysis Section 3

3.0 BACKGROUND

This section provides background information useful for interpreting and applying the analysis contained in this document. We first present information for understanding goals and values represented in the county related to natural salmonid production, followed by a description of relevant aspects of the environment.

3.1 Goals and Values Many statements regarding goals and values for aquatic resources within the boundaries of Pierce County have been issued by various organizations and governments. We reviewed a wide variety of documents published by these entities in order to identify major themes within the range of statements issued (e.g., Pierce County 1999; Tetra Tech/KCM 2000; Tri- County 2000; Tri-County Executive Committee 1999). We also solicited input on this matter from the many entities that attended EDT workshops during the project. Although there exists no common, agreed upon set of goal statements within the county, these themes reflect the values and visions held for fish and closely related resources (Table 3-1).

We do not suggest that these are definitive statements of goals and values for these resources within the county—but they provide a basis for developing more specific and comprehensive goals with regard to salmon conservation and recovery actions. A recurring theme is to maintain, or recover, wild populations of salmonids at sustainable, viable levels. This core value is reflected throughout the county, as seen in the increasing emphasis given to issues affecting the performance of naturally produced salmonids.

We recognize that the values reflected in Table 3-1 are held in the context of a very diverse set of socio-economic, and cultural values and goals within the communities of Pierce County. The Pierce County region is a major population and industrial center within Washington State, and the dominant goals represented in the county reflect that. This fact does not diminish the potential application of the EDT results—it makes them all the more important. The challenge facing entities operating within Pierce County is to strategically prioritize conservation and restoration measures to provide a balanced approach for realizing the diversity of goals and values across the county.

The values reflected in Table 3-1 should not be interpreted to mean that only naturally produced salmonids are valued within the county. In fact, a large percentage of salmon produced there, the majority in some years, are propagated in hatcheries. Hatchery salmon produced at the Voight Creek, Diru Creek, and Chambers Creek hatcheries return large numbers of adult fish that contribute to local fisheries. Tribal fisheries are to a large extent sustained by this production. While the Co-Managers place high value on these fisheries, they recognize that naturally produced populations are essential for the long-term health of the resource.

It should be noted that fish population performance needed to maintain, or achieve, sustainable, viable status is not well defined at this time. The federal agencies responsible for listing chinook and bull trout in the region have not yet identified delisting criteria or described a method for developing such criteria.

June 2001 Mobrand Biometrics, Inc. Page 3-1 Pierce County Watershed Analysis Section 3

Table 3-1. Themes of the goals and values pertaining to salmon recovery in WRIAs 10, 12, and 15.

Maintain, or recover, wild populations of salmonids at sustainable, viable levels. Provide and protect functioning and sustainable ecosystems where selected habitats and species will be enhanced to provide a net gain of habitat function beyond existing conditions. Develop and enhance open water, emergent, and riparian freshwater habitat to promote habitat values associated with freshwater wetland systems. Enhance nearshore and intertidal habitats for salmonid and benthic resources. Protect water quality and beneficial uses of water for fish, shellfish, macroinvertebrate, wildlife and human populations. Revise, refine and implement existing regulations and programs to protect and enhance salmonid habitat. Conserve resource lands, minimize urban sprawl, maintain the integrity of open space corridors and encourage economic development within the capacity of the natural resources. Provide fishery opportunities to tribal and non-tribal fishers consistent with treaty obligations, while limiting overall fishing mortality to ensure the long-term health and viability of naturally produced populations.

3.2 The Environment General environmental descriptions of Water Resource Inventory Areas (WRIAs) 10, 12, and 15 are provided below. More specific information on watershed origin, migration barriers, riverine habitat, and fisheries distribution for each WRIA follows the general description.

3.2.1 General Characteristics 3.2.1.1 Geographic Setting The Puyallup River Basin (WRIA 10) is the largest of the four Pierce County WRIAs (Figure 3-1—WRIA Basins). Draining an area of approximately 1,065 square miles, it compasses all or parts of more than a dozen cities and towns, including the state's third largest, Tacoma (Kerwin 1999). Other municipalities include Fife, Puyallup, Sumner, Edgewood, Milton, Bonney Lake, Orting, Buckley, South Prairie, Wilkeson, Enumclaw, and Carbonado. Much of the higher elevation lands are under federal ownership, managed by the National Park Service and U.S. Forest Service. Also included in WRIA 10 is the Hylebos Basin, a small watershed that encompasses much of Federal Way. These basins drain to Commencement Bay, a natural deep water embayment of approximately 5,700 acres in size.

June 2001 Mobrand Biometrics, Inc. Page 3-2

Figure 3-1. WRIA basins 10, 12, and 15.

Pierce County Watershed Analysis Section 3

The Tacoma Basin (WRIA 12) is a small lowland area of the central Puget Sound region lying between the Nisqually River Basin to the south and the Puyallup River Basin to the north (Figure 3-1—WRIA Basins). The lower portion of the WRIA is within the city limits of Tacoma, University Place, Lakewood, and Steilacoom. Much of the upper portion contains low elevation timberlands that are gradually being developed as urban growth increases.

The Kitsap Basin (WRIA 15) is located in central Puget Sound and consists of 13 subbasins draining portions of Kitsap, Pierce, and Mason counties (Figure 3-1—WRIA Basins). The Kitsap Basin includes several islands and the southern portion of the Kitsap Peninsula, which is

surrounded by marine inlets. The focus of this analysis is on Carr Inlet and those portions of Case Inlet, Henderson Bay, and Colvos Passage that are located within the Pierce County area of WRIA 15, at the southeastern end of the Kitsap Peninsula, and west of Tacoma. The WRIA is within the city limits of Gig Harbor and the communities of Purdy, Vaughn, Home, and Longbranch.

3.2.1.2 Climate Climate conditions within WRIAs 12 and 15 and the lower reaches of WRIA 10 reflect cool dry summers and mild wet winters. Average winter temperatures range between 40° F and 44°F. Annual low temperatures in the low to mid-30° F range usually occur during December or January. Summer temperatures average around 60° F, with highs in the mid- 70° F range during July and August. An average annual precipitation of approximately 40 inches falls almost exclusively as rain in WRIA 12 and the lower portions of WRIA 10, with only 0.3 inches recorded as average annual snowfall. Annual average precipitation increases to approximately 50 inches in WRIA 15 due to a more maritime climate.

Climate conditions vary considerably in the upper reaches and headwaters of the Puyallup River Basin. Originating on the slopes of Mt. Rainier, the upper Puyallup River watershed experiences an average annual minimum temperature of 30.1° F, with the average low of 20.3° F occurring during the month of January. The average annual maximum temperature in the upper watershed is 45.1° F, with the highest average temperature of 61.7° F occurring in August. Average annual precipitation at these high elevations is 116.93 inches of rainfall and 682.4 inches of snowfall, resulting in an average snow depth of 178 inches during the month of April (WRCC 2000).

3.2.1.3 Land Use Much of land or water surface area contained by the three WRIAs have undergone major alterations due to land use activity compared to their pristine states. The lower half of WRIA 10 and most of WRIA 12 bear little resemblance to their historic condition. Extensive alterations to land forms, river courses, stream channels and estuaries have occurred as a result of urban, industrial, and agricultural development. In contrast, the high elevation areas of WRIA 10 remain for the most part pristine within Mt. Rainier National Park.

June 2001 Mobrand Biometrics, Inc. Page 3-4 Pierce County Watershed Analysis Section 3

The upper elevations of streams in WRIA 10 are highly mountainous with steep gradient channels and cascading stream flows (Figure 3-2--Topography). State and private timber interests, National Forests, and Mount Rainier National Park encompass most of the upper watershed, resulting in heavily forested stream banks (Figure 3-3—Land Cover Classification) (R2 1999). Logging is the most significant development factor in the upper Puyallup watershed.

Agricultural, forestry and rural low-density housing and communities dominate WRIA 15 and the mid-section of the Puyallup River Basin. In WRIA 15 the majority of development has occurred along the moderate or low bank waterfront areas. Flood control features dominate the lowermost 25 miles of the mainstem Puyallup River in the form of levees, dikes, channelization or stream straightening (Williams et al. 1975; Kerwin 1999). The lowest eight miles of the Puyallup River Basin are predominantly open farmland with municipal and residential developments and heavy industrialization on the Commencement Bay waterfront. The entire WRIA 12 watershed has been heavily urbanized with residential, commercial and industrial developments and falls under multiple municipal jurisdictions throughout its course.

The Port of Tacoma, located on the Puyallup River estuary (including the Hylebos) and Commencement Bay, is today the third largest commercial shipping terminus in the western United States. Major alterations to the natural estuarine and nearshore features have occurred, due to extensive filling and dredging (U.S. EPA 1989).

3.2.1.4 Salmonid Species Six anadromous salmonid and two char species are found within the Puyallup River Basin, the Tacoma Basin, and that portion of the Kitsap Basins located within Pierce County. These species include chinook (Oncorhynchus tshawytscha), coho (Oncorhynchus kisutch), chum (Oncorhynchus keta) and pink salmon (Oncorhynchus gorbuscha); steelhead (Oncorhynchus mykiss) and sea-run cutthroat trout (Oncorhynchus clarki clarki); and native char bull trout (Salvelinus confluentus) and Dolly Varden (Salvelinus malma). Sockeye salmon (Oncorhynchus nerka) are considered indigenous to the Puyallup River Basin and are occasionally observed during Washington Department of Fish and Wildlife (WDFW) spawning surveys in South Prairie Creek; however, their number are considered insignificant relative to other anadromous species in the basin (R2 1999).

June 2001 Mobrand Biometrics, Inc. Page 3-5

Figure 3-2. Topography in WRIA basins 10, 12, and 15.

Figure 3-3. Land cover classification in WRIA basins 10, 12, and 15.

Pierce County Watershed Analysis Section 3

3.2.1.4.1 Anadromous Salmonids Chinook: Chinook salmon in the lower Puget Sound region are generally separated into two dominant life-history types: an ocean type that is typical of fall-run stocks and a stream type that is characteristic of spring-run fish. Adult spring chinook utilize principally the higher elevations portions of streams and rivers, while fall chinook generally utilize mainstem riverine habitat for spawning. Adult spring chinook enter the rivers and streams beginning in mid- to late March and continuing through July, with peak entry occurring between August and September. Spring chinook spawning can begin as early as late July with the majority spawning in late August to early September. Fall chinook spawning begins in early September and is usually completed by mid-November. Following emergence from the gravel, spring chinook juveniles characteristically remain in the riverine system for one year, migrating seaward between early March and late July. Fall chinook juveniles generally rear in the system about three to six months prior to seaward migration, with this occurring mainly from late February through early August. However, some chinook emigration can occur throughout the year (Williams 1975). Chinook salmon can reside anywhere from two to five years in the ocean, although jacks may spend less than a year at sea before returning to spawn. Overall abundance of chinook salmon in Puget Sound has declined substantially from historic levels and short- and long-term trends in escapement are predominately decreasing (R2 1999). The National Marine Fisheries Service (NMFS) listed the Puget Sound chinook salmon as Threatened under the Endangered Species Act (ESA) on March 24, 1999. Coho: Coho salmon generally spawn in small, low-gradient streams or stream reaches. Adult coho begin entering Puget Sound rivers and streams in late July and continue into December. Spawning commences about mid-October and continues in some areas until mid-January. Following emergence from the gravel, the juvenile coho characteristically remain in the system more than a year, migrating to sea early in their second year of freshwater life. The bulk of migration occurs between early March and early August. Coho spend approximately 18 months at sea, although jack males may return after only six months in the ocean. Although NMFS considers that coho salmon populations in Puget Sound are abundant, they have also stated that it is difficult to distinguish between self-sustaining, native stocks and artificially propagated stocks. Puget Sound coho salmon are therefore considered a candidate species for future listing under the ESA.

Chum: Chum salmon typically spawn in coastal streams not far from tidewater, occasionally within the tidal zone. Fall run adult chum begin entering rivers and streams in late September with spawning commencing in mid-November and continuing in some areas into February. Soon after emerging from the gravel the young chum move quickly seaward to estuary habitat, with this migration usually complete by mid-July. Chum salmon spend two to five years in the ocean before returning to their natal stream. This stock is considered healthy and not at risk of extinction or likely to become endangered in the foreseeable future.

Pink: Pink salmon utilize coastal streams in near proximity to tidal influences. Adult pinks enter the Puget Sound streams from mid-July into October, during odd years. Spawning commences in mid- to late September and is usually completed by early November. Soon after fry emerge from the gravel, their seaward migration begins, with this movement usually completed by the end of June. Almost without exception, pink salmon mature at two years of age, at which time they return to freshwater to spawn. On October 4, 1995, NMFS

June 2001 Mobrand Biometrics, Inc. Page 3-8 Pierce County Watershed Analysis Section 3

determined that an ESA listing of Puget Sound odd year runs of pink salmon was not warranted.

Steelhead: Steelhead can be anadromous or freshwater resident (rainbow or redband) trout and under some circumstances, apparently yield offspring of the opposite form. The most widespread run type of steelhead is the winter (ocean-maturing) steelhead. Summer (stream- maturing) steelhead are less common. A life-history variation in steelhead from other anadromous species is their ability to spawn more than once, however, repeat spawning is rare for populations north of Oregon, and more than two spawning migrations are uncommon (Busby et. al. 1996). Adult winter steelhead migrate to their natal streams during December through March and spawn through May, whereas summer steelhead enter freshwater during June or July and over-winter in freshwater until they spawn the following March or April. Wild steelhead juveniles typically spend two years in their freshwater environment prior to ocean migration, although some migrate seaward at age three (Hymer et. al. 1992). Hatchery steelhead smolts are reared and released at age one, and this difference in age is used by biologists to distinguish them from wild steelhead. Steelhead smolts begin their outmigration in late April through June. Washington coastal and Puget Sound steelhead most commonly spend two years in the ocean prior to spawning. Protection of Puget Sound steelhead stocks under the ESA is considered unnecessary at this time by the NMFS.

Sea-run Cutthroat: Coastal cutthroat trout belong to the same genus as Pacific salmon and steelhead, but are generally smaller, rarely over-winter at sea, and do not usually make extensive ocean migrations. Like steelhead, coastal cutthroat trout populations may contain both migratory and non-migratory individuals within the same population. Coastal cutthroat trout have the ability to spawn more than once, with some individuals known to spawn each year for more than six years. Coastal cutthroat trout in the Puget Sound region are considered to be "late entering," as adults return to freshwater from December through March. Spawning typically starts in December and continues through June, with a peak in February. Spawning takes place in the upper reaches of small, low gradient streams and in the upper reaches of small tributaries of moderate-size streams. Juvenile cutthroat trout migrate to sea between the ages of one and six; however, the majority is reported to migrate at age two, three, or four. Outmigration begins as early as March and peaks in mid-May. Coastal cutthroat trout return to freshwater each year to feed and over-winter (Johnson et al. 1999). On April 5, 1999, the NMFS concluded that available scientific data did not warrant listing of the Puget Sound populations of coastal cutthroat trout under the ESA.

3.2.1.4.2 Native Char Bull Trout: Bull trout are usually associated with cold water streams with temperatures below 15° C year round. Complex forms of cover habitat including large woody debris, undercut banks, boulders, and pools are associated with all life stages of bull trout. Bull trout exhibit both resident and migratory life histories. Resident bull trout complete their life cycles in the tributary streams in which they spawn and rear. Migratory bull trout spawn in tributary streams, and juvenile fish rear from one to four years before migrating to either a lake, river, or in certain coastal areas, saltwater to mature. Bull trout typically spawn from August to November; however, migratory bull trout may begin spawning as early as April. Biologists have reported repeat and alternate year spawning. Time from egg deposition to emergence may exceed 200 days. Fry normally emerge from the gravel beginning in early

June 2001 Mobrand Biometrics, Inc. Page 3-9 Pierce County Watershed Analysis Section 3

April through May depending upon water temperatures and increasing stream flows. Bull trout normally reach sexual maturity in four to seven years and can live 12 or more years (USFWS 1999). On November 1, 1999, the U.S. Fish and Wildlife Service determined that bull trout were threatened under the ESA throughout their range in the United States, including the Puget Sound population.

Dolly Varden: Bull trout and Dolly Varden were previously considered a single species, but were formally recognized as separate species by the American Fisheries Society in 1980. Although bull trout and Dolly Varden co-occur in several northwestern Washington River drainages, there is little evidence of introgression, and the two species appear to be maintaining distinct genomes (USFWS 1999). Bull trout and Dolly Varden are difficult to distinguish in the field and are believed to have similar life histories. The WDFW has made no effort to separate them when inventorying populations, as morphological methodologies have not yet been widely applied (WDFW 2000b). At present all bull trout/Dolly Varden stocks in Washington are considered native, as there is no cultured production within the state (WDFW 1998b). Although Dolly Varden are currently classified as a game fish in Washington, they are considered a species of concern by the WDFW.

3.2.2 WRIA 10 – Hylebos and Puyallup Systems 3.2.2.1 The Streams 3.2.2.1.1 Origin The Hylebos Creek system includes the mainstem Hylebos Creek (9 miles in length) and four unnamed tributaries (Figure 3-4—Hylebos Basin). Hylebos Creek originates at Lake Geneva and Lake Killarney, which are located approximately four miles northeast of Milton. Unnamed tributary 10.0013 begins approximately five miles north of Milton, flows south along the I-5 freeway, and joins Hylebos Creek at RM 5.1. At this point, the Hylebos Creek is joined by unnamed tributary 10.0009 (from Surprise Lake) and turns northwest, flowing through the Hylebos waterway and entering Commencement Bay (Williams et al. 1975).

The Puyallup River and its major tributaries, the Carbon and White rivers, are located entirely in Pierce County, with the exception of a small portion north of the mainstem White River, which is located in King County (Figure 3-5—Puyallup River Basin) (R2 1999). The Puyallup River Basin is fed by five glaciers at high elevations on the rugged west and north slopes of Mount Rainier (14,408 feet mean sea level); however, the primary source for the mainstem Puyallup River is the Puyallup and Tahoma glaciers (Williams et al. 1975). The glacial source of these rivers provides cold, nutrient-poor water, while also providing higher summer flows than other low elevation streams in WRIA 10 (R2 1999). The Puyallup River flows for approximately 54 miles from the steep mountainous terrain, through the foothills, and onto the Puget Sound lowlands before discharging in Commencement Bay (0 feet mean sea level). There are 728 identified rivers and streams in the Puyallup River Basin providing 1,287 linear miles of drainage. The Puyallup River Basin provides over 1,287 linear miles of drainage (R2 1999). Each drainage offers suitable habitat for both anadromous and resident fish.

June 2001 Mobrand Biometrics, Inc. Page 3-10

Figure 3-4. Hylebos Basin—WRIA 10.

Figure 3-5. Puyallup River Basin—WRIA 10.

Pierce County Watershed Analysis Section 3

3.2.2.1.2 Migration Barriers Puget Sound Energy's Electron power canal diversion at RM 41.8 blocks approximately 10 miles of the upper Puyallup River watershed to anadromous fish until 2000, when a fish ladder was installed for passage. The U.S. Army Corps of Engineers Mud Mountain Dam, a flood control project, is located at approximately RM 29.5 on the White River—the dam is not equipped for upstream passage on site. The diversion dam for Puget Sound Energy's White River Hydroelectric Project is located downstream of Mud Mountain Dam at RM 24.3 on the White River. A fish trap and haul facility are located at this site, where upstream migrants are trapped and hauled upstream of Mud Mountain Dam. The water diverted at this site flows through a series of flumes to Lake Tapps, where it then flows through a tunnel and penstocks to a powerhouse at Dieringer. Throughout the Puyallup, White, and Carbon rivers, and many of the smaller tributaries, numerous flood control diversions, dikes, and stream channelization projects act as barriers to anadromous migration.

3.2.2.1.3 Riverine Habitat Puyallup River (Headwaters to Confluence with the White River): From its origin on the Tahoma and Puyallup glaciers, the upper Puyallup River to RM 37.0 and most of its 14 tributaries exhibit steep mountainous characteristics with numerous cascades and rapids and boulder-rubble bottoms. The major tributaries in this reach, the upper North Puyallup and Mowich rivers, reveal glacial character with much braiding and fast riffle-rapid flow. Lower reaches of most tributaries offer moderate to moderately steep gradients producing a few rapids and cascades, but mostly fast riffles and some shallow pools. Below RM 37.0 the river flows first through a narrow, steep-sloped, densely forested valley, then enters a deep, steep-walled canyon. Just below the Electron powerhouse at RM 31.2, the valley bottom broadens, alternately widening and narrowing over the next seven miles. Much of the channel in this reach is contained within relatively narrow, artificially contoured, heavily armored banks. Six tributaries enter the Puyallup River between RM 37.0 and RM 25.0. From RM 25.0 to the confluence with the White River at RM 10.4, ten small to moderate tributaries and one major tributary, the Carbon River, enter the Puyallup River system. The valley floor in this reach varies from one to two miles with steeply rising hillsides of 400 to 500 feet on both sides. The river meanders within the confined area and contains medium to large gravel and pool-riffle-glide areas. There is little stream cover due to the broad flood plain gravel bars. All of the smaller tributaries contain very little gradient, are comprised essentially of drainage and groundwater seepage, and provide little access for spawning and rearing due to low summer flows.

Carbon River: From its headwaters on the Carbon and Russel glaciers, the Carbon River flows approximately 33 miles to its confluence with the Puyallup River at RM 17.9. Major tributaries to the Carbon River include Chenuis, Evans, Lilly, Voights and South Prairie creeks. Steep gradient, a confined channel, and numerous cascades with a few falls exist over the upper four miles of the Carbon River. Below this, the channel is less confined with considerable braiding and channel splitting until it enters a steep canyon with numerous rapids and some sharp cascades separated by relatively short, lesser gradient pool-riffle stretches. At approximate RM 8.0, the river channel widens allowing channel splitting and large gravel bars. Excellent pool-riffle conditions exist in a moderate gradient, offering good spawning gravel with mixed rubble. Extensive levees along the southwest side of the river from RM 3.8 to its mouth provide flood control for the town of Orting. The lower eight

June 2001 Mobrand Biometrics, Inc. Page 3-13 Pierce County Watershed Analysis Section 3 miles of South Prairie Creek, the largest tributary to the Carbon River, provides major spawning habitat, as do portions of many of its tributaries.

White River: The White River flows approximately 74 miles from its headwaters on the Emmons and Fryingpan glaciers to its confluence at RM 10.4 with the Puyallup River. Over 55 identified tributaries ranging from small streams to larger river systems feed into the White River. The upper river basin is highly mountainous with slopes often rising over 6,000 feet from the river's edge. Much of the channel is unstable with considerable splitting and braiding and a number of falls, cascades, and rapids. The streambed is composed of boulder and bedrock with some rubble-patch gravel stretches. The Greenwater River enters the White River at RM 45.8. The Greenwater heads in a high mountain valley that contains at least two small lakes. The river drops rapidly for the first 10 miles, then moderates for the remaining 11 miles to its confluence with the White River. There are more than 30 tributaries to the Greenwater River, adding an additional 88 linear miles to the system. The mainstem White River, from RM 31.0 to 35.5 is inundated by the Mud Mountain Dam reservoir. The Clearwater River flows into the White River at RM 35.3. The river gradient in this reach is approximately 50 feet per mile, with fast-moving flows over a streambed of boulder and rubble with some gravel and considerable silt and fines. The main river and tributary creeks in this reach all show the effects of heavy flood flows and runoff. Below Mud Mountain Dam, Puget Sound Energy's White River Hydroelectric Project forms a barrier at RM 24.5, where water is diverted into Lake Tapps for power generation and returned to the river at RM 3.5. The river below the White River Hydroelectric Project Dam meanders across a relatively broad valley floor. The river has been highly channelized and diked for flood control, and considerable channel splitting occurs each year. This reach does, however, provide considerable pool-riffle habitat suitable for anadromous and resident fish use.

Puyallup River (Confluence with the White River to mouth): The lower Puyallup River is located below the confluence of the White River near the town of Puyallup, where it flows northwesterly through Tacoma and into Commencement Bay. The river is generally contained in formal channelized banks and extensive levees with little to no bank cover or overhanging vegetation. Tidal influence extends about seven miles upstream from the mouth, and the bottom is composed largely of gravel and rubble mixed with deposits of sand and fine materials. Hylebos Creek and Wapato Creek are two independent streams in the basin. Hylebos Creek originates about four miles northeast of Milton and flows into Commencement Bay through the Hylebos waterway. Wapato Creek originates north of the town of Puyallup and flows into the industrial waterway in Commencement Bay.

3.2.2.2 Environmental Issues 3.2.2.2.1 Water Quantity Nearly all of the smaller streams in the basin suffer from summer low flows. In the mainstem Puyallup River above the confluence of the Carbon River, and in the mainstem White River between the White River Hydroelectric Project and Sumner, low flow conditions are further impacted by the diversion of water for electric power production. Seasonal flooding is common in all of the major drainages, and some of the smaller drainages within the basin also experience flooding. The major causes of flooding are rapid

June 2001 Mobrand Biometrics, Inc. Page 3-14 Pierce County Watershed Analysis Section 3 run-off from glaciers during periods of heavy rainfall and warm temperatures, and run-off from heavily logged areas (Kerwin 1999).

3.2.2.2.2 Water Quality The Puyallup River from Kings Creek (RM 31.6) to its headwaters and the White River from Mud Mountain Dam (RM 27.1) to its headwaters are classified as AA streams. The Puyallup River from its mouth to RM 1.0 is a Class B stream reach. All other streams and stream reaches in the Puyallup River Basin are Class A. All lakes and reservoirs are categorized as Lake Class (WAC 1992).

The mainstem of the Puyallup, Carbon, and White rivers experience extremely high turbidity during the early spring and again in the fall due to their glacial origins. In the upper mountainous portions of these rivers anadromous fish production may be limited by the extreme cold water temperatures, shifting braided channels and high turbidity. In the lower reaches of the Puyallup River Basin rivers and streams, water quality is effected by heavy domestic, agricultural and industrial developments. This has resulted in reduced summer flows with high water temperatures, stream sedimentation, and poor water quality from agricultural run-off and wastewater discharges. Many of the stream reaches suffer from combinations of high fecal coliform levels, low dissolved oxygen levels, phosphorous, nitrogen and various toxic materials (R2 1999).

The Puyallup River estuary and its receiving waters, Commencement Bay, have been severely degraded due to discharge of hazardous substances, including use of fill containing such substances (U.S. EPA 1999; Kerwin 1999). In 1982, the federal government ranked Commencement Bay amongst the most hazardous waste sites in the U.S. and portions of the area were officially designated as Superfund sites under CERCLA. The Commencement Bay Nearshore/Tideflats was later added to the National Priorities List after fish and sediments were found with elevated levels of toxic substances.

3.2.2.2.3 Physical Habitat Large segments of the Puyallup, Carbon, and White rivers have been extensively altered through channel realignments and diking. The construction of revetments and levees along the lower and middle (in the case of the Puyallup) portions of these rivers has eliminated connections with side- and off-channel aquatic habitats (Kerwin 1999). These flood control measures have also altered riverine processes that form pools, side channels, and other habitat features.

Many of the streams in the basin have undergone tremendous losses of wood structure, due to logging, agriculture, and other land use activities. These losses have had cascading effects on habitat conditions, including reducing pool quantity and quality and increasing the vulnerability of stream channels to instability (Kerwin 1999).

3.2.2.3 Salmonid Use 3.2.2.3.1 Anadromous Fish Six anadromous salmonid and two char (resident) species are known to occur in the Puyallup River Basin: chinook, coho, chum, and pink salmon; steelhead; sea-run cutthroat trout; and Dolly Varden and bull trout. Sockeye salmon are considered indigenous to the basin and are

June 2001 Mobrand Biometrics, Inc. Page 3-15 Pierce County Watershed Analysis Section 3 occasionally observed during WDFW spawning surveys in South Prairie Creek; however, the number are considered insignificant relative to other anadromous species inventoried in the basin (R2 1999).

Three chinook stocks have been identified in the Puyallup River Basin including White River spring chinook, White River summer/fall chinook, and Puyallup River fall chinook salmon. Chinook arriving at the adult trap on the White River at Buckley on or before August 15th annually are considered spring chinook, while those arriving after August 15th are considered fall chinook. Fall chinook enter the Puyallup River Basin in mid-July. Adult spring chinook utilize principally the high mountain streams of the White River including its tributaries of Huckleberry Creek and the West Fork White, Greenwater and Clearwater rivers. Limited numbers of spring chinook probably also use the upper reaches of the mainstem Puyallup and the upper Carbon rivers. The heaviest concentrations of adult fall chinook are found in the mainstem Puyallup, the lower White, and Carbon rivers and their tributaries Kapowsin, South Prairie, and Voights creeks. Voights Creek Salmon Hatchery on the Carbon River recorded a total escapement to the hatchery during 1999 of 3,484 adult and 60 jack chinook salmon (WDFW 2000a). Since 1989 escapement has ranged from a low of 2,975 in 1991 to a high of 5,166 in 1990, with an average annual escapement during this period of 4,290 chinook salmon (WDFW 1998a).

Puyallup River coho and White River coho salmon have been identified as two distinct spawning stocks. The Puyallup River stock is considered depressed due to low spawning escapements and a short-term severely declining trend, while the White River stock is considered healthy with a short-term increasing trend. Virtually all accessible streams and tributaries draining the Puyallup River Basin are utilized by coho salmon. The independent Hylebos and Wapato creek drainages also receive spawning coho. Because of the difficulty in distinguishing between self-sustaining, native coho, and artificially propagated stocks with this population, the status of wild coho salmon stocks is uncertain (R2 1999). Between the period of 1987 and 1996, coho escapement ranged from a low of 9,317 in 1988 to a high of 62,609 in 1994. Complete data are not yet available for 1997 through 1999. The average annual escapement over this period for coho salmon was 35,459 (WDFW 1998a).

Three fall chum salmon stocks have been identified in WRIA 10: the Puyallup/Carbon river, Fennel Creek, and Hylebos Creek stocks. The non-native Fennel Creek stock is considered healthy, but the status of the other two stocks is unknown based on a lack of data. Stream survey data show the presence of spawning chum in the mainstem Puyallup, White, and Carbon rivers; South Prairie, Canyonfalls, Fennel, and Clark creeks; and the two independent drainages of Hylebos and Wapato creeks. The lower Puyallup River and Commencement Bay are exceedingly important to early chum rearing and to successful transition into marine waters (R2 1999). Between the years of 1990 and 1999, chum salmon escapement ranged from a low of 638 in 1997 to a high of 4,088 in 1995. Average annual escapement of chum salmon is 1,940 adults (WDFW 1998a).

In odd numbered years Puyallup River pink salmon stock spawn almost exclusively in sections of the mainstem Puyallup, lower Carbon, and lower White rivers. Pink salmon are known to use Fennel and Kapowsin creeks in the Puyallup River Basin and both South Prairie and Voights creeks in the Carbon River Basin. The majority of spawning, however, comes from South Prairie Creek. Pink runs are not well established elsewhere in WRIA 10

June 2001 Mobrand Biometrics, Inc. Page 3-16 Pierce County Watershed Analysis Section 3

(R2 1999). In the odd numbered years between the years 1979 and 1997, pink salmon escapement to the Puyallup River has averaged 21,800, with a low return of 2,732 in 1997 and the high return of 50,273 adults in 1989 (WDFW 1998a).

Three winter steelhead stocks have been identified in WRIA 10; the mainstem Puyallup, White and Carbon river stocks. These wild, native stocks are treated separately due to geographical spawning isolation, but there is no evidence that they are genetically distinct. No summer steelhead stocks have been identified in the Puyallup River Basin.

Little is known about the status or occurrence of sea-run cutthroat trout in the Puyallup River Basin.

3.2.2.3.2 Native Char Three separate stocks of bull trout/Dolly Varden have been identified in WRIA 10; Puyallup, White, and Carbon river char. These stocks are considered distinct based on their geographic isolation, but the status of these stocks is unknown due to a lack of sufficient data. Native char have been observed in the mainstem Puyallup River, Mowich Creek, White River, West Fork White River, and the mainstem Carbon River. Genetic sampling of char trapped in the White River identified most fish at this location to be bull trout.

3.2.3 WRIA 12 – Chambers-Clover System 3.2.3.1 The Streams 3.2.3.1.1 Origin The Tacoma Basin contains two relatively small subbasins, Chambers/Clover Creek and Sequalitchew Creek (Figure 3-6—WRIA 12 Tacoma Basin). Clover Creek originates from springs and groundwater drainage at a 400-foot elevation approximately six miles east of Spanaway, where it flows for approximately 14.5 miles northwesterly through McChord Air Force Base to drain into Steilacoom Lake. Chambers Creek is formed at the outlet to Steilacoom Lake where it flows for approximately four miles west through a narrow, steep- sided ravine to enter Chambers Bay. A fish weir is located near the mouth of Chambers Creek where all migrating fish are trapped, with some retained for artificial propagation. A formal dam controls the flow at the outlet of Steilacoom Lake. This structure was recently redesigned to allow for fish passage. Major tributaries to Chambers/Clover Creek are the North Fork Clover Creek at approximate river mile (RM) 12, and both Flett and Leach creeks at approximate RM 2.5. These three tributaries add an additional 8.5 miles of stream habitat in varying conditions to that offered by Chambers/Clover Creek (R2 1999).

Sequalitchew Creek is formed from the overflow from American Lake, which drains into Sequalitchew Lake where the outlet creates the creek. Sequalitchew Creek flows for 3.05 miles on the Fort Lewis Military Reservation through two large marsh areas and converges with southern Puget Sound marine waters on the northern edge of the Nisqually reach (Williams et al. 1975).

June 2001 Mobrand Biometrics, Inc. Page 3-17

Figure 3-6. Tacoma Basin—WRIA 12 (Chambers-Clover Creek).

Pierce County Watershed Analysis Section 3

3.2.3.1.2 Migration Barriers Several structures partially or entirely block migration of anadromous salmonids at different locations within the basin. At the stream mouth, a small dam is a complete migration block, though it is equipped with a fish trap for sorting fish to be released upstream. WDFW operates the facility in conjunction with its hatchery operations in the area. Coho, chum, steelhead, and cutthroat are currently passed upstream to spawn. The agency has stopped passing chinook in conjunction with a policy of not allowing chinook passage in this drainage. Chinook that return to the facility are hatchery fish and it is believed that the stream did not support a self-sustaining run historically (Chuck Baranski, WDFW, personal communications).

Other barriers in the drainage include a structure on Flett Creek and several others in the Spanaway Creek area. Passage improvements have been made in recent years on the dam at the outlet of Lake Steilacoom, as well as on one located just upstream of the lake. Little or no access to upstream migrating salmonids occurs in Spanaway Creek due to barriers.

3.2.3.1.3 Riverine Habitat Although Clover Creek remains near its natural state in its uppermost reach, its character changes dramatically as it reaches McChord Air Force Base. Here the creek has been placed in straight sections of formal channel for several miles and is contained in a culvert as it flows below the base runways and the Interstate-5 freeway. There are three tributaries to Clover Creek: the North Fork, Spanaway, and Morey creeks. The lower reach of the North Fork, near its confluence with Clover Creek, contains good instream and riparian habitat although rearing habitat potential is limited as flow ceases during the summer. Water from Spanaway and Morey creeks has been diverted to create artificial ponds on the properties of riparian landowners. Chambers Creek has widths to 25 feet and varies in depth from 6 inches to 2 feet. It contains excellent gravel and good pool-riffle ratios with a moderate gradient. A canyon section from RM 0.5 to 1.75 contains steep hillsides and a narrow confined valley. Leach Creek varies in width from 6 to 15 feet and from 6 inches to 2 feet in depth and contains good gravel and pool-riffle balance. Flett Creek also contains good gravel and pool-riffle balance in the lower half of the stream. Bank erosion, lack of mature riparian vegetation, habitat modifications, water diversions and reductions in streamflow limit fish production throughout the course of Chambers/Clover Creek and its tributaries.

3.2.3.2 Environmental Issues 3.2.3.2.1 Water Quantity Winter flows in the upper reaches of Clover Creek average around 80 cubic feet per second (cfs), with peaks up to 200 cfs common during storm events. Because the soil types covering the majority of the basin are well drained, it is estimated that 50 to 60 percent of this precipitation recharges groundwater.

Winter flows in the lower reaches of Chambers Creek, below the confluence of Leach and Flett creeks, are heavily influenced by precipitation, averaging from 60 to 200 cfs with peaks up to 450 cfs. There are several large lakes plus many small lakes, ponds and marsh areas that directly or indirectly provide some seepage to Chambers/Clover Creek during the summer months. However, flow is often intermittent during this time in the middle and upper reaches of Clover Creek. Flows in the lower reaches of Chambers Creek average

June 2001 Mobrand Biometrics, Inc. Page 3-19 Pierce County Watershed Analysis Section 3 around 30 cfs during the late summer. Average yearly discharges of 7.5 cfs occur in Flett Creek and about 9.5 cfs in Leach Creek.

3.2.3.2.2 Water Quality The waters of Chambers/Clover Creek are rated as Class A (excellent) by the Washington Department of Ecology, and therefore should support migration, rearing, spawning and harvesting of native species. Because upper reaches of Clover Creek and its tributaries flow intermittently during part of the summer, low dissolved oxygen levels and high water temperatures associated with low streamflow are likely to limit fish production.

3.2.3.2.3 Water Quality Physical habitat for fish in the Chambers-Clover Creek system, particularly in Clover Creek, has been dramatically altered through development and urbanization. Long stretches of Clover Creek flow through culverts or are routed through a pavement lined channel. Little or no wood is found in much of the stream system.

3.2.3.3 Salmonid Use 3.2.3.3.1 Anadromous Fish Chinook salmon returning to the fish trap at Chambers Dam at the head of Chambers Bay are assumed to be entirely hatchery fish returning to Garrison Springs Hatchery located in the basin. Currently, no chinook are passed above the weir to spawn naturally in the stream.

The stock of coho salmon in Chambers Creek has been influenced by hatchery-origin strays and outplants from programs through Puget Sound. Thousands of juvenile coho salmon have until recently been released into Clover Creek annually. Adult coho returning to the fish trap at Chambers Dam have to a large extent consisted of hatchery fish produced in nearby hatcheries, most notably at the Sequalitchew facility (Keith Keown, WDFW, personal communications).

Both fall and winter run chum salmon exist in the Chambers Creek drainage. The status of the winter run is healthy, while the fall run status in unknown. Some chum salmon collected at the fish weir are spawned at the Garrison Springs Hatchery (2 percent in 1998). In 1999, 2,176 adult chum salmon were collected at the fish weir (WDFW 2000a). Between 1990 and 1999 the annual average escapement of chum salmon to Chambers Creek was 3,596 adults. A low return of 751 was seen in 1997 and a high of 7,880 in 1995 (WDFW 1998a).

The steelhead trout (O.mykiss) run in Chambers Creek originated from artificial production at the Garrison Springs Hatchery. The status of any wild, naturally spawning population or the distribution of steelhead spawning areas in the Chambers Creek drainage is unknown. The hatchery does not currently culture steelhead, and trends in abundance are unknown. During 1999, 28 steelhead were trapped at the fish weir at the Chambers Creek mouth (WDFW 2000a).

There are no complete records of sea-run cutthroat trout (O. clarki clarki) entering Chambers Creek. During 1998, six were trapped at the fish weir, while in 1999, none were recorded (WDFW 2000a). The distribution of sea-run cutthroat spawning habitat in the Chambers Creek drainage is unknown.

June 2001 Mobrand Biometrics, Inc. Page 3-20 Pierce County Watershed Analysis Section 3

3.2.3.3.2 Native Char It is unlikely that Chambers/Clover Creek provides the specific physical characteristics required to sustain successful populations of bull trout. The Coastal-Puget Sound population of bull trout was listed as threatened under the ESA by the U.S. Fish and Wildlife Service on November 1, 1999.

3.2.4 WRIA 15 – Kitsap Peninsula Systems 3.2.4.1 The Streams 3.2.4.1.1 Origin The Kitsap Basin (WRIA 15) consists of 13 subbasins located in central Puget Sound draining portions of Kitsap, Pierce, and Mason counties. The major streams include Rocky, Lackey, Minter, Purdy, and Crescent creeks with many smaller named and unnamed streams and tributaries (Figure 3-7—WRIA 15 Kitsap Basin).

The Kitsap Basin topography in this area is low, with flat-topped hills and ridges peaking at approximately 400 feet. All of the streams in this southeastern portion of the Kitsap Basin are rather small and represent typical lowland type streams with generally moderate to steep gradients, typically 15 percent or less, and steep shoreline slopes of approximately 45 percent (R2 1999). Many streams originate from lakes, groundwater run-off, or swamp-like basins (Williams et al. 1975).

3.2.4.1.2 Migration Barriers There are no artificial barriers to upstream migration on subbasins in WRIA 15 that lie within Pierce County. Intermittent barriers are occasionally created on all streams by debris accumulation or beaver activity (Williams et al. 1975).

3.2.4.1.3 Riverine Habitat Case Inlet: Rocky Creek, the major stream within Pierce County in this subbasin, flows south from its headwaters into Rocky Bay, which in turn flows into Case Inlet. Only the lower three miles of this five-mile long creek lie within the county boundary. The stream gradient is moderate, with few quiet water areas and dense stream bank cover, which offers excellent spawning and rearing habitat for chinook, coho, and chum salmon and steelhead trout. A major tributary enters Rocky Creek at River Mile 0.4, but due to insufficient water flow provides only limited spawning and rearing habitat to coho salmon in its lower reaches (Williams et al. 1975).

Henderson Bay: Minter Creek (6.3 miles long) is the largest stream in the portion of this subbasin that lies within Pierce County, followed by Purdy, Lackey, and McCormick creeks. All of these creeks flow directly into Henderson Bay, which in turn discharges into Carr Inlet to the south. These streams have a moderate gradient, stable channels with substrate composed mostly of gravel suitable for spawning, and moderately dense and brushy stream bank cover in spite of heavy development in many areas. Pool and riffle habitat is most predominant and supports spawning and rearing for chinook, coho and chum salmon and The lower two miles of Crescent Creek support coho and chum salmon although portions are unsuitable due to sediment accumulation and a lack of stream bank attributable to heavy development, and low summer flows (Williams et al. 1975).

June 2001 Mobrand Biometrics, Inc. Page 3-21

Figure 3-7. Kitsap Basin—WRIA 15.

Pierce County Watershed Analysis Section 3

3.2.4.2 Environmental Issues 3.2.4.2.1 Water Quantity The streams in WRIA 15 are small, less than 50 cubic feet per second (cfs) mean annual flow, and drain low elevation basins. The highest mean monthly streamflows occur in January, and the lowest occur in August. With no supporting glaciers or snow packs to provide continuous melt during the summer months, many of the smaller streams experience low or intermittent flows. Seasonal flooding occurs infrequently, and the effects are usually minimal (Williams et al. 1975).

3.2.4.2.2 Water Quality All streams in Pierce County's WRIA 15 are rated as Class A (excellent) by the Washington Department of Ecology, and therefore should support migration, rearing, spawning, and harvesting of native species (WAC 1992). Water quality and aquatic insect production is highly conducive to anadromous fish habitation. Low flows during the late summer and early fall can produce water temperatures that exceed the 18.0° C standard in the smaller streams. Surface runoff reflects seasonal precipitation patterns.

3.2.4.3 Salmonid Use 3.2.4.3.1 Anadromous Fish At least four species of anadromous salmonids utilize the streams in the lower Kitsap Basin including chinook, coho, and chum salmon and steelhead trout. Sea-run cutthroat trout likely use the same drainages, but little information exists regarding their abundance or distribution.

Chinook salmon are found spawning in the small streams of the Kitsap Peninsula but are suspected to be largely the result of stray hatchery fish (Chuck Baranski, WDFW, personal communications). Chinook are no longer passed above the weir on Minter Creek by WDFW.

Coho salmon are the most widely distributed salmon species in WRIA 15. The major coho- producing streams within the Pierce County portion of this WRIA include Rocky, Artondale, Lackey, Minter, Purdy, McCormick, and Crescent creeks. Due to hatchery influences, coho salmon within the Kitsap Basin are collectively managed with other regional coho populations as a single south Puget Sound tributary stock (R2 1999). In 1999 a total escapement of 9,363 adult and 61 jack coho salmon was recorded at the Minter Creek Hatchery (WDFW 2000a). From 1988 through 1997 the average annual return of coho salmon to Minter Creek was 15,615. 1989 saw the low return at 5,727, while a high return of 29,550 was recorded in 1987 (WDFW 1998a).

Two races of chum salmon, a summer and a fall run, exist on the Kitsap Basin. The only noteworthy runs of summer chum salmon within Pierce County are located in Rocky Creek. The Rocky Creek chum salmon are managed together with other Case Inlet tributary populations as a single stock. The Case Inlet stock is native, but natural production has been supplemented with intra-basin plants from the Coulter Creek Hatchery in Mason County. Fall chum salmon exist in Minter, Lackey, and Rocky creeks. The Carr Inlet stock is composed of offspring from Hood Canal and is considered a stock of mixed origin (R2 1999). Between 1990 and 1999, the average annual return of the Case Inlet summer chum

June 2001 Mobrand Biometrics, Inc. Page 3-23 Pierce County Watershed Analysis Section 3 stock was 37,567, with a low return of 14,928 in 1997 and a high return of 80,404 in 1998. During this same time period, the average annual return of fall chum stock was 20,938, with a low return of 1,556 in 1990 and a high return of 48,238 in 1998.

Minter Creek is the only Kitsap Basin stream within Pierce County that historically supported runs of pink salmon (Williams et al. 1975). Current runs are presently insignificant, with average annual returns in odd years of 135. During this period, a low return of 3 was experienced in 1987 and the high return of 371 in 1979 (WDFW 1998a).

A single run of winter steelhead from the Case/Carr Inlet stock uses streams in Pierce County WRIA 15. A small population of wild, native winter-run steelhead trout uses Rocky, Dutcher, Artondale, Minter, Purdy, McCormick, and Lackey creeks. Escapement of Case/Carr Inlet stocks is not monitored, so trends in abundance are unknown (R2 1999).

3.2.4.3.2 Native Char Complex forms of cover habitat including large woody debris, undercut banks, boulders and pools are associated with all life stages of bull trout. It is unlikely that streams within the Pierce County portion of WRIA 15 provide the specific physical characteristics required to sustain successful populations of bull trout. The Coastal-Puget Sound population of bull trout was listed as threatened under the ESA by the U.S. Fish and Wildlife Service on November 1, 1999.

June 2001 Mobrand Biometrics, Inc. Page 3-24 Pierce County Watershed Analysis Section 4

4.0 RESOURCE ASSESSMENT

The purpose of the assessment is to diagnose the environmental impediments to achieving the goals and values associated with the salmon resources of Pierce County, drawing conclusions at basin, subbasin, and stream reach scales. It provides a comprehensive, analytically derived, limiting factors analysis of each watershed, from which we formulated strategic priorities for conservation and restoration measures.

When interpreting and using the information presented in the following sections, the reader should be aware of the following:

Section 4 presents results from the EDT analysis of three Scenarios:

1. historic environmental conditions as an approximation of maximum potential 2. current environmental conditions with harvest and genetic fitness loss, as an approximation of current potential 3. current environmental conditions with no harvest and no loss in fitness.

Scenario 2 is the EDT assessment of the current situation—if this assessment is accurate, it should be consistent with recently observed population numbers. The difference between Scenarios 1 and 3 is an approximation of the loss in potential due to environmental factors alone. To the extent that environmental losses can be mitigated, it also approximates the maximum benefits of habitat restoration.

The productivity, abundance, and diversity parameters computed for each scenario are relative indices of population performance potential. Regarding the precision of the results presented, a good rule of thumb is to consider the parameter values given as accurate to within one significant digit. Because of the way the scenarios are constructed, we also have confidence in the ranking of the scenarios. We have tried to make the conclusions in the text consistent with this level of precision; however, the numbers shown in tables and figures are for the most part raw model output numbers. The reader should be aware of this and interpret the results accordingly. For example, in Figure 4-4 the historic abundance shown as 1,609 should be interpreted as “one to two thousand,” and the current abundance of 92 as “about a hundred.” However, even though the difference between current with harvest and current without harvest is small (97 vs. 107), the latter abundance potential is greater, thus the rank is accurate even though numeric difference is insignificant.

This section is divided by basin, as follows: Hylebos Puyallup White Chambers-Clover Kitsap

June 2001 Mobrand Biometrics, Inc. Page 4-1 Pierce County Watershed Analysis Section 4

We separated the results for the Puyallup-White Basin into the Puyallup and White basins because salmon populations for these two areas are identified and managed separately. The results for the Kitsap Basin are also separated by independent watershed entering Puget Sound.

The assessment results are presented for each species by basin, arranged into the following topics: Population performance summary 10. Strategic priorities for restoration and protection measures 11. Data uncertainties 12. Assessment conclusions

The results for population performance measures (capacity or abundance, productivity, and life history diversity) are presented in both tabular and graphic displays. The reader should note that a high level of precision is not intended by the numeric outputs shown—we rounded performance values in a manner to make comparisons as simple as possible and, hopefully, to minimize confusion. Abundance estimates are rounded to the nearest integer only.

A short explanation of how we address uncertainty in the assessment is warranted. The issue is: how certain are we that the factors affecting salmonid performance are correctly identified? This issue has two aspects. The first involves certainty about the data and information used in the analysis, and the second involves certainty about the analysis itself and how it is used to draw conclusions. Our conclusions regarding the first aspect are summarized in this report in the sections devoted to the individual basins. The second aspect is addressed in other sections of this document— Sections 1, 2, and 6—and also in other related reports that can be found on the Web at http://www.edthome.org.

4.1 Hylebos Basin 4.1.1 Chinook Salmon 4.1.1.1 Population Performance Summary for Hylebos Chinook Based on modeling, we conclude that the Hylebos system currently cannot sustain a naturally reproducing population of chinook. The average spawning population size of chinook was estimated to be less than ten fish, with a population productivity approaching a value of one adult return per parent spawner (Figure 4-1). The life history diversity value indicates that very few life history pathways can be successfully used. None of these performance measures is consistent with population sustainability.7 Removing all harvest and genetic loss effects in the analysis did not result in any significant improvement in average run size and productivity. Substantial improvements in environmental condition are needed to sustain a relatively small population.

7 Although effective population size (Ne) is not equivalent to the average annual number of spawners, it would be close in this case. Under any suggested approach for determining sustainability, a Hylebos population of this size would not be considered sustainable.

June 2001 Mobrand Biometrics, Inc. Page 4-2 Pierce County Watershed Analysis Section 4

Hylebos Fall Chinook Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 477 12.4 100% Current with harvest and fitness loss 7 1.2 4% Current without harvest and fitness loss 16 1.5 10%

Chinook spawner abundance 200 477 h

s 150 of fi r 100

50 Numbe

0 Historic Current-with harv Current-no harv Scenario

Chinook productivity 15 r e

awn 10 r sp e

s p 5 rn u t e R 0 Historic Current-with harv Current-no harv Scenario

Chinook life history divesity 100%

75% t

50% Percen 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-1. Hylebos chinook (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-3 Pierce County Watershed Analysis Section 4

These results are consistent with the near absence of sightings of adult chinook in Hylebos Creek in recent years. Adult chinook are rarely observed in the basin, and it is generally believed that the few fish seen in some years are strays from the Puyallup system (Don Nauer, WDFW, personal communication). Long-time residents of the watershed, however, report observing adult chinook more frequently in years past.

The analysis does suggest that the Hylebos basin might have supported a small run of chinook historically. The model estimated an average run size of about 500 fish under pristine conditions with a productivity of about twelve returning adults per parent spawner. However, the EDT model may present an overly favorable view of chinook for small independent drainages to Puget Sound, like the Hylebos. Chinook salmon spawning is almost always associated with mainstem rivers or their larger tributaries; however, the EDT model does not limit watershed size for utilization by chinook, although it does reduce adult migration success as stream size diminishes.

The WDFW would consider a Hylebos population in this case to be what it calls a "Category 3 population"—one associated with an independent small drainage to Puget Sound that is not self-sustaining over the long-term. Chinook observed in these drainages are considered by the agency likely to be hatchery strays, naturally-produced strays from other systems, or progeny of one of these two groups (Chuck Baranski, WDFW, personal communication). The Puget Sound Technical Review Team (TRT) also does not consider such small drainages as having self-sustaining chinook populations (Puget Sound TRT 2001).

4.1.1.2 Strategic Priorities for Hylebos Chinook We assessed strategic priorities for chinook in the Hylebos drainage for the sake of completeness for the analysis, although the potential importance of the drainage for chinook reproduction is questioned8. The relative importance of geographic areas within the drainage to Hylebos chinook for both restoration and protection measures is displayed in Figure 4-2. The drainage is divided into 11 geographic areas, from the estuary to the headwaters (Table 4-1). The three most important areas are the same for both restoration and protection measures—encompassing the entirety of the lower portion of the drainage, the West and East Forks, and the mainstem below the forks. This is largely the result of reduced production potential moving up the drainage as a function of stream size and corresponding spawning distribution.

8 The Hylebos waterway, technically considered part of the Hylebos estuary here, is likely used by migrating juveniles from other chinook populations, particularly those produced in the Puyallup and White rivers. We assume that chinook originating in other drainages will use portions of the Hylebos estuary at some times of the year—the analysis presented in this document does not explicitly address such use. Historically, the Hylebos estuary would have often been directly joined to the larger Puyallup River estuary by cross delta channels and therefore extensively used by Puyallup chinook (Simenstad 2000; Graeber 1999).

June 2001 Mobrand Biometrics, Inc. Page 4-4 Pierce County Watershed Analysis Section 4

Hylebos Chinook Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 8 C Puget Sound 4 B Commencement Bay 4-8 B-C1/ Hylebos estuary 5 B Lower mainstem Hylebos 1 A Surprise Lake drainage 9 D Lower WF Hylebos 3 A Upper WF Hylebos 6 B Lower NF Hylebos 9 D West branch NF Hylebos 9 D Upper NF Hylebos 9 D Lower EF Hylebos 2 A Trib 0016 9 D Upper EF Hylebos 9 D

Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA Commencement Bay NA Hylebos estuary NA Lower mainstem Hylebos 3 A Surprise Lake drainage 6 C Lower WF Hylebos 2 A Upper WF Hylebos 4 B Lower NF Hylebos 6 C West branch NF Hylebos 6 C Upper NF Hylebos 6 C Lower EF Hylebos 1 A Trib 0016 6 C Upper EF Hylebos 6 C

1/ A range of values is shown associated with differing assumptions about survival conditions within the estuary and bay.

Figure 4-2. Relative importance of geographic areas for restoration and protection measures for Hylebos chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-5 Pierce County Watershed Analysis Section 4

Table 4-1. Geographic areas applied in identifying strategic priorities in Hylebos Basin.

Area Description Hylebos estuary Entrance to Hylebos Waterway upstream to RM 3.85 on mainstem Hylebos Creek (extent of tidal influence). Lower mainstem Hylebos Mainstem Hylebos Creek from the upper end of tidal influence to the confluence of the East and West forks.. Surprise Lake drainage Tributary to Hylebos Creek draining out of Surprise Lake. Lower WF Hylebos West Fork Hylebos Creek from the confluence with East Fork to the junction of the North Fork and West Branch (of West Fork). Upper WF Hylebos West Branch of West Fork, including Brook Lake Tributary. Lower NF Hylebos North Fork (which joins the West Branch to form West Fork) to the left bank branch of North Fork.. West branch NF Hylebos Left bank tributary to North Fork (note: this is mislabeled throughout report as "West branch of NF, it should read "East branch"). Upper NF Hylebos Upper North Fork, upstream of the left bank tributary. Lower EF Hylebos East Fork Hylebos Creek from the confluence with West Fork to the junction with Tributary 0016. Trib 0016 Tributary 0016 in its entirety. Upper EF Hylebos East Fork Hylebos Creek upstream of the junction with Tributary 0016.

Charts showing how benefit categories were identified for Hylebos chinook are provided in Appendix D—see the first pair of charts in the Hylebos chinook section.

The results state that the highest priority should be given to restoring habitat conditions within the lower portion of the drainage if chinook is the focus of actions. The fact that these areas are given higher priority than the estuary does not diminish the importance of estuarine habitat here—rather, it simply indicates that the condition of the lower Hylebos system is so poor for chinook utilization that it has a higher restoration benefit in this case than the estuary.

It should be noted that two restoration benefit category grades are identified for Commencement Bay (Figure 4-2). The range of grades between B and C is a result of applying different assumptions about timing and residency of juvenile chinook moving through the bay, as well as differing assumptions about survival conditions in the estuary and bay. We analyzed a range of survival conditions in the estuary, reflecting evidence of contamination by PCBs and DDTs within Hylebos Waterway (Collier et al. 1998). Though still uncertain, these findings suggest that performance loss might occur from relatively short exposure to these substances.9 The reader should refer to Section 4.2.1 on Puyallup chinook, where this matter is described in more detail.

9 Ewing (1999) presents further evidence that relatively brief exposure to pesticides, particularly combinations of such substances, may reduce performance.

June 2001 Mobrand Biometrics, Inc. Page 4-6 Pierce County Watershed Analysis Section 4

The principal attribute classes or factors that rank highest for chinook restoration benefit are flow, sediment load, channel (or substrate) stability, channel landscape (in the estuary10)and habitat diversity (Figure 4-3), though other factors rank nearly as high. Figure 4-3 summarizes strategic priorities for formulating restoration focused measures in the basin if chinook is the focus of actions. We recognize that some attribute conditions, such as fine sediment load, are not necessarily caused within the geographic area showing those conditions—the cause may be located far upstream in the drainage associated with land use. Hence identification of specific actions to rectify environmental conditions needs to consider their origin and pattern of dispersal in the watershed.

Hylebos Chinook Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority ch) y it hat y1/ LOCATION her sp) y poaching

(miles from mouth) / quant abilit t ure load a ion (ot ion (w/ (shaded rows characterize tributaries or ent it it diversit ion t ent relevant sites) icals a pet pet perat hogens hdrawals structions b Benefit category Channel st Chem Com Com Flow Food Habit Harassm O Oxygen Pat Predat Sedim Tem Wit Key habit 7.2 Top of East Fork In Tributary 0016 6.8 ' as EF)

5.6 WF enters (cont

r In Surprise Lk trib 4.4 3.8 End of tidal influence

lebos C Hylebos estuary (waterway) y

H Hylebos Creek mouth

.) 8.2 Top of West Branch r In Brook Lake Trib

st B 7.6 We

s In North Fork Hylebos a '

t - Upstream of East Br. n - East branch (0013A) - Downstream of East Br.

rk (co NF enters o 6.4 st F

We 5.2 West Fork joins East Fork

1/ "Channel Stability" within estuary Key to Strategic Priority (Benefit Category letter shown) refers to "Channel Landscape", which D & E C B A represents the presence of the estuarine Indirect or General Low Medium High zones.

Figure 4-3. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Hylebos watershed for chinook salmon.

10 The attribute "channel landscape" applies to the estuary and represents the relative composition of estuarine zones described by Cowardin et al. (1979) and Hayman et al. (1996).

June 2001 Mobrand Biometrics, Inc. Page 4-7 Pierce County Watershed Analysis Section 4

Reach specific strategic priorities for Hylebos chinook are provided in Appendix D – Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in all types of watershed planning related to salmon conservation and recovery.

4.1.2 Coho Salmon 4.1.2.1 Population Performance Summary for Hylebos Coho Hylebos coho show a sharp reduction in population performance measures between historic and current conditions (Figure 4-4). The average spawning population size of coho was estimated to be approximately 100 fish under existing conditions, with a productivity of about four returning adults per parent spawner. Removing all harvest and genetic loss effects in the analysis did not result in any significant improvement in average run size and productivity. The historic abundance suggests potential for improvement throughout habitat restoration.

These results are consistent with the expected production for a stream of this size and condition. There are no estimates of spawning population sizes or smolt yields of coho in the Hylebos system based on field observations. Two streams of comparable basin size in the Puget Sound region that are monitored for coho abundance using weirs are Big Beef and Snow creeks. Neither basin is urbanized, but both have undergone significant alterations through logging and rural development. Coho spawning escapements have ranged between near zero (Snow Creek) to over 1,000 fish (Big Beef Creek) in the past decade (Lestelle et al. 1993b; Seiler 2001). Snow Creek, with major environmental change in its basin, has spawning escapements averaging 100 coho or less.

We consider the performance estimates for Hylebos Creek to be reasonable characterizations of coho performance under average existing and historic conditions. In any given year, spawner abundances could be significantly less or greater than the average measures listed in Figure 4-4, due to environmental variation. The results depict a population that has experienced a major loss in performance due to environmental alteration—but one that appears to have significant potential for recovery through watershed actions. We find that coho salmon would make an excellent indicator species for formulating watershed action plans for the Hylebos drainage to address salmonid conservation and recovery needs. Coho salmon utilize nearly the entire drainage during multiple life stages.

4.1.2.2 Strategic Priorities for Hylebos Coho The relative importance of geographic areas within the Hylebos drainage to coho for restoration or protection benefits reflects a wide range of environmental alterations across the basin (Figure 4-5). The greatest benefits through restoration measures would potentially be achieved by restoring reaches within the lower portion of the basin, including the mainstem below the forks, Surprise Lake drainage, and the lower sections of both the East and West forks. The upper West Fork (or called West Branch in Figure 4-6), which includes the entire subbasin upstream of the second Highway 99 crossing, is also ranked high for restoration potential. See Table 4-1 for a description of geographic areas.

The highest benefits associated with habitat protection (i.e., preserving or maintaining what currently exists) are associated with the entire West Fork and the lower North Fork. The

June 2001 Mobrand Biometrics, Inc. Page 4-8 Pierce County Watershed Analysis Section 4

Hylebos Coho Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 1,609 24.0 100% Current with harvest and fitness loss 92 4.3 58% Current without harvest and fitness loss 107 4.9 63%

Coho spawner abundance 800 1,609 600 fish f 400 er o mb

Nu 200

0 Historic Current-with harv Current-no harv Scenario

Coho productivity 25

5 Returns per spawner 0 Historic Current-with harv Current-no harv Scenario

Coho life history divesity

100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-4. Hylebos coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-9 Pierce County Watershed Analysis Section 4

Hylebos Coho Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 13 D Puget Sound 11 C Commencement Bay 11-14 C-D1/ Hylebos estuary 6 B Lower mainstem Hylebos 2 A Surprise Lake drainage 3 A Lower WF Hylebos 3 A Upper WF Hylebos 5 A Lower NF Hylebos 6 B West branch NF Hylebos 9 B Upper NF Hylebos 8 B Lower EF Hylebos 1 A Trib 0016 9 B Upper EF Hylebos 12 C

Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA Commencement Bay NA Hylebos estuary NA Lower mainstem Hylebos 8 C Surprise Lake drainage 9 C Lower WF Hylebos 2 A Upper WF Hylebos 1 A Lower NF Hylebos 2 A West branch NF Hylebos 6 B Upper NF Hylebos 5 B Lower EF Hylebos 4 A Trib 0016 7 B Upper EF Hylebos 9 C

1/ A range of values is shown associated with differing assumptions about survival conditions within the estuary and bay.

Figure 4-5. Relative importance of geographic areas for restoration and protection measures for Hylebos coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed. upper West Fork (the area upstream of the second Highway 99 crossing) offers the greatest potential benefit to habitat protection measures. These areas have a relatively high amount of spring influence, tending to buffer the effects of development that have already occurred there. Consideration should be given to placing a high priority on protecting existing groundwater sources and wetlands located there, as well as minimizing additional increases in rates of runoff.

Charts showing how benefit categories were identified for Hylebos coho are provided in Appendix D (the first pair of charts under Hylebos coho).

June 2001 Mobrand Biometrics, Inc. Page 4-10 Pierce County Watershed Analysis Section 4

Hylebos Coho Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority ch) y it hat y1/ LOCATION her sp) / y poaching

(miles from mouth) / quant abilit t ure load a als (shaded rows characterize tributaries or ion (w ion (ot it it diversit ion relevant sites) t a hogens hdraw w mpet mpet rassment mperat xygen hemicals hannel st o o abit a ood e Benefit category C C C C Flo F H H Obstructions O Pat Predat Sediment T Wit Key habit 7.2 Top of East Fork In Tributary 0016 6.8

5.6 WF enters In Surprise Lk trib 4.4 3.8 End of tidal influence Hylebos estuary (waterway)

Hylebos Cr (cont' as EF) Hylebos Creek mouth

8.2 Top of West Branch In Brook Lake Trib 7.6

In North Fork Hylebos - Upstream of East Br. - East branch (0013A) - Downstream of East Br. NF enters 6.4

West Fork (cont' as Br.) 5.2 West Fork joins East Fork

Figure 4-6. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Hylebos watershed for coho salmon.

Lower benefit potential associated with restoration of the estuary and Commencement Bay is due to the short exposures of this species to these areas compared to the amount of time spent in other areas. The sensitivity of the results to length of exposure to the estuary and bay is discussed further under "Information Uncertainties" and "Puyallup Chinook."

The principal attribute classes or factors that rank highest for coho restoration benefit are generally flow, sediment load, channel (or substrate) stability, habitat diversity, water quality, and habitat types (e.g. pools, backwater pools, and off-channel habitat). Figure 4-6 summarizes strategic priorities for formulating restoration focused measures in the basin if coho is the focus of actions. As noted previously, we recognize that some attribute conditions, such as fine sediment load, are not necessarily caused within the geographic area showing those conditions—the cause may be located far upstream in the drainage associated

June 2001 Mobrand Biometrics, Inc. Page 4-11 Pierce County Watershed Analysis Section 4

with land use. Identification of specific actions to rectify environmental conditions needs to consider their origin and pattern of dispersal in the watershed.

Two obstructions to migration are seen in Figure 4-6, the most significant being caused by the Highway 99 crossing on West Fork. This structure is scheduled for fish passage improvement measures in 2002 (Don Nauer, WDFW, personal communication).

Reach specific strategic priorities for Hylebos coho are provided in Appendix D – Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in all types of watershed planning related to salmon conservation and recovery.

4.1.3 Inferences to Bull Trout for Hylebos Basin Native char species (bull trout and Dolly Varden) have not been found in Hylebos Creek, though they do inhabit the Puyallup-White systems nearby (WDFW 1998b). In Western Washington, these species are typically associated with larger stream systems than the Hylebos. Spawning in those river systems occurs in colder and higher gradient streams than those in the Hylebos system. We conclude that no consideration should be given to char species in identifying strategic priorities for salmonid species in the Hylebos Basin.

4.1.4 Data/Information Uncertainties for Hylebos Basin The data and information used to characterize the environment were brought into the analysis through two processes: one involving all stream reaches upstream of tidewater and one that addressed the estuary and marine areas. Each process and the information used have a different level of uncertainty.

The characterization of the Hylebos watershed was assembled by a team of three resource specialists, all having worked extensively in the watershed (see Appendix C).11 The procedure required assigning a "level of proof" to each attribute rating for each reach, using a scale of 1-4, where a value of 1 meant that empirical data were used and a high level of confidence was placed in the rating and a 4 represented an educated guess with low confidence. Values of 2 and 3 were intermediate, where a 2 represented a relatively high level of confidence, based on a combination of personal observation and "weight of evidence" and a 3 drew on a theoretical application.

The large majority of ratings applied to the Hylebos upstream of tidewater were assigned a level of proof of 2. Relatively few ratings were assigned values of 1, 3 or 4. The large majority of ratings used to characterize historic conditions were assigned a level of proof of 3. We find that these levels of proof are reasonable, reflecting a good understanding of watershed processes and documented indices of environmental change (such as described in May et al. 1997). Recommended field verification of: Based on the conclusions of the assessment, we fine sediment within riffles identify three attributes in particular that should bed scour be field verified as opportunity occurs: fine distribution of springs and upwelling storm-runoff patterns

11 Ratings were subsequently checked by MBI for consistency with how the approach was developed, reviewed by other resource specialists who have worked on the Hylebos, and again by the team who first assembled the ratings.

June 2001 Mobrand Biometrics, Inc. Page 4-12 Pierce County Watershed Analysis Section 4

sediment, bed (substrate) scour, and the extent of spring sources.12 In addition, a stream flow gauging station has recently been established, which will aid understanding of flow characteristics in the watershed. Other attributes should also be field verified, as part of an on-going monitoring plan associated with action effectiveness monitoring.

Ratings for the Hylebos estuary and Commencement Bay were obtained through another project that addressed these and other estuaries and bays in Puget Sound, supplemented with a review by estuarine experts as part of this project. There is a high level of uncertainty about the overall contribution of the estuary and bay to population performance of Hylebos salmon. Until further information becomes available, however, we believe that the basic conclusions of the assessment are reasonable with regard to the relative contributions of estuarine/bay areas and freshwater reaches to population performance.

A multi-agency effort was recently initiated (led by David Johnson of WDFW) to formulate Level-Two type attributes for estuaries and near shore areas and corresponding rules for deriving survival related factors. We expect this work to advance the level of understanding about the effect of estuaries and marine areas on population performance.

4.1.5 Hylebos Basin Conclusions Most of the Hylebos watershed has undergone extensive alterations over the past 150 years— first by logging, followed by various kinds of development and urbanization. Major changes have occurred to salmon habitat throughout its length. Salmon population performance has been dramatically reduced corresponding to these changes.

The watershed is currently unsuited for a self-sustaining chinook population. We question whether a chinook population was ever sustainable over the long-term due to the basin's small size. The estuary, including the area containing Hylebos Waterway, was historically used by juvenile chinook originating in other streams, as it still is today.

The Hylebos watershed was historically highly suited to coho salmon, and it appears that a population is still present, though at relatively low numbers. We find that this species would make an excellent indicator species for formulating watershed action plans to address salmonid conservation and recovery needs. Attributes to target for restoration: Conservation and restoration measures can be fine sediment load developed following a set of strategic priorities for flow characteristics geographic areas within the basin (Figure 4-7). These habitat types (e.g., pools) priorities identify the strategic importance of different habitat structure (e.g., wood) watershed areas for restoring (including only partial streambed stability water quality recovery) or protecting conditions for salmonid performance. Attributes were identified as targets in Attributes to target for protection: planning new action measures. groundwater sources riparian function (e.g., wetlands)

12 In reference to springs, of particular relevance would be identification of spring sources that promote upwelling within the channel substrate associated with high dissolved oxygen. These sites are important for a variety of reasons, one of which is to ameliorate the effect of fine sediment on incubating eggs. The likely presence of such areas in the Hylebos is an important assumption made in this analysis.

June 2001 Mobrand Biometrics, Inc. Page 4-13 Pierce County Watershed Analysis Section 4

Hylebos Basin Overview Strategic Assessment of Geographic Areas

Chinook Coho strategic strategic priority priority

Geographic Area otection otection r r Restoration P Restoration P Hylebos estuary Lower mainstem Hylebos Surprise Lake drainage Lower WF Hylebos Upper WF Hylebos Lower NF Hylebos West branch NF Hylebos Upper NF Hylebos Lower EF Hylebos Trib 0016 Upper EF Hylebos

Key to Strategic Priority (Benefit Category letter shown)

D & E C B A Indirect or General Low Medium High

Figure 4-7. Overview of strategic priorities for restoration and protection measures by geographic area within the Hylebos watershed.

4.2 Puyallup Basin (excluding White) 4.2.1 Chinook Salmon 4.2.1.1 Population Performance Summary for Puyallup Chinook Puyallup chinook show a sharp reduction in population performance measures between historic and current conditions (Figure 4-8). The average spawning population size was estimated to be approximately 1,200 fish under existing conditions, with a productivity of less than three returning adults per parent spawner. The model projected a historic run size in excess of 30,000 adult chinook. Removing all harvest and genetic loss effects in the analysis increased productivity and doubled average spawner abundance.13 Differences between the historic performance measures and those shown for the current condition without harvest (and genetic effects) in Figure 4-8 are due to environmental effects alone.

13 Puyallup chinook are harvested in both marine and freshwater waters, though fishery impacts have been reduced markedly in recent years. The fisheries that take place in the Puyallup River target hatchery salmon (chinook and coho) produced at Voight and Diru Creek hatcheries located in the basin—naturally produced chinook are harvested at the same time.

June 2001 Mobrand Biometrics, Inc. Page 4-14 Pierce County Watershed Analysis Section 4

Puyallup Chinook Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 31,618 14.1 100% Current with harvest and fitness loss 1,166 2.9 19% Current without harvest and fitness loss 2,419 3.7 33%

Chinook spawner abundance 10,000 31,618 8,000

6,000

4,000

Number of fish 2,000

0 Historic Current-with harv Current-no harv Scenario

Chinook productivity 15 r

Returns per spawne 0 Historic Current-with harv Current-no harv Scenario

Chinook life history diversity 100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-8. Puyallup chinook (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-15 Pierce County Watershed Analysis Section 4

The analysis indicates that different population segments, or subpopulations, have experienced differential environmental effects. The model estimated that the relative distribution of spawners in the basin between historic and current scenarios has changed dramatically, due largely to land use patterns (Figure 4-9). The lower Puyallup subpopulation (i.e., all spawning occurring downstream of the Carbon River) has shown the greatest loss. The upper Puyallup subpopulation, those produced upstream of the Electron Dam site, show no current production due to lack of passage at the dam.14 The percentage of the population produced in the South Prairie watershed shows a significant increase over historic condition, due to this subpopulation being less impacted by land use than other areas.

These differential land use effects on population performance are also seen in the pattern of productivity estimates for the subpopulations (Figure 4-10). The South Prairie Creek subpopulation was estimated to have the highest remaining productivity in the basin, as well as the highest historically. The productivity estimates for all subpopulations except South Prairie are between 1 and 2, indicating that these population segments may in effect be functionally extirpated. This is particularly true for the lower Puyallup subpopulation. It bears noting that the historic productivity values for the Carbon, mid Puyallup, and upper Puyallup, are particularly low for a pristine watershed (compared to other population modeling that we have done for Western Washington chinook). This is due to an assumed effect of extremely heavy glacial silt in these areas within the analysis.

Little information exists for chinook abundance in the Puyallup system for evaluating the modeled estimates of average abundance for the basin. The extremely heavy silt load in the Puyallup and Carbon rivers make spawning counts difficult to obtain—and the estimates derived from them of questionable use. The only reliable estimates of spawner abundance based on empirical observation are for South Prairie Creek. Since 1994, spawner abundance estimates in this stream have ranged between 600 to 1,400, averaging 1,000 fish (Chuck Baranski, WDFW, personal communication).15 The modeled estimate for South Prairie Creek is approximately half of this. Even though we regard the modeled estimate to be reasonably close to the independent estimate, it is worth noting that the South Prairie subpopulation is likely supplemented to some extent by the stray Voight Creek Hatchery fish. Even with only periodic straying occurrences, the subpopulation there likely would be increased over a level corresponding to a total absence of straying. In some years, straying could be considerable.16 We conclude that the modeled estimates of performance for Puyallup chinook are reasonable and suitable to be used in the context of evaluating strategic priorities for planning.

14 Passage has been provided, beginning in 2000. Modeling was done to represent current levels of adult fish performance, which still reflect lack of passage. 15 Based on surveys of redd (spawning nest) counts for 1994 through 2000. 16 Fish marking at the Voight Creek Hatchery had not been done over a long period of years until 1997, when it was initiated on a large scale. This marking program will enable the Co-Managers to evaluate stray rates of Voight Creek fish.

June 2001 Mobrand Biometrics, Inc. Page 4-16 Pierce County Watershed Analysis Section 4

Historic vs Current Distribution of Spawning (from model) Historic Current

Lower Puy Lower Puy S Prairie 46% 8% 41% S Prairie 9%

Upper Puy Carbon 0% 20% Upper Puy Mid Puy Carbon 6% 16% Mid Puy 35% 19%

Figure 4-9. Estimated spawning distributions of Puyallup chinook based on modeling. Lower Puyallup consists of all areas downstream of the Carbon River; Mid Puyallup includes all areas upstrem of the Carbon River but downstream of Electron Dam; Upper Puyallup encompasses areas upstream of Electron Dam; and South Prairie includes all areas within its drainage.

Puyallup Chinook Productivity

4 Returns per spawner 0 Lower Puy S Prairie Carbon Mid Puy Upper Puy Sub-Population Historic Current

Figure 4-10. Productivity estimates for Puyallup chinook subpopulations based on modeling.

June 2001 Mobrand Biometrics, Inc. Page 4-17 Pierce County Watershed Analysis Section 4

4.2.1.2 Strategic Priorities for Puyallup Chinook The relative importance of geographic areas within the Puyallup Basin (excluding the White system) to chinook for restoration or protection benefits reflects a wide range of environmental conditions (Figure 4-11). The greatest benefits of restoration measures would be obtained by restoring segments of the Puyallup estuary and the lower to middle sections of the Puyallup and Carbon rivers. The two highest ranked geographic areas for restoration potential benefit encompass the Puyallup mainstem between the White River and the town of Orting. See Table 4-2 for a description of geographic areas.

Puyallup Chinook Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 211/ D1/ Puget Sound 61/ B1/ Commencement Bay 5-272/ A-D2/ Clear Creek 7 B Puyallup estuary 5 A Clarks Creek 9 B Misc lower Puyallup tribs below White 28 E Puyallup mainstem below White 2 A Misc lower Puyallup tribs below Carbon 28 E Fennel and Canyon Falls 23 D Puyallup mainstem below Carbon R 1 A Lower Carbon mainstem 4 A Lower Voight Cr 18 C Upper Voight Cr 13 C Lower South Prairie mainstem 11 B Wilkeson Creek 14 C Middle South Praire mainstem 16 C Misc middle South Praire tribs 28 E Upper South Prairie mainstem 25 D Top South Prairie 28 E Carbon canyon area 15 C Misc upper Carbon tribs 28 E Upper Carbon mainstem 10 B Horsehaven Creek 24 D Lower Kapowsin Creek 25 D Upper Kapowsin Creek 28 E Mid Puyallup mainstem Orting area 2 A Miscel mid Puyallup tribs below Canyon 28 E Miscel mid Puyallup tribs below Elect Dam 28 E Mid Puyallup mainstem Electron area 20 C Electron Dam 8 B Mowich River 16 C Misc Upper Puyallup tribs 19 C Upper Puyallup mainstem 12 B Top Upper Puyallup 21 D

1/ Due to the large size of this geographic area, its rank and benefit category is skewed high (toward higher benefit) compared to other areas. The rank shown supposes that the entirety of the area would be restored. 2/ A range of values is shown associated with differing assumptions about survival conditions within the estuary and bay.

Figure 4-11. Relative importance of geographic areas for restoration measures for Puyallup chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-18 Pierce County Watershed Analysis Section 4

Table 4-2. Geographic areas applied in identifying strategic priorities in Puyallup Basin.

Area Description Puyallup estuary The entirety of the Puyallup estuary, from the "neo-delta" at the river mouth extending to the upstream end of tidal influence (near mouth of Clarks Cr). Clear Creek Clear Creek, including tributaries. Clarks Creek Clarks Creek, including tributaries. Misc lower Puyallup tribs below White Small unnamed tributaries to the mainstem Puyallup River downstream of the White R. Puyallup mainstem below White Mainstem Puyallup River between the upstream end of tidal influence (near Clarks Cr) to the White R. Misc lower Puyallup tribs below Miscellaneous tributaries to the mainstem Puyallup R. between the junctions of Carbon the White and Carbon rivers (does not include Fennel and Canyon Falls). Fennel and Canyon Falls Fennel and Canyon Falls creeks, including tributaries. Puyallup mainstem below Carbon R Mainstem Puyallup River between the junctions of the White and Carbon rivers. Lower Carbon mainstem Mainstem Carbon River from its mouth to the beginning of the canyon at approximately RM 10. Lower Voight Cr Voight Creek, including tributaries, from its mouth to the hatchery rack site. Upper Voight Cr Upper Voight Creek, including tributaries, from the hatchery rack site (including the rack) upstream. Lower South Prairie mainstem Mainstem South Prairie Creek from its mouth upstream to the junction with Wilkeson Cr (not including Wilkeson Cr). Wilkeson Creek Wilkeson Creek and tributaries. Middle South Prairie mainstem Mainstem South Prairie Creek from the junction with Wilkeson Cr upstream to Beaver Cr. Misc middle South Prairie tribs Upper South Prairie mainstem Mainstem South Prairie Creek from the junction with Beaver Cr upstream to falls just downstream of diversion dam. Top South Prairie Prairie Creek and tributaries upstream of falls (located downstream of diversion dam). Carbon canyon area Mainstem Carbon River through the canyon reach, beginning at RM 10 and extending to RM 16. Misc upper Carbon tribs Miscellaneous tributaries to the Carbon River upstream of the canyon (approximately RM 16). Upper Carbon mainstem Mainstem Carbon River upstream of RM 16. Horsehaven Creek Horsehaven Creek and tributaries. Lower Kapowsin Creek Kapowsin Creek and tributaries, up to and including Lake Kapowsin. Upper Kapowsin Creek Kapowsin Creek and tributaries upstream of Lake Kapowsin. Mid Puyallup mainstem Orting area Mainstem Puyallup River from the junction with Carbon River upstream to the Electron Powerhouse. Misc mid Puyallup tribs below Canyon Miscellaneous tributaries to the Puyallup River from the Carbon River upstream to and including Kings Creek (excludes Kapowsin and Horsehaven crs). Misc mid Puyallup tribs below Miscellaneous tributaries to the Puyallup River upstream of Kings Creek to Electron Dam Electron Dam. Mid Puyallup mainstem Electron area Mainstem Puyallup River from the Electron Powerhouse extending upstream to Electron Dam. Electron Dam Site of Electron Dam. Mowich River Mowich River and tributaries. Misc Upper Puyallup tribs Miscellaneous tributaries to the Puyallup River upstream of Electron Dam to the junction with the North Puyallup River. Upper Puyallup mainstem Mainstem Puyallup River from Electron Dam upstream to the junction with North Puyallup River. Top Upper Puyallup Puyallup River and tributaries upstream of the confluence with North Puyallup River.

June 2001 Mobrand Biometrics, Inc. Page 4-19 Pierce County Watershed Analysis Section 4

Four geographic areas were assigned an "A" grade for habitat protection measures (i.e., preserving or maintaining what currently exists)(Figure 4-12). These include the lower and middle portions of South Prairie Creek, the lower Carbon River, and mid Puyullup mainstem in the vicinity of Orting. The highest priority (by ranking) is given to lower South Prairie Creek, which scored high in all three population performance measures. Although South Prairie Creek has undergone some alterations, it still contains relatively high quality habitat features. Consideration should be given to placing high priority on maintaining (or improving as opportunity exists) flow, substrate, stream bank, and riparian characteristics in the South Prairie system.

Puyallup Chinook Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA Commencement Bay NA Clear Creek 12 C Puyallup estuary NA Clarks Creek 13 C Misc lower Puyallup tribs below White 19 E Puyallup mainstem below White 5 B Misc lower Puyallup tribs below Carbon 19 E Fennel and Canyon Falls 16 D Puyallup mainstem below Carbon R 4 A Lower Carbon mainstem 2 A Lower Voight Cr 15 C Upper Voight Cr 18 D Lower South Prairie mainstem 1 A Wilkeson Creek 8 B Middle South Praire mainstem 3 A Misc middle South Praire tribs 19 E Upper South Prairie mainstem 13 C Top South Prairie 19 E Carbon canyon area 9 B Misc upper Carbon tribs 19 E Upper Carbon mainstem 5 B Horsehaven Creek 17 D Lower Kapowsin Creek 11 B Upper Kapowsin Creek 19 E Mid Puyallup mainstem Orting area 10 B Miscel mid Puyallup tribs below Canyon 19 E Miscel mid Puyallup tribs below Elect Dam 19 E Mid Puyallup mainstem Electron area 5 B Electron Dam 18 D Mowich River 18 D Misc Upper Puyallup tribs 18 D Upper Puyallup mainstem 18 D Top Upper Puyallup 18 D

Figure 4-12. Relative importance of geographic areas for protection measures for Puyallup chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-20 Pierce County Watershed Analysis Section 4

Charts showing how benefit categories were identified for Puyallup chinook are provided in Appendix D—see the first pair of charts under Puyallup chinook.

It should be noted that a range (A-D) of restoration benefit category grades are identified for Commencement Bay (Figure 4-12). This range, which virtually encompasses the full range available, is due to uncertainty about two sets of critical assumptions regarding use survival conditions and use of the bay by juveniles. The first set of assumptions represented survival effects on standardized durations for chinook life stages within the bay and estuary. The estuarine team found a high degree of uncertainty in this, suggesting that we model a range of survival assumptions. The second set of assumptions defined travel rates along the bay in the presence of restored habitats. For current conditions, we modeled juvenile migration patterns consistent with those described in reports for Commencement Bay (e.g., Pacific International Engineering 1999, 2000, and 2000b; Port of Tacoma and Puyallup Tribe of Indians 1999). We expect, however, that this pattern may not hold as shoreline habitat is restored—under those conditions, juveniles may have a greater tendency to reside longer along these habitats, experiencing greater growth and hence more of a survival benefit than would occur with the movement pattern currently seen. We therefore modeled a reasonable range of patterns under the restoration scenario—the range of grades is based on those results.

The principal attribute classes or factors that rank highest for chinook restoration benefit are generally channel (or substrate) stability and habitat diversity in the areas of highest importance to restoration (Figure 4-13). This reflects the benefit that would occur if side channels and backwaters were reopened and restored for use, primarily for fry colonization and juvenile rearing. Areas outside the main rivers would benefit from attention to these attributes, as well as to sediment loading. In the estuary, channel landscape (function of estuarine zones), habitat diversity, and habitat types should be principal targets. Other effects occurring in the estuary are related to the presence of toxic substances and competition with hatchery fish.

Reach specific strategic priorities for Puyallup chinook are provided in Appendix D – Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in all types of watershed planning related to salmon conservation and recovery.

4.2.2 Coho Salmon 4.2.2.1 Population Performance Summary for Puyallup Coho Puyallup coho show a loss in population performance measures between historic and current conditions comparable to that seen for chinook (Figure 4-14). The average spawning population size was estimated to be approximately 1,100 fish under existing conditions, with a productivity of about four returning adults per parent spawner. The model projected a historic run size of approximately 30,000 adult coho. Removing all harvest and genetic loss effects in the analysis increased productivity and tripled average spawner abundance.17

17 Puyallup chinook are harvested in both marine and freshwater waters, though fishery impacts have been reduced markedly in recent years. The fisheries that take place in the Puyallup River target hatchery salmon (chinook and coho) produced at Voight and Diru Creek hatcheries located in the basin—naturally produced coho are harvested at the same time.

June 2001 Mobrand Biometrics, Inc. Page 4-21 Pierce County Watershed Analysis Section 4

Figure 4-13. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Puyallup watershed for chinook.

June 2001 Mobrand Biometrics, Inc. Page 4-22 Pierce County Watershed Analysis Section 4

Puyallup Coho Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 29,634 21.7 91% Current with harvest and fitness loss 1,092 4.2 23% Current without harvest and fitness loss 3,130 10.1 29%

Coho spawner abundance 10,000 29,634 8,000

6,000

4,000

Number of fish 2,000

0 Historic Current-with harv Current-no harv Scenario

Coho productivity 20 21.7 15

5 Returns per spawner 0 Historic Current-with harv Current-no harv Scenario

Coho life history diversity

100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-14. Puyallup coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-23 Pierce County Watershed Analysis Section 4

Differences between the historic performance measures and those shown for the current condition without harvest (and genetic effects) in Figure 4-14 are due to environmental effects alone.

Scant information exists for naturally produced coho abundance in the Puyallup system for evaluating the modeled estimates of average abundance for the basin. Because fishery management has focused on hatchery production, little assessment has been done for coho spawning in the basin. Based on model performance for the White Basin, which appears to have provided a reasonable characterization of population response there, we consider the Puyallup estimates as reasonable. It may be important to recognize that the actual coho spawner abundance in the system may reflect hatchery strays from Voight Creek Hatchery to some extent, as noted for South Prairie Creek chinook. Furthermore, until relatively recently, hatchery coho fry were being annually released into various Puyallup tributaries, and these fish may still be contributing to abundance, though one or more generations removed. We expect it will take more than one generation to remove the contribution of that added productivity.

4.2.2.2 Strategic Priorities for Puyallup Coho The relative importance of geographic areas within the Puyallup Basin (excluding the White system) to coho for restoration or protection benefits reflects a wide range of environmental conditions (Figure 4-15 and Figure 4-16). Areas of highest priority for restoration are located across the basin, reflecting the contribution of many different areas in the pristine state of the basin to overall production potential. Three of the highest priority areas are in the extreme lower end of the basin: the estuary, Clear Creek, and Clarks Creek. The other high priority areas generally reflect locations where off-channel and side channel habitat has been heavily affected by land use. See Table 4-2 for a description of geographic areas.

The highest priorities identified for habitat protection measures (i.e., preserving or maintaining what currently exists) are all within either the South Prairie or Kapowsin systems (Figure 4-16). Consideration should be given to placing high priority on maintaining (or improving as opportunity exists) flow, substrate, stream bank, and riparian characteristics in these areas.

Charts showing how benefit categories were identified for Puyallup coho are provided in Appendix D—see the first pair of charts under Puyallup coho.

The principal attribute classes or factors that rank highest for coho restoration benefit are generally channel (or substrate) stability, habitat diversity, habitat types (e.g., pool frequency, back water pools), and, within clear water tributaries, fine sediment loading and flow patterns (Figure 4-17). Attributes having the most potential to affect production within mainstem areas are principally substrate stability and habitat diversity. Both of these attribute classes, or

June 2001 Mobrand Biometrics, Inc. Page 4-24 Pierce County Watershed Analysis Section 4

Puyallup Coho Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 33 D Puget Sound 171/ B1/ Commencement Bay 342/ D2/ Clear Creek 1 A Puyallup estuary 5 A Clarks Creek 3 A Misc lower Puyallup tribs below White 31 D Puyallup mainstem below White 22 C Misc lower Puyallup tribs below Carbon 26 C Fennel and Canyon Falls 18 B Puyallup mainstem below Carbon R 12 B Lower Carbon mainstem 8 A Lower Voight Cr 24 C Upper Voight Cr 11 B Lower South Prairie mainstem 9 A Wilkeson Creek 15 B Middle South Praire mainstem 2 A Misc middle South Praire tribs 23 C Upper South Prairie mainstem 27 C Top South Prairie 35 E Carbon canyon area 30 D Misc upper Carbon tribs 32 D Upper Carbon mainstem 14 B Horsehaven Creek 28 C Lower Kapowsin Creek 21 B Upper Kapowsin Creek 12 B Mid Puyallup mainstem Orting area 4 A Miscel mid Puyallup tribs below Canyon 29 C Miscel mid Puyallup tribs below Elect Dam 24 C Mid Puyallup mainstem Electron area 19 B Electron Dam 10 B Mowich River 15 B Misc Upper Puyallup tribs 20 B Upper Puyallup mainstem 6 A Top Upper Puyallup 7 A

Figure 4-15. Relative importance of geographic areas for restoration measures for Puyallup coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-25 Pierce County Watershed Analysis Section 4

Puyallup Coho Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA Commencement Bay NA Clear Creek 11 B Puyallup estuary NA Clarks Creek 19 C Misc lower Puyallup tribs below White 23 C Puyallup mainstem below White 14 B Misc lower Puyallup tribs below Carbon 21 C Fennel and Canyon Falls 12 B Puyallup mainstem below Carbon R 8 B Lower Carbon mainstem 7 B Lower Voight Cr 19 C Upper Voight Cr 25 D Lower South Prairie mainstem 1 A Wilkeson Creek 5 A Middle South Praire mainstem 2 A Misc middle South Praire tribs 6 A Upper South Prairie mainstem 14 B Top South Prairie 27 E Carbon canyon area 24 D Misc upper Carbon tribs 14 B Upper Carbon mainstem 14 B Horsehaven Creek 8 B Lower Kapowsin Creek 4 A Upper Kapowsin Creek 3 A Mid Puyallup mainstem Orting area 22 C Miscel mid Puyallup tribs below Canyon 18 C Miscel mid Puyallup tribs below Elect Dam 10 B Mid Puyallup mainstem Electron area 13 B Electron Dam 26 D Mowich River 26 D Misc Upper Puyallup tribs 26 D Upper Puyallup mainstem 26 D Top Upper Puyallup 26 D

Figure 4-16. Relative importance of geographic areas for protection measures for Puyallup coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-26 Pierce County Watershed Analysis Section 4

Figure 4-17. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Puyallup watershed for coho.

June 2001 Mobrand Biometrics, Inc. Page 4-27 Pierce County Watershed Analysis Section 4 factors, represent a suite of attributes including wood loading, channel confinement, and the presence of side channel networks and off-channel habitats. These features within mainstem rivers are especially important to coho performance.

Reach specific strategic priorities for Puyallup coho are provided in Appendix D – Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in all types of watershed planning related to salmon conservation and recovery.

4.2.3 Inferences to Bull Trout for Puyallup Basin Native char species (bull trout and Dolly Varden) are present in the Puyallup system (WDFW 1998b). The WDFW considers that there are two separate populations of bull trout present, one found in the Carbon system and one in the Puyallup outside the Carbon. Limited information exists on distribution. No information is available on life history patterns and abundance for these populations in the basin. We did not model performance for bull trout.

We suggest that plans developed for chinook and coho on the basis of the information contained in this document would be beneficial to bull trout. The ranges of distribution overlap for all three species, though this range likely is much more limited currently than it was historically. Life history requirements are similar, though generally more limited for bull trout. A more direct application of the results contained herein would be to focus attention on the upper portion of the basin, i.e., upstream of the confluence of the Puyallup and Carbon rivers. This area is expected to encompass the primary habitats of bull trout in the basin. We expect that strategic priorities identified within this geographic range for coho salmon would be largely applicable to bull trout.

4.2.4 Data/Information Uncertainties for Puyallup Basin The data and information used to characterize the environment were brought into the analysis through two processes, one involving all stream reaches upstream of tidewater and one that addressed the estuary and marine areas. Each process and the information used have a different level of uncertainty.

The principal characterization of the Puyallup watershed was assembled by a team of resource specialists, all having worked extensively in the watershed (see Appendix C). MBI staff worked with the team to apply its ratings to other streams due to time constraints of the team. All of the ratings were subsequently reviewed by team members. As part of the process, a "level of proof" was assigned to each rating for each reach, using a scale of 1-4, where a value of 1 meant that empirical data were used and a high level of confidence was placed in the rating and a 4 represented an educated guess with low confidence. Values of 2 and 3 were intermediate, where a 2 represented a relatively high level of confidence, based on a combination of personal observation and "weight of evidence" and a 3 drew on a theoretical application.

The large majority of ratings applied to the Puyallup upstream of tidewater were assigned levels of proof of 2 and 3 for current conditions. Comparatively few ratings were assigned values of 1and 4. The large majority of ratings used to characterize historic conditions were assigned levels of proof of 3, followed by 4. We find that these levels of proof are

June 2001 Mobrand Biometrics, Inc. Page 4-28 Pierce County Watershed Analysis Section 4

reasonable, reflecting a good understanding of watershed processes and documented indices of environmental change (such as described in May et al. 1997).

Based on the conclusions of the assessment, we identify four attributes in particular that should be Recommended field verification of: fine sediment within riffles field verified as opportunity occurs: fine sediment, bed scour bed (substrate) scour, and the extent of spring distribution of springs and upwelling sources.18 storm-runoff patterns

Considerable uncertainty also existed about how runoff patterns have changed in some streams, notably those in rural or forested areas under timber management. We find that it would be helpful to engage additional help from hydrologists to address possible changes in flow patterns in these areas. We note in particular the conclusions drawn by Mastin (1998) for South Prairie Creek, who stated that the runoff pattern has not been materially affected by logging in the upper watershed. The extent that these results can be applied to other drainages is in question. Other attributes will also need refinement, but we suggest that this should be part of an on-going monitoring plan associated with action effectiveness monitoring.

The need to better assess bed scour in the basin reflects a larger need that exists across the Pacific Northwest. In recent years, more attention is being directed to this need, but too little is still being done. One notable study underway is being conducted by the Washington Department of Ecology in the upper White Basin (led by Joanne Schuett-Hames). We recommend that this study be continued and, if possible, expanded to the Puyallup Basin outside the White.

The extent of fine sediment loading also requires additional attention in the basin. There is a need to verify sediment levels in many of the tributaries to the Puyallup and Carbon. We see an important need to also address a major uncertainty with respect to the effect of glacial sand and silt in the mainstem rivers. For this analysis, we assumed that these sediments have a high effect on incubating eggs in these rivers.19 Virtually no work has been done on this topic anywhere, however. We recommend that this issue be addressed in the Puyallup-White system because of the particularly high sediment load carried in these rivers.

As noted previously, there is a high level of uncertainty about the overall contribution of the estuary and bay to population performance of Puyallup salmon, particularly as applicable under restored conditions. Until further information becomes available, however, we believe that the basic conclusions of the assessment are reasonable with regard to the relative contributions of estuarine/bay areas and freshwater reaches to population performance.

18 In reference to springs, of particular relevance would be identification of spring sources that promote upwelling within the channel substrate associated with high dissolved oxygen. These sites are important for a variety of reasons, one of which is to ameliorate the effect of fine sediment on incubating eggs. It would be helpful to identify tributaries that have such flow contributions. 19 One of the few biologists in the Pacific Northwest who has looked at the effects of glacial sands and silts on salmon species is Jeff Koenings, Director of WDFW. We consulted him on this topic as it relates to this project. He expects that these sediment particles strongly affect salmon survival, including effects on survival from egg to fry. He confirms that this is a major uncertainty that should be addresed, particularly for rivers carrrying high sediment loads as the Puyallup River.

June 2001 Mobrand Biometrics, Inc. Page 4-29 Pierce County Watershed Analysis Section 4

4.2.5 Puyallup Basin Conclusions Much of the Puyallup watershed has undergone extensive alterations over the past 150 years, first by logging, followed by various kinds of development and urbanization. Major changes have occurred to salmon habitat throughout much of its length. Salmon population performance has been dramatically reduced corresponding to these changes. Historically the watershed was capable of producing large numbers of chinook and coho, as well as other salmonid species, including native char species. The extreme glacial nature of its two mainstem rivers, the Puyallup and Carbon, pose natural constraints on the system to a greater extent than those that operate in other non-glacial rivers. The effect of these natural conditions, operating in conjunction with man-made factors in the system, have contributed to the overall decline in performance within this system.

We find that both coho and chinook make excellent indicator species for formulating watershed action plans to address salmonid conservation and recovery needs in this basin. Both species should be considered since both utilize the system somewhat differently. We suggest that plans formulated around both species, though in particular those for coho, would be Attributes to target for restoration: streambed stability beneficial to native char species. hydro-modified channel confinement habitat structure (e.g., wood) Conservation and restoration measures can be fine sediment load (in clear streams) developed following a set of strategic priorities for flow characteristics (urban streams) geographic areas within the basin (Figure 4-18). These habitat types (e.g., off-channel areas) priorities identify the strategic importance of different water quality (urban streams and estuary) estuarine function in estuary areas in the watershed for either restoring (including only partial recovery) or protecting conditions for Attributes to target for protection: salmonid performance. Attributes were identified groundwater sources here to be targeted in planning new action measures. riparian function (e.g., wetlands) habitat structure 4.3 White Basin 4.3.1 Chinook Salmon 4.3.1.1 Population Performance Summary for White Chinook White River chinook exhibit a sharp decline in population performance measures between historic and current conditions (Figure 4-19). The comparison between historic and current condition scenarios for the White Basin is partly hypothetical due to the rerouting of the river in the early 1900s.20 A comparison is still valid because historic performance would have been similar regardless of whether the river connected to the Green River or the Puyallup River.

20 The White River changed course in 1906 during a flood event, shifting its course from flowing north to join with the Green River, eventually entering Elliot Bay, to flowing south and joining the Puyallup River, then entering Commencement Bay. The new course was made permanent through flood control measures.

June 2001 Mobrand Biometrics, Inc. Page 4-30 Pierce County Watershed Analysis Section 4

Puyallup Basin Overview Strategic Assessment of Geographic Areas

Chinook Coho strategic strategic priority priority

Geographic Area Restoration Protection Restoration Protection Puyallup estuary Clear Creek Clarks Creek Misc lower Puyallup tribs below White Puyallup mainstem below White Misc lower Puyallup tribs below Carbon Fennel and Canyon Falls Puyallup mainstem below Carbon R Lower Carbon mainstem Lower Voight Cr Upper Voight Cr Lower South Prairie mainstem Wilkeson Creek Middle South Praire mainstem Misc middle South Praire tribs Upper South Prairie mainstem Top South Prairie Carbon canyon area Misc upper Carbon tribs Upper Carbon mainstem Horsehaven Creek Lower Kapowsin Creek Upper Kapowsin Creek Mid Puyallup mainstem Orting area Miscel mid Puyallup tribs below Canyon Miscel mid Puyallup tribs below Elect Dam Mid Puyallup mainstem Electron area Electron Dam Mowich River Misc Upper Puyallup tribs Upper Puyallup mainstem Top Upper Puyallup

Key to Strategic Priority (Benefit Category letter shown)

D & E C B A Indirect or General Low Medium High

Figure 4-18. Overview of strategic priorities for restoration and protection measures by geographic area within the Puyallup basin (excluding White basin).

June 2001 Mobrand Biometrics, Inc. Page 4-31 Pierce County Watershed Analysis Section 4

White Chinook Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 24,327 8.8 100% Current with harvest and fitness loss 1,069 2.4 20% Current without harvest and fitness loss 1,679 2.8 33%

Chinook spawner abundance 10,000 24,327 8,000

6,000

4,000

Number of fish 2,000

0 Historic Current-with harv Current-no harv Scenario

Chinook productivity 15 r

Returns per spawne 0 Historic Current-with harv Current-no harv Scenario

Chinook life history diversity 100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-19. White chinook (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-32 Pierce County Watershed Analysis Section 4

The average spawning population size was estimated to be approximately 1,100 fish under existing conditions, with a low productivity of 2.4 returning adults per parent spawner. This run size is for the entire White River system. The model projected a historic run size of about 24,000 adult chinook. Removing all harvest and genetic loss effects in the analysis increased spawner abundance to roughly 1,700 fish with a productivity of 2.8 returns per spawner.21 Differences between the historic performance measures and those shown for the current condition without harvest (and genetic effects) in Figure 4-19 are due to environmental effects alone. Environmental effects seen here are meant to reflect existing conditions along the White River, recognizing that major improvements have been made in recent years at Mud Mountain Dam, in flow regulation at Puget Sound Energy's diversion dam near Buckley, and to fish screening facilities on the diversion canal (Russ Ladley, Puyallup Tribe Fisheries, personal communication).

The analysis shows a shift in distribution above and below the PSE Diversion Dam between historic and current conditions (Figure 4-20). We estimate that nearly 60% of the historic chinook production in the basin would have originated below the dam, whereas we estimate that only 34% originates there currently. This shift is due to a greater effect of land use factors (including hydro and flood control activities) operating downstream of the dam compared to upstream.

Historic vs Current Distribution of Spawning (from model) Historic Current

Below Below PSE PSE Dam Dam 34% 57%

Above PSE Above Dam PSE Dam 43% 66%

Figure 4-20. Estimated spawning distributions of White chinook upstream and downstream of PSE Diversion Dam based on modeling.

21 White chinook are currently harvested almost entirely within marine waters. Harvest in freshwater has been reduced to near zero due to the extremely depressed status of the population. We assume that some genetic fitness has been lost due to the extremely low run sizes that occurred in the recent past, combined with the use of captive brood stock measures used in recovery activities (Muckleshoot Tribe et al. 1996).

June 2001 Mobrand Biometrics, Inc. Page 4-33 Pierce County Watershed Analysis Section 4

White River subpopulations exhibit different productivity values, depending on whether the subpopulation spawns in clear water tributaries or in the mainstem White or West Fork, similar to the pattern described for Puyallup chinook. The White River and West Fork carry extremely high loads of glacial sand and silt that originate on Mt. Rainier. We assumed that this material affects environmental quality for incubating and rearing fish.

Fish trapping data for the Buckley Fish Trap located at the PSE Diversion Dam provide a means of comparing known fish abundance to estimates derived using the model. The model estimated that approximately 66% of the population originates upstream of Mud Mountain Dam currently, or approximately 700 spawners on average. This number equates to roughly 850 adults trapped at the PSE Dam in the modeling routine. For comparison, the average number of adult chinook actually trapped between 1992 and 2000 was roughly 600 per year, ranging from 300 to 1,500 fish. In addition, the model projected numbers of spawners to subbasins in the upper basin, e.g., Greenwater and Clearwater rivers, that correspond well with observations reported in Ladley and Smith (1998). We conclude that the modeled estimates of performance for Puyallup chinook are reasonable and suitable to be used in the context of evaluating strategic priorities for planning.

4.3.1.2 Strategic Priorities for White Chinook The relative importance of geographic areas within the White Basin to chinook for restoration or protection benefits reflects a wide range of environmental conditions and changes to the basin associated with land use (Figures 4-21 and 4-22). The greatest benefits of restoration measures were projected to be associated with several geographic areas widely distributed in the basin, including the estuary, the lower White River, Mud Mountain Dam and reservoir, and the Clearwater River. The effect associated with Mud Mountain Dam is due to habitat essentially lost within the impoundment reach.These areas have been altered by very different types of land use activities. The geographic area ranked highest for potential restoration benefit is the estuary. See Table 4-3 for a description of geographic areas.

The areas that ranked highest for protection (Figure 4-22) are also associated with various types of land use activities, including hydro and flood control measures. (It should be noted that the "A" grade assigned to the reach below Mud Mountain Dam is an artifact of the modeling process, which allowed for spawning to occur there when, in actuality, few if any fish spawn there.) The high grade given to the mainstem reach downstream of PSE Diversion Dam should not be construed to mean it would be best to "protect" all conditions within the channel as they currently exist. Flow limitations there do continue to cause some adverse effects on the population, but conditions could be worse for some attribute conditions.

Charts showing how benefit categories were identified for White chinook are provided in Appendix D—see the first pair of charts under White chinook.

It should be noted that a range (A-D) of restoration benefit category grades are identified for Commencement Bay (Figure 4-21). This range, which virtually encompasses the full range available, is due to uncertainty about survival conditions and use of the bay by juveniles. See discussion on this matter in Section 4.2.1.2 for Puyallup chinook.

June 2001 Mobrand Biometrics, Inc. Page 4-34 Pierce County Watershed Analysis Section 4

White Chinook Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 201/ C1/ Puget Sound 61/ B1/ Commencement Bay 5-222/ A-D2/ Puyallup estuary 1 A Puyallup mainstem below White 16 C White mainstem below Powerhouse 4 A White mainstem to Bowman Cr 8 B Misc Lower White tribs 26 E White mainstem reservation 6 B Reservation tribs 20 C Boise Cr 14 B White mainstem to PSE Dam 9 B TPU XXX and PSE Dam 12 B White mainstem to Mud Mountain 11 B Red and Scatter Crs 22 D Mud Mountain and reservoir 2 A Clearwater tribs 26 E Clearwater mainstem 5 A White mainstem to Greenwater 3 A Lower Greenwater mainstem 10 B Middle Greenwater mainstem 12 B Greenwater tribs 26 E Upper Greenwater mainstem 16 C Top Greenwater mainstem 26 E Misc White tribs near Greenwater 26 E WF White tribs 25 D WF White mainstem 15 C White mainstem to Huckleberry Cr 19 C Huckleberry tribs 26 E Huckleberry mainstem 16 C Misc Upper White tribs 26 E Upper White mainstem 24 D

1/ Due to the large size of this geographic area, its rank and benefit category is skewed high (toward higher benefit) compared to other areas. The rank shown supposes that the entirety of the area would be restored. 2/ A range of values is shown associated with differing assumptions about survival conditions within the estuary and bay.

Figure 4-21. Relative importance of geographic areas for restoration measures for White chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

The principal attribute classes or factors that rank highest for chinook restoration benefit are generally channel (or substrate) stability and habitat diversity in the freshwater areas of highest importance to restoration (Figure 4-23). This reflects the benefit that would occur if side channels and backwaters were reopened and restored for use, primarily for fry colonization and juvenile rearing. Tributaries outside the main rivers would benefit from attention to these attributes, as well as to sediment loading. In the estuary, channel landscape (function of estuarine zones), habitat diversity, and habitat types should be principal targets. Other effects occurring in the estuary are related to the presence of toxic substances and competition with hatchery fish.

June 2001 Mobrand Biometrics, Inc. Page 4-35 Pierce County Watershed Analysis Section 4

Table 4-3. Geographic areas applied in identifying strategic priorities in Puyallup Basin.

Area Description Puyallup estuary The entirety of the Puyallup estuary, from the "neo-delta" at the river mouth extending to the upstream end of tidal influence (near mouth of Clarks Cr). Puyallup mainstem below White Mainstem Puyallup River between the upstream end of tidal influence (near Clarks Cr) to the White R. White mainstem below Powerhouse Mainstem White River from its mouth upstream to the tailrace of the PSE Powerhouse. White mainstem to Bowman Cr Mainstem White River from the tailrace of the PSE Powerhouse upstream to Bowman Cr. Misc Lower White tribs Miscellaneous tributaries to White R. between its mouth and Bowman Cr (includes Bowman Creek). White mainstem reservation Mainstem White River from Bowman Cr upstream to Second Creek (upstream boundary is within the Muckleshoot Reservation).

Reservation tribs Miscellaneous tributaries to the White River within the Muckleshoot Reservation (includes Pussyfoot and Second Crs). Boise Cr Boise Creek, including tributaries. White mainstem to PSE Dam Mainstem White River from Second Creek to PSE Diversion Dam. TPU XXX and PSE Dam Instream structures associated with the Tacoma Power Utilities pipeline crossing and the PSE Diversion Dam. White mainstem to Mud Mountain Mainstem White River from PSE Diversion Dam to Mud Mountain Dam. Red and Scatter Crs Red and Scatter creeks. Mud Mountain and reservoir Mud Mountain Dam and its reservoir (extends roughly upstream to Clearwater River). Clearwater tribs Tributaries to Clearwater River. Clearwater mainstem Mainstem Clearwater River. White mainstem to Greenwater Mainstem White River from the upstream end of the Mud Mountain Dam impoundment to the Greenwater River. Lower Greenwater mainstem Mainstem Greenwater River from its mouth to Midnight Creek. Middle Greenwater mainstem Mainstem Greenwater River from Midnight Creek to RM 7.5. Greenwater tribs Tributaries to Greenwater River. Upper Greenwater mainstem Mainstem Greenwater River from RM 7.5 to impassable barrier to anadromous migration below lake. Top Greenwater mainstem Mainstem Greenwater River from impassable barrier upstream. Misc White tribs near Greenwater Miscellaneous small tributaries to White River near the confluence with the Greenwater River. WF White tribs Tributaries to West Fork White River. WF White mainstem Mainstem West Fork White River. White mainstem to Huckleberry Cr Mainstem White River from Greenwater River to Huckleberry Creek. Huckleberry tribs Tributaries to Huckleberry Creek. Huckleberry mainstem Mainstem Huckleberry Creek. Misc Upper White tribs Miscellaneous tributaries to the White River upstream of Huckleberry Creek. Upper White mainstem Mainstem of White River to its head.

June 2001 Mobrand Biometrics, Inc. Page 4-36 Pierce County Watershed Analysis Section 4

White Chinook Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA Commencement Bay NA Puyallup estuary NA Puyallup mainstem below White 11 B White mainstem below Powerhouse 14 C White mainstem to Bowman Cr 14 C Misc Lower White tribs 22 E White mainstem reservation 5 A Reservation tribs 16 D Boise Cr 9 B White mainstem to PSE Dam 1 A TPU XXX and PSE Dam 21 E White mainstem to Mud Mountain 5 A Red and Scatter Crs 16 D Mud Mountain and reservoir 18 D Clearwater tribs 22 E Clearwater mainstem 4 A White mainstem to Greenwater 2 A Lower Greenwater mainstem 8 B Middle Greenwater mainstem 13 C Greenwater tribs 22 E Upper Greenwater mainstem 11 B Top Greenwater mainstem 22 E Misc White tribs near Greenwater 22 E WF White tribs 19 D WF White mainstem 19 D White mainstem to Huckleberry Cr 5 A Huckleberry tribs 22 E Huckleberry mainstem 2 A Misc Upper White tribs 22 E Upper White mainstem 9 B

Figure 4-22. Relative importance of geographic areas for protection measures for White chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

Reach specific strategic priorities for White chinook are provided in Appendix D – Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in all types of watershed planning related to salmon conservation and recovery.

4.3.2 Coho Salmon 4.3.2.1 Population Performance Summary for White Coho White coho show a loss in population performance measures between historic and current conditions comparable to that seen for chinook (Figure 4-24). The average spawning population size was estimated to be approximately 1,900 fish under existing conditions, with a productivity of approximately four returning adults per parent spawner. The model projected a historic run size of approximately 25,000 adult coho. Removing all harvest and

June 2001 Mobrand Biometrics, Inc. Page 4-37 Pierce County Watershed Analysis Section 4

Figure 4-23. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the White watershed for chinook.

June 2001 Mobrand Biometrics, Inc. Page 4-38 Pierce County Watershed Analysis Section 4

White Coho Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 24,630 19.2 91% Current with harvest and fitness loss 1,902 3.8 32% Current without harvest and fitness loss 2,650 4.5 36%

Coho spawner abundance 10,000 24,630 8,000

6,000

4,000

Number of fish 2,000

0 Historic Current-with harv Current-no harv Scenario

Coho productivity 20 ner 15

10 rns per spaw

u 5 t e R 0 Historic Current-with harv Current-no harv Scenario

Coho life history diversity

100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-24. White coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-39 Pierce County Watershed Analysis Section 4

genetic loss effects in the analysis resulted in a slight increase in productivity and abundance.22 Differences between the historic performance measures and those shown for the current condition without harvest (and genetic effects) in Figure 4-24 are due to environmental effects alone. Environmental effects seen here are meant to reflect existing conditions along the White River, recognizing that major improvements have been made in recent years at Mud Mountain Dam, in flow regulation at Puget Sound Energy's diversion dam near Buckley, and to fish screening facilities on the diversion canal (Russ Ladley, Puyallup Tribe Fisheries, personal communication).

As seen for White chinook, the analysis shows a shift in distribution above and below the PSE Diversion Dam between historic and current conditions (Figure 4-25). We estimate that approximately 33% of the historic coho production in the basin would have originated below the dam; whereas, we estimate that only 16% originates there currently. This shift is due to a greater effect of land use factors (including hydro and flood control activities) operating downstream of the dam compared to upstream.

Historic vs Current Distribution of Spawning (from model) Historic Current

Below PSE Dam Below PSE 16% Dam 34%

Above PSE Dam Above 66% PSE Dam 84%

Figure 4-25. Estimated spawning distributions of White River coho in areas upstream and downstream of PSE Diversion Dam based on modeling.

22White River coho are harvested both in marine waters, as well as in the lower Puyallup River and estuary. The combined harvest rate by all fisheries is believed to have dropped significantly in recent years, due to fishery reductions in marine areas as well as in the Puyallup River. There is a question, however, about harvest impacts in the Puyallup River relative to those operating on Puyallup coho. Some evidence exists suggesting that at least one of the three brood lines for White coho is harvested at a lower rate than the other two brood lines due to earlier river entry for these fish (Russ Ladley, Puyallup Tribe, personal communication). We assumed a slightly lower rate of harvest within the river on White coho than for Puyallup coho.

June 2001 Mobrand Biometrics, Inc. Page 4-40 Pierce County Watershed Analysis Section 4

Fish trapping data for the Buckley Fish Trap located at the PSE Diversion Dam provide a means of comparing known fish abundance to estimates derived using the model. The model estimated that approximately 1,600 fish would arrive at the Buckley Trap to be transported upstream of Mud Mountain. For comparison, the average number of adult coho actually trapped and hauled between 1992 and 2000 was approximately 5,000 fish. However, there is a clear pattern for the three brood lines of coho: one is performing at a much higher level than the other two (Figure 4-26).23 The average number of coho trapped and hauled for the two weak brood lines (i.e., those spawning in 1992, 1993, 1995, 1996, etc.) has averaged 1,500 beginning in 1992.

Adult Coho Trapped and Hauled Past PSE Diversion Dam 10,000 21,806

8,000

6,000 er model estimate mb

Nu 4,000

2,000

0 67 70 73 76 79 82 85 88 91 94 97 00 Year

Figure 4-26. Numbers of adult coho trapped and hauled at PSE Diversion Dam. Dashed line represents the estimated number arriving to the trap site based on modeling.

Russ Ladley (Puyallup Tribe Fisheries, personal communication) hypothesizes that the differences in performance between the two weak brood lines and the one strong brood line are due largely to luck. He reports that the strong brood line has missed the major flood events of the past decade during the egg incubation stage. Further, he suggests that slightly earlier run timing of this brood line results in a lower fishery impact (both marine and freshwater) compared to the other two brood lines. There is also indication that marine survival conditions for some populations that returned in 2000 were unusually favorable, contributing to the very large run size at PSE Diversion Dam in that year. We conclude from Figure 4-26 that the model analysis reflects performance conditions that the two weak brood lines have experienced in the recent past. We conclude that the modeled estimates of

23Adult coho in the Pacific Northwest are almost always three years old at the time of spawning. Hence, over a period of three consecutive years, the spawning population is comprised of three separate brood lines or families.

June 2001 Mobrand Biometrics, Inc. Page 4-41 Pierce County Watershed Analysis Section 4

performance for Puyallup chinook are reasonable and suitable to be used in the context of evaluating strategic priorities for planning.

4.3.2.2 Strategic Priorities for White Coho The relative importance of geographic areas within the White system to coho for restoration or protection benefits reflects a wide range of environmental conditions and changes to the environment over the past century (Figure 4-27 and Figure 4-28). Areas of highest priority for restoration tend to be concentrated in two vicinities: one located along the Muckeshoot Reservation and extending up to the PSE Diversion Dam including Boise Creek and one concentrated around the Clearwater and Greenwater rivers. In addition, the estuary is ranked quite high for restoration benefit. See Table 4-3 for a description of geographic areas. Figure 4-27. Relative importance of geographic areas for restoration measures for White White Coho Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 28 D Puget Sound 221/ C1/ Commencement Bay 312/ D2/ Puyallup estuary 7 A Puyallup mainstem below White 26 D White mainstem below Powerhouse 24 C White mainstem to Bowman Cr 21 C Misc Lower White tribs 11 A White mainstem reservation 2 A Reservation tribs 3 A Boise Cr 14 A White mainstem to PSE Dam 5 A TPU XXX and PSE Dam 20 C White mainstem to Mud Mountain 4 A Red and Scatter Crs 27 D Mud Mountain and reservoir 10 A Clearwater tribs 6 A Clearwater mainstem 1 A White mainstem to Greenwater 9 A Lower Greenwater mainstem 13 A Middle Greenwater mainstem 14 A Greenwater tribs 18 B Upper Greenwater mainstem 12 A Top Greenwater mainstem 32 D Misc White tribs near Greenwater 19 C WF White tribs 23 C WF White mainstem 7 A White mainstem to Huckleberry Cr 17 B Huckleberry tribs 30 D Huckleberry mainstem 16 B Misc Upper White tribs 29 D Upper White mainstem 25 C

1/ Due to the large size of this geographic area, its rank and benefit category is skewed high (toward higher benefit) compared to other areas. The rank shown supposes that the entirety of the area would be restored. 2/ We modeled one set of survival conditions for coho salmon for Commencement Bay, unlike the range reported for chinook. The Benefit Category for coho might be increased by one grade if a range of survival conditions is modeled. coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-42 Pierce County Watershed Analysis Section 4

White Coho Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA Commencement Bay NA Puyallup estuary NA Puyallup mainstem below White 8 A White mainstem below Powerhouse 12 B White mainstem to Bowman Cr 17 C Misc Lower White tribs 22 C White mainstem reservation 2 A Reservation tribs 7 A Boise Cr 3 A White mainstem to PSE Dam 6 A TPU XXX and PSE Dam 28 E White mainstem to Mud Mountain 13 B Red and Scatter Crs 25 D Mud Mountain and reservoir 18 C Clearwater tribs 27 E Clearwater mainstem 1 A White mainstem to Greenwater 5 A Lower Greenwater mainstem 9 A Middle Greenwater mainstem 14 B Greenwater tribs 19 C Upper Greenwater mainstem 22 C Top Greenwater mainstem 26 E Misc White tribs near Greenwater 21 C WF White tribs 3 A WF White mainstem 24 D White mainstem to Huckleberry Cr 14 B Huckleberry tribs 19 C Huckleberry mainstem 9 A Misc Upper White tribs 16 C Upper White mainstem 11 A

Figure 4-28. Relative importance of geographic areas for protection measures for White coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

Generally, the same areas that ranked high for restoration benefit also ranked high for potential protection benefit (Figure 4-28). This indicates that while these areas have undergone significant environmental change, they generally still reflect the best conditions available in the basin for coho performance.

Charts showing how benefit categories were identified for White coho are provided in Appendix D—see the first pair of charts under White coho.

The principal attribute classes or factors that rank highest for coho restoration benefit are generally channel (or substrate) stability, habitat diversity, habitat types (e.g., pool frequency, back water pools), and, within clear water tributaries, fine sediment loading and flow patterns (Figure 4-29). Attributes having the most potential to affect production within mainstem

June 2001 Mobrand Biometrics, Inc. Page 4-43 Pierce County Watershed Analysis Section 4

Figure 4-29. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the White watershed for coho.

June 2001 Mobrand Biometrics, Inc. Page 4-44 Pierce County Watershed Analysis Section 4 areas are principally substrate stability and habitat diversity. Both of these attribute classes, or factors, represent a suite of attributes, including wood loading, channel confinement, and the presence of side channel networks and off-channel habitats. These features within mainstem rivers are especially important to coho performance.

Reach specific strategic priorities for White coho are provided in Appendix D – Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in all types of watershed planning related to salmon conservation and recovery.

4.3.3 Inferences to Bull Trout for White Basin Native char species (bull trout and Dolly Varden) are present in the White system (WDFW 1998b). The WDFW considers that the White River population of bull trout is distinct from the Puyallup and Carbon river populations. Limited information exists on distribution. No information is available on life history pattens and abundance for these populations in the basin. We did not model performance for bull trout.

We suggest that plans developed for chinook and coho on the basis of the information contained in this document would be beneficial to bull trout. The ranges of distribution overlap for all three species, though this range likely is much more limited currently than it was historically. Life history requirements are similar, though generally more limited for bull trout. A more direct application of the results contained herein would be to focus attention on the upper portion of the basin, i.e., upstream of Mud Mountain Dam.This area is expected to encompass the primary habitats of bull trout in the basin. We expect that strategic priorities identified within this geographic range for coho salmon would be largely applicable to bull trout.

4.3.4 Data/Information Uncertainties for Puyallup Basin The data and information used to characterize the environment were brought into the analysis through two processes, one involving all stream reaches upstream of tidewater and one that addressed the estuary and marine areas. Each process and the information used have a different level of uncertainty.

The principal characterization of the White watershed was assembled by the same team that worked on the Puyallup system (Appendix C). The team worked together to rate attributes on many reaches scattered around the basin, but it could not address every reach due to time constraints. MBI staff applied the team's ratings to the remainder of the reaches based on input received from team members. These ratings were subsequently reviewed by team members. As part of the process, a "level of proof" was assigned to each rating for each reach, using a scale of 1-4, where a value of 1 meant that empirical data were used and a high level of confidence was placed in the rating, and a 4 represented an educated guess with low confidence. Values of 2 and 3 were intermediate, where a 2 represented a relatively high level of confidence, based on a combination of personal observation and "weight of evidence," and a 3 drew on a theoretical application.

The large majority of ratings applied to the White upstream of tidewater were assigned levels of proof of 2 and 3 for current conditions. Comparatively fewer ratings were assigned values of 1 and 4. The large majority of ratings used to characterize historic conditions were assigned levels of proof of 3, followed by 4. We find that these levels of proof are

June 2001 Mobrand Biometrics, Inc. Page 4-45 Pierce County Watershed Analysis Section 4

reasonable, reflecting a good understanding of watershed processes and documented indices of environmental change (such as described in May et al. 1997).

Based on the conclusions of the assessment, we identify four attributes in particular that Recommended field verification of: fine sediment within riffles should be field verified as opportunity bed scour occurs: fine sediment, bed (substrate) scour, distribution of springs and upwelling extent of spring sources (in select areas, storm-runoff patterns such as Boise Creek), and storm runoff patterns.24 Considerable uncertainty also exists about how runoff patterns have changed in some streams, notably those in rural or forested areas under timber management. We find that it would be helpful to engage additional help from hydrologists to address possible changes in flow patterns in these areas. We note in particular the conclusions drawn by Mastin (1998) for South Prairie Creek, who stated that the runoff pattern has not been materially affected by logging in the upper watershed. The extent that these results can be applied to other drainages is in question. Other attributes will also need refinement, but we suggest that this should be part of an on-going monitoring plan associated with action effectiveness monitoring.

The need to better assess bed scour in the basin reflects a larger need that exists across the Pacific Northwest. In recent years, more attention is being directed to this need but too little is still being done. One notable study underway is being conducted by the Washington Department of Ecology in the upper White Basin (led by Joanne Schuett-Hames). We recommend that this study be continued and, if possible, expanded to other areas in the White system.

The extent of fine sediment loading also requires additional attention in the basin. There is a need to verify sediment levels in key tributaries to the White, such as in the Greenwater and Clearwater rivers, as well as in tributaries downstream of PSE Diversion Dam. We also see an important need to address a major uncertainty with respect to the effect of glacial sand and silt in the White River and West Fork, as described for the Puyallup and Carbon rivers. For this analysis, we assumed that these sediments have a high effect on incubating eggs.25 We recommend that this issue be addressed in the Puyallup-White system because of the particularly high sediment load carried in these rivers.

As noted previously, there is a high level of uncertainty about the overall contribution of the estuary and bay to population performance of Puyallup-White salmon, particularly as

24 Of particular relevance would be identification of spring sources that promote upwelling within the channel substrate associated with high dissolved oxygen. These sites are important for a variety of reasons, one of which is to ameliorate the effect of fine sediment on incubating eggs. It would be helpful to identify tributaries that have such flow contributions. We assumed that Boise Creek, for example, exhibits substantial areas of upwelling, based on conversations with Jeanne Stypula (King County Department of Natural Resources, personal communication). 25 One of the few biologists in the Pacific Northwest who has looked at the effects of glacial sands and silts on salmon species is Jeff Koenings, Director of WDFW. We consulted him on this topic as it relates to this project. He expects that these sediment particles strongly affect salmon survival, including effects on survival from egg to fry. He confirms that this is a major uncertainty that should be addressed, particularly for rivers carrying high sediment loads as the White River.

June 2001 Mobrand Biometrics, Inc. Page 4-46 Pierce County Watershed Analysis Section 4 applicable under restored conditions. Until further information becomes available, however, we believe that the basic conclusions of the assessment are reasonable with regard to the relative contributions of estuarine/bay areas and freshwater reaches to population performance.

A multi-agency effort was recently initiated (led by David Johnson of WDFW) to formulate Level 2-type attributes for estuaries and near shore areas and corresponding rules for deriving survival related factors. We expect this work to advance the level of understanding about the effect of estuaries and marine areas on population performance.

4.3.5 White Basin Conclusions As in the Puyallup Basin, much of the White River watershed has undergone extensive alterations over the past 150 years, first by logging, followed by various kinds of development. Urbanization is taking place in the lower portion of the basin. Salmon population performance has been dramatically reduced corresponding to these changes. Historically the watershed was capable of producing large numbers of chinook and coho, as well as other salmonid species, including native char species. The extreme glacial nature of the mainstem White River and West Fork pose natural constraints on the system to a greater extent than those that operate in other non-glacial rivers. The effect of these natural conditions, operating in conjunction with man-made factors in the system, have contributed to the overall decline in performance within this system.

June 2001 Mobrand Biometrics, Inc. Page 4-47 Pierce County Watershed Analysis Section 4

White Basin Overview Strategic Assessment of Geographic Areas

Chinook Coho strategic strategic priority priority

Geographic Area estoration estoration R Protection R Protection Puyallup estuary Puyallup mainstem below White White mainstem below Powerhouse White mainstem to Bowman Cr Misc Lower White tribs White mainstem reservation Reservation tribs Boise Cr White mainstem to PSE Dam TPU XXX and PSE Dam White mainstem to Mud Mountain Red and Scatter Crs Mud Mountain and reservoir Clearwater tribs Clearwater mainstem White mainstem to Greenwater Lower Greenwater mainstem Middle Greenwater mainstem Greenwater tribs Upper Greenwater mainstem Top Greenwater mainstem Misc White tribs near Greenwater WF White tribs WF White mainstem White mainstem to Huckleberry Cr Huckleberry tribs Huckleberry mainstem Misc Upper White tribs Upper White mainstem

Key to Strategic Priority (Benefit Category letter shown)

D & E C B A Indirect or General Low Medium High

Figure 4-30. Overview of strategic priorities for restoration and protection measures by geographic area within the White basin.

June 2001 Mobrand Biometrics, Inc. Page 4-48 Pierce County Watershed Analysis Section 4

4.4 Chambers-Clover Basin 4.4.1 Chinook Salmon 4.4.1.1 Population Performance Summary for Chambers-Clover Chinook We modeled chinook performance in the Chambers-Clover system, although it is uncertain whether the basin produced this species on a sustained basis historically. WDFW has classified Chambers-Clover chinook as a "Category 3 population"—one associated with an independent small drainage to Puget Sound that is not self-sustaining over the long-term. Chinook observed in these types of drainages are considered by the agency likely to be hatchery strays, naturally-produced strays from other systems, or the progeny of one of these two groups (Chuck Baranski, WDFW, personal communication). Streams like Chambers- Clover Creek are believed to have generally responded as sink populations, or satellites to the larger source populations in the major rivers of Puget Sound. Stray fish from the core populations would have colonized smaller nearby streams, where small runs could have been sustained under favorable climate and flow conditions. In drought years, entry to these small streams by adult chinook is difficult—in these years the runs may not have been supported historically. The Puget Sound Technical Review Team (TRT) has not identified these small drainages as having self-sustaining chinook populations (Puget Sound TRT 2001). Currently, WDFW, in operating a trap on Chambers Dam at the head of tidewater, does not pass captured chinook upstream to the free flowing stream. Therefore, a self-sustaining population is not currently present in the basin.26

The modeling results suggest that a small run of chinook could be produced within the Chambers-Clover Basin under current conditions (Figure 4-31), at least during years that permit chinook to penetrate the watershed.27 Modeling projected a historic run of nearly 2,000 fish, with a productivity of approximately 20 returns per spawner. However, we find that the EDT model may present an overly favorable view of chinook for small drainages in Puget Sound, like Chambers-Clover and the other independent basins modeled as part of this project. Chinook salmon spawning is almost always associated with mainstem rivers or their larger tributaries; however, the EDT model does not limit watershed size for utilization by this species, although it does reduce adult migration success as stream size diminishes.

4.4.1.2 Strategic Priorities for Chambers-Clover Chinook We assessed strategic priorities for chinook in the Chambers-Clover drainage for the sake of completeness, although the potential importance of the drainage for chinook reproduction is questioned.28 The relative importance of geographic areas within the drainage to

26 The Garrison Springs State Hatchery located below the outlet of Lake Steilacoom releases large numbers of chinook into Chambers Creek. The stock is not native to this basin. Returning fish are transported from the trap located on Chambers Dam at the head of tidewater directly to the hatchery.

27 Although effective population size (Ne) is not equivalent to the average annual number of spawners, it would be close in this case. Under accepted approaches for determining sustainability, it is unlikely that a Chambers- Clover population would be considered viable (see McElhany et al. 2000). 28 We recognize that despite whether chinook could be self-sustaining in the Chambers-Clover basin, its estuary has importance to other chinook populations in South Puget Sound. It is likely that the Chambers-Clover estuary is used to some extent by migrating juveniles produced in nearby rivers.

June 2001 Mobrand Biometrics, Inc. Page 4-49 Pierce County Watershed Analysis Section 4

Chambers-Clover Chinook Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 1,993 20.4 100% Current with harvest and fitness loss 67 2.6 41% Current without harvest and fitness loss 172 4.7 66%

Chinook spawner abundance 1,500 1,993 1,200

900

600

Number of fish 300

0 Historic Current-with harv Current-no harv Scenario

Chinook productivity 25

Returns per spawner 0 Historic Current-with harv Current-no harv Scenario

Chinook life history diversity 100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-31. Chambers-Clover chinook (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-50 Pierce County Watershed Analysis Section 4

Chambers-Clover chinook for both restoration and protection measures is displayed in Figure 4-32. The drainage is divided into 13 geographic areas, from the estuary to the headwaters (Table 4-4). We assumed that chinook might be able to access six of these areas, though it is highly unlikely that our upper limit to migration (Lower Clover area) would be accessible in low flow years. The most important area for both restoration and protection benefits is the area encompassing mainstem Chambers Creek (the mainstem downstream of Steilcoom Lake).

Chambers-Clover Chinook Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 7 D Puget Sound 41/ B1/ Chambers Bay 32/ B2/ Chambers mainstem 1 A Leach 5 C Flett 6 D Steilacoom Lake 9 D Ponce de Leon 10 D Lower Clover (to Spanaway) 2 B Morey 10 D Lower Spanaway 10 D Spanaway Lake 10 D Upper Spanaway 10 D NF Clover 10 D Upper Clover 10 D

Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA Chambers Bay NA Chambers mainstem 1 A Leach 2 B Flett 2 B Steilacoom Lake 5 C Ponce de Leon 6 C Lower Clover (to Spanaway) 4 C Morey 6 C Lower Spanaway 6 C Spanaway Lake 6 C Upper Spanaway 6 C NF Clover 6 C Upper Clover 6 C

1/ Due to the large size of this geographic area, its rank and benefit category may be skewed high (toward higher benefit) compared to other areas. The rank shown supposes that the entirety of the area would be restored.

2/ Only one set of survival conditions was modeled for the estuary; it is unlikely that the rank and grade would change as described for Puyallup and White chinook.

Figure 4-32. Relative importance of geographic areas for restoration and protection measures for Chambers-Clover chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-51 Pierce County Watershed Analysis Section 4

Table 4-4. Geographic areas applied in identifying strategic priorities in Chambers-Clover Basin.

Area Description Chambers Bay Chambers Bay, located immediately downstream of Chambers Dam at the mouth of Chambers Creek Chambers mainstem Mainstem Chambers Creek from its mouth at Chambers Bay to Steilacoom Lake Leach Leach Creek Flett Flett Creek Steilacoom Lake Steilacoom Lake Ponce de Leon Ponce de Leon Creek Lower Clover (to Spanaway) Clover Creek from its mouth at Steilacoom Lake to Spanaway Creek. Morey Morey Creek Lower Spanaway Spanaway Creek from its mouth to Spanaway Lake. Spanaway Lake Spanaway Lake Upper Spanaway The Spanaway Creek system upstream of Spanaway Lake. NF Clover North Fork Clover Creek and tributaries. Upper Clover Clover Creek and its tributaries upstream of Spanaway Creek (excluding North Fork Spanaway Cr)

Charts showing how benefit categories were identified for Chambers-Clover chinook are provided in Appendix D—see the first pair of charts in the Chambers-Clover chinook section.

The principal attribute classes or factors that rank highest for chinook restoration benefit are sediment load, flow, channel (or substrate) stability, channel landscape (in the estuary29), competition with hatchery fish, and habitat diversity (Figure 4-33), in addition to several other factors. Figure 4-33 summarizes strategic priorities for formulating restoration focused measures in the basin if chinook is the focus of actions. We recognize that some attribute conditions, such as fine sediment load, are not necessarily caused within the geographic area showing those conditions—the cause may be located far upstream in the drainage associated with land use. Hence identification of specific actions to rectify environmental conditions needs to consider their origin and pattern of dispersal in the watershed.

Reach specific strategic priorities for Chambers-Clover chinook are provided in Appendix D, Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in all types of watershed planning related to salmon conservation and recovery.

4.4.2 Coho Salmon 4.4.2.1 Population Performance Summary for Chambers-Clover Coho Chambers-Clover coho show a sharp reduction in population performance measures between historic and current conditions (Figure 4-34). The average spawning population size

29 The attribute "channel landscape" applies to the estuary and represents the relative composition of estuarine zones described by Cowardin et al. (1979) and Hayman et al. (1996).

June 2001 Mobrand Biometrics, Inc. Page 4-52 Pierce County Watershed Analysis Section 4

Figure 4-33. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the White watershed for chinook.

June 2001 Mobrand Biometrics, Inc. Page 4-53 Pierce County Watershed Analysis Section 4

Chambers-Clover Coho Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 4,947 27.5 100% Current with harvest and fitness loss 110 6.5 30% Current without harvest and fitness loss 156 9.4 36%

Coho spawner abundance 1,200 4,947 900

600

300 Number of fish

0 Historic Current-with harv Current-no harv Scenario

Coho productivity 30 25 20 spawner 15 s per

n 10 5 Retur 0 Historic Current-with harv Current-no harv Scenario

Coho life history diversity

100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-34. Chambers-Clover coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-54 Pierce County Watershed Analysis Section 4

of coho was estimated to be approximately 100 fish under existing conditions, with a productivity of approximately six returning adults per parent spawner. Removing all harvest and genetic loss effects in the analysis resulted in a small increase in average performance. The model estimated historic average abundance to be approximately 5,000 fish, suggesting that this basin was once highly suited to this species.

No independent estimates exist of the number of natural coho currently being produced in the Chambers-Clover system. Adult coho do enter the trap operated by WDWF at Chambers Dam at the head of tidewater, but the large majority of these fish are believed to have been hatchery fish through 1998. Approximately 500-2,000 adult coho were captured and passed upstream at the trap site between 1985 and 1998. Releases of hatchery coho smolts are no longer made in the Chambers system, but large releases had been made in the Sequalitchew system (an independent drainage to Puget Sound immediately south of Chambers Creek) until the late 1990s. It is assumed that the large majority of coho captured at the Chambers Dam trap through this period were produced at the Sequalitchew facility. In addition, hatchery coho fry were routinely released throughout the Chambers-Clover basin through 1998. These releases have now been stopped. Adults produced by these juveniles returned to the basin in years through 2000. The number of coho returning to the trap site sharply declined in 1999, when approximately 200-400 fish were captured in that year and in 2000 (Rich Eltrich, WDFW, personal communication). It appears that coho returns are declining in response to reductions in hatchery fish releases. In this light, we consider the performance estimates for Chambers-Clover Creek to be reasonable characterizations of coho performance under average existing and historic conditions.

4.4.2.2 Strategic Priorities for Chamber-Clover Coho The relative importance of geographic areas within the Chambers-Clover system to coho for restoration or protection benefits reflects the many alterations that have occurred in the basin over the past 150 years (Figure 4-35). Areas of highest priority for restoration are located upstream of Steilacoom Lake and include all of the Clover maintem, North Fork Clover Creek, and Spanaway Creek. Restoration of others areas would contribute substantial benefits to coho as well. See Table 4-3 for a description of geographic areas.

The three highest ranked areas for protection benefits were the Chambers Creek mainstem, lower Spanaway Creek, and upper Clover Creek. It should be noted that upper Clover Creek contains areas that have been severely impacted by development, as well as areas having flow and habitat conditions highly suited for coho production. Hence, this area ranks high for both restoration and protection benefits.

Charts showing how benefit categories were identified for Chambers-Clover coho are provided in Appendix D—see the first pair of charts in the Chambers-Clover coho section.

The principal attribute classes or factors that rank highest for coho restoration benefit are generally sediment load, channel (or substrate) stability, habitat diversity, habitat types (e.g., pool frequency, back water pools), water quality characteristics, and obstructions to fish passage (Figure 4-36). Particularly severe water flow conditions exist in upper Clover Creek downstream of the North Fork and in the North Fork. At least three obstructions to fish passage exist in lower Spanaway Creek and Morey Creek. It has been suggested that more

June 2001 Mobrand Biometrics, Inc. Page 4-55 Pierce County Watershed Analysis Section 4 Chambers-Clover Coho Relative Importance Of Geographic Areas For Restoration Measures

Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Figure 4-35. Relative importance of geographic areas for restoration and protection measures for Chambers-Clover coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed. than three barriers to passage may exist there—though it is difficult to assess due to difficulties of accessing private property located there.

Reach specific strategic priorities for Chambers-Clover coho are provided in Appendix D – Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in all types of watershed planning related to salmon conservation and recovery.

4.4.3 Inferences to Bull Trout for Chambers-Clover Basin Native char species (bull trout and Dolly Varden) have not been found in Chambers-Clover Creek (Don Nauer, WDFW, personal communication). In Western Washington, these

June 2001 Mobrand Biometrics, Inc. Page 4-56 Pierce County Watershed Analysis Section 4

Figure 4-36. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Chambers-Clover watershed for coho.

June 2001 Mobrand Biometrics, Inc. Page 4-57 Pierce County Watershed Analysis Section 4

species are typically associated with larger stream systems than Chambers-Clover. Spawning in those river systems occurs in colder and higher gradient streams than those in the Chambers-Clover system. We conclude that no consideration should be given to char species in identifying strategic priorities for salmonid species in this stream system.

4.4.4 Data/Information Uncertainties for Chambers-Clover Basin The data and information used to characterize the environment were brought into the analysis through two processes: one involving all stream reaches upstream of tidewater and one that addressed the estuary and marine areas. Each process and the information used have a different level of uncertainty.

The principal characterization of the Chambers-Clover watershed was assembled by staff of Pierce County (Appendix C), all having worked extensively in the watershed (see Appendix C).30 The procedure required assigning a "level of proof" to each attribute rating for each reach, using a scale of 1-4, where a value of 1 meant that empirical data were used and a high level of confidence was placed in the rating and a 4 represented an educated guess with low confidence. Values of 2 and 3 were intermediate, where a 2 represented a relatively high level of confidence, based on a combination of personal observation and "weight of evidence" and a 3 drew on a theoretical application.

The large majority of ratings applied to the Chambers-Clover upstream of tidewater were assigned a level of proof of 2 and 3. Fewer ratings were assigned values of 1 and 4. The large majority of ratings used to characterize historic conditions were assigned levels of proof of 3 and 4. We find that these levels of proof are reasonable, reflecting a good understanding of watershed processes and documented indices of environmental change (such as described in May et al. 1997).

Based on our assessment of the uncertainty Recommended field verification of: that exists in this basin, we conclude that fine sediment within riffles field verification should concentrate on two bed scour attribute classes: complete identification of distribution of springs and upwelling fish passage barriers, particularly in the storm-runoff patterns Spanaway and Morey subbasins, and flow characteristics in the Clover and Spanaway areas with respect to spring influences (upwelling sources).31 We also find that additional information may be needed to better characterize water quality conditions for fish life in the basin, such as in the lakes. Other attributes may also need to be field verified, but we suggest that this should be part of an on-going monitoring plan associated with action effectiveness monitoring. We found that although empirical data are lacking for some attributes in the basin, there is ample evidence that the characterization made by the team is reasonably accurate.

30 Ratings were subsequently checked by MBI for consistency with the approach, reviewed by other resource specialists who have worked on the Chambers-Clover system, and again by the team who first assembled the ratings. 31 In reference to springs, of particular relevance would be identification of spring sources that promote upwelling within the channel substrate associated with high dissolved oxygen. These sites are important for a variety of reasons, one of which is to ameliorate the effect of fine sediment on incubating eggs. The likely presence of such areas in the upper basin is an important assumption made in this analysis.

June 2001 Mobrand Biometrics, Inc. Page 4-58 Pierce County Watershed Analysis Section 4

Ratings for Chambers Bay were extrapolated from another project that addressed other estuaries and bays in Puget Sound. There is a high level of uncertainty about the overall contribution of the estuary and bay to Chambers-Clover fish, as well as to other migrating salmon juveniles that may utilize these areas. Until further information becomes available, however, we believe that the basic conclusions of the assessment are reasonable with regard to the relative contributions of estuarine/bay areas and freshwater reaches to population performance.

4.4.5 Chambers-Clover Basin Conclusions The large majority of the Chambers-Clover watershed has undergone extensive alterations over the past 150 years—first by logging, followed by various kinds of development and urbanization. Major changes have occurred to salmon habitat throughout its length. Salmon population performance has been dramatically reduced corresponding to these changes.

The watershed currently does not support a self-sustaining chinook population. We question whether a chinook population was ever sustainable over the long-term due to the basin's small size. The estuary was historically used by juvenile chinook originating in other streams, as it still is today.

The Chambers-Clover watershed was historically highly suited to coho salmon, and it appears that a population is still present, though at relatively low numbers. We find that this species would make an excellent indicator species for formulating watershed action plans to address salmonid conservation and recovery needs. Attributes to target for restoration: Conservation and restoration measures can be fine sediment load developed following a set of strategic priorities for flow characteristics geographic areas within the basin (Figure 4-37). habitat types (e.g., pools) These priorities identify the strategic importance of habitat structure (e.g., wood) streambed stability different areas in the watershed for either restoring water quality (including only partial recovery) or protecting conditions for salmonid performance. Attributes Attributes to target for protection: were identified to be targeted in planning new groundwater sources action measures. riparian function (e.g., wetlands)

4.5 Kitsap Watersheds The 5 independent watersheds examined on the Kitsap Peninsula were (from east to west): 1) Crescent Creek, 2) Donkey Creek, 3) Burley Creek, 4) Minter Creek, and 5) Rocky Creek. We completed an assessment of chinook and coho in all of these watersheds except Donkey Creek. We concluded, based on small watershed size, that chinook assessment in Donkey Creek is not warranted.

June 2001 Mobrand Biometrics, Inc. Page 4-59 Pierce County Watershed Analysis Section 4

Chambers-Clover Basin Overview Strategic Assessment of Geographic Areas

Chinook Coho strategic strategic priority priority

Key to Strategic Priority (Benefit Category letter shown)

D & E C B A Indirect or General Low Medium High

Figure 4-37. Overview of strategic priorities for restoration and protection measures by geographic area within the Chambers-Clover basin.

4.5.1 Chinook Salmon 4.5.1.1 Population Performance Summaries We concluded from our analysis that the four watersheds modeled for chinook performance cannot currently sustain a naturally reproducing population of chinook over the long-term. Combined, the average spawning population size was less than 200 fish. Individually, population size ranged from less than 10 fish to 70 fish, and population productivity varied from 2 to 3 adult returns per spawner (Figures 4-38 to 4-41).

June 2001 Mobrand Biometrics, Inc. Page 4-60 Pierce County Watershed Analysis Section 4

Crescent Fall Chinook Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 92 21.2 100% Current with harvest and fitness loss 7 2.9 68% Current without harvest and fitness loss 32 7.7 100%

Chinook spawner abundance 100 90 80 sh 70 fi f 60 50 er o 40

mb 30

Nu 20 10 0 Historic Current-with harv Current-no harv Scenario

21.2 Chinook productivity 15

Returns per spawner 0 Historic Current-with harv Current-no harv Scenario

Chinook life history diversity 100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-38. Crescent chinook (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-61 Pierce County Watershed Analysis Section 4

Burley Fall Chinook Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 760 15.1 100% Current with harvest and fitness loss 47 2.2 32% Current without harvest and fitness loss 252 5.0 99%

Chinook spawner abundance 1,000 900 800 sh 700 fi f 600 500 er o 400

mb 300

Nu 200 100 0 Historic Current-with harv Current-no harv Scenario

15.1 Chinook productivity 15

5 Returns per spawner 0 Historic Current-with harv Current-no harv Scenario

Chinook life history diversity 100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-39. Burley chinook (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-62 Pierce County Watershed Analysis Section 4

Minter Fall Chinook Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 529 18.8 100% Current with harvest and fitness loss 24 1.8 20% Current without harvest and fitness loss 197 4.9 100%

Chinook spawner abundance 600 500 sh 400

of fi 300 200

Number 100 0 Historic Current-with harv Current-no harv Scenario

18.8 Chinook productivity 15

10 spawner

s per 5 n Retur 0 Historic Current-with harv Current-no harv Scenario

Chinook life history diversity 100%

75%

cent 50% Per 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-40. Minter chinook (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-63 Pierce County Watershed Analysis Section 4

Rocky Fall Chinook Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 477 15.8 100% Current with harvest and fitness loss 69 2.9 82% Current without harvest and fitness loss 335 8.6 100%

Chinook spawner abundance 600 500 sh 400

of fi 300 200

Number 100 0 Historic Current-with harv Current-no harv Scenario

15.8 Chinook productivity 15

10 spawner

s per 5 n Retur 0 Historic Current-with harv Current-no harv Scenario

Chinook life history diversity 100%

75%

cent 50% Per 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-41. Rocky chinook (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-64 Pierce County Watershed Analysis Section 4

The EDT analysis suggests that these watersheds might have supported a small run of chinook historically. The model estimated a combined average run size of about 1,900 fish under pristine conditions, with a productivity of about 15-20 returning adults per parent spawner 32. However, the EDT model may give an overly favorable view of chinook for small independent drainages to Puget Sound. Chinook salmon spawning is almost always associated with large mainstem rivers or their larger tributaries—the EDT model does not limit watershed size for utilization by chinook, although it does reduce adult migration success as stream size diminishes.

The WDFW consider these populations to be what it calls a "Category 3 population"—one associated with an independent small drainage to Puget Sound that is not self-sustaining over the long-term. Chinook observed in these drainages are considered by the agency to likely be strays, naturally-produced strays from other systems, or progeny of one of these two groups (Chuck Baranski, WDFW, personal communication). The Puget Sound Technical Review Team (TRT) also does not consider such small drainages as having self-sustaining chinook populations (Puget Sound TRT 2001).

Crescent Creek: Based on model results, the average spawning population size of chinook, after taking into account harvest and loss in genetic fitness, was estimated to be less than 10 fish, with a population productivity of 3 adult returns per parent spawner (Figure 4-38). Clearly, these performance measures indicate a population that is not sustainable. Chinook performance was modeled for the lower 1 mile of Crescent Creek. We limited chinook distribution because of the small size of the stream upstream of this location. The life history diversity value is high for this watershed largely because of the limited distribution of the analysis (e.g., low number of potential life history pathways). Removing all harvest and genetic loss effects from the analysis resulted in a minor increase in average run size and a moderate increase in productivity. This suggests that effects of the environment on performance (mostly capacity) are too severe to sustain chinook in this system under existing conditions.

These model results are consistent with the low number of sightings of adult chinook in Crescent Creek (WCC, 2001). Chinook observed are thought to be the result of annual hatchery plants in the stream.

The analysis suggests that historically Crescent Creek never supported many chinook salmon. The model estimated an average run size of about 90 fish under pristine conditions with a productivity of about 21 returning adults per parent spawner.

Burley Creek: Based on model results, the average spawning population size of chinook, after taking into account harvest and loss genetic fitness, was estimated to be less than 50 fish, with a population productivity of approximately 2 adult returns per parent spawner (Figure 4-39) The life history diversity value indicates that less than half of the historic life history pathways can be successfully used. These performance measures are consistent with a low probability of population sustainability. Removing all harvest and genetic loss effects

32 Although effective population size (Ne) is not equivalent to the average annual number of spawners, it would be close in this case.

June 2001 Mobrand Biometrics, Inc. Page 4-65 Pierce County Watershed Analysis Section 4

from the analysis resulted in a moderate increase in average run size and productivity. The analysis suggests that the environment is not too severely degraded to support chinook.

At first glance, these results appear inconsistent with observations of adult chinook in Burley Creek in recent years (from 1986 to 1999 escapement estimate for Burley Creek varied from 75 adults to 1,300 adults average ~500 adults; reported in WCC, 2001). However, adult chinook abundance in Burley Creek is thought to be mostly strays from South Puget Sound hatcheries (Chuck Baranski, WDFW, personal communication).

The analysis does suggest that the Burley Creek might have supported a small run of chinook historically. The model estimated an average run size of about 800 fish under pristine conditions with a productivity of about 15 returning adults per parent spawner.

Minter Creek: Adult chinook were not passed upstream of the Minter Creek Hatchery weir in fall of 2000 (Chuck Baranski, WDFW, personal communication). Supposedly, this policy has been in place for several years. However, the WCC Limiting Factors report (2001) identified chinook distribution up to the Minter-Kitsap Road and mentioned that modifications are planned at the hatchery weir to improve upstream access of adults. We modeled a hypothetical current condition using the assumption that fish are passed upstream of the weir. The average spawning population size was estimated to be about 20 fish, with productivity approximately 2 returns per parent spawner (Figure 4-40). The life history diversity value indicates that only 20% of the historic life history pathways can be successfully used. None of these performance measures are consistent with population sustainability.

Historically, Minter Creek likely supported a small run of adult chinook. Although the small population size suggests that this stock was not sustainable over the long-term. The model estimated an average run size of 500 fish under pristine conditions with a productivity of 19 returning adults per spawner.

Rocky Creek: Based on model results, the average spawning population size of chinook, after taking into account harvest and loss in genetic fitness, was estimated to be less than 70 fish, with a population productivity of approximately 3 adult returns per parent spawner (Figure 4-41). The life history diversity value indicates that 80% of the historic life history pathways can be successfully used. Removing all harvest and genetic loss effects from the analysis resulted in a substantial increase in average run size and productivity. The analysis suggests that the environment is in relatively good condition. Adult abundance is 70% of historic, and productivity is about half of historic. Of the four Kitsap watersheds modeled for chinook performance, Rocky Creek had the highest performance measures.

4.5.1.2 Strategic Priorities for Kitsap Chinook We assessed strategic priorities for chinook in the Kitsap watersheds for the sake of completeness for the analysis, although the potential importance of these drainages for chinook production and status as individual populations for ESA recovery is questioned (Puget Sound TRT 2001).

Charts showing how benefit categories were identified for Kitsap chinook are provided in Appendix D—see the first pair of charts under Kitsap chinook.

June 2001 Mobrand Biometrics, Inc. Page 4-66 Pierce County Watershed Analysis Section 4

Reach specific strategic priorities for Kitsap chinook are provided in Appendix D—Stream Reach Analysis for Species Performance. The reach analysis document is a reference tool to be used in all types of watershed planning related to salmon conservation and recovery.

Crescent Creek: The relative importance of geographic areas within the drainage to Crescent chinook for both restoration and protection measures is displayed in Figure 4-42. The drainage is divided into seven geographic areas, from the estuary to Crescent Lake (Table 4- 5). Potential chinook distribution is presumed to be in Crescent Creek in the lower and mid geographic areas. Thus, both geographic areas rank high for restoration and protection. The Crescent Creek estuary, including portions of Gig Harbor, ranks high for restoration measures. Crescent Chinook Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 6B Puget Sound 4A South Puget Sound 5B Crescent Estuary 3A Lower Crescent Cr 1A Mid Crescent Cr 1A Crescent Cr upstream Crescent Valley Rd 7D Up Crescent (downstream driveway culvert) 7D 136th St driveway culvert - Crescent Creek 7D Upper Crescent Creek to Lake 7D

Crescent Chinook Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area) Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA South Puget Sound NA Crescent Estuary NA Lower Crescent Cr 2A Mid Crescent Cr 1A Crescent Cr upstream Crescent Valley Rd 7D Up Crescent (downstream driveway culvert) 7D 136th St driveway culvert - Crescent Creek 7D Upper Crescent Creek to Lake 7D

Figure 4-42. Relative importance of geographic areas for restoration and protection measures for Crescent chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-67 Pierce County Watershed Analysis Section 4

Table 4-5. Geographic areas applied in identifying strategic priorities in Crescent Basin.

Area Description Crescent Estuary Gig Harbor and tidally influenced section of Crescent Creek (to upper extent of tidal influence). Lower Crescent Creek Top of estuary to beginning of streamside residential (CM 0.7); includes portion of Gig Harbor City Park. Mid Crescent Creek CM 0.7 to approximately 0.1 mile upstream of last crossing Crescent Valley Rd. Crescent Cr above Crescent CM 1.6 to CM 2.5; area of mixed agriculture/forest Valley Rd Upper Crescent Creek CM 2.5 to 136th St Driveway Culvert (CM 3.1); area mostly forest with (downstream 136th St occasional residential. Driveway Culvert) 136th St Driveway Culvert Driveway culvert just downstream of Crescent Lake outlet (culvert identified in WCC Limiting Factors report (2001); field verified February 2001. Upper Crescent Creek to Driveway culvert (136th St) to Crescent Lake outlet. Lake If chinook are the focus of actions, then priority should be given to lower Crescent Creek (downstream of the Crescent Valley Road) and the estuary/Gig Harbor. The relative importance of the estuary and nearshore environment should not be surprising as juvenile chinook are highly dependent on these habitats for early rearing (Healey 1982).

The principal attribute classes or factors that rank highest for chinook restoration benefit are channel (or substrate stability), channel landscape (in estuary), and loss of key habitat (Figure 4-43). Figure 4-43 summarizes strategic priorities for formulating restoration measures in the basin if chinook is the focus species for actions. Several of the attribute conditions are the results of streamside development (loss of key habitat and, to an extent, loss of channel stability and habitat diversity). The cause for others may not be within the geographic areas showing those conditions—the cause may be located upstream in the drainage associated with land use (sediment load is a good example). Hence identification of specific actions to rectify the environmental conditions needs to consider their origin and pattern of dispersal in the watershed.

Burley Creek: The relative importance of geographic areas within Burley Creek to Burley chinook for both restoration and protection measures is displayed in Figure 4-44. The drainage is divided into 8 geographic areas, from the estuary to the headwaters (Table 4-6). Again the estuary (including portions of Burley Lagoon) and the lower river rank high for both restoration and protection.

Habitat quality in the lower most geographic area (Lower Burley Creek) is generally poor (urban area from mouth to ~CM 0.75) (WCC 2001). This area ranked highest for restoration measures. The geographic area immediately upstream (Lower-mid Burley Creek) is in relatively good condition (largely intact riparian zone). This geographic area ranked highest for protection measures.

June 2001 Mobrand Biometrics, Inc. Page 4-68 Pierce County Watershed Analysis Section 4

Crescent Chinook Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority ch) y it hat y1/ LOCATION her sp) / y poaching

(miles from mouth) / quant abilit t ure load a als (shaded rows characterize tributaries or ion (w ion (ot it it diversit ion relevant sites) t a hogens hdraw w mpet mpet rassment mperat xygen hannel st hemicals o o abit a e ood Benefit category C C C C Flo F H H Obstructions O Pat Predat Sediment T Wit Key habit 3.57 Crescent Lake Outlet

3.26 136th St Driveway Culvert 2.88

2.57

2.26 k e e

r 1.95 C

t n

e 1.64 c s e r 1.33 C

1.02 Crescent Valley Rd

0.71

0.4 End of tidal influence

0.0 Crescent Creek Estuary Gig Harbor

Figure 4-43. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Crescent watershed for chinook salmon.

The results state that the highest priority should be given to protecting and restoring habitat conditions within the lower portion of the drainage if chinook are the focus of actions. For simply protection, the highest priority should be given to protecting habitat conditions upstream of this area to approximately tributary 15.0058. The fact that these areas are ranked higher than the estuary does not diminish the importance of estuarine habitat—although protection priority was not analyzed for the estuary, we can infer from the lower restoration ranking that the Burley estuary is largely intact and should be given high priority for protection. Also, the WCC Limiting Factors report for WRIA 15 described the upper third of Burley estuary as mostly undeveloped (WCC 2001).

June 2001 Mobrand Biometrics, Inc. Page 4-69 Pierce County Watershed Analysis Section 4

Burley Chinook Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 7 C Puget Sound 2 A South Puget Sound 6 C Burley Estuary/Lagoon 4 A Lower Burley Creek 1 A Lower-Mid Burley Creek 2 A Upper-Mid Burley Creek 9 D Upper Burley Creek 9 D Little Bear Creek 5 B Unnamed Tributary 15.0058 8 C Unnamed Tributary 15.0059 5 B

Burley Chinook Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA South Puget Sound NA Burley Estuary/Lagoon NA Lower Burley Creek 2 A Lower-Mid Burley Creek 1 A Upper-Mid Burley Creek 6 D Upper Burley Creek 6 D Little Bear Creek 3 A Unnamed Tributary 15.0058 4 B Unnamed Tributary 15.0059 4 B

Figure 4-44. Relative importance of geographic areas for restoration and protection measures for Burley chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

Table 4-6. Geographic areas applied in identifying strategic priorities in Burley Basin.

Area Description Burley Estuary/Lagoon Entrance to Burley Lagoon (area inside causeway) upstream to upper extent of tidal influence. Lower Burley Creek Mainstem Burley Creek from the upper end of tidal influence to ~0.5 miles upstream of confluence with Little Bear Creek. Lower-Mid Burley Creek Mainstem Burley Creek from ~0.5 miles upstream of confluence with Little Bear Creek to confluence with tributary 15.0059. Burley-Ollala Road crossing is within this area. Upper-Mid Burley Creek Mainstem Burley Creek from confluence with tributary 15.0059 to Holman Road. Area of agriculture development. Upper Burley Creek Mainstem Burley Creek from Holman Road to headwaters (SR 16 crossing). Includes Burley Road crossing and Mullenix Road crossing. Little Bear Creek Little Bear Creek in its entirety Unnamed Tributary 15.0058 Tributary 15.0058 in its entirety. Unnamed Tributary 15.0059 Tributary 15.0059 in its entirety.

June 2001 Mobrand Biometrics, Inc. Page 4-70 Pierce County Watershed Analysis Section 4

The principal attribute classes or factors that rank highest for chinook restoration benefit are sediment load, channel (or substrate) stability, channel landscape (in the estuary) and habitat diversity (Figure 4-45). Figure 4-45 summarizes strategic priorities for formulating restoration focused measures in the basin if chinook are the focus of actions.

Burley Chinook Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority ch) y it hat y1/ LOCATION her sp) / y poaching

(miles from mouth) / quant abilit t ure load a als (shaded rows characterize tributaries or ion (w ion (ot it it diversit ion relevant sites) t a hogens hdraw w mpet mpet rassment mperat xygen hannel st hemicals o o abit a e ood Benefit category C C C C Flo F H H Obstructions O Pat Predat Sediment T Wit Key habit Headwaters Burley Creek 6.3 Mullenix Rd Crossing

5.5 Burley Rd Crossing

4.7 In unnamed tributary 15.0059 - Upstream SR 16 - SR 16 Road Crossing - Downstream SR 16 4.4 In unnamed tributary 15.0058 - Upstream SR 16 - SR 16 Road Crossing

Burley Creek - Downstream SR 16 3.6 Burley-Olalla Rd

In Little Bear Creek - Upper Little Bear Cr - Lower Little Bear Cr

1.6 End of tidal influence Burley Creek Estuary Burley Creek mouth

Figure 4-45. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Burley watershed for chinook salmon.

Minter Creek: The relative importance of geographic areas within Minter Creek to Minter chinook for both restoration and protection measures is displayed in Figure 4-46. The drainage is divided into 9 geographic areas, from the estuary to the headwaters (Table 4-7). The hatchery weir was modeled as a geographic area. Other barriers to fish passage in Minter Creek (upper most culvert on 118th Street and the culvert at Pine Road were included with their respective geographic areas). Again the estuary (including portions of Minter Bay) and the lower river rank high for both restoration and protection. We assumed adult chinook were passed immediately upstream of the hatchery weir, thus we assumed this structure had

June 2001 Mobrand Biometrics, Inc. Page 4-71 Pierce County Watershed Analysis Section 4

Minter Chinook Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 7C Puget Sound 3A South Puget Sound 6C Minter Estuary 2A Minter Hatchery Dam 5B Lower Minter Creek 1A Mid Minter Creek 3A Upper Minter Creek below Pine Rd 8D Upper Minter Creek above Pine Rd 8D Lower Huge Creek 8D Upper Huge Creek 8D Little Minter Creek 8D

Minter Chinook Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area) Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA South Puget Sound NA Minter Estuary NA Minter Hatchery Dam NA Lower Minter Creek 1A Mid Minter Creek 2A Upper Minter Creek below Pine Rd 3D Upper Minter Creek above Pine Rd 3D Lower Huge Creek 3D Upper Huge Creek 3D Little Minter Creek 3D

Figure 4-46. Relative importance of geographic areas for restoration and protection measures for Minter chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

Table 4-7. Geographic areas applied in identifying strategic priorities in Minter Basin.

Area Description Minter Estuary Includes portion of Minter Bay to upstream extent of tidal influence in Minter Creek (approx. location of Hatchery Weir). Minter Creek Hatchery Dam Hatchery Dam and Weir Lower Minter Creek Mainstem Minter Creek from Hatchery Dam to confluence with Huge Creek. Mid Minter Creek Mainstem Minter Creek from confluence of Huge Creek to upper most 118th Street culvert. Upper Minter Creek below Pine Mainstem Minter Creek form 118th Street to Pine Road. Upper Minter Creek above Pine Mainstem Minter Creek form Pine Road to headwaters. Lower Huge Creek From confluence with Minter Creek to county line. Upper Huge Creek County line to headwaters (wetland). Little Minter Creek From Confluence with Minter Creek to headwaters; includes culvert at 118th Street.

June 2001 Mobrand Biometrics, Inc. Page 4-72 Pierce County Watershed Analysis Section 4

a minor effect on chinook survival. Barrier culverts identified in the WCC Limiting Factors report (WCC 2001) in Minter Creek were upstream of the chinook distribution, and thus they do not affect strategic priorities for Minter chinook.

The principal attribute classes or factors that rank highest for chinook restoration benefit are sediment load and habitat diversity (Figure 4-47). Attribute factors that rank high for the reach upstream of Huge Creek (area that Minter Creek flows alongside 118th Street) are sediment load, habitat diversity, and loss of key habitat.

Minter Chinook Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority ch) y it hat y1/ LOCATION her sp) / y poaching

(miles from mouth) / quant abilit t ure load a als (shaded rows characterize tributaries or ion (w ion (ot it it diversit ion relevant sites) t a hogens hdraw w mpet mpet rassment mperat xygen hannel st hemicals o o abit a e ood Benefit category C C C C Flo F H H Obstructions O Pat Predat Sediment T Wit Key habit 8 Minter Creek Headwaters

6.6 Pine Rd culvert 5.8

4.6 Upper most 118th St Culvert 4

In Huge Creek - Headwaters - to County line nter Creek

i - Huge Cr to gage station

M 2.8 In Little Minter Creek - Upstream of culvert - 118th St culvert 2.2

Hatchery Dam 1.2 End of tidal influence Minter Creek Estuary Henderson Bay

Figure 4-47. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Minter watershed for chinook salmon.

June 2001 Mobrand Biometrics, Inc. Page 4-73 Pierce County Watershed Analysis Section 4

Rocky Creek: The relative importance of geographic areas within Rocky Creek to Rocky chinook for both restoration and protection measures is displayed in Figure 4-48. The drainage is divided into 7 geographic areas, from the estuary to the headwaters (including Fork Muck Creek) (Table 4-8). The impassible culvert crossing 144th Street (WCC 2001) was included within the Mid Rocky Creek geographic area.

Rocky Chinook Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 6B Puget Sound 1A South Puget Sound 4A Rocky Estuary 4A Lower Rocky Cr 2A Mid Rocky Cr 7D Fern Lake and Dam 7D Rocky Cr above Fern Lk 7D Lower Fork Muck Cr 3A Upper Fork Muck Cr 7D

Rocky Chinook Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA South Puget Sound NA Rocky Estuary NA Lower Rocky Cr 1A Mid Rocky Cr 3D Fern Lake and Dam 3D Rocky Cr above Fern Lk 3D Lower Fork Muck Cr 2A Upper Fork Muck Cr 3D

Figure 4-48. Relative importance of geographic areas for restoration and protection measures for Rocky chinook salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-74 Pierce County Watershed Analysis Section 4

Table 4-8. Geographic areas applied in identifying strategic priorities in Rocky Basin.

Area Description Rocky Estuary Includes portion of Rocky Bay to upstream extent of tidal influence in Rocky Creek. Lower Rocky Creek Mainstem Rocky Creek from upper extent tidal influence to 144th Street culvert (approximate confluence with unnamed tributary 15.0021. Fork Muck Creek enters this area. Mid Rocky Creek Mainstem Rocky Creek from 144th Street Culvert to Fern Lake outlet (144th culvert identified as passage barrier in WCC Limiting Factors report (2001). Fern Lake and Dam Fern Lake Rocky Creek above Fern Lake Mainstem Rocky Creek upstream of Fern Lake (headwaters). Lower Fork Muck Creek Fork Muck Creek from confluence with Rocky Creek to Wright-Bliss Road. Upper Fork Muck Creek From Wright-Bliss Road to headwaters.

Lower Rocky Creek and lower Fork Muck Creek rank high for protection measures. These geographic areas have experienced minor degradation of habitat conditions, but they are largely intact. Rocky estuary ranked low for restoration measures, which is a reflection of the relatively good condition of the estuary. Although not analyzed for protection, we can infer a high protection rank for Rocky estuary.

The principal attribute classes or factors that rank highest for chinook restoration benefit are sediment load, habitat diversity, and channel stability (Figure 4-49). Most attribute factors were classified low to medium priority (compare to the medium to high classifications shown in the more heavily developed streams such as Crescent, Burley, or Minter creeks).

4.5.2 Coho Salmon 4.5.2.1 Population Performance Summary for Kitsap Coho Kitsap coho show a sharp reduction in population performance measures between historic and current conditions (Figures 4-50 to 4-54). Excluding Donkey Creek, the average spawning population size of coho was estimated to be approximately 100-600 fish under existing conditions, with a productivity of about 5-9 returning adults per parent spawner. Donkey Creek abundance is estimated to be 0 adult coho. Removing all harvest and genetic loss effects from the analysis resulted in a moderate increase in performance for most stocks (we modeled a 28% exploitation rate for these stocks, which includes a South Puget Sound fishery). Loss of genetic fitness was 15% for most stocks—the exception was Minter Creek. We assumed a 30% loss for this stock based on the long-term hatchery program associated with this stock. The model estimated historic average abundance to be in near 8,000 spawners on average for all stocks. Individual stock abundance ranged from 50 (Donkey Creek) to 3,600 (Minter Creek).

We consider the performance estimates for Kitsap watersheds to be reasonable characterizations of coho performance under average existing and historic conditions. In any given year, spawner abundances could be significantly less or greater than the average measures listed in Figures 4-50 to 4-54 , due to environmental variation. The results depict a group of populations that have experienced a major loss in performance due to environmental alteration—but that appear to have significant potential for recovery through watershed actions. We find that coho salmon would make an excellent indicator species for

June 2001 Mobrand Biometrics, Inc. Page 4-75 Pierce County Watershed Analysis Section 4

Rocky Chinook Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority

LOCATION (miles from mouth) (shaded rows characterize tributaries or relevant sites) Benefit category Channel stability1/ Chemicals Competition (w/ hatch) Competition (other sp) Flow Food Habitat diversity Harassment/poaching Obstructions Oxygen Pathogens Predation Sediment load Temperature Withdrawals Key habitat quantity 5.6 Rocky Creek Headwaters

5.1

4.6 Fern Lake Dam

4.1

3.6

3.1 144th Street Culvert 2.9

Rocky Creek In Fork Muck Creek - Upstream Wright-Bliss Rd - below Wright-Bliss Rd 1.5

1.0 End of tidal influence

0.5 Rocky Creek Estuary

Rocky Bay

Figure 4-49. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Rocky watershed for chinook salmon. formulating watershed action plans for these drainages to address salmonid conservation and recovery needs. Coho salmon utilize nearly the entire drainage during multiple life stages.

A description of model results by watershed follows.

Crescent Creek: Based on model results, the average spawning population size of coho after taking into account harvest and loss of genetic fitness, was estimated to be 100 fish, with a productivity of 9 adult returns per spawner (Figure 4-50). Removing all harvest and genetic fitness loss effects from the analysis resulted a moderate increase in abundance.

These results are consistent with expected production for a stream of this size and condition. There are no estimates of spawning population sizes or smolt yields of coho in the Crescent

June 2001 Mobrand Biometrics, Inc. Page 4-76 Pierce County Watershed Analysis Section 4

Crescent Coho Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 903 32.3 100% Current with harvest and fitness loss 103 8.7 100% Current without harvest and fitness loss 168 14.1 100%

Coho spawner abundance 1,000 900 800 700 600 500 er of fish

b 400 300

Num 200 100 0 Historic Current-with harv Current-no harv Scenario

32.3 Coho productivity 30

10 Returns per spawner 0 Historic Current-with harv Current-no harv Scenario

Coho life history diversity

100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-50. Crescent coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-77 Pierce County Watershed Analysis Section 4

Donkey Coho Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 53 23.1 100% Current with harvest and fitness loss 0 0.0 0% Current without harvest and fitness loss 0 0.0 0%

Coho spawner abundance 100 90 80 70 60 of fish 50 er

b 40 30

Num 20 10 0 Historic Current-with harv Current-no harv Scenario

Coho productivity 30

10 Returns per spawner 0 Historic Current-with harv Current-no harv Scenario

Coho life history diversity

100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-51. Donkey coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-78 Pierce County Watershed Analysis Section 4

Burley Coho Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 1,089 22.5 100% Current with harvest and fitness loss 134 4.8 71% Current without harvest and fitness loss 228 7.6 82%

Coho spawner abundance 1,200 1,000 800 600 er of fish b 400

Num 200 0 Historic Current-with harv Current-no harv Scenario

Coho productivity 30

10 Returns per spawner 0 Historic Current-with harv Current-no harv Scenario

Coho life history diversity

100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-52. Burley coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-79 Pierce County Watershed Analysis Section 4

Minter Coho Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 3,597 33.1 100% Current with harvest and fitness loss 480 8.5 68% Current without harvest and fitness loss 905 16.7 68%

Coho spawner abundance 4,000

sh 3,000

2,000

1,000 Number of fi

0 Historic Current-with harv Current-no harv Scenario

33.1 Coho productivity 30

20 spawner s per

n 10 Retur 0 Historic Current-with harv Current-no harv Scenario

Coho life history diversity

100%

75%

cent 50% Per 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-53. Minter coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-80 Pierce County Watershed Analysis Section 4

Rocky Coho Summary Of Projected Performance Measures Under Three Scenarios

Scenario Abundance Productivity Diversity index Historic 2,328 23.6 100% Current with harvest and fitness loss 593 9.2 75% Current without harvest and fitness loss 958 15.0 75%

Coho spawner abundance 2,500

2,000 fish f 1,500 er o

b 1,000 m

Nu 500

0 Historic Current-with harv Current-no harv Scenario

Coho productivity 30 ner

10 Returns per spaw 0 Historic Current-with harv Current-no harv Scenario

Coho life history diversity

100%

75%

50% Percent 25%

0% Historic Current-with harv Current-no harv Scenario

Figure 4-54. Rocky coho (naturally produced) performance measures based on modeling results.

June 2001 Mobrand Biometrics, Inc. Page 4-81 Pierce County Watershed Analysis Section 4

system based on field observations. The model assumes an approximate 5% smolt to adult return rate (marine survival and harvest). Back calculating to smolts results in approximately 2,000 smolts produced from Crescent Creek. Observed smolt abundance data from streams of comparable size on the Olympic Peninsula are consistent with this estimate (Lestelle et al. 1993a); these streams have not experienced the urban development of Crescent Creek, but they have experienced significant alternations through logging. Adult abundance in coastal streams is less than that estimated for Crescent Creek because of lower marine survival rates and higher exploitation rates.

The model estimated an average run size of 900 fish under historic pristine conditions with a productivity of 30 returning adults per spawner. Much of the loss in performance from the historic condition can be attributed to loss of habitat quality and quantity.

Donkey Creek: Based on model results, the average spawning population size of coho after taking into account harvest and loss of genetic fitness, was estimated to be 0 fish, with a productivity of less than 1 adult return per spawner (Figure 4-51). Zero production from Donkey Creek is based on the conclusion that the culvert under Harborview Drive is a complete barrier to adult coho (Figure 4-55). Water is backed up in the culvert to supply water to a series of streamside chum incubators. We assumed that this water supply is set up in early October, which is prior to coho adult migration into the drainage.

Figure 4-55. Flow diversion in Donkey Creek – Harborview Drive culvert.

The model estimated an average run size of 50 fish under historic pristine conditions with a productivity of 23 returning adults per spawner. The analysis suggests that Donkey coho was not an independent population (based on low abundance). Adult strays from the larger, nearby Crescent Creek, probably maintained the population in this drainage.

June 2001 Mobrand Biometrics, Inc. Page 4-82 Pierce County Watershed Analysis Section 4

Burley Creek: Based on model results, the average spawning population size of coho after taking into account harvest and loss of genetic fitness, was estimated to be 130 fish, with a productivity of 5 adult returns per spawner (Figure 4-52). The life history diversity value indicates about two-thirds of the historic life history pathways can be successfully used. Removing all harvest and genetic fitness loss effects from the analysis resulted in potential abundance of about 200 fish.

Model results for smolt production are 2,000 – 3,000 fish. It is difficult to compare these results with other Puget Sound streams. Burley Creek is about half the size of Big Beef Creek (based on linear stream length—7 and 12 miles, respectively). Big Beef Creek has a long record of smolt abundance data. Average coho smolt abundance from Big Beef Creek is 25,000 fish (Seiler 2001). However, Big Beef Creek has extensive wetlands in the upper third of the watershed, which greatly increases smolt potential. Furthermore, Big Beef Creek has not experienced the effect of urban and agricultural development to the same degree as Burley Creek.

Burley coho show a sharp reduction in population performance measures between historic and current conditions (Figure 4-52). The model estimated an average run size of 1,100 fish under historic pristine conditions with a productivity of 23 returning adults per spawner.

Minter Creek: Based on model results, the average spawning population size of coho after taking into account harvest and loss of genetic fitness, was estimated to be 480 fish, with a productivity of 8.5 adult returns per spawner (Figure 4-53). The life history diversity value indicates that about 70% of the historic life history pathways can be successfully used. Removing all harvest and genetic fitness loss effects from the analysis resulted in a near doubling of abundance. Because Minter Creek has an extensive history of hatchery propagation, we assumed a higher loss of genetic fitness for this population than for others in the areas.

Model results for smolt production are approximately 10,000 fish.

The model estimated an average run size of 3,600 fish under historic pristine conditions with a productivity of 33 returning adults per spawner. Salo and Bayliff (1958) analyzed results of production monitoring of coho from Minter Creek for 1938 to 1955. Female spawning escapements of 500-1,000 produced 25,000-35,000 smolts (Salo and Bayliff assumed the freshwater capacity to be in this range). Based on our model results, freshwater capacity for the pristine condition was 30,000 to 35,000 smolts. The Minter Creek watershed was first logged off from 1900 to 1935. However, Salo and Bayliff described the watershed as in good condition at the time of their study (second-growth timber established and no serious erosion problems). They made no mention of the adult barriers that are currently present in the watershed (Little Minter Creek and at Pine Road).

Rocky Creek: Based on model results, the average spawning population size of coho after taking into account harvest and loss of genetic fitness was estimated to be 600 fish, with a productivity of 9 adult returns per spawner (Figure 4-54). The life history diversity value indicates that about 75% of the historic life history pathways can be successfully used. Removing all harvest and genetic fitness loss effects from the analysis resulted in a modest

June 2001 Mobrand Biometrics, Inc. Page 4-83 Pierce County Watershed Analysis Section 4

improvement in abundance. Model results for smolt production are approximately 8,000 – 9,000 fish.

The model estimated an average run size of 2,300 fish under historic pristine conditions with a productivity of 23 returning adults per spawner. Smolt production with pristine conditions was 21,000 fish.

4.5.2.2 Strategic Priorities for Kitsap Coho Crescent Creek: The relative importance of geographic areas within the Crescent drainage to coho for restoration or protection benefits reflects a wide range of environmental alterations across the basin (Figure 4-56). The greatest benefits through restoration measures would potentially be achieved by restoring reaches just downstream of the Crescent Valley road to Crescent Lake. The driveway culvert separating the Upper Crescent Creek areas did not rank high because it is a thought to be a barrier to upstream juvenile migration—we modeled juvenile movement predominately downstream. See Table 4-5 for a description of geographic areas.

Crescent Coho Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 9D Puget Sound 7C South Puget Sound 8D Crescent Estuary 6C Lower Crescent Cr 5C Mid Crescent Cr 3B Crescent Cr upstream Crescent Valley Rd 1A Up Crescent (downstream driveway culvert) 2A 136th St driveway culvert - Crescent Creek 10 D Upper Crescent Creek to Lake 3B

Crescent Coho Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA South Puget Sound NA Crescent Estuary NA Lower Crescent Cr 5B Mid Crescent Cr 1A Crescent Cr upstream Crescent Valley Rd 3A Up Crescent (downstream driveway culvert) 2A 136th St driveway culvert - Crescent Creek NA Upper Crescent Creek to Lake 3A

Figure 4-56. Relative importance of geographic areas for restoration and protection measures for Crescent coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-84 Pierce County Watershed Analysis Section 4

The highest benefits associated with habitat protection (i.e., preserving or maintaining what currently exists) are associated with the Upper Crescent geographic areas. These areas have relatively intact riparian condition, and instream structure. Lower benefit potential associated with restoration of the estuary is due to the short exposures of this species to this area compared to the amount of time spent in other areas.

The principal attribute classes or factors that rank highest for coho restoration benefit are generally sediment load, channel (or substrate) stability, habitat diversity, and key habitat (e.g. pools, backwater pools, and off-channel habitat). Figure 4-57 summarizes strategic priorities for formulating restoration focused measures in the basin if coho is the focus of actions.

Crescent Coho Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority ch) y it hat y1/ LOCATION her sp) / y poaching

(miles from mouth) / quant abilit t ure load a als (shaded rows characterize tributaries or ion (w ion (ot it it diversit ion relevant sites) t a hogens hdraw w mpet mpet rassment mperat xygen hannel st hemicals o o abit a e ood Benefit category C C C C Flo F H H Obstructions O Pat Predat Sediment T Wit Key habit 3.57 Crescent Lake Outlet

3.26 136th St Driveway Culvert 2.88

2.57

2.26

1.95

1.64

1.33 Crescent Creek

1.02 Crescent Valley Rd

0.71

0.4 End of tidal influence

0.0 Crescent Creek Estuary Gig Harbor

Figure 4-57. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Crescent watershed for coho salmon.

June 2001 Mobrand Biometrics, Inc. Page 4-85 Pierce County Watershed Analysis Section 4

Donkey Creek: All geographic areas within Donkey Creek resulted in nearly the same restoration benefits (Figure 4-58). They all included removal of the assumed passage barrier at Harborview Drive. Conversely, modeling protection priorities had no response because of the barrier.

Donkey Coho Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 4B Puget Sound 4B South Puget Sound 4B Gig Harbor Bay 4B Lower Donkey Cr 4B Mid Donkey Cr 1A Upper Donkey Cr 3A Donkey Cr Headwaters 2A

Donkey Coho

Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA South Puget Sound NA Gig Harbor Bay NA Lower Donkey Cr D Mid Donkey Cr D Upper Donkey Cr D Donkey Cr Headwaters D

Figure 4-58. Relative importance of geographic areas for restoration and protection measures for Donkey coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

The principal attribute classes or factors that rank highest for coho restoration benefit are obstructions (Harborview, a private driveway, and the 96th Street culvert), sediment load, and habitat diversity (Figure 4-59). Recall, channel stability within the estuary refers to channel landscape, which represents the presence of the estuarine zones. This attribute factor ranks high for the Donkey Creek estuary. The estuary has been almost entirely confined to a culvert.

Burley Creek: Nearly all mainstem Burley Creek geographic areas ranked high for restoration activities. Upper Burley Creek (upstream of CM 2.5 to Holman Road) ranked high for restoration activities, as this was the agricultural area (Figure 4-60). Note that the Lower – Mid Burley Creek area ranked high because of its larger size, and it also included a portion of the agricultural area. The unnamed tributaries 15.0058 and 15.0059 ranked high for improving population life history diversity but contributed little to increasing capacity. Restoration of small tributaries often results in a large improvement to population diversity

June 2001 Mobrand Biometrics, Inc. Page 4-86 Pierce County Watershed Analysis Section 4

Donkey Coho Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority

LOCATION gory e

(miles from mouth) t a

(shaded rows characterize tributaries or c t i relevant sites) f ne thdrawals i Be Channel stability1/ Chemicals Competition (w/ hatch) Competition (other sp) Flow Food Habitat diversity Harassment/poaching Obstructions Oxygen Pathogens Predation Sediment load Temperature W Key habitat quantity Donkey Creek Headwaters 0.9 96th Street Culvert 0.8

0.7 Driveway Culvert

0.45

er Creek 0.35 Mint 0.25 Harborview Drive Culvert 0.2

0.1 End of tidal influence

Gig Harbor

Figure 4-59. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Crescent watershed for coho salmon.

but contributes only a small amount to abundance. It does not diminish the importance of these streams. Life history diversity is included as a performance measure because we think it is important to the long-term sustainability of the population. High life history diversity signifies a population able to use all areas of its historic geographic range and temporal life history pathways.

The highest benefits associated with habitat protection are associated with the area upstream of CM 0.75 to CM 2.5 (i.e., Lower – Mid Burley Creek). The area immediately upstream had little protection benefit.. Little Bear Creek had a high protection benefit for maintaining population life history diversity.

The principal attribute classes or factors that rank highest for coho restoration benefit are sediment load, flow, and habitat diversity (Figure 4-61). Two obstructions to migration are seen – tributary 15.0059 and tributary 15.0058. The obstruction in tributary 15.0058 is SR 16 and the obstruction in tributary 15.0059 is SR 16 and an instream pond (WCC 2001). Burley Road and Burley-Ollala Road are barriers to upstream juvenile migration and were not a priority if coho are the focus of actions.

June 2001 Mobrand Biometrics, Inc. Page 4-87 Pierce County Watershed Analysis Section 4

Burley Coho Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 11 D Puget Sound 8 C South Puget Sound 10 D Burley Estuary/Lagoon 5 B Lower Burley Creek 3 A Lower-Mid Burley Creek 1 A Upper-Mid Burley Creek 2 A Upper Burley Creek 4 B Little Bear Creek 6 B Unnamed Tributary 15.0058 8 C Unnamed Tributary 15.0059 7 C Burley Coho Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA South Puget Sound NA Burley Estuary/Lagoon NA Lower Burley Creek 2 A Lower-Mid Burley Creek 1 A Upper-Mid Burley Creek 5 B Upper Burley Creek 4 B Little Bear Creek 2 A Unnamed Tributary 15.0058 6 B Unnamed Tributary 15.0059 7 C

Figure 4-60. Relative importance of geographic areas for restoration and protection measures for Burley coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

Minter Creek: The greatest benefits through restoration measures would potentially be achieved by re-establishing access of adult coho to the upper reaches of Minter Creek (i.e., Upper Minter Creek above Pine Road) (Figure 4-62). Lower and mid Minter Creek geographic areas rank 2 and 3. However, potential benefits were nearly similar for all other areas. Little Minter Creek was high for potential benefits to improving life history diversity. This was based on eliminating the adult passage barrier at 118th Street. The highest benefits associated with habitat protection are associated with the area upstream of 118th Street and downstream of Pine Road and all of Huge Creek.

The principal attribute classes or factors that rank highest for coho restoration benefit are obstructions, and habitat diversity (Figure 4-63). Most reaches had low to medium attribute priority. The exception was the section of Minter Creek that flowed alongside 118th Street. Restoring habitat diversity is a high priority for this reach.

June 2001 Mobrand Biometrics, Inc. Page 4-88 Pierce County Watershed Analysis Section 4

Burley Coho Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority ch) y it hat y1/ LOCATION her sp) / y poaching

(miles from mouth) / quant abilit t ure load a als (shaded rows characterize tributaries or ion (w ion (ot it it diversit ion relevant sites) t a hogens hdraw w mpet mpet rassment mperat xygen hannel st hemicals o o abit a e ood Benefit category C C C C Flo F H H Obstructions O Pat Predat Sediment T Wit Key habit Headwaters Burley Creek 4.7 Mullenix Rd Crossing

3.9 Burley Rd Crossing

3.1 In unnamed tributary 15.0059 - Upstream SR 16 - SR 16 Road Crossing - Downstream SR 16 2.8 In unnamed tributary 15.0058 - Upstream SR 16 - SR 16 Road Crossing

Burley Creek - Downstream SR 16 2 Burley-Olalla Rd

In Little Bear Creek - Upper Little Bear Cr - Lower Little Bear Cr

0.0 End of tidal influence Burley Creek Estuary Burley Creek mouth

Figure 4-61. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Burley watershed for coho salmon.

Rocky Creek: The greatest benefits through restoration measures would potentially be achieved by re-establishing access of adult coho to the upper reaches of Rocky Creek (i.e., Rocky Creek above 144th Street) (Figure 4-64). Fork Muck Creek above the Wright-Bliss Road also ranked higher for restoration priority. The highest benefits associated with habitat protection are associated with the Fork Muck Creek and lower Rocky Creek. Coho access must first be re-established into the upper basin, then this area also would presumably rank high for protection.

The principal attribute classes or factors that rank highest for coho restoration benefit are obstructions, sediment load, and habitat diversity (Figure 4-65).

June 2001 Mobrand Biometrics, Inc. Page 4-89 Pierce County Watershed Analysis Section 4

Minter Coho Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 12 D Pu get Sound 9C South Puget Sound 11 D Minter Estuary 5B Mint er Hatchery Dam 6B Lower Minter Creek 2A Mid Minter Creek 3A Up per Minter Creek below Pine Rd 4B Upper Minter Creek above Pine Rd 1A Lower Huge Creek 8C Upper Huge Creek 9C Litt le Minter Creek 6B

Minter Coho Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Ma rine outside Puget Sound NA Puget Sound NA South Puget Sound NA Mint er Estuary NA Minter Hatchery Dam NA Lower Minter Creek 4A Mid Minter Creek 5A Up per Minter Creek below Pine Rd 1A Up per Minter Creek above Pine Rd 6D Lower Huge Creek 2A Upper Huge Creek 2A Litt le Minter Creek 6D

Figure 4-62. Relative importance of geographic areas for restoration and protection measures for Minter coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-90 Pierce County Watershed Analysis Section 4

Minter Coho Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority ch) y it hat y1/ LOCATION her sp) / y poaching

(miles from mouth) / quant abilit t ure load a als (shaded rows characterize tributaries or ion (w ion (ot it it diversit ion relevant sites) t a hogens hdraw w mpet mpet rassment mperat xygen hannel st hemicals o o abit a e ood Benefit category C C C C Flo F H H Obstructions O Pat Predat Sediment T Wit Key habit 5.2 Minter Creek Headwaters

4.4 Pine Rd culvert 4

2.8 Upper most 118th St Culvert 2.4

In Huge Creek - Headwaters - to County line nter Creek

i - Huge Cr to gage station

M 1.6 In Little Minter Creek - Upstream of culvert - 118th St culvert 1.2

Hatchery Dam 0.4 End of tidal influence Minter Creek Estuary Henderson Bay

Figure 4-63. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Minter watershed for coho salmon.

June 2001 Mobrand Biometrics, Inc. Page 4-91 Pierce County Watershed Analysis Section 4

Rocky Coho Relative Importance Of Geographic Areas For Restoration Measures

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound 10 C Puget Sound 5B South Puget Sound 9C Rocky Estuary 8C Lower Rocky Cr 4B Mid Rocky Cr 1A Fern Lake and Dam 7B Rocky Cr above Fern Lk 2A Lower Fork Muck Cr 5B Upper Fork Muck Cr 2A

Rocky Coho Relative Importance Of Geographic Areas For Protection Measures ("NA" indicates that no analysis was done for the area)

Combined Benefit Geographic Area Capacity Productivity Diversity Index rank category Marine outside Puget Sound NA Puget Sound NA South Puget Sound NA Rocky Estuary NA Lower Rocky Cr 2A Mid Rocky Cr 4D Fern Lake and Dam 4D Rocky Cr above Fern Lk 4D Lower Fork Muck Cr 2A Upper Fork Muck Cr 1A

Figure 4-64. Relative importance of geographic areas for restoration and protection measures for Rocky coho salmon. Areas are ranked and assigned to benefit categories according to potential (A is highest) to affect population performance. Contribution of performance measures to rankings are graphed.

June 2001 Mobrand Biometrics, Inc. Page 4-92 Pierce County Watershed Analysis Section 4

Rocky Coho Restoration Strategic Priority Summary

Reach Priority Attribute Class Priority ch) y it hat y1/ LOCATION her sp) / y poaching

(miles from mouth) / quant abilit t ure load a als (shaded rows characterize tributaries or ion (w ion (ot it it diversit ion relevant sites) t a hogens hdraw w mpet mpet rassment mperat xygen hannel st hemicals o o abit a e ood Benefit category C C C C Flo F H H Obstructions O Pat Predat Sediment T Wit Key habit 4.4 Rocky Creek Headwaters

3.9

3.4 Fern Lake Dam

2.9

2.4

1.9 144th Street Culvert 1.65 nter Creek i

M In Fork Muck Creek - Upstream Wright-Bliss Rd - below Wright-Bliss Rd 1

0.5 End of tidal influence

Rocky Creek Estuary

Rocky Bay

Figure 4-65. Summary of restoration strategic priorities for environmental factors (attribute classes) corresponding to geographic areas within the Rocky watershed for coho salmon.

June 2001 Mobrand Biometrics, Inc. Page 4-93 Pierce County Watershed Analysis Section 4

4.5.3 Inference for Bull Trout for Kitsap Watersheds Native char species (bull trout and Dolly Varden) have not been found in Kitsap watersheds (WCC 2001). In Western Washington, these species are typically associated with larger systems than Kitsap streams. Spawning in those river systems occurs in colder and higher gradient streams than those in WRIA 15. We conclude that no consideration should be given to char species in identifying strategic priorities for salmonid species in the Kitsap Watersheds.

4.5.4 Data/Information Uncertainties for Kitsap Watersheds The data and information used to characterize the environment were brought into the analysis through two processes, one involving all stream reaches upstream of tidewater and one that addressed the estuary and marine areas. Each process and the information used have a different level of uncertainty.

The characterization of the Kitsap watersheds was assembled by Greg Blair and Larry Lestelle (Mobrand Biometrics, Inc.) in consultation with Pierce County resource specialists. The Crescent and Donkey creek assessments relied heavily on habitat assessments completed for the Gig Harbor Basin (Pierce County Water Programs 2000). Other source documents used in the characterization were the WRIA 15 Habitat Limiting Factors Analysis report (WCC 2001)and the Kitsap Peninsula Salmonid Refugia Study (CTC 2000). Finally, Greg Blair visited each stream at least once (several more than once) to visually observed key features of the watersheds. However, streams were not intensively surveyed for this assessment.

The procedure required assigning a "level of proof" to each attribute rating for each reach, using a scale of 1-4, where a value of 1 meant that empirical data were used and a high level of confidence was placed in the rating, and a 4 represented an educated guess with low confidence. Values of 2 and 3 were intermediate, where a 2 represented a relatively high level of confidence, based on a combination of personal observation and "weight of evidence," and a 3 drew on a theoretical application.

The large majority of ratings applied to the Kitsap streams were assigned a level of proof of 2-3. Relatively few ratings were assigned values of 1or 4. The large majority of ratings used to characterize historic conditions were assigned a level of proof of 3.

Ratings for Henderson Bay and Puget Sound proper were obtained through another project that addressed estuaries and bays in Puget Sound. We supplemented this information with ratings for the various estuaries and smaller bays in WRIA 15. All of these estuaries are small and do not have the well-developed estuarine zones found in larger, pristine estuaries in Puget Sound. We assumed that the nearshore environment within the bays has a more important role during early marine rearing of juvenile salmon than in larger estuaries. There is a high level of uncertainty about the overall contribution of the estuaries and bays to population performance of Kitsap salmon. Until further information becomes available, however, we believe that the basic conclusions of the assessment are reasonable with regard to the relative contributions of estuarine/bay areas and freshwater reaches to population performance.

June 2001 Mobrand Biometrics, Inc. Page 4-94 Pierce County Watershed Analysis Section 4

Finally, attributes in the freshwater environment that should be field verified are identified for each watershed. Other attributes should also be field verified, but we suggest that this should be part of an on-going monitoring plan associated with action effectiveness monitoring.

Crescent Creek: Based on the conclusions of Crescent Creek recommended field the assessment, we identify three attributes in verification of: particular that should be field verified as fine sediment within riffles opportunity occurs: bed (substrate) scour, fine bed scour sediment, water quality and quantity, and the water quality extent of spring sources. In addition, a stream distribution of springs and upwelling flow gauging station has been recently established, which will be useful in better understanding flow and water quality characteristics in the watershed. This assessment did not have a long-term continuous set of records on streamflow available.

Donkey Creek: We assumed the water supply Donkey Creek recommended field structure for chum egg incubation was a total verification of: barrier to November-December migrating Adult access upstream of Harborview coho. Someone knowledgeable about the Drive Culvert chum project should verify this assumption.

Burley Creek: Based on the conclusions of the Burley Creek recommended field assessment, we identified several verification of: attributes/geographic areas in particular that Habitat quality features should be field verified as opportunity occurs: fine sediment bed (substrate) scour and fine sediment, and bed scour habitat diversity features (quantity of wood, riparian function, and channel hydromodifications) for reaches upstream of CM 2.0.

Minter Creek: Based on the conclusions of Minter Creek recommended field the assessment, we identified several verification of: attributes/geographic areas in particular that Inventory marsh/pond habitat should be field verified as opportunity occurs: extent spring sources extent of wetland/pond habitat along the coho access above Pine Road mainstem Minter Creek upstream of 118th water quality and quantity in Little Minter Street and in Huge Creek, extent of spring Creek sources in Huge and Minter creeks, adult coho access upstream of Pine Road, and water quality and quantity in Little Minter Creek. Our assessment of Minter Creek was particularly sensitive to the assumption that there is extensive pond habitat available for juvenile coho rearing. We were unable to find any reference to these features other than from topographic maps and a short description in Salo and Bayliff (1958).

June 2001 Mobrand Biometrics, Inc. Page 4-95 Pierce County Watershed Analysis Section 4

Rocky Creek: Based on the conclusions of the Rocky Creek recommended field assessment, we identify several verification of: attributes/geographic areas in particular that Inventory fish passage barrier at 144th should be field verified as opportunity occurs: fine sediment within riffles adult coho access upstream of 144th Street, bed scour extent of spring sources, fine sediment, and extent spring sources bed (substrate) scour. Lower Rocky Creek habitat features in Fork Muck Creek above Wrigh-Bliss Road appeared to be particularly sensitive to increased bed load (scour and fill during storm events). The upper basin appeared to be more sensitive to fine sediment inputs. Overall conditions of riparian cover, habitat types, water quality and quantity, and instream wood should be verified within Fork Muck Creek.

4.5.5 Kitsap Watersheds Conclusions The Kitsap watersheds included in this assessment are diverse. All of this area was logged at the turn of the century. Crescent, Donkey, and Burley watersheds have experienced the greatest change due to rural/urban development. Yet even these watersheds are relatively intact, compared to streams in WRIA 10 and 12. Rocky Creek was the most intact of the streams. Changes that have occurred in this watershed are largely due to logging practices.

These watersheds are currently marginal for a self-sustaining chinook population. We question whether chinook populations were ever historically sustainable over the long-term due to the small size of these watersheds. We concluded that these watersheds were historically important to South Puget Sound coho production, and coho populations are still present in most systems, although abundance is much less than it was historically. We find that this species would make an excellent indicator species for formulating watershed action plans to address salmonid conservation and recovery needs.

Conservation and restoration measures can be developed following a set of strategic priorities for geographic areas within the basin. These priorities identify the strategic importance of different areas in the watershed for either restoring (including only partial recovery) or protecting conditions for salmonid performance (Figure 4-66).

June 2001 Mobrand Biometrics, Inc. Page 4-96 Pierce County Watershed Analysis Section 4

Crescent Basin Overview Donkey Basin Overview Strategic Assessment of Geographic Areas Strategic Assessment of Geographic Areas Chinook Coho Chinook Coho strategic strategic strategic strategic priority priority priority 1/ priority

Geographic Area Geographic Area otection otection otection otection r r r r Restoration P Restoration P

Restoration P Restoration P Crescent Estuary Gig Harbor Lower Crescent Cr Lower Donkey Creek Mid Crescent Cr Mid Donkey Creek Crescent Cr abv Crescent Valley Rd Upper Donkey Creek Up Crescent (downstream driveway culvert) Donkey Creek Headwaters 136th St driveway culvert - Crescent Creek Upper Crescent Creek to Lake 1/ Chinook were not modeled in Donkey Creek

Burley Basin Overview Minter Basin Overview Strategic Assessment of Geographic Areas Strategic Assessment of Geographic Areas

Chinook Coho Chinook Coho strategic strategic strategic strategic priority priority priority priority

Geographic Area Geographic Area ation ation Restoration Protection Restoration Protection otection otection

r r Minter Estuary

Restor P Restor P Minter Hatchery Dam Burley Est/Lagoon Lower Minter Creek Lower Burley Cr Mid Minter Creek Lower-Mid Burley Cr Upper Minter Creek below Pine Rd Upper-Mid Burley Cr Upper Minter Creek above Pine Rd Upper Burley Cr Lower Huge Creek Upper Huge Creek Little Bear Cr Little Minter Creek Unnamed Trib 0058 Unnamed Trib 0059

Rocky Basin Overview Strategic Assessment of Geographic Areas

Chinook Coho strategic strategic priority priority

Geographic Area otection otection r r Restoration P Restoration P Rocky Estuary Lower Rocky Cr Mid Rocky Cr Fern Lake and Dam Rocky Cr above Fern Lk Lower Fork Muck Cr Upper Fork Muck Cr

Key to Strategic Priority (Benefit Category letter shown) D & E C B A Indirect or General Low Medium High

Figure 4-66. Overview of strategic priorities for restoration and protection measures by geographic area within Kitsap watersheds.

June 2001 Mobrand Biometrics, Inc. Page 4-97

Pierce County Watershed Analysis Section 5

5.0 ANALYSIS OF ACTIONS

The purpose of this step in the analysis is to identify candidate actions and analyze them for their potential benefit to the fish populations of interest. These candidate actions should be considered as a starting point for an adaptive planning process where routine updates of the diagnosis and associated watershed goals may lead to additions and/or modifications of actions. As new information regarding action effectiveness becomes available, attention can be refocused on the most successful methods for achieving watershed goals.

We identified candidate actions in the two ways, described in Section 2 and Appendix E. We solicited candidate actions from entities and agencies working in the basins, and we added to these solicited actions others that we considered relevant based on the results of the diagnosis.

We analyzed groups of actions sequentially with the second group added to the first, and so on. This grouping of actions was considered a useful way to incorporate and compare a large number of actions. The procedure for grouping actions is described in detail in Appendix C. In brief, the results of the strategic assessment (Section 4) were used to assign each action a priority level by applying a set of salmonid conservation and restoration principles described in Lestelle et al. (1996). The principles were developed to be used within a context of depressed or listed populations. These principles, adapted here to the Pierce County application, serve as a guideline for prioritizing, or sequencing, actions. The principles, in order of priority, are as follows: 13. Maintain (or protect) habitat quality and quantity that support the existing core (or most productive) life history patterns of the population; 14. Improve (or restore—even partly) habitat quality and quantity that support the existing core (or most productive) life history patterns of the population; 15. Maintain or restore habitat quality and quantity that support secondary life history patterns (less productive population components) of the population; and 16. Improve habitat quality, or reconnect habitat segments as needed, to recover lost life history patterns that existed in the historic population.33

Reconnection of habitat segments—for example, by removal of barriers—would be given a higher priority (as high as number 1) if conditions upstream of the barrier could be immediately colonized and would serve to support a core or secondary group of life history patterns. Removal of the Elwha dams, for example, would be assigned the highest priority because the dams would open up the core habitat to be used by the existing population.34 These principles provided the basis for organizing the actions into four Action Scenarios corresponding to each priority level. We learned, in the process of identifying candidate

33 Principles 4 and 5 here are condensed from those described in Lestelle et al. (1996), where each of these was described by two separate principles. In total, six priorities were presented in that document. 34 It is important to note that the principles are meant only to serve as a guideline in watershed planning. For a successful recovery program, actions across all four levels may be required. Also, opportunities for actions at lower priority levels may present themselves—these should not be ignored if attention at that point in time is being directed mainly toward a higher priority level.

June 2001 Mobrand Biometrics, Inc. Page 5-1 Pierce County Watershed Analysis Section 5 actions, that many actions are already in various stages of implementation. We incorporated these actions into a future scenario represented by existing or status quo management (Future Status Quo Scenario). This characterization became a third baseline reference condition for comparing the effect of new actions (in addition to the current and historic conditions). We included in this characterization projected patterns of human population growth in the watersheds and associated effects on environmental conditions. A technical work team consisting of planners in the region provided likely growth patterns and likely changes in the percent of impervious surfaces in the watersheds. We used this information to project a set of future Level 2 attributes, from which we modeled species performance responses.

We defined effectiveness of actions relative to the difference between current and historic conditions. For example, an effectiveness of 10% for a specific attribute would mean that the objective of the action would be to recover 10% of the difference between the historic and existing conditions. A procedure used by the Columbia River Multi-Species Framework Project (NWPPC in press) was used to assess cumulative effects of multiple actions affecting the same attribute.

Assumptions about action effectiveness for the Pierce County project were made with the aid of a working group of civil engineers and biologists (Appendix C). These assumptions become objectives for the actions that can, if implemented, be monitored for effectiveness. The working group identified the length of time it would take to achieve full effect, ranging from 0-2 years to over 50 years. We analyzed benefits for a period 20-25 years from implementation.

The projections of benefits presented here provide a basis for comparing groups of actions and evaluating the amount of projected change in population performance under the various sets of assumptions. They represent hypotheses about population response to assist in developing and implementing action plans.

Due to overlapping, cumulative effects of actions on populations originating in the White and Puyallup river systems, we present the results for these areas combined under the heading "Puyallup-White Basin." Projections of population response are shown separately for each population.

5.1 Hylebos Basin 5.1.1 Identification and Ordering of Actions for Hylebos Basin A total of 37 actions were assembled and analyzed for Hylebos salmon (Table 5-1). Of these, 13 are in the process of being implemented. The Status Quo Future Scenario reflects the implementation of these 13, in addition to some expected alterations in environmental conditions due to continued population growth. Six of these 13 are being implemented in the Hylebos estuary (including the waterway) or along the Commencement Bay shoreline. Two of the 13 are obstruction removals, most notably the improvement of a problem

June 2001 Mobrand Biometrics, Inc. Page 5-2 Pierce County Watershed Analysis Section 5

Table 5-1. Actions grouped by Action Scenario analyzed for Hylebos salmon performance. Action Scenarios are Status Quo Future (SQ) and four scenarios (1 to 4) with new actions. Actions are organized based on application of strategic priorities described in the text. See Appendix E for action descriptions. The Action Code box is shaded for actions applied to the estuary or Commencement Bay.

Action code Action name Action scenario City of Federal Way Environmentally Sensitive Areas Pierce-5 SQ Stream Code Amendments Adoption of the 1998 King County Surface Water Pierce-6 SQ Design Manual (KCSWDM) Pierce-8 Belmore Detention Pond SQ

Pierce-9 Kitts Corner Regional Detention Facility SQ

Pierce-10 So. 364th Street Culvert Replacement SQ

Pierce-12 S. 356th Street Regional Detention Pond SQ

Pierce-158 Culvert replacement on West Hylebos at Highway 99 SQ

Pierce-148 Tahoma salt marsh restoration SQ

Pierce-152 Protection of Olympic View resource area SQ

Pierce-154 Mowitch Restoration Project SQ

Pierce-155 Squally Beach Restoration Project SQ

Pierce-156 Yowkwala Restoration Project SQ

Pierce-157 Skookum Wulge Beach Habitat Restoration Project SQ City of Federal Way Shoreline Management Pierce-4 1 Amendments Pierce-13 High Efficiency Street Sweeper Acquisition 1

Pierce-64 West Hylebos Creek Habitat Acquisition 1 Property tax reductions for landowners protecting Pierce-65 1 sensitive areas Pierce-67 Public education 1

Pierce-71 Hylebos Stream Team/watershed steward 1

Pierce-11 S. 360th Regional Stormwater Detention Facility 2

Pierce-53 Stream nutrient project 2

Pierce-55 Gravel restoration - Brooklake 2

Pierce-56 Gravel Restoration – Spring Valley 2

June 2001 Mobrand Biometrics, Inc. Page 5-3 Pierce County Watershed Analysis Section 5

Action code Action name Action scenario Restore stream connections to small ponds in Spring Pierce-57 2 Valley area Pierce-58 Re-connect Justus pond to Hylebos Creek 2 Adjust Federal Way regional R/D ponds to Pierce-59 2 supplement base flows Pierce-66 Impervious surface reduction 2

Pierce-68 Establish 200-foot no-impact buffer zones 2

Pierce-69 Low-density zoning near prime habitat 2

Pierce-18 East Branch Hylebos Conifer Underplanting 3

Pierce-19 East Branch Hylebos LWD Enhancement 3 Invasive weed control/native planting at Hylebos Pierce-54 3 confluence Pierce-70 Scientific Study 3

Pierce-60 Redirect creek from between I-5 & Pac Hwy S. 4

Pierce-61 Surprise Lake Channel Restoration 4

Pierce-63 SR 167/Fife habitat restoration 4

Pierce-139 Intensive Restoration of Lower Hylebos Below Forks 4

culvert on Highway 99. The other 24 actions were organized into the four Action Scenarios for the analysis.

5.1.2 Projections of Benefits for Hylebos Basin Benefits to salmon performance were projected to be substantial for both chinook and coho when all actions are implemented (Figures 5-1 and 5-2). This is striking when benefits are expressed as a percent improvement over current conditions. As noted earlier, however, we question whether the responses exhibited by chinook could be realized given preference of this species for larger basins. Still, the results indicate that the actions would improve performance potential for chinook.

Coho appear to be highly responsive to the actions analyzed. Notably, we project substantial improvements in performance under the Status Quo Scenario due to the effect of the 13 actions scheduled for implementation. Passage improvements are a significant reason for this response. It should be noted that productivity responses do increase as new groups of actions are added, while total abundance does. This is due to recovery of lost life histories-- fish are added to the population but with less productive life histories.

June 2001 Mobrand Biometrics, Inc. Page 5-4 Pierce County Watershed Analysis Section 5

Hylebos Fall Chinook Summary Of Projected Performance Measures Under Status Quo and Expanded Action Scenarios

Scenario Abundance Productivity Diversity index Historic 477 12.4 100% Current with harvest and fitness loss 7 1.2 4% Future with status quo mgmt (SQ) 17 2.3 25% Action Scenario-1 (S-1) 21 2.5 32% Action Scenario-2 (S-2) 25 2.7 36% Action Scenario-3 (S-3) 33 2.6 50% Action Scenario-4 (S-4) 59 2.8 67%

Chinook spawner abundance 200 477

sh 150 fi f 100 er o

mb 50 Nu

0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Chinook productivity 15 r e

awn 10 r sp e

s p 5 rn

Retu 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Chinook life history diversity 100%

75%

50% rcent e P 25%

0% Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Figure 5-1. Hylebos chinook (naturally produced) performance measures under historic, current, and five action scenarios.

June 2001 Mobrand Biometrics, Inc. Page 5-5 Pierce County Watershed Analysis Section 5

Hylebos Coho Summary Of Projected Performance Measures Under Status Quo and Expanded Action Scenarios

Scenario Abundance Productivity Diversity index Historic 1609 24.0 100% Current with harvest and fitness loss 92 4.3 58% Future with status quo mgmt (SQ) 196 8.6 72% Action Scenario-1 (S-1) 264 10.2 81% Action Scenario-2 (S-2) 369 12.6 83% Action Scenario-3 (S-3) 442 11.8 86% Action Scenario-4 (S-4) 517 11.2 97%

Coho spawner abundance 800 1609 600

400

200 Number of fish

0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho productivity 25

Returns per spawner 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho life history diversity 100%

75%

50% Percent 25%

0% Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Figure 5-2. Hylebos coho (naturally produced) performance measures under historic, current, and five action scenarios.

June 2001 Mobrand Biometrics, Inc. Page 5-6 Pierce County Watershed Analysis Section 5

5.1.3 Summary for Hylebos Basin A graphic comparison of all actions related to the Hylebos Basin is seen in Figure 5-3. The display compares actions in terms of effects on attributes, extent of effect dispersal, technical feasibility, likelihood of outcome, community support, and cost. Details associated with the latter four items are provided in Appendix E.

Figure 5-3 Comparison of action components for the Hylebos Basin.

June 2001 Mobrand Biometrics, Inc. Page 5-7 Pierce County Watershed Analysis Section 5

5.2 Puyallup-White Basins 5.2.1 Identification and Ordering of Actions for Puyallup-White Basins A total of 61 actions were assembled and analyzed for salmon populations in the Puyallup and White basins (Table 5-2). Fifty of these actions applied to White River salmon populations, while 40 applied to Puyallup salmon. Twenty-four of the total actions are funded and in the process of being implemented. The Status Quo Future Scenario reflects the implementation of these 24 funded actions as well as expected alterations in environmental conditions due to continued population growth and the recent installation of fish passage facilities at Electron Dam on the Puyallup River in 2000.

Eight of the 24 actions now being implemented are in the Puyallup estuary or along the Commencement Bay shoreline. A total of ten actions were analyzed for the estuary and bay (two proposed, in addition to the eight being implemented). Table 5-2 identifies which actions are applied to the estuary and bay.

5.2.2 Projections of Benefits for Puyallup-White Basins Benefits projected for the actions generally increase incrementally with the addition of each action set to the modeled populations(Figures 5-4 to 5-7). For each population in each river, we projected substantial increases in abundance under the Status Quo Future Scenario— indicating that actions already, or soon to be, implemented could more than offset any detrimental effects of new developments in these basins over the next approximately 25 years. These patterns seem to be contrary to the generally prevailing notion in the region— that fish populations in developed watersheds will continue to decline.

There is clearly large uncertainty associated with these estimates. The projections are based on three important sets of assumptions: 1) that actions would be effective to the extent assumed, 2) that future development of the watersheds would have the assumed level of detrimental effects on the environment, and 3) that other conditions associated with salmon survival, such as climate and ocean regimes, would not worsen. We suggest, however, that the projections provide an initial set of hypotheses for proceeding with salmon conservation and recovery measures in the basins. We further note that the actions are not small and inexpensive—they represent a substantial commitment on the part of agencies and organizations, as well as by the citizens of Pierce County in general.

5.2.3 Summary for Puyallup-White Basins A graphic comparison of all actions related to the Puyallup-White Basins is presented in Figure 5-8. The display compares actions in terms of effects on attributes, extent of dispersal of effect, technical feasibility, likelihood of outcome, community support, and cost. Details associated with the latter four items are provided in Appendix E.

June 2001 Mobrand Biometrics, Inc. Page 5-8 Pierce County Watershed Analysis Section 5

Table 5-2. Actions grouped by Action Scenario analyzed for Puyallup and White salmon performance. Action Scenarios are Status Quo Future (SQ), and four scenarios (1 to 4) with new actions. Actions are organized based on application of strategic priorities described in text. See Appendix E for action descriptions. The Action Code box is shaded for actions applied to the estuary or Commencement Bay.

Action code Action name Action scenario

Pierce-35 4 Culverts – Foothills Trail SQ

Pierce-37 Butte Pit Wetland Mitigation Site SQ

Pierce-39 Foothills Trail Vegetation Installation SQ

Pierce-44 Puyallup River Setback Levee SQ

Pierce-45 Puyallup River Setback Levee SQ

Pierce-46 Puyallup River Setback Levee SQ

Pierce-47 SWM Fee Credit Program SQ

Pierce-48 Wetland Mitigation Banking SQ

Pierce-49 Stormwater Regulations SQ

Pierce-50 Puyallup River Setback Levee SQ

Pierce-51 Salmon Recovery Land Acquisition SQ

Pierce-67 Public education SQ

Pierce-71 Stream Team/watershed steward SQ

Pierce-72 Forest Practices SQ

Pierce-76 NW Forest Plan SQ

Pierce-80 White River Pipeline Crossing SQ

Pierce-159 Swan Creek Stream Restoration SQ

Pierce-148 Tahoma salt marsh restoration SQ

Pierce-149 Middle waterway wetland restoration SQ

Pierce-150 Foss shoreline redevelopment SQ

Pierce-151 Mitigation for St. Paul waterway fill SQ

Pierce-152 Protection of Olympic View resource area SQ

Pierce-156 Yowkwala Restoration Project SQ

Pierce-157 Skookum Wulge Beach Habitat Restoration Project SQ

June 2001 Mobrand Biometrics, Inc. Page 5-9 Pierce County Watershed Analysis Section 5

Action code Action name Action scenario

Pierce-77 White River pH 1

Pierce-146 Added conservation areas 1

Pierce-14 Map and protect Channel Migration 2

Pierce-20 Greenwater and White Confluence 2

Pierce-29 Game Farm Wilderness Park 2

Pierce-34 Trans Canada Levee Removal 2

Pierce-40 Foothills Trail Culvert Replacement 2

Pierce-42 Foothills Trail Wetland Mitigation Site 2

Pierce-73 Lower Carbon River Setback (Project 11 2

Pierce-74 White River bypass flow increase 2

Pierce-75 Oxbow Reconnections 2

Pierce-81 South Fork Road Setback Levee 2

Pierce-82 Old Soldier’s Home Setback levee 2

Pierce-143 Nutrient enhancement 2

Pierce-153 Union Pacific Parcel-RM 2.8 2

Pierce-1 Subbasin A Project Group A-2 3

Pierce-2 Subbasin A Project Group A-3 3

Pierce-3 Subbasin Z Project Group Z-1 3

Pierce-16 Boise Creek LWD Enhancement 3

Pierce-21 Greenwater SR 410 Bridge Replacement 3

Pierce-22 Lower Boise Creek Relocation and 3

Pierce-24 Upper Boise Creek Revegetation 3

Pierce-25 Floodplain Acquisition with Levee Breaching 3

Pierce-27 Floodplain Acquisition Near Stuck River 3

Pierce-28 White River Estates Riparian Buffer 3

Pierce-30 Relocation of Mobile Home Park and 3

Pierce-31 Pacific Park Buffer Restoration 3

June 2001 Mobrand Biometrics, Inc. Page 5-10 Pierce County Watershed Analysis Section 5

Action code Action name Action scenario

Pierce-32 Roegner Park Riverbank Restoration 3

Pierce-33 Riparian Buffer Restoration (Segale) 3

Pierce-78 White River Hydroelectric Facility 3

Pierce-79 Estuary Marsh Corridor Acquisition 3

Pierce-141 Off-channel habitat creation/enhancement 3

Pierce-142 LWD enhancement 3

Pierce-144 Alter hydromodifications 3

Pierce-145 Addition of runoff detention/retention system 3

Pierce-23 Red Creek and White River Confluence 4 Blockage (e.g., dam or culvert) removal or passage Pierce-140 provision 4

June 2001 Mobrand Biometrics, Inc. Page 5-11 Pierce County Watershed Analysis Section 5

Puyallup Chinook Summary Of Projected Performance Measures Under Status Quo and Expanded Action Scenarios

Scenario Abundance Productivity Diversity index Historic 31,618 14.1 100% Current with harvest and fitness loss 1,166 2.9 19% Future with status quo mgmt (SQ) 1,951 3.4 29% Action Scenario-1 (S-1) 1,921 3.4 30% Action Scenario-2 (S-2) 2,813 3.7 33% Action Scenario-3 (S-3) 3,280 3.9 35% Action Scenario-4 (S-4) 3,345 3.9 38%

Chinook spawner abundance 10,000 31,618 8,000 sh 6,000

4,000

Number of fi 2,000

0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Chinook productivity 15

Returns per spawner 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Chinook life history diversity 100%

75%

50% Percent 25%

0% Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Figure 5-4. Puyallup chinook (naturally produced) performance measures under historic, current, and five action scenarios.

June 2001 Mobrand Biometrics, Inc. Page 5-12 Pierce County Watershed Analysis Section 5

White Chinook Summary Of Projected Performance Measures Under Status Quo and Expanded Action Scenarios

Scenario Abundance Productivity Diversity index Historic 24,327 8.8 100% Current with harvest and fitness loss 1,069 2.4 20% Future with status quo mgmt (SQ) 1,471 2.7 24% Action Scenario-1 (S-1) 1,507 2.8 24% Action Scenario-2 (S-2) 2,272 2.9 31% Action Scenario-3 (S-3) 2,745 3.2 35% Action Scenario-4 (S-4) 2,753 3.2 35%

Chinook spawner abundance 10,000 24,327 8,000 sh i f 6,000

4,000 mber of u

N 2,000

0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Chinook productivity 15

Returns per spawner 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Chinook life history diversity 100%

75%

50% Percent 25%

0% Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Figure 5-5. White chinook (naturally produced) performance measures under historic, current, and five action scenarios.

June 2001 Mobrand Biometrics, Inc. Page 5-13 Pierce County Watershed Analysis Section 5

Puyallup Coho Summary Of Projected Performance Measures Under Status Quo and Expanded Action Scenarios

Scenario Abundance Productivity Diversity index Historic 29,634 21.7 91% Current with harvest and fitness loss 1,092 4.2 23% Future with status quo mgmt (SQ) 2,024 6.3 33% Action Scenario-1 (S-1) 2,173 6.2 34% Action Scenario-2 (S-2) 3,292 6.9 39% Action Scenario-3 (S-3) 3,824 7.4 42% Action Scenario-4 (S-4) 3,843 7.4 43%

Coho spawner abundance 10,000 29,634 8,000 sh 6,000 er of fi

b 4,000

Num 2,000

0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho productivity 20

Returns per spawner 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho life history diversity 100%

75%

50% Percent 25%

0% Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Figure 5-6. Puyallup coho (naturally produced) performance measures under historic, current, and five action scenarios.

June 2001 Mobrand Biometrics, Inc. Page 5-14 Pierce County Watershed Analysis Section 5

White Coho Summary Of Projected Performance Measures Under Status Quo and Expanded Action Scenarios

Scenario Abundance Productivity Diversity index Historic 24,630 19.2 91% Current with harvest and fitness loss 1,902 3.8 32% Future with status quo mgmt (SQ) 2,361 5.1 37% Action Scenario-1 (S-1) 2,436 5.4 37% Action Scenario-2 (S-2) 3,850 6.3 42% Action Scenario-3 (S-3) 4,488 7.2 48% Action Scenario-4 (S-4) 4,655 7.1 50%

Coho spawner abundance 10,000 24,630 8,000 sh i f 6,000

4,000

Number of 2,000

0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho productivity 20 ner 15

rns per spaw 5 u

Ret 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho life history diversity 100%

75%

50% Percent 25%

0% Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Figure 5-7. White coho (naturally produced) performance measures under historic, current, and five action scenarios.

June 2001 Mobrand Biometrics, Inc. Page 5-15 Pierce County Watershed Analysis Section 5

Figure 5-8. Comparison of action components for the Puyallup-White basins.

June 2001 Mobrand Biometrics, Inc. Page 5-16 Pierce County Watershed Analysis Section 5

Puyallup-White Basins Comparison of Actions -- Continued

Space- Feasibility, support, Time Effect on Attribute Class (Factor) cost Effect y r d / 1 ect ect ty f f hoo i i ility s ef ef b r suppo e l kel l a e t y i Action Action t t u v ur a of s ons i t Action abbreviation i f l s Scenario code at o ct e sal cal t d u n r ome l s t ta ment n i c assment a b sper ow h a Di Year C Chemi Fl Food H Har Obst Sedi Temper Key habi Technical feasibilit Out Communi Cost 3 Pierce-1 Project Group A-2 3 Pierce-2 Project Group A-3 3 Pierce-3 Project Group Z-1 3 Pierce-16 Boise Cr LWD 3 Pierce-21 SR 410 Bridge 3 Pierce-22 Boise Cr Relocate 3 Pierce-24 Boise Cr Reveg 3 Pierce-25 Floodplain Acquisition 3 Pierce-27 Floodplain Acquisition 3 Pierce-28 White River Estates 3 Pierce-30 Relocate Mobile Homes 3 Pierce-31 Pacific Park Buffer 3 Pierce-32 Roegner Park Restore 3 Pierce-33 Riparian Restoration 3 Pierce-78 White R Hydroelec 3 Pierce-141 Off-channel habitat 3 Pierce-142 LWD enhancement 3 Pierce-144 Alter hydromodifications 3 Pierce-145 Additional det/retention 3 Pierce-79 Marsh Corridor Acquire 4 Pierce-23 Red Cr Confluence 4 Pierce-140 Blockage removal 1/ "Channel Stability" within estuary refers to "Channel Blank - no effect on attribute; information not available for feasibility, support, cost Landscape", which represents the presence of the estuarine zones. Low effect on attribute; Cost < $100K 2/ The Swan Creek action (Pierce-159) was not Moderate effect on attribute; High uncertainty in feasibility; Conflicts in support; Cost $100K - $700K analyzed due to an oversight. High effect on attribute; Some uncertainty in feasibility; Uncertain support; Cost $700K - $3M Full effect on attribute; High feasibility; High community support; Cost > $3M

Degree of dispersal indicates extent of geographic range of effect in watershed; Years to full effect corresponds to immediate and up to 50 yrs.

Figure 5-8 (continued from previous page).

5.3 Chambers-Clover Basin 5.3.1 Identification and Ordering of Actions for Chambers-Clover Basin A total of 40 actions were assembled and analyzed for Chambers-Clover salmon (Table 5-3). Twenty-eight of these actions, the large majority, are in the process of being implemented. The Status Quo Future Scenario reflects the implementation of these 28 as well as expected alterations in environmental conditions due to continued population growth. No actions were analyzed for Chambers Bay. The twelve actions analyzed for consideration for future implementation were organized into the four Action Scenarios for the analysis.

5.3.2 Projections of Benefits for Chambers-Clover Basin Benefits to salmon performance were projected to be substantial for both chinook and coho when all actions are implemented (Figures 5-9 and 5-10). This is particularly evident on the

June 2001 Mobrand Biometrics, Inc. Page 5-17 Pierce County Watershed Analysis Section 5

Table 5-3. Actions grouped by Action Scenario analyzed for Chambers-Clover salmon performance. Action Scenarios are Status Quo Future (SQ), and four scenarios (1 to 4) with new actions. Actions are organized based on application of strategic priorities described in text. See Appendix E for action descriptions.

Action code Action name Action scenario

Pierce-47 SWM Fee Credit Program SQ

Pierce-48 Wetland Mitigation Banking SQ

Pierce-49 Stormwater Regulations SQ

Pierce-72 Forest Practices SQ

Pierce-101 City Hall Infiltration System SQ

Pierce-102 Hipkins Road Storm Drainage SQ

Pierce-103 Weller Road Storm Drainage SQ

Pierce-104 Steilacoom Blvd/Weller Road SQ

Pierce-105 Gravelly Lake / Nyanza Road Storm SQ

Pierce-107 Clover Creek Fish Ladder SQ

Pierce-108 Flett Creek Fish Passage Barrier SQ

Pierce-109 City of Lakewood Site Development SQ

Pierce-110 City of Lakewood NPDES Phase II SQ

Pierce-111 Broback Parcel Detention Pond SQ

Pierce-113 Amendments to Critical areas SQ

Pierce-119 North Fork Clover Creek “E1” Detention SQ

Pierce-120 North Fork Clover Creek “W1” Detention SQ

Pierce-121 Repetitive Loss Buyout SQ

Pierce-123 136th St. E. – “A” St E. to “B” St. E. SQ

Pierce-129 Spanaway/Morey Cr. Habitat Acquisition SQ

Pierce-130 Clover Creek Habitat Acquisition SQ

Pierce-131 Third detention pond on Clover Creek SQ

Pierce-132 Clover Creek Water Wheel Dam Replacement SQ

Pierce-134 Stream gage monitoring SQ

Pierce-135 Parkland Prairie Land Purchase SQ

June 2001 Mobrand Biometrics, Inc. Page 5-18 Pierce County Watershed Analysis Section 5

Action code Action name Action scenario

Pierce-136 Preserve wildlife habitat SQ

Pierce-137 Planting Native plants along Clover Creek SQ

Pierce-138 Restore native plant species SQ

Pierce-127 Reissue of Pierce County NPDES Phase 1

Pierce-133 Remove concrete structure 1 Blockage (e.g., dam or culvert) removal or passage Pierce-140 1 provision Pierce-112 Buffer Acquisition Program 2

Pierce-117 Fircrest Project 2

Pierce-118 Tacoma Project 2

Pierce-141 Off-channel habitat creation/enhancement 2

Pierce-142 LWD enhancement 2

Pierce-143 Nutrient enhancement 2

Pierce-106 Leach Creek at Bridgeport Way 3

Pierce-116 Leach Creek Restoration 3 Restore natural substrates and retain flow through Pierce-147 4 mid Clover reaches

June 2001 Mobrand Biometrics, Inc. Page 5-19 Pierce County Watershed Analysis Section 5

Chambers-Clover Chinook Summary Of Projected Performance Measures Under Status Quo and Expanded Action Scenarios

Scenario Abundance Productivity Diversity index Historic 1,993 20.4 100% Current with harvest and fitness loss 67 2.6 41% Future with status quo mgmt (SQ) 131 3.1 93% Action Scenario-1 (S-1) 156 3.1 93% Action Scenario-2 (S-2) 399 4.2 100% Action Scenario-3 (S-3) 408 4.3 100% Action Scenario-4 (S-4) 408 4.3 100%

Chinook spawner abundance 1,500 1,993 1,200

900

600

Number of fish 300

0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Chinook productivity 25

Returns per spawner 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Chinook life history diversity 100%

75% t

50% Percen 25%

0% Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Figure 5-9. Chambers-Clover chinook (naturally produced) performance measures under historic, current, and five action scenarios.

June 2001 Mobrand Biometrics, Inc. Page 5-20 Pierce County Watershed Analysis Section 5

Chambers-Clover Coho Summary Of Projected Performance Measures Under Status Quo and Expanded Action Scenarios

Scenario Abundance Productivity Diversity index Historic 4,947 27.5 100% Current with harvest and fitness loss 110 6.5 30% Future with status quo mgmt (SQ) 464 9.5 46% Action Scenario-1 (S-1) 663 9.9 60% Action Scenario-2 (S-2) 1,034 12.5 67% Action Scenario-3 (S-3) 1,059 12.3 71% Action Scenario-4 (S-4) 1,078 12.5 71%

Coho spawner abundance 2,500 4,947 2,000

1,500

1,000

Number of fish 500

0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho productivity 30 25 20 15 10 5

Returns per spawner 0 Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Coho life history diversity 100%

75% t

50% Percen 25%

0% Historic Current SQ S-1 S-2 S-3 S-4 Scenario

Figure 5-10. Chambers-Clover coho (naturally produced) performance measures under historic, current, and five action scenarios.

June 2001 Mobrand Biometrics, Inc. Page 5-21 Pierce County Watershed Analysis Section 5

basis of percent improvement over current conditions. As noted earlier, however, we question whether the responses exhibited by chinook could be realized, given the preference of this species for larger basins. Still, the results indicate that the actions would improve performance potential for chinook in the basin.

Coho are shown to be highly responsive to the actions analyzed. Notably, we project substantial improvements in performance under the Status Quo Scenario due to the effect of the 28 actions now in the process of being implemented. Passage improvements are a significant reason for this response. As noted previously, there is considerable uncertainty in these projections, projections that hinge on assumptions made about action effectiveness and growth patterns in the basin. The projections do provide, however, an initial set of hypotheses for proceeding with salmon conservation and recovery measures in the basins. We further note that while some of the actions would be costly and require significant effort and commitment, others, such as barrier removal in Spanaway Creek would provide substantial benefit with relatively little cost.

5.3.3 Summary for Chambers-Clover Basin A graphic comparison of all actions related to the Chambers-Clover Basin is presented in Figure 5-11. The display compares actions in terms of effects on attributes, extent of dispersal of effect, technical feasibility, likelihood of outcome, community support, and cost. Details associated with the latter four items are provided in Appendix E.

5.4 Kitsap Basins We analyzed no specific actions for the Kitsap basins. We received no information about candidate actions being considered for future implementation in that area aside from the countywide programs and regulatory changes. Planning should proceed by focusing on the strategic priorities identified for the area under Section 4 Resource Assessment.

June 2001 Mobrand Biometrics, Inc. Page 5-22 Pierce County Watershed Analysis Section 5

Figure 5-11. Comparison of action components for the Chambers-Clover Basin.

June 2001 Mobrand Biometrics, Inc. Page 5-23

Pierce County Watershed Analysis Section 6

6.0 ADAPTIVE IMPLEMENTATION PROCESS

In this section, we address the implications of uncertainty and how to proceed in the face of it. Our understanding of ecosystems and their response to interventions is inevitably incomplete, and our ability to measure progress toward management goals accurately and timely is limited. Here we suggest that adaptive management, supported by the EDT model, provides the means to proceed with implementation while managing and containing risks due to these kinds of uncertainties.

6.1 Uncertainty Our knowledge about the biology of the salmonid species we seek to manage, the present and future condition of their environment, and the effects of human interventions is imperfect. Research progressively increases our understanding of the salmonid species and their habitat requirements and sensitivities, but some uncertainty always remains. Annual returns of chinook salmon, for example, vary greatly from year to year. We can explain some of the variation in terms of parent spawning escapements, ocean and fresh water survival conditions, and so on, but much of the variation remains unexplained.

It is particularly difficult to detect, with confidence, the effects of habitat improvements based on observed run size trends. It has been estimated that, because of inherent variability, it would take 30 years to detect a 50% improvement in average production, if we were to use adult run size as the response variable (Lichatowich and Cramer 1979). This time frame is obviously unsatisfactory. We need to be able to track the progress of restoration programs less erratically. The EDT method can help accomplish this. It is an example of a “weight of evidence” approach, where a working hypothesis is carefully constructed based on available knowledge, information, and data. As we obtain new information, we modify the working hypothesis.

The working hypothesis helps us distinguish the stochastic variation and the inter-annual environmental fluctuations from the underlying long-term changes in habitat suitability. It allows us to see the long-term trends through the noise of annual variation (Figure 6-1). The working hypothesis reflects the long-term changes in the average environmental conditions, which are the target of restoration (Walters, 1997). As a part of the adaptive management process, the working hypothesis is tested for consistency with population trends.

When we talk about risk due to uncertainty, we are referring both to the obscuring factors of inter-annual variation and to our imperfect understanding of how restoration actions affect the average, long-term, survival conditions for the diagnostic species. A key to understanding the latter component of risk is the implied cause and effect relationship between actions and stakeholder values. This relationship is made explicit when the specific assumptions in this linkage are stated. Once these are identified, we can analyze the uncertainties associated with all assumptions needed to form the logical conclusion: that the action will lead to achievement of a specified set of objectives without adverse impact on other values. The EDT framework provides the logical linkage between cause and effect—the model and reports like this one document the assumptions.

June 2001 Mobrand Biometrics Inc. Page 6-1 Pierce County Watershed Analysis Section 6

Production Potential as measured through EDT Goal Indicators of Progress Objective ) l e ) d f e n z i o o i r t o c (Mo t uns u s a s c R i od . re d g . g Pr In e o ( r P

Observed Production (e.g. Runsize)

1999 20xx

Figure 6-1. Hypothetical salmon production time series. The working hypothesis explains cause and effect based on all available information. It is updated as better information becomes available through monitoring of key habitat attributes, hypothesis testing (e.g., point-to-point survival of treatment groups), and the literature.

Assumptions and working hypotheses are logical statements about presumed relationships and conditions of the ecosystem and its function. These assumptions are always present in the management of natural resources because knowledge is imperfect. Explicitly stated assumptions enable those persons engaged in the management process—or the general public—to consider them and use them as a basis for learning and improving future decision making. Assumptions also need to be explicitly disclosed to enable questioning. For example: Are the assumptions reasonable, i.e., are they consistent with existing information? Do the assumptions pose significant risk? Can the assumptions/hypotheses be tested?

The process of identified and disclosed assumptions associated with each candidate action helps provide accountability to the overall planning process itself and to the public in considering potential benefits and risks of those proposed actions. It is a step that is essential if adaptive management is to become a reality.

There are five categories of assumptions associated with the planning process and with the conceptual framework:

Actions to attributes: These assumptions refer to the relationship between actions and their impact on environmental conditions or attributes. They state the effectiveness of actions.

Attributes to performance: These assumptions involve the effects of environmental attributes on the elements of biological performance (productivity, capacity, and life history diversity). There are two distinct subcategories of these assumptions: one set deals with the baseline environmental conditions; the second set deals with the rules for translating physical environmental attribute conclusions into survival parameters for the diagnostic species.

June 2001 Mobrand Biometrics Inc. Page 6-2 Pierce County Watershed Analysis Section 6

Performance to objectives: These assumptions refer to the relationship between biological performance and stakeholder values or program objectives.

Conceptual framework: These assumptions involve conceptual or theoretical bases for understanding the ecosystem and its processes.

Monitoring and evaluation (M&E): These assumptions involve our ability to monitor and evaluate changes in the ecosystem: the feasibility to make observations from which conclusions can be drawn about the validity of the other categories of assumptions.

The adaptive implementation process requires the identification of all of the assumptions that are made in these five categories. It is therefore unavoidable that the lists of these assumptions, which are documented in this report, will be long. Once these lists are initially completed, they need to be checked against one another to ensure that they do not conflict. If one set of assumptions is used to rationalize one suite of actions, and a conflicting set of alternative assumptions is used for another suite, the program is internally inconsistent and obviously cannot succeed.

The following classification of uncertainties is useful in that it provides some organization to what might otherwise appear as a tangled web. The classification is based on three qualities or attributes of uncertainty: degree of uncertainty, consequences of error, and resolvability.

An uncertainty is classified as ACCEPTED when either the probability or the consequences of error are insignificant (Figure 6-2). All others are labeled CRITICAL. CRITICAL uncertainties may in turn be RESOLVABLE or UNRESOLVABLE.

ACCEPTED

LITERATURE UNCERTAINTY UNRESOLVABLE REVIEW

CRITICAL non-PROJECT STUDIES RESOLVABLE

PROJECT M&E

Figure 6-2. Uncertainty classification. Resolution of uncertainties may be through literature review, studies outside the scope of the project, or studies that are a part of the project. The monitoring and evaluation section below presents different ways in which uncertainties may be addressed within the project.

6.2 Adaptive Management Because of uncertainty, it is necessary to incorporate flexibility into the implementation of watershed plans so that unsuccessful strategies and unattainable objectives can be replaced

June 2001 Mobrand Biometrics Inc. Page 6-3 Pierce County Watershed Analysis Section 6 with more suitable ones. We also need, however, stability and accountability to ensure that sound strategic decisions are made that lead toward achievement of long-term resource goals. What is called for is a structured decision process based on adaptive management principles (Figure 6-3).

Goals Stock Management Policies t n e

u Indicators of Progress

eq (Objectives) Fr

ss Strategies Le d n Action Sequencing a y c i l o P

e Implementation r o Monitoring M & Evaluation

Figure 6-3. Mind map of the adaptive management feedback loops. The outermost loops require more policy involvement and are less frequent than the inner loops.

The adaptive management process begins with informed policy decisions that set in motion implementation of a set of strategies to meet basin goals. As Figure 6-3 indicates, adaptive management decisions vary in degree of policy involvement and in the frequency by which they need to be revisited. For example, broad basin goals may be reviewed every ten years, near-term objectives and strategies annually, and action components perhaps on a real-time, daily basis.

Since knowledge about effectiveness of strategies and about future events is imperfect, there is a need to regularly revisit the policy decisions in light of new information or unforeseen circumstances. In order for this iterative decision process to be effective, it needs to be supplied with current and reliable information. Hence, a monitoring and evaluation program is needed. The EDT framework provides the means for prioritizing monitoring activities, updating databases, and reevaluating strategy effectiveness.

Hilborn and Walters (1992) use the terms active and passive adaptive management to distinguish between decisions based on the outcome of specifically designed grand experiments (active) versus those based on accumulation of information obtained more

June 2001 Mobrand Biometrics Inc. Page 6-4 Pierce County Watershed Analysis Section 6

Annually Updated Project Status Report Initial Goals are Modified Goals based on initial expectations from the preferred Modified "The alternative Policy Decisions Performance Community" Indicators

Initial Strategies Modified are based on Strategies preferred alternative

Annual Project Evaluation Report

Implement Options forModifying Options for Meeting Strategies Strategies to meet Monitoring ModifiedGoals Current Goals Plans

Effectiveness Monitoring

Population Validation Implementation Habitat Attribute Performance Monitoring Monitoring (QC+) Monitoring Monitoring (e.g. stock status) Active A.M. Active A.M. Passive& Active A.M. Passive A.M.

Technical Recommendations Technical Evaluation Surprise! (e.g. Bayes decision rules)

Figure 6-4. The elements of a simple adaptive management process. The diagram displays the annual cycle of activities that support informed policy decisions. The process revolves around two annually updated reports indicated by the shaded boxes in the diagram. opportunistically. The adaptive management process proposed here includes elements of both, recognizing that the grand experiments are usually very difficult and costly to carry out on a broad scale.

Policy and technical roles and responsibilities must be clearly and explicitly spelled out. The Project Status Report (PSR) is the blueprint for action. In addition to goals, objectives, and strategies the PSR should also identify biological, regulatory, budgetary, and other constraints within which the adaptive decision process must operate. The EDT model—and its associated documentation of information, data sources, and assumptions—is a supplement to this report. The Project Evaluation Report conveys technical recommendations and proposes amendments to the PSR.

6.3 Monitoring and Evaluation35 Monitoring and evaluation (M&E) need to be an integral part of adaptive watershed management. The purpose of M&E is to guide decision making toward implementation of measures and actions that effectively contribute to achieving objectives while controlling risk. It is a cornerstone of adaptive management, which is the imperative safety net of

35 Much of the material for this section is drawn from Lestelle et al. (1996).

June 2001 Mobrand Biometrics Inc. Page 6-5 Pierce County Watershed Analysis Section 6 watershed stewardship in the face of uncertainty. M&E questions can be organized into three categories: Implementation, Effectiveness and Validation (Figure 6-5).

Is Framework OK?

3 5 Was Performance Improved? Were Values 1 2 1) For spr. chinook? Enhanced? a) life history diver.? Were Actions Did Environmental 1) For spr. chin.? b) productivity? Implemented Attributes Change 2) Were other c) abundance? as Plannned? as Presumed? values affected 2) In general? consistent with a) biodiversity? expectations? b) reprod. success? c) abundance

Figure 6-5. Monitoring questions. Implementation monitoring addresses question 1, effectiveness monitoring addresses question 2, and validation monitoring addresses questions 3, 4, and 5.

Implementation monitoring addresses quality control—it seeks to determine if indeed the action was implemented as designed. Quality control can and should accompany all implemented actions to assure effectiveness and validate the conclusions upon which the actions are based. Without quality control, we have little confidence in any inferences drawn from our monitoring results. Quality control standards and procedures will be specified in contract work plans.

Effectiveness monitoring asks whether the actions were effective in altering (or protecting the current condition of) the environment. Actions are taken with the intent to modify (improve) a specified set of environmental attributes—these modifications are in turn expected to improve or maintain the performance of the biological system, furthering progress toward the goal of enhancing values (e.g., greater abundance of important species). Monitoring plans must be statistically well designed to account for variation due to causes other than the action being taken, as environmental attributes are notoriously variable.

Validation monitoring consists of experimental testing of hypotheses regarding the response of populations to environmental factors. Hypotheses stated or implicit in the framework that form the rationale for the contemplated action should be examined. This examination is a part of the benefit-risk analysis step in the planning process. The benefit- risk analysis, along with an assessment of the monitoring feasibility and cost forms the basis for prioritizing the research that would be undertaken. It should be noted also that some of the critical hypotheses may allow broad inferences, with the implication that the research may have more global value and/or might be more appropriately conducted elsewhere.

June 2001 Mobrand Biometrics Inc. Page 6-6 Pierce County Watershed Analysis Section 6

Validation monitoring also addresses the need to continually and progressively improve the theories that guide our decision-making. Fundamental to the evaluation of watershed actions is the validity of the conceptual framework, which forms the basis for interpreting all of our observations. Both general and specific aspects of the framework should be reviewed and tested. Current literature on related subjects should be reviewed and new ideas from a broad range of interests and expertise explored.

Validation monitoring includes, for example, monitoring of the conditions of the diagnostic species itself (e.g., run size, spawning escapement), which provides information essential to tracking the condition of the population over time. While erratic and imprecise as short-term indicators, trends in stock status are invaluable for assessing long-term prognoses for populations and their environment. The maintenance of a sense of history, in terms of conditions that reflect values and benefits to the community, is important as a long- term guide for setting public policy and for detecting and responding to more gradual and insidious changes in the watershed.

A fully functional adaptive management process continually incorporates the results of all three types of monitoring. The M&E plan then serves both as an information repository and as a tool for updating the working hypothesis.

The proposed initial adaptive management implementation steps are: 17. Establish a decision making infrastructure that can effectively implement adaptive management on an ecosystem scale. 18. Build a work plan around the process outlined in Figure 6-4. 19. Use this report as the initial draft of a Project Status Report, updating and adding sections as needed.

June 2001 Mobrand Biometrics Inc. Page 6-7

Pierce County Watershed Analysis

LITERATURE CITED

Beverton, Raymond J. H. and Holt, Sidney J. 1957. On the Dynamics of Exploited Fish Populations. Chapman & Hall, London.

Busby, Peggy J., Wainwright, Thomas C., Bryant, Gregory J., Lierheimer, Lisa J., Waples, Robin S., Waknitz, F. William, and Lagomarsino, Irma V. 1996. Status review of west coast steelhead from Washington, Oregon, and California. NMFS-NWFSC-27. NOAA Technical Memorandum. National Marine Fisheries Service, Seattle, Washington.

Collier, T. K., Johnson, L. L., Meyers, M. S., Stehr, C. M., Krahn, M. M., and Stein, J. E. 1998. Fish injury in the Hylebos Waterway in Commencement Bay, Washington. U.S. Department of Commerce. NOAA Technical Memorandum NMFS-NWFSC- 36.

Concurrent Technologies Corporation (CTC). 2000. Kitsap Peninsula salmonid refugia study. Concurrent Technologies Corporation,

Cowardin, L. M., Carter, V., Golet, F. C., and LaRoe, E. T. 1979. Classification of wetlands and deepwater habitats of the United States. U.S. Fish and Wildlife Service. Biological Services Program FWS/OBS-79-31.

Ecological Work Group (EWG). An ecological framework for the multi-species planning process. Northwest Power Planning Council, http://www.edthome.org/framework/ecoframework.htm.

Ewing, Richard D. 1999. Diminishing returns: salmon decline and pesticides. Report. Oregon Pesticide Education Network (OPEN), http://www.pond.net/!fish1ifr/salpest.pdf.

Graeber, William. 1999. Draft Puyallup River delta estuary landscape restoration plan: an estuary wide ecological assessment and decision making framework for long-term ecosystem restoration and protection within the context of an emerging salmon recovery regime. Unpublished document. Washington Department of Natural Resources, Olympia, WA.

Hayman, R. A., Beamer, E. M., and McClure, R. E. 1996. FY 1995 Skagit River Chinook Restoration Research. Progress Report No. 1. Skagit System Cooperation, LaConner, WA.

Healy, M. C. 1982. Juvenile Pacific salmon in estuaries: the life support system. Pages 315- 341 in Kennedy, Victor S., Estuarine Comparisons. Academic Press, Cambridge, Maryland.

Hilborn, R. and Walters, C. 1992. Quantitative fisheries stock assessment: choice, dynamics & uncertainty. Routledge, Chapman and Hall, New York.

Hilborn, Ray and Mangel, Marc. 1997. The ecological detecttive: confronting models with

June 2001 Mobrand Biometrics, Inc. Page L-1 Pierce County Watershed Analysis

data. Princeton Universtiy Press, Princeton, New Jersey.

Hymer, J., Pettit, R., Wastel, M., Hahn, P., and Hatch, K. 1992. Stock summary reports for Columbia River anadromous salmonids for the coordinated information system, Volume III: Washington subbasins below McNary Dam. Report. Bonneville Power Administration, Portland, Oregon.

Johnson, Orlay W., Ruckelshaus, Mary H., Grant, W. Stewart, Waknitz, F. William Garrett Ann M., Bryant, Gregory J., Neely, Kathleen, and Hard, Jeffrey. 1999. Status review of coastal cutthroat trout from Washington, Oregon, and California. NOAA Technical Memorandum NMFS-NWFSC-37. National Marine Fisheries Service, Seattle, WA.

Kerwin, J. 1999. Salmon habitat limiting factors. Water Resource Inventory Area 10. Washington State Conservation Commission, Olympia, WA.

Ladley, R. and Smith, B. 1998. Migratory behavior of chinook salmon in the lower Puyallup and White rivers. Available from Puyallup Tribal Fisheries, Puyallup, WA.

Lestelle, L. C., Blair, G. R., and Chitwood, S. A. 1993. Approaches to supplementing coho salmon in the Queets River, Washington. Pages 104-119 in Berg, L. and Delaney, P. W., editors. Proceedings of the coho workshop. British Columbia Department of Fisheries and Oceans, Vancouver, British Columbia.

Lestelle, L. C., Rowse, M. L., and Weller, C. 1993. Evaluation of natural stock improvement measures for Hood Canal coho salmon. Technical Report 93-1. Point No Point Treaty Council, Kingston, Washington.

Lestelle, Lawrence C., Mobrand, Lars E., Lichatowich, James A., and Vogel, Thomas S. 1996. Applied ecosystem analysis - a primer, EDT: the ecosystem diagnosis and treatment method. Project number 9404600. Report. Bonneville Power Administration, Portland, Oregon.

Lichatowich, J. and Cramer, S. 1979. Parameter selection and sample sizes in studies of anadromous salmonids. Oregon Department of Fish and Wildlife, Portland, Oregon.

Lichatowich, J. A. and Mobrand, L. E. 1995. Analysis of chinook salmon in the Columbia River from an ecosystem perspective. Project number 92-18; Contract No. DE- AM79-92BP25105. Final Report. Bonneville Power Administration, Portland, Oregon.

Lichatowich, James, Mobrand, Lars E., Lestelle, Larry, and Vogel, Tom. 1995. An approach to the diagnosis and treatment of depleted Pacific salmon populations in freshwater ecosystems. Fisheries 20(1): 10-18.

Mastin, M. C. 1998. Flood potential of South Prairie Creek, Pierce County, Washington. U.S. Geological Survey. Water-Resources Investigations Report 98-4009.

May, C. W., Welch, E. B., Horner, R. R., Karr, J. R., and Mar, B. W. 1997. Quality indices

June 2001 Mobrand Biometrics, Inc. Page L-2 Pierce County Watershed Analysis

for urbanization effects in Puget Sound lowland streams. Water Resources Series Technical Report No. 154. Final report. Department of Civil Engineering, University of Washington, Seattle, Washington.

McElhany, Paul, Ruckleshaus, Mary H., Ford, Michael J., Wainwright, Thomas C., and Bjorkstedt, Eric P. 2000. Viable salmonid populations and the recovery of evolutionarily significant units. U.S. Department of Commerce, Seattle, Washington.

Mobrand, Lars, Lestelle, Larry, and Blair, Greg. 1998. Recovery of a Columbia River watershed from an ecosystem perspective: a case study using the EDT method. Contract #94AM33243. Final report to Bonneville Power Administration. Mobrand Biometrics, Inc., Vashon, Washington.

Mobrand, Lars, Lestelle, Lawrence, and Blair, Gregory. 1998. Completion report for Habitat - Coho Production Model project. Mobrand Biometrics, Inc., Vashon, Washington.

Mobrand, Lars E., Lichatowich, James A., Lestelle, Lawrence C., and Vogel, Thomas S. 1997. An approach to describing ecosystem performance "through the eyes of salmon". Canadian Journal of Fisheries and Aquatic Sciences 54: 2964-2973.

Muckleshoot Indian Tribe, Puyallup Tribe of Indians, and Washington Department of Fish and Wildlife. 1996. Recovery plan for White River spring chinook salmon. Report. Available from Washington Department of Fish and Wildlife, Olympia, WA.

Northwest Power Planning Council. In preparation. Columbia River Multi-Species Framework project coarse screening analysis, Volume 1: Progress report. Volume 2: Methods and Appendices (A-E). Report.

Pacific International Engineering. Salmon life history report. Draft report prepared for Port of Tacoma. Pacific International Engineering,

Pacific International Engineering. 1999. Milwaukee habitat area juvenile salmon utilization study, 1999. Draft report prepared for Port of Tacoma. Sitcum Waterway Remedication Project. Pacific International Engineering, Tacoma, WA.

Pacific International Engineering. 2000. Clear Creek habit monitoring report, 1999. Draft Report prepared for Port of Tacoma. Sitcum Waterway Remediation Project. Pacific International Engineering, Tacoma, WA.

Pierce County. 1999. Pierce County Endangered Species Act response executive summary.

Pierce County Water Programs. 2000. Gig Harbor Basin study characterization report. Draft Report.

Port of Tacoma and Puyallup Tribe of Indians. 1999. Commencement Bay juvenile salmonid sampling: 1980-1995. Data compilation by Pacific International Engineering for Port of Tacoma.

June 2001 Mobrand Biometrics, Inc. Page L-3 Pierce County Watershed Analysis

Puget Sound Technical Recovery Team (TRT). April, 2001. Independent populations of chinook salmon is Puget Sound. Report. Puget Sound TRT, http://www.nwfsc.noaa.gov/cbd/trt/trt_puget.htm.

R2 Resource Consultants. 1999. Pierce County Endangered Species Act Response, Appendix A-D, Salmon Habitat Conditions Review. Report.

Salo, E. and Bayliff, W. 1958. Artificial and natural production of silver salmon, Oncorhynchus kisutch, at Minter Creek, Washington. Research bulletin 4. Report. Washington Department of Fisheries, Olympia, Washington.

Seiler, Dave. 2001. 2001 Wild coho forecasts for Puget Sound & Washington coastal systems. Washington Department of Fish & Wildlife, Olympia, WA.

Simenstad, Charles A. 2000. Commencement bay aquatic ecosystem assessment: ecosystem-scale restoration for juvenile salmon recovery. Unpublished report prepared for City of Tacoma, Washington Department of Natural Resources, and U.S. Environmental Protection Agency. Tacoma, Wa.

Tetra Tech/KCM, Inc. 2000. City of Auburn 2000 comprehensive drainage plan. Report.

Tri-County. 2000. Draft Tri-County Watershed Assessment Framework.

Tri-County Executive Committee. 1999. Tri-County Endangered Species Act response work plan. Report.

U.S. Fish and Wildlife Service. 1999. Endangered and threatened wildlife and plants; determination of threatened status for Bull Trout in the coterminous United States; final rule. Federal Register 64(210): 58909-58933.

Walters, Carl. 1997. Information requirements for salmon management. Pages 61-68 in Stouder, Deanna J., Bisson, Peter A., and Naiman, Robert J., editors. Pacific salmon & their ecosystems: status and future options. Chapman & Hall , San Francisco.

Washington Administration Code (WAC). 1992. 173-201A. November 25, 1992,

Washington Department of Fish and Wildlife. 1998. 1998 Washington Salmonid Stock Inventory--Appendix Bull Trout and Dolly Varden. Washington Department of Fish and Wildlife, Olympia, WA.

Washington Department of Fish and Wildlife. 1998. WDF catch-escapement run size calculation summary: Puget Sound escapement estimates. Report. http://www.nwifc.wa.gov/release data.

Washington Department of Fish and Wildlife. 2000. 1998/1999 final hatchery escapement & broodstock report. WDFW, Olympia, WA.

Washington Department of Fish and Wildlife. 2000. Final Bull Trout and Dolly Varden management plan. Report. WDFW, Olympia, WA.

June 2001 Mobrand Biometrics, Inc. Page L-4 Pierce County Watershed Analysis

Washington State Conservation Commission (WCC). 2001. Salmon and steelhead habitat limiting factors water resource inventory, area 15, East Kitsap watershed, November 2000.

Western Regional Climate Center (WRCC). 2000. Period of record monthly climate summary. http://www.wrcc.dri.edu.

Williams, Walter R., Laramie, Richard M., and Ames, James J. 1975. A catalog of Washington streams and salmon utilization: volume 1 Puget Sound region. Report. Washington Department of Fisheries, Olympia, WA.

June 2001 Mobrand Biometrics, Inc. Page L-5