SEPA
U.S. Environmental Protection Agency, Region 10
www.epa.gov
Prepared by:
U.S. Environmental Protection Agency
Region 10
FT""
Columbia River Cold
Water Refuges Plan
EPA-910-R-21-001
January 2021

•S3
HLZi'.
...

-------
Columbia River Cold Water Refuges Plan
Final January 2021
TABLE	OF CONTENTS
ACKNOWLEDGEMENTS	XII
EXECUTIVE SUMMARY	1
1	INTRODUCTION	3
1.1	Regulatory Background	3
1.2	Types of Cold Water Refuges	4
1.3	Overview of Columbia River Cold Water Refuges Plan	5
2	COLD WATER REFUGES IN THE LOWER COLUMBIA RIVER	7
2.1	Columbia River Temperatures	7
2.2	Tributary Temperatures Compared to Columbia River Temperatures	10
2.3	Tributaries Providing Cold Water Refuge	13
2.4	Twelve Primary Cold Water Refuges	18
3	SALMON AND STEELHEAD USE OF COLD WATER REFUGES	34
3.1	Salmon and Steelhead Migration Timing and Columbia River Temperatures	34
3.2	Columbia River Temperatures that Trigger Cold Water Refuge Use	35
3.3	Examples of Salmon and Steelhead Use of Cold Water Refuges	37
3.4	Number of Steelhead in Cold Water Refuges	40
3.5	Number of Fall Chinook in Cold Water Refuges	45
3.6	Summary of the Number of Steelhead and Fall Chinook in Cold Water Refuges	47
3.7	Historic Steelhead Use of Cold Water Refuges	47
3.8	Deschutes River Cold Water Refuge Use	49
3.9	Use of CWR by Specific Populations of Steelhead and Fall Chinook	51
4	TEMPERATURE AND FISH HARVEST IMPACTS ON MIGRATING SALMON AND
STEELHEAD	55
4.1	Adverse Temperature Effects to Migrating Adult Salmon and Steelhead	55
4.2	Relationship Between Temperature and Migration Survival of Adult Steelhead and
Fall Chinook Salmon	56
4.3	Fishing Harvest of Salmon and Steelhead in Cold water Refuges	58
4.4	Snake River Steelhead and Fall Chinook Migration Survival Rates in the Lower
Columbia and Lower Snake Rivers	59
4.5	Energy Loss and Pre-Spawning Mortality of Fall Chinook Salmon from Exposure to
Warm Migration Temperatures	62
4.6	Increased Mortality and Shift in Run Timing of Sockeye and Summer Chinook from
Warm Migration Temperatures	64

-------
Columbia River Cold Water Refuges Plan
Final January 2021
5	HISTORIC AND FUTURE TRENDS IN COLUMBIA RIVER TEMPERATURES	71
5.1	Historic Temperature Conditions of the Lower Columbia River	71
5.2	Future Temperature Conditions of the Lower Columbia River and its Tributaries	73
6	SUFFICIENCY OF COLD WATER REFUGES IN THE LOWER COLUMBIA RIVER
77
6.1	CWR Sufficiency Assessment Framework	77
6.2	HexSim Model	78
6.3	Assessment oF Sufficiently Distributed CWR	85
7	ACTIONS TO PROTECT & RESTORE COLD WATER REFUGES	89
7.1	Cold Water Refuge Watershed Snapshots	90
7.2	Characteristics of Primary Cold Water Refuge Tributaries	90
7.3	Cowlitz River (River Mile 65) - Protect and Enhance	93
7.4	Lewis River (River Mile 84) - Protect and Enhance	99
7.5	Sandy River (River Mile 117) - Protect and Enhance	104
7.6	Tanner Creek (River Mile 141) - Protect and Enhance	110
7.7	Eagle Creek (River Mile 143) - Protect and Enhance	115
7.8	Herman Creek (River Mile 147.5) - Protect and Enhance	120
7.9	Wind River (River Mile 151) - Protect and Enhance	125
7.10	Little White Salmon River (River Mile 158.7) - Protect and Enhance	130
7.11	White Salmon River (River Mile 165) - Protect and Enhance	136
7.12	Hood River (River Mile 166) - Protect and Enhance	142
7.13	Klickitat River (River Mile 177) - Protect and Enhance	148
7.14	Fifteenmile Creek (River Mile 188.9) - Restore	154
7.15	Deschutes River (River Mile 201) - Protect and Enhance	159
7.16	Umatilla River (River Mile 284.7) - Restore	166
7.17	Summary of Actions to Protect and Restore Cold Water Refuges	172
7.18	Action to Address Fishing in Cold Water Refuges	175
8	UNCERTAINTIES AND ADDITIONAL RESEARCH NEEDS	176
9	SUMMARY AND RECOMMENDATIONS	180
10	REFERENCES	186
11	CHAPTER 7 BIBLIOGRAPHY	191
Cowlitz River	191
Lewis	River	192

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Sandy River	192
Tanner Creek	193
Eagle Creek	194
Herman Creek	194
Wind River	194
Little White Salmon River	195
White Salmon River	196
Hood River	196
Klickitat River	197
Fifteenmile Creek	198
Deschutes River	198
Umatilla River	199
12	APPENDICES	201
12.1	Lower Columbia River Temperature Variation	201
12.2	Evaluation of the Potential Cold Water Refugia Created by Tributaries within the
Lower/Middle Columbia River based on NorWeST Temperature Model	201
12.3	Screening Approach to Identify the 23 Tributaries that Currently Provide CWR in the
Lower Columbia River	201
12.4	Location of Upstream Extent of 23 CWR Areas Used by Migrating Salmon and
Steelhead	201
12.5	Volume of Cold Water Refuge Associated with the 23 Tributaries Providing CWR in
the Lower Columbia River and Selection of the 12 Primary CWR	201
12.6	Columbia River Cold Water Refuge Assessment Plume Modeling Report	201
12.7	Estimating the Potential Cold Water REfugia Volume within Tributaries that
Discharge into the Columbia River	201
12.8	Estimates of Plume Volume for Five Tributary/Columbia River Confluence Sites
Using USEPA Field Data Collected in 2016	201
12.9	Estimated CWR Volume for the Wnd River and Little White Salmon River/Drano
Lake	201
12.10	Estimated CWR Volume in Herman Creek Cove	201
12.11	Supplement to Estimated CWR Volume in Herman Creek Cove	201
12.12	Tributary and Columbia River Measured Temperature Data Summary	201
12.13	Estimated Number of Steelhead and Fall Chinook Using CWR in the Bonneville
Reservoir Reach	201
12.14	Water Temperature Estimates of the Columbia River and Tributaries in 2040 and
2080	201
iv

-------
Columbia River Cold Water Refuges Plan
Final January 2021
12.15	Stream Temperature Predictions Under Varying Shade and Climate Scenarios in the
Columbia River Basin	201
12.16	Assessment of Climate Change Impacts on Temperatures of the Columbia and
Snake Rivers	202
12.17	Water Temperature Estimates of the Lower/Middle Columbia River and Tributaries in
2040 and 2080 based on the NorWeST Model	202
12.18	Predicted Maximum Temperatures Using the NorWeST Model in 12 Primary Cold
Water Tributaries and 2 "Restore" Tributaries	202
12.19	Comparison of NorWeST Future Temperature Estimates to a Continuation of
Historical Warming Trends in the Lower Columbia River	202
12.20	Watershed Snapshot Assumptions and Approaches	202
12.21	HexSim Migration Corridor Simulation Model Preliminary Results	202
12.22	Comparison of NorWeST Temperature Estimates to Monitoring Data in the Twelve
Primary CWR	202
12.23	Comparison between NHDPIus modeled August mean flow conditions and available
flow data collected at the primary Cold Water Refugia (CWR) streams	202
LIST OF TABLES
Table 2-1 23 tributaries providing cold water refuge in the Lower Columbia River	15
Table 2-2 Estimates for the volume of water in tributary confluence areas that is more than
2°C cooler than the Columbia River	17
Table 2-3 Twelve primary CWR tributaries (highlighted in bold and color)	19
Table 3-1 Estimated number of steelhead in cold water refuges each year (1999-2016)
(Appendix 12.13)	43
Table 3-2 Estimated number of steelhead in each Bonneville reach cold water refuge
(Appendix 12.13)	44
Table 3-3 Distribution of radio-tagged steelhead in the Bonneville reach cold water refuges
on August 31 (Combined 2000/2001 Data Set) (M. Keefer, personal
communication, September 11, 2017)	44
Table 3-4 Estimated steelhead density in cold water refuges (Appendix 12.13)	45
Table 3-5 Deschutes River mouth steelhead PIT-tag detections by calendar year and
Distinct Population Segment (DPS) (NMFS 2017a)	49
Table 3-6 Percent of Snake River (SR) steelhead using Deschutes cold water refuges and
number of steelhead using Deschutes cold water refuges (NMFS 2017a)	50
Table 4-1 Summary of temperature effects to migrating adult salmon and steelhead in the
Lower Columbia River (EPA 2003; McCullough 1999, Richter and Kolmes 2005)
	56
Table 5-1 Future temperature conditions of the Lower Columbia River tributaries (Appendix
12.17)	76
Table 6-1 Adult salmon and steelhead survival estimates after correction for harvest and
straying based on PIT-tag conversion rate analysis from Bonneville (BON) to
v

-------
Columbia River Cold Water Refuges Plan
Final January 2021
McNary (MCN) dams, McNary to Lower Granite (LGR) dams, and Bonneville to
Lower Granite dams (NMFS 2017b)	86
Table 7-1 Location and characteristics of primary cold water refuges	92
Table 7-2 Water Availability Analysis, 5/20/20, Sandy River at mouth, Oregon Water
Resources Department	107
Table 7-3 Water Availability Analysis, Eagle Creek at mouth, 5/20/20, Oregon Water
Resources Department	118
Table 7-4 Water Availability Analysis, Herman Creek at mouth, 5/20/20, Oregon Water
Resources Department	123
Table 7-5 Water Availability Analysis, 5/20/20 Hood River at river mile 0.75, 5/23/18,
Oregon Water Resources Department	145
Table 7-6 Water Availability Analysis, 5/20/20 for the Deschutes River confluence with the
Columbia River	162
Table 7-7 Water Availability Analysis, 5/20/20 for the Umatilla River confluence with the
Columbia River	169
LIST OF FIGURES
Figure 1-1 Map of the Columbia Basin, with the Columbia River Cold Water Refuges Plan
scope circled in red (USACE)	6
Figure 2-1 Current August mean water temperature in the Columbia River and tributaries
(2011-2016) (Appendix 12.14)	7
Figure 2-2 Longitudinal profile of the August mean Columbia River temperature from
McNary Dam to the Bonneville Dam (DART)	8
Figure 2-3 Lower Columbia River temperature from early July to mid-September, 6-year
average 2011-2016 (DART)	9
Figure 2-4 Seasonal temperature profiles downstream of Bonneville Dam, 10-year average
2009-2018 (DART)	10
Figure 2-5 191 tributary confluences with the Lower Columbia River (white dots), with
predicted stream temperatures from the NorWeST database [predicted August
mean stream temperature for the 1993-2011 period]	11
Figure 2-6 Columbia mainstem and tributary temperature difference (August mean water
temperatures from USFS NorWeST)	12
Figure 2-7 Modeled August mean stream temperatures for tributaries in the Lower Columbia
River (1993-2011) (USFS NorWeST). Circle sizes illustrate relative tributary
August mean flow (1971-2000) (NHDPIus)	13
Figure 2-8 Twelve primary cold water refuge tributaries (purple and green) to the Lower
Columbia River as well as the 11 non-primary cold water refuge tributaries
(white)	20
Figure 2-9 Cowlitz River Cold Water Refuge and Associated Temperatures	22
Figure 2-10 Lewis River Cold Water Refuge	23
Figure 2-11 Sandy River Cold Water Refuge and Associated Temperatures	24
Figure 2-12 Tanner Creek Cold Water Refuge and Associated Temperatures	25
Figure 2-13 Eagle Creek Cold Water Refuge and Associated Temperatures	26
Figure 2-14 Herman Creek and Cove Cold Water Refuge and Associated Temperatures ....27
Figure 2-15 Wnd River Cold Water Refuge and Associated Temperatures	28
vi

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 2-16 Little White Salmon River and Drano Lake Cold Water Refuge and Associated
Temperatures	29
Figure 2-17 White Salmon River Cold Water Refuge and Associated Temperatures	30
Figure 2-18 Hood River Cold Water Refuge and Associated Temperatures	31
Figure 2-19 Klickitat River Cold Water Refuge and Associated Temperatures	32
Figure 2-20 Deschutes River Cold Water Refuge and Associated Temperatures	33
Figure 3-1 Salmon and steelhead Bonneville Dam passage and temperature (DART)	35
Figure 3-2 Steelhead use of cold water refuge (black dots and 'Used tributaries' axis)
(Keeferet. al. 2009)	36
Figure 3-3 Fall Chinook use of cold water refuge (Goniea et. al. 2006)	36
Figure 3-4 Temperature profile of a steelhead using cold water refuges (Keefer & Caudill
2017)	37
Figure 3-5 Temperature profile of a steelhead using cold water refuges (Keefer & Caudill
2017)	38
Figure 3-6 Temperature profile of a fall Chinook salmon using cold water refuges (Keefer &
Caudill 2017)	39
Figure 3-7 Temperature profile of a summer Chinook salmon using cold water refuges
(Keefer & Caudill 2017)	39
Figure 3-8 Steelhead passage at Bonneville Dam and The Dalles Dam (Appendix 12.13)..41
Figure 3-9 Estimated number of steelhead in Bonneville reach cold water refuges (Appendix
12.13)	42
Figure 3-10 Proportion of 219 radio-tagged steelhead in Bonneville cold water refuges (M.
Keefer, personal communication, August 31, 2017)	42
Figure 3-11 Accumulation of fall Chinook in the Bonneville reach and the number of fall
Chinook in cold water refuges (2008-2017 average) (Appendix 12.13)	46
Figure 3-12 Accumulation of fall Chinook in the Bonneville reach and the number of fall
Chinook in cold water refuges (2013) (Appendix 12.13)	47
Figure 3-13 Steelhead passage at Bonneville Dam and The Dalles Dam, 1957-1966 (DART)
	48
Figure 3-14 Current versus 1950s water temperatures in the Lower Columbia River (DART)
	49
Figure 3-15 Estimated number of PIT-tagged Snake River steelhead and estimated total
number of Snake River steelhead (estimated by tag expansion) present in
Deschutes River cold water refuges by month 2013-2015 (NMFS 2017a)	51
Figure 3-16 Percent of population-specific steelhead that used cold water refuges for >12
hours (solid circles) and associated median passage time from Bonneville Dam
to the John Day Dam for those that used and did not use (clear circles) CWR.
TUC, Tucannon River; HAN, Hanford Reach; LFH, Lyons Ferry Hatchery; UCR,
Upper Columbia River; WWR, Walla Walla River; CWR, Clearwater River; SAL,
Salmon River; SNK, Snake River above Lower Granite Dam; YAK, Yakima River;
IMR, Imnaha River; GRR, Grande Ronde River; UMA, Umatilla River; JDR, John
Day River. (Keefer etal. 2009)	52
Figure 3-17 Median timing distributions (median, quartiles, and 10th and 90th percentiles) at
Bonneville Dam for steelhead that successfully returned to tributaries or
hatcheries. Vertical dotted lines show mean first and last dates that Columbia
River water temperatures were 19°C; the shaded area shows dates with mean
temperatures >21 °C. (Keefer et al. 2009)	53

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 3-18 Mean composition of upriver bright fall-run Chinook salmon at Bonneville Dam
using five-day intervals based on release dates of radio-tagged fish. 1998 and
2000-2004. MCB-BPH = mid-Columbia River bright-Bonneville Pool hatchery
stock. (Jepson et al. 2010)	54
Figure 4-1 Estimated survival rate of adult steelhead between Bonneville Dam and McNary
Dam (FPC, October 31, 2016 Memo)	57
Figure 4-2 Estimated survival rate of adult fall Chinook between Bonneville Dam and
McNary Dam (FPC, May 8, 2018 Memo)	57
Figure 4-3 Adjusted survival estimates of adult Snake River steelhead between Bonneville
Dam (BON) and McNary Dam (MCN) and between Bonneville Dam and Lower
Granite Dam (LGR) for the whole run (NMFS, 2019)	60
Figure 4-4 Adjusted survival estimates of adult Snake River fall Chinook between Bonneville
Dam and Lower Granite Dam for the whole run (NMFS, 2019)	61
Figure 4-5 The proportion of simulated fish that had energy densities greater than the 4 kJ/g
threshold needed for sufficient energy to spawn (Plumb, 2018)	63
Figure 4-6 Standardized, simulated spawning initiation date distributions for PIT-tagged,
hatchery-origin Snake River fall Chinook salmon adults, 2010-2015 (Conner et. al
2018)	64
Figure 4-7 Sockeye passage and river temperature at Bonneville Dam (FPC, August 26,
2015 Memo)	65
Figure 4-8 Weekly survival estimates from Bonneville Dam to McNary Dam in 2015 for
Upper Columbia River Sockeye (blue bars), Snake River sockeye that migrated
in-river as juveniles (orange bars), and Snake River sockeye that were
transported as juveniles (yellow-orange bars) with water temperatures (red line)
at The Dalles Dam (NMFS 2016)	66
Figure 4-9 Estimated relationship between Bonneville Dam forebay temperature and
Bonneville Dam to McNary Dam survival by return year for Snake and Upper
Columbia adult sockeye (FPC Memo 2015)	67
Figure 4-10 Median sockeye salmon migration date (A), July mean temperature (B), and
June mean flow (C) at Bonneville Dam (Crozier et al. 2011)	68
Figure 4-11 Daily average temperature (°F) in the Bonneville Dam forebay from June 1 to
July 31 by return year (FPC 2016)	69
Figure 4-12 Hatchery Snake River summer Chinook adult reach survival with 95% confidence
intervals by return year (FPC 2016)	69
Figure 4-13 Summer Chinook run timing past Bonneville Dam (1994-2018) (DART)	70
Figure 4-14 Trends in summer Chinook run distribution past Bonneville Dam (1994-2018)
(DART)	70
Figure 5-1 Trend in Columbia River August temperatures at Bonneville Dam (National
Research Council 2004)	71
Figure 5-2 Simulated monthly mean temperatures at Bonneville Dam (current) (EPA 2020)
	72
Figure 5-3 Simulated monthly mean temperatures at Bonneville Dam (free flowing) (EPA
2020)	73
Figure 5-4 Current August mean water temperature in the Columbia River and tributaries
(2011-2016) (Appendix 12.14)	74
Figure 5-5 Estimated 2040 August mean water temperature in the Columbia River and
tributaries (Appendix 12.14)	75
VIII

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 5-6 Estimated 2080 August mean water temperature in the Columbia River and
tributaries (Appendix 12.14)	75
Figure 6-1 Simulated energy loss for Grande Ronde summer steelhead from Bonneville
Dam to the Snake River under various scenarios (Appendix 12.21)	80
Figure 6-2 Simulated arrival date at the Snake River for Grande Ronde summer steelhead
with and without CWR use under current conditions (Appendix 12.21)	81
Figure 6-3 Simulated cumulative degree days above 21 °C for Grande Ronde summer
steelhead between Bonneville Dam and the Snake River under different
scenarios (Appendix 12.21)	83
Figure 6-4 Simulated cumulative degree days above 22°C for Grande Ronde summer
steelhead between Bonneville Dam and the Snake River under different
scenarios (Appendix 12.21)	84
Figure 6-5 Simulated cumulative degree days above 21 °C under 2017 Columbia River
temperatures for Grande Ronde summer steelhead between Bonneville Dam and
the Snake River under different scenarios (Appendix 12.21)	84
Figure 7-1 12 primary and 2 "restore" cold water refuge tributary locations	89
Figure 7-2 Cowlitz River land cover	94
Figure 7-3 Cowlitz River land ownership	94
Figure 7-4 Cowlitz River shade difference between potential maximum and current shade 95
Figure 7-5 Map of Cowlitz River Dams	95
Figure 7-6 Lewis River land ownership	100
Figure 7-7 Lewis River land cover	100
Figure 7-8 Lewis River shade difference between potential maximum and current shade. 101
Figure 7-9 Map of Lewis River dams	101
Figure 7-10 Sandy River land ownership	105
Figure 7-11 Sandy River land cover	105
Figure 7-12 Sandy River shade difference between potential maximum and current shade 106
Figure 7-13 Sandy River Delta Dam pre-removal - white line indicates location of former dam
(USACE, 2015)	106
Figure 7-14 Tanner Creek land cover	111
Figure 7-15 Tanner Creek land ownership	111
Figure 7-16 Tanner Creek shade difference between potential maximum and current shade
	112
Figure 7-17 Eagle Creek Fire Burn Severity map in the Tanner Creek Watershed. (Peter
Leinenbach and USFS)	112
Figure 7-18 Eagle Creek land ownership	116
Figure 7-19 Eagle Creek land cover	116
Figure 7-20 Eagle Creek shade difference between potential maximum and pre-2017 fire
shade	117
Figure 7-21 Eagle Creek Fire Burn Severity map in the Eagle Creek Watershed (Peter
Leinenbach and USFS)	117
Figure 7-22 Herman Creek land cover	121
Figure 7-23 Herman Creek land ownership	121
Figure 7-24 Herman Creek shade difference between potential maximum and current shade
	122
Figure 7-25 Wind River land cover	126
Figure 7-26 Wind River land ownership	126
ix

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 7-27 Wind River shade difference between potential maximum shade and current
shade	127
Figure 7-28 Wind River Basin - Water rights and availability, Washington Department of
Ecology	127
Figure 7-29 Little White Salmon River Basin land ownership	131
Figure 7-30 Little White Salmon River Basin land cover	131
Figure 7-31 Difference between potential stream shade conditions and current stream shade
	132
Figure 7-32 White Salmon River Basin land cover	137
Figure 7-33 White Salmon River Basin land ownership	137
Figure 7-34 White Salmon River shade difference potential maximum and current shade ..138
Figure 7-35 Hood River land cover	143
Figure 7-36 Hood River land ownership	143
Figure 7-37 Hood River shade difference between potential maximum and current shade .144
Figure 7-38 Estimated flow diversions in the Hood River Basin in 2006	145
Figure 7-39 Klickitat River land cover	149
Figure 7-40 Klickitat ownership	149
Figure 7-41 Klickitat River shade difference between potential maximum and current shade
	150
Figure 7-42 Water Availability in WRIA 30 (Washington Department of Ecology, Revised
2012)	150
Figure 7-43 Fifteenmile Creek land ownership	155
Figure 7-44 Fifteenmile Creek land cover	155
Figure 7-45 Fifteenmile Creek shade difference between potential maximum and current
shade	156
Figure 7-46 Land cover in the Deschutes Basin	160
Figure 7-47 Land ownership in the Deschutes Basin	160
Figure 7-48 Deschutes River shade difference between potential maximum and current
shade	161
Figure 7-49 Aerial view of the confluence of the Umatilla and Columbia Rivers; yellow pin
denotes upstream extent of refuge	166
Figure 7-50 Umatilla River and Columbia River water temperatures (Appendix 12.12)	167
Figure 7-51 Land ownership in the Umatilla Basin	167
Figure 7-52 Land cover in the Umatilla Basin	168
Figure 7-53 Umatilla River shade difference between potential maximum and current shade
	168
x

-------
Columbia River Cold Water Refuges Plan
Final January 2021
ACRONYMS/ABBREVIATIONS
Acronyms/Abbreviations
Definition
°C
Degrees Celsius
°F
Degrees Fahrenheit
cfs
cubic feet per second
CWR
Cold Water Refuge
CTUIR
Confederated Tribes of the Umatilla Indian Reservation
DART
Data Access in Real Time
EPA
U.S. Environmental Protection Agency
ESA
Endangered Species Act
FERC
Federal Energy Regulatory Commission
GIS
Geographic Information System
HCP
Habitat Conservation Plan
ISWR
Instream Water Right
LCFRB
Lower Columbia Fish Recovery Board
NHD
National Hydrography Dataset
NMFS
National Marine Fisheries Service
NorWeST
Northwest Stream Temperature database
NPCC
Northwest Power and Conservation Council
ODEQ
Oregon Department of Environmental Quality
ODF
Oregon Department of Forestry
ODFW
Oregon Department of Fish and Wildlife
OWRD
Oregon Water Resources Department
PIT-tag
Passive Integrated Transponder-tag
SWSL
Surface Water Source Limitation
TMDL
Total Maximum Daily Load
USACE
U.S. Army Corps of Engineers
USBR
United State Bureau of Reclamation
USFS
United States Forest Service
USGS
United States Geological Survey
WDFW
Washington Department of Fish and Wldlife
WDNR
Washington Department of Natural Resources
WQMP
Water Quality Management Plan
WRIA
Water Resource Inventory Area
xi

-------
Columbia River Cold Water Refuges Plan
Final January 2021
ACKNOWLEDGEMENTS
The U.S. Environmental Protection Agency (EPA) is indebted to many people for their help in
developing this Columbia River Cold Water Refuges Plan. Development of this Plan has been a
collaborative team effort.
EPA Region 10 CWR Team: John Palmer (Lead), Dru Keenan, Jenny Wu, Peter Leinenbach,
Ben Cope, Rochelle Labiosa, Gretchen Hayslip, Alexandra Clayton, Jonell Deacon, Keyyana
Blount, Martin Merz, Miranda Magdangal, David Gruen, Dylan Laird, Martin Jacobsen, Abigail
Conner, Lindsay Guzzo, Sally Goodman, Andrea Lindsay, Mary Lou Soscia, Jennifer Byrne,
Christine Psyk, and Angela Chung
EPA Office of Research and Development CWR Team: Joe Ebersole, Marcia Snyder, Nathan
Schumaker, Randy Comeleo, and Jonathan Halama (Corvallis, OR); Naomi Detenbeck, and
Matthew Fuller (Narragansett, Rl)
Contributors: Matt Keefer (University of Idaho); Debra Sturdevant, James McConaghie, Don
Butcher, Gene Foster, Bonnie Lamb, Smita Mehta, Tonya Dombrowski (Oregon Department of
Environmental Quality); Melissa Gildersleeve, Chad Brown, Ben Rau, Jim Pacheco, Paul
Pickett, Mike Gallagher (Washington Department of Ecology); Ritchie Graves, Blane Bellerud,
Anne Mullan, Josie Thompson, Spencer Hovekamp, and Scott Carlon (National Marine
Fisheries Service); Dan Isaak, Brian Staab, Diane Hopster, Robin Shoal, Bengt Coffin (United
States Forest Service); Art Martin, Rod French, Tucker Jones, Anna Stevenson, Spencer
Sawaske, Erin Andyke, Erick Van Dyke, Derrek Faber (Oregon Department of Fish and
Wildlife); Dan Rawding, Thomas Beuhrens (Washington Department of Fish and Wildlife); John
Plumb, Pat Connelly, Christian Torgersen, Jason Dunham, Krista Jones, and Ian Jezorek
(United States Geological Survey); Brian Maschhoff; Jessica Olson (Columbia River Gorge
Commission); Margaret Filardo (Fish Passage Center); Catherine Corbett, Chris Collins, and
Keith Marcoe (Lower Columbia Estuary Partnership); Lynn Palensky, Laura Robinson, and
Leslie Bach (Northwest Power and Conservation Council); Denis Lofman (Columbia River
Estuary Study Taskforce); Lowell Dickson (Washington Department of Natural Resources); Bill
Sharp, Joe Zendt, Shuba Pandit, and Tom Iverson (Yakama Nation); Laura Gephart, Dianne
Barton, and Jeff Fryer (Columbia River Inter-Tribal Fisheries Commission); Scott O'Daniel, Gary
James, Robin Harris (Confederated Tribes of the Umatilla Indian Reservation); Chris Brun, Brad
Houslett, Ryan Gerstenberger (Confederated Tribes of the Warm Springs Reservation); David
Moskowitz (The Conservation Angler); Sarah Cloud, Greg McMillan, and Ben Kirsch (Deschutes
River Alliance); Nina Bell (Northwest Environmental Advocates); Miles Johnson (Columbia
Riverkeeper); Scott Levy (Bluefish); Dan Turner and Mike Langeslay (United States Army Corps
of Engineers); Agnes Lut (Bonneville Power Administration); Megan Hill and Lori Campbell
(Portland General Electric); Tova Tilinghast (Underwood Conservation District), Jess Groves
and Sally Moore (Port of Cascade Locks), Bruce Aylward (AMP Insights), Holly Coccoli, Cindy
Thiemann (Watershed Council Hood River Watershed Group), Steve Hood (Sandy River
Watershed Council), Dave McClure (Klickitat County), Josh Epstein and Gardner Johnston
(Interfluve)
xii

-------
Columbia River Cold Water Refuges Plan
Final January 2021
EXECUTIVE SUMMARY
The Columbia River Cold Water Refuges Plan focuses on the lower 325 miles of the Columbia
River from the Snake River to the ocean. Cold water refuges (CWR) are locations migrating
adult salmon and steelhead temporarily use to escape warm summer river temperatures. CWR
serve an increasingly important role to some salmon species as the Lower Columbia River has
warmed over the past 50 years and will likely continue to warm in the future. The Plan:
•	Describes available CWR in the Lower Columbia River,
•	Characterizes how salmon and steelhead use CWR,
•	Assesses the amount of CWR needed to attain Oregon's Clean Water Act
CWR narrative water quality standard,
•	Identifies actions to protect and restore CWR, and
•	Recommends future CWR studies.
Fish Use of CWR
Adult salmon and steelhead commonly use CWR in the Lower Columbia River from mid-July
through mid-September when river temperatures typically exceed 20°C (68°F). August is the
warmest month, with average river temperatures of 21-21.5°C (70-71 °F); the warmest days
commonly reach 22.5°C (72.5°F). Daily average river temperatures are similar throughout the
entire 325-mile stretch of the Lower Columbia River, with slightly cooler temperatures near
McNary Dam and the warmest temperatures near the John Day and The Dalles Dams.
CWR are found where cooler tributary rivers flow into the Columbia River. A CWR tributary is at
least 2°C cooler than the Columbia River. EPA identified 23 tributaries that provide CWR. EPA
defined 12 of these as "primary" CWR tributaries because of their size, accessibility, and
documented or presumed use by migrating salmon and steelhead. Of the 12 primary CWR, four
are below Bonneville Dam (Cowlitz River, Lewis River, Sandy River, and Tanner Creek);
seven are between Bonneville Dam and The Dalles Dam (Eagle Creek, Wind River, Herman
Creek, White Salmon River, Little White Salmon River, Hood River, and Klickitat River);
and one is between The Dalles Dam and the John Day Dam (Deschutes River). There are no
primary CWR between the John Day Dam and McNary Dam.
Salmon and steelhead use of the eight primary CWR above Bonneville Dam is well-documented
from scientific tagging studies. Less is known about the use of the four CWR below Bonneville
Dam. The largest CWR with well-documented use are the Little White Salmon River (Drano
Lake), Deschutes River, Klickitat River, Herman Creek (Herman Creek Cove), and the White
Salmon River.
Among the various Columbia River salmon runs, CWR are used most often by adult summer
steelhead and fall Chinook salmon because their migration timing corresponds with the warmest
river temperatures. Using CWR allows fish to escape warm Columbia River temperatures and
complete their upstream migration when river temperatures are cooler. EPA modeling (HexSim)
simulated fish migrating between Bonneville Dam and McNary Dam and showed that CWR use
allows summer steelhead to reduce the time exposed to stressful temperatures by 50 percent.
Other modeling studies have predicted that use of CWR by early migrating fall Chinook salmon
allows them to retain enough energy to successfully spawn in the fall. Summer steelhead often
will use CWR for several weeks, while fall Chinook salmon generally use CWR for a few days to
a week.
1

-------
Columbia River Cold Water Refuges Plan
Final January 2021
EPA Findings and Recommendations
Oregon's Clean Water Act CWR narrative standard stipulates that the Lower Columbia River
"must have coldwater refugia that's sufficiently distributed so as to allow salmon and steel head
migration without significant adverse effects from higher water temperature elsewhere in the
waterbody," and coldwater refugia is "at least 2°C colder" than the river. To assess attainment
with Oregon's CWR standard, EPA evaluated the total amount of CWR, the extent to which fish
use CWR, the density of fish in CWR, the distribution of CWR, the health benefits of CWR use,
and the overall importance of adult migration risk factors in the recovery of ESA-listed salmon
and steelhead. EPA made this assessment under current Lower Columbia River temperatures,
while recognizing increased use of CWR is likely to occur as the Columbia River continues to
warm due to climate change. EPA has concluded that attainment of Oregon's CWR standard
will depend on maintaining the volume of the 12 primary CWR and increasing CWR in the
Umatilla River.
To protect the 12 primary CWR tributaries and the Umatilla River from future warming and to
retain the existing CWR volume, this Plan describes an array of existing programs, plans, and
regulations. These include: the aquatic protection prescriptions for timber harvest on national
forest land, state forest land, and private forest land; stream buffer protections associated with
the Management Plan for the Columbia River Gorge National Scenic Area] Wld and Scenic
River management plans, county land use and shoreline regulations; established minimum flow
requirements; and state water quality provisions to protect existing cold water.
This Plan highlights recommended restoration actions found in the Northwest Power and
Conservation Council subbasin plans, salmon recovery plans, TMDL implementation plans, and
water management plans. Such plans exist for all of the 12 primary CWR tributaries and the
Umatilla River. Implementation of these plans will help reduce river temperatures in their
watersheds. The identified restoration actions serve to improve fish habitat and to cool river
temperatures that will help maintain CWR volume in light of predicted tributary warming due to
climate change.
Recommended restoration actions to maintain and increase CWR include:
•	Restoring riparian vegetation to provide river shading,
•	Restoring stream morphology and floodplain connectivity to reduce channel widths
and create pools and groundwater connectivity, and
•	Restoring summer river flows that are more resistant to warming and increase CWR
volume.
To address identified uncertainties, this Plan recommends future studies to track fish use of
CWR, to assess the benefits of CWR use, and to assess density effects and the carrying
capacity of CWR. This Plan identifies immediate monitoring priorities to track CWR use, stream
temperature, and flow trends, including:
•	Installing PIT-tag detectors in Little White Salmon/Drano Lake and Herman Creek
Cove,
•	Re-establishing USGS flow gauges, including temperature gauges, near the mouth of
Little White Salmon River and Wind River, and
•	Installing and operating long-term annual summer temperature monitoring at or near
the USGS flow gauge sites near the mouth of the Cowlitz, Lewis, Sandy, White Salmon,
Hood, Klickitat, and Umatilla Rivers.
2

-------
Columbia River Cold Water Refuges Plan
Final January 2021
1 INTRODUCTION
Approximately 700,000 to two million adult salmon and steelhead return from the ocean and
migrate up the Columbia River each year past the Bonneville Dam. Roughly 40% of these fish
that migrate during the summer months when Columbia River water temperatures reach or
exceed 20°C may endure adverse effects in the form of disease, stress, decreased spawning
success, and lethality (	). To minimize their exposure to warm temperatures in the
Columbia River, many salmon and steelhead temporarily move into areas of cooler water, which
are called cold water refuges (CWR). In the Lower Columbia River, these CWR are primarily
where cooler tributary rivers flow into the Columbia River.
This Plan characterizes Columbia River water temperatures, the amount of available CWR in
the Lower Columbia River (mouth to Snake River), and the extent to which salmon and
steelhead use the CWR. The plan also assesses the amount of CWR needed to support
migrating adult salmon and steelhead, highlights recommended actions to protect and restore
the CWR, and recommends future studies and monitoring.
1.1 REGULATORY BACKGROUND
Both the States of Oregon and Washington have established temperature water quality
standards for the Lower Columbia River to protect migrating salmon and steelhead, which
include a 20°C (68°F) numeric criterion1 for limiting the maximum water temperatures. The
State of Oregon also includes a narrative temperature standard that stipulates that the Lower
Columbia River:
"must have coldwater refugia that's sufficiently distributed so as to allow salmon and steelhead
migration without significant adverse effects from higher water temperatures elsewhere in the
water body."
Oregon standards define coldwater refugia as
"those portions of a water body where, or times during the diel temperature cycle when, the
water temperature is at least 2 degrees Celsius colder than the daily maxium temperature of the
adjacent well mixed flow of the water body (OAR 340-041-0002(10))."
Under the Clean Water Act, the U.S. Environmental Protection Agency must approve (or
disapprove) new or revised state water quality standards. In 2004, EPA approved the State of
Oregon's temperature water quality standards for the Lower Columbia River, including the 20°C
maximum numeric criterion and the coldwater refugia narrative provision noted above. As part
of the approval process, EPA consulted with the National Marine Fisheries Services (NMFS) per
the requirements of the Endangered Species Act to ensure EPA's approval would not
jeopardize the continued existence of ESA listed species.
The ESA consultation on the Oregon Lower Columbia River temperature standards noted above
(among other standards) was initially completed in 2004, but was invalidated by the United
States District Court of Oregon in 2012. In accordance with a court order, NMFS issued a new
Biological Opinion in November 2015. In that Opinion, NMFS concluded that Oregon's Lower
1 Oregon's 20°C numeric criterion is based on a 7-day average daily maximum. Washington's 20°C numeric criterion
is based on a daily maximum.
3

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Columbia River temperature standards are likely to jeoparize the survival and recovery of ESA
listed salmon and steelhead because the coldwater refugia narrarative standard, to date, did not
appear to be an effective means for minimizing the adverse effects likely to be experienced by
migrating salmon and steelhead under the 20°C numeric criterion.
To avoid jeopardizing ESA listed salmon and steelhead, the NMFS 2015 Biological Opinion
included a reasonable and prudent alternative for EPA to develop this Columbia River Cold
Water Refuges Plan.
The EPA recently issued the Columbia and Lower Snake Rivers Temperature Total Daily
Maximum Load (TMDL) ( ) that addresses exceedances of the 20°C numeric criteria2 on
the Lower Columbia River as well as Oregon's coldwater refugia narrative criteria. The
Columbia River Temperature TMDL calculates how much various sources are contributing to
exceedances of the 20°C numeric criteria and establishes temperature targets for cold water
refuge tributaries to attain Oregon's CWR narrative criteria. The Columbia River Temperature
TMDL establishes temperature targets for cold water refuge tributaries consistent with the
scientific analysis sumarized in this Columbia River Cold Water Refuges Plan. The states of
Oregon and Washington are responsible for the development of management plans to
implement the Columbia River Temperature TMDL. This Columbia River Cold Water Refuges
Plan, specifically actions and recommendations listed in Chapters 7, 8, and 9 of the Plan, can
serve as a reference to the states in the establishment of the management plans to meet the
CWR targets established in the TMDL. This Plan, however, does not address actions to cool the
mainstem Columbia River to attain the 20°C numeric criteria.
Lastly, EPA is issuing this Plan as a result of the reasonable and prudent alternative in the 2015
NMFS Biological Opinion. The Plan is not a regulation and does not impose binding
requirements on any party, including EPA, other federal agencies, the states, or private entities.
1.2 TYPES OF COLD WATER REFUGES
Cold water refuges are created in several ways. Tributary streams that are colder than the river
they flow into provide CWR for migrating fish in the confluence area of the tributary (plume
CWR) and in the lower section of the tributary (stream CWR). Fish can enter these tributary
areas to reside in water temperatures cooler than the river, minimizing their heat exposure. This
is the main type of CWR in the Lower Columbia River.
CWR can also be formed by inflowing groundwater colder than the river channel, including river
water that submerges into the gravels then re-emerges colder than the river (referred to as
hyporheic flow) (Torgersen, C. et al. 2012). CWR can occur in stratified reservoirs, where
warmer surface water can be avoided by fish residing in cooler water at depth. Additionally, if a
river's temperature varies throughout the day, with warmer temperatures during the daylight
hours and cooler temperatures at night due to the difference in solar heating, the cooler
nighttime conditions serve as CWR relative to the warmer daytime temperatures. These other
types of CWR are minor in scope in the Lower Columbia River, and there is no evidence that
2 EPA's Columbia and Lower Snake Rivers Temperature TMDL also addresses exceedances of other numeric
criteria in the Columbia and Snake Rivers.
4

-------
Columbia River Cold Water Refuges Plan
Final January 2021
they serve a significant role for salmon and steelhead in the Lower Columbia River (Appendix
12.1; High etal. 2006).
1.3 OVERVIEW OF COLUMBIA RIVER COLD WATER REFUGES PLAN
This Plan is focused on the Lower Columbia River between the mouth and river mile 309
(Oregon-Washington border), where the Oregon cold water narrative criteria applies (Figure
1-1). Since the Snake River entry at river mile 325 is near the Oregon-Washington border, EPA
extended some of the analyses in the plan to the Snake River.
The following is a brief summary of the chapters in the plan.
•	Chapter 1 provides introductory and background information.
•	Chapter 2 characterizes the existing temperature conditions in the Lower Columbia
River and identifies tributaries that provide CWR, including the location and size of each
CWR.
•	Chapter 0 describes how various salmon and steelhead species use CWR, including
the Columbia River temperatures that trigger CWR use and the number of salmon and
steelhead that reside in CWR during the warmest time of year.
•	Chapter 4 summarizes the adverse effects warm river temperatures have on migrating
adult salmon and steelhead and the relationship of river temperature to survival rates
and the loss of energy reserves.
•	Chapter 5 assesses how much the Columbia River has warmed over the past century
and the extent to which the Columbia River is predicted to continue to warm due to
climate change.
•	Chapter 6 assesses whether there is a sufficient amount of CWR to support healthy
salmon and steelhead populations and attain Oregon's coldwater refugia narrative
criteria.
•	Chapter 7 analyzes the watersheds of CWR tributaries and recommends actions to
protect, restore, and enhance them.
•	Chapter 8 summarizes scientific uncertainties related to CWR in the Lower Columbia
River and recommends research studies to address those uncertainties.
•	Lastly, Chapter 9 includes the plan's overall conclusions and recommendations.
5

-------
Columbia River Cold Water Refuges Plan
Final January 2021
PftfKj Ore-##' ^
i
• _unloV
Jf Gotoo'
Y)r^£**
'island
ranapom
Lucky
Artd«r*e«i
Rjrxh

Mica
[ll'VC
Canada
U
S
I # Corps oi Engineers
Dams
I
|
finlmon
FaH
M dn#r

IQ P,ims ounod by
O^cin
I ABu roou of Reclamation
CA 1 NV



Dams
1



Figure 1-1 Map of the Columbia Basin, with the Columbia River Cold Water Refuges Plan
scope circled in red (USACE)
6

-------
Columbia River Cold Water Refuges Plan
Final January 2021
2 COLD WATER REFUGES IN THE LOWER COLUMBIA RIVER
2.1 COLUMBIA RIVER TEMPERATURES
The Columbia River enters the State of Washington from Canada and warms as it moves
through Washington towards the Pacific Ocean. Figure 2-1 illustrates this longitudinal warming
in the warm summer month of August, when river temperatures are at their seasonal peak.
When the river enters Washington from Canada, average August river temperatures generally
fluctuate between 17-18'C from year to year. Throughout the Lower Columbia River where the
river serves as the border between Washington and Oregon, average August temperatures are
between 21-22 'C. This warm lower section of the river is the corridor through which ali
Columbia Basin salmon must begin their migration and is the focus of EPA's Cold Water
Refuges Plan.
Figure 2-1 Current August mean water temperature in the Columbia River and tributaries
(2011-2016) (Appendix 12.14)
7

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Temperature data from the four Lower Columbia River dams show the longitudinal temperature
regime in the Lower Columbia River (Figure 2-2). At McNary Dam, the most upstream of the
four Lower Columbia River dams, the average August temperature is 21 °C. The Columbia River
then warms by 0.5°C in the 80-mile pool between McNary Dam and John Day Dam. The highest
average August temperatures in the Lower Columbia River and the entire Columbia River occur
near the John Day Dam, reaching 21.5°C on average in August. Temperatures decrease slightly
at The Dalles Dam and the Bonneville Dam (Figure 2-2).
21.6
21.5
21.4
21.3
U

-------
Columbia River Cold Water Refuges Plan
Final January 2021
23
22
21
Daily Average River Temperature at the Lower Columbia River Dams (2011-2016)
• • • t •
•••«:• *..• •*i.
• •

• ••«••••
. *.
-	, * s1 * s • .*	*		
««*• ..... •
»|! .•*
i
19	« _•	•«!.
«
~. t •
»• •
11*1
18
17
1-Jul	15-Jul	29-Jul	12-Aug	26-Aug	9-Sep	23-Sep
Date
—«~—McNary —^»—John Day —^The Dalles —•—Bonneville —•—Water Quality Standard
Figure 2-3 Lower Columbia River temperature from early July to mid-September, 6-year
average 2011-2016 (DART)
Figure 2-1 through Figure 2-3 illustrate data averaged across multiple years, which illustrate
patterns for typical years but do not illustrate annual variability. The temperature regime can be
very different between years primarily due to different air temperatures. Figure 2-4 depicts
observed data downstream of Bonneville Dam for 10 individual years (2009-2018) to illustrate
the seasonal temperature range. The 10-year average of these Bonneville Dam daily average
temperatures (thick black line) reaches 20°C in mid-July, rises to 21-22°C in August, then falls
below 20°C in early September. The gray, red, and blue lines illustrate the variability in the
Lower Columbia River temperature regime, showing that magnitude, timing, and duration of
peak warming can vary between years. The red line depicts 2015 temperatures, which were
unusually warm early in the summer contributing to high rates of sockeye salmon mortality.
During this 10-year timeframe, mid-July temperatures ranged from about 17.5°C in 2011 (blue
line) to 22.5°C in 2015 (red line), a spread of 5°C. In mid-August, temperatures have less
interannual variability, ranging from 20-22°C.
9

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 2-4 Seasonal temperature profiles downstream of Bonneville Dam, 10-year average
2009-2018 (DART)
There is little daily variation in the temperature of the Columbia River. Since the river is so large,
it does not respond quickly to the air temperature differential between night and day as smaller
rivers and creeks tend to do. The Lower Columbia River dams are 'run of river' so the reservoirs
generally do not thermally stratify like deeper storage reservoirs. However, due to heating of the
upper surface layer in the John Day and McNary Reservoirs in the summer, the upper part of
these two reservoirs can be substantially warmer than the main channel temperature. During
warm periods, the upper surface layer of these two reservoirs, especially near the dam
forebays, can be 3-6 CC warmer than the main channel temperature (Appendix 12.1).
2.2 TRIBUTARY TEMPERATURES COMPARED TO COLUMBIA RIVER
TEMPERATURES
The National Hydrography Dataset (NHD) identifies 191 tributaries that flow directly into the
Columbia River between the mouth of the Columbia River and the confluence with the Snake
River (Appendix 12.2). Current August mean water temperatures for these rivers were obtained
from a Spatial Stream Network model developed by the U.S. Forest Service (LJSFS) called
NorWeST. The NorWeST database houses temperature data assembled from over 100
resource agencies across the western United States, and where data are unavailable, provides
10

-------
Columbia River Cold Water Refuges Plan
Final January 2021
modeled temperature estimates based on nearby temperature measurements and other factors
(Isaak et al. 2017). Figure 2-5 illustrates these 191 tributary confluences (white dots) along with
the predicted August mean temperature of the tributary.
O Tributary Confluence
Predicted Current
Stream Temperature (*C)
	Less Than 13.1
	13.2-15,2
	15.3 -16.7
16.8-18.0
18.1-193
- 19 4-205
	20.6-21.8
	219-239
^—24 0 - 300
E		
] Miles
	
Figure 2-5 191 tributary confluences with the Lower Columbia River (white dots), with
predicted stream temperatures from the NorWeST database [predicted August mean stream
temperature for the 1993-2011 period]
EPA compared the predicted August mean temperature of these 191 tributaries to the August
mean temperature of the Columbia River. Figure 2-6 illustrates the August mean temperature
difference between the Columbia River and its tributaries. The largest tributaries in Figure 2-6
are displayed for geographical reference.
11

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 2-6 Columbia mainstem and tributary temperature difference (August mean water
temperatures from USFS NorWeST)
Each of the 191 tributaries is color coded in Figure 2-6, with purple identifying tributaries that
are more than 4 C cooler than the Columbia River and green and yellow identifying tributaries
that are between 2-4°C and 0-2°C colder than the Columbia River, respectively. Red identifies
tributaries that are warmer than the Columbia River, As can be seen in the Figure 2-6, most of
the coolest tributaries (purple and green) are located within and downstream (west) of the
Cascade mountain range.
In addition to the temperature analysis described above, the average (1971-2000) August flows
for the 191 tributaries to the Lower Columbia River were derived from the Extended Unit Runoff
Method model in NHDPIusV2, a national surface water database. It is important to note that
there is a very large range of stream flows within these tributaries, ranging from <1 cfs to
8591 cfs (August mean). Figure 2-7 illustrates the relative flow (size of circle), tributary and
Columbia River temperature (position along y-axis), and temperature relative to the Columbia
River (color).
12

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Willamette (not refuge)
Lewis
Tanner Creek



> 4*C
COoVf !K»n th* rraintffm








item



betwe



o



r» than th« nwtosttm



Ob**rv*dCu«


v*r Watff Trmf









25 50 75 100 125 150 175 200 225 250
Lower and Middle Columbia River (River Mile derived from NHDPlus)
Deschutes
Cowlitz
Figure 2-7 Modeled August mean stream temperatures for tributaries in the Lower Columbia
River (1993-2011) (USFS NorWeST). Circle sizes illustrate relative tributary August mean flow
(1971-2000) (NHDPlus).
2.3 TRIBUTARIES PROVIDING COLD WATER REFUGE
Whether a tributary will provide cold water refuge (CWR) depends upon its temperature relative
to the Columbia River and the size and accessibility of its confluence area to migrating salmon
and steelhead. Using the information described in section 2.2 and other information noted
below, EPA conducted a screening analysis to identify tributaries that provide CWR for salmon
and steelhead in the Lower Columbia River. The first screen in the analysis was based on: 1)
the tributary's August mean temperature being 2 C colder than the Columbia River; and 2) the
tributary's August mean flow being greater than 10 cubic feet per second (cfs). EPA used 10 cfs
as an approximate minimum flow needed to form a cool water plume in the Columbia River,
which would attract salmon and steelhead use (Appendix 12.3).
From this list of tributaries, EPA excluded tributaries that were inaccessible to migrating salmon
and steelhead and excluded several tributaries where field flow data indicated flow was
significantly less than 10 cfs. EPA added the Umatilla River to the list, because although its
August mean temperature difference is less than 2 C cooler than the Columbia River, it is 2°C
cooler in late August/September and is the only CWR in the John Day Reservoir, so its location
13

-------
Columbia River Cold Water Refuges Plan
Final January 2021
is important. EPA also included two tributaries (Germany Creek and Bridal Veil Creek) on the
list with August mean flows between 7-10 cfs that are especially cold. This screening approach
resulted in listing 23 tributaries that currently provide CWR in the Lower Columbia River, as
noted in Table 2-1 (Appendix 12.3).
In Table 2-1 the August mean Columbia River mainstem temperatures (2005-2014) reflect data
in DART from the nearest mainstem dam. The August mean tributary temperatures are from the
NorWeSt model (1993-2011).The tributary flows are either from NHD Plus (1971-2000), or if
available, USGS gauge data. Although this information has varying time frames due to the
availably of the data, these data are intended to represent long term average temperature and
flow conditions.
14

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Tributary Name
River
Mile
August Mean
Mainstem
Temperature
(DART)
August Mean
Tributary
Temperature
(NorWeST)
August Mean
Temperature
Difference
August Mean
Tributary Flow (NHD
& USGS*)


°C
°C
°C
cfs
Skamokawa Creek (WA)
30.9
21.3
16.2
-5.1
23
Mill Creek (WA)
51.3
21.3
14.5
-6.8
10
Abernethy Creek (WA)
51.7
21.3
15.7
-5.6
10
Germany Creek (WA)
53.6
21.3
15.4
-5.9
8
Cowlitz River (WA)
65.2
21.3
16.0
-5.4
3634
Kalama River (WA)
70.5
21.3
16.3
-5.0
314*
Lewis River (WA)
84.4
21.3
16.6
-4.8
1291*
Sandy River (OR)
117.1
21.3
18.8
-2.5
469
Washougal River (WA)
117.6
21.3
19.2
-2.1
107*
Bridal Veil Creek (OR)
128.9
21.3
11.7
-9.6
7
Wahkeena Creek (OR)
131.7
21.3
13.6
-7.7
15
Oneonta Creek (OR)
134.3
21.3
13.1
-8.2
29
Tanner Creek (OR)
140.9
21.3
11.7
-9.6
38
Bonneville Dam
Eagle Creek (OR)
142.7
21.2
15.1
-6.1
72
Rock Creek (WA)
146.6
21.2
17.4
-3.8
47
Herman Creek (OR)
147.5
21.2
12.0
-9.2
45
Wind River (WA)
151.1
21.2
14.5
-6.7
293
Little White Salmon River (WA)
158.7
21.2
13.3
-7.9
248*
White Salmon River (WA)
164.9
21.2
15.7
-5.5
715*
Hood River (OR)
165.7
21.4
15.5
-5.9
374
Klickitat River (WA)
176.8
21.4
16.4
-5.0
851*
The Dalles Dam
Deschutes River (OR)
200.8
21.4
19.2
-2.2
4772*
John Day Dam
Umatilla River1 (OR)
284.7
20.9
20.8
-0.1
*
1^
00
McNary Dam
1 The Umatilla is 2°C cooler than the Columbia River in late August and September.
Table 2-1 23 tributaries providing cold water refuge in the Lower Columbia River
15

-------
Columbia River Cold Water Refuges Plan
Final January 2021
EPA estimated the volume in cubic meters (m3) of water that is at least 2°C colder than the
Columbia River for each of the 23 tributaries listed in Table 2-1. The purpose of estimating the
CWR volume is to compare the relative size and importance of the refuges and to assess the
density of fish in CWR. EPA used a combination of monitoring and modeling techniques to
estimate the volume of CWR in tributary confluence areas (plume CWR) and in the lower
portion of the CWR tributaries (stream CWR) used by salmon and steelhead. As part of
estimating the stream CWR volume in the lower portion of a given CWR tributary, EPA
estimated how far upstream salmon or steelhead are likely to go when using it as a CWR.
These 'upstream extent' estimates are based on Passive Integrative Transponder-tag (PIT-tag)
and radio tag information, discussions with field biologists, stream depth measurements,
satellite images, and field observations (Appendix 12.4). To estimate the volume of the plume
extending into the Columbia River that remained 2°C colder than the Columbia River itself
(plume CWR), EPA used a CORMIX plume model or in some cases (Herman Creek Cove, Little
White Salmon (Drano Lake), and the Wnd River delta) took direct measures of embayment
areas to calculate the volumes (Appendix 12.5 through 12.11). The 23 tributaries and their
associated plume CWR and stream CWR are listed in Table 2-2.
16

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Tributary Name
River
Mile
August Mean
Mainstem
Temperature
(DART)
August Mean
Tributary
Temperature
(NorWeST)
August Mean
Temperature
Difference
August Mean
Tributary Flow
(NHD&
USGS*)
Plume CWR
Volume
(> 2 C A)
Stream
CWR
Volume
(> 2=C A)
Total CWR
Volume
(> 2 C A)


C
C
C
cfs
mj
mj
mj
Skamokawa Creek (WA)
30.9
21.3
16.2
-5.1
23
450
1,033
1,483
Mill Creek (WA)
51.3
21.3
14.5
-6.8
10
110
446
556
Abernethy Creek (WA)
51.7
21.3
15.7
-5.6
10
81
806
887
Germany Creek (WA)
53.6
21.3
15.4
-5.9
8
72
446
518
Cowlitz River (WA)
65.2
21.3
16.0
-5.4
3634
870,000
684,230
1,554,230
Kalama River (WA)
70.5
21.3
16.3
-5.0
314*
14,000
27,820
41,820
Lewis River (WA)
84.4
21.3
16.6
-4.8
1291*
120,000
493,455
613,455
Sandy River (OR)
117.1
21.3
18.8
-2.5
469
9,900
22,015
31,915
Washougal River1 (WA)
117.6
21.3
19.2
-2.1
107*
740
32,563
33,303
Bridal Veil Creek (OR)
128.9
21.3
11.7
-9.6
7
120
0
120
Wahkeena Creek (OR)
131.7
21.3
13.6
-7.7
15
220
0
220
Oneonta Creek (OR)
134.3
21.3
13.1
-8.2
29
820
54
874
Tanner Creek (OR)
140.9
21.3
11.7
-9.6
38
1,300
413
1,713
Bonneville Dam
Eagle Creek (OR)
142.7
21.2
15.1
-6.1
72
2,100
888
2,988
Rock Creek1 (WA)
146.6
21.2
17.4
-3.8
47
530
1,178
1,708
Herman Creek (OR)
147.5
21.2
12.0
-9.2
45
168,000
1,698
169,698
Wind River (WA)
151.1
21.2
14.5
-6.7
293
60,800
44,420
105,220
Little White Salmon River (WA)
158.7
21.2
13.3
-7.9
248*
1,097,000
11,661
1,108,661
White Salmon River (WA)
164.9
21.2
15.7
-5.5
715*
72,000
81,529
153,529
Hood River (OR)
165.7
21.4
15.5
-5.9
374
28,000
0
28,000
Klickitat River (WA)
176.8
21.4
16.4
-5.0
851*
73,000
149,029
222,029
The Dalles Dam
Deschutes River (OR)
200.8
21.4
19.2
-2.2
4772*
300,000
580,124
880,124
John Day Dam
Umatilla River1 (OR)
284.7
20.9
20.8
-0.1
*
1^
00
0
10,473
10,473
McNary Dam
1 Only provide intermittent cold water refugia; CWR volume represents volume when river is greater than 2°C colder than Columbia River.
Table 2-2 Estimates for the volume of water in tributary confluence areas that is more than 2°C cooler than the Columbia River
17

-------
Columbia River Cold Water Refuges Plan
Final January 2021
2.4 TWELVE PRIMARY COLD WATER REFUGES
Of the 23 tributaries in Table 2-1 and Table 2-2, EPA identified 12 as particularly important
primary CWR areas based on CWR volume, stream temperatures, field observations, and
documented or presumed use by salmon and steelhead (Appendix 12.5). The 12 primary CWR
are bolded in Table 2-3 and displayed in Figure 2-8. In both Table 2-3 and Figure 2-8, primary
CWR tributaries that are >4°C cooler than the Columbia are highlighted in purple, and primary
CWR tributaries with temperatures 2-4°C cooler than the Columbia are highlighted in green.
The 12 primary tributaries constitute 98% of the total CWR volume in the Lower Columbia River,
are easily accessible, are sufficiently deep to provide cover, and have documented or presumed
use by migrating salmon and steelhead. The other 11 non-primary CWR tributaries have small
CWR volume (less than 2,000 m3), have substantial periods of time when the tributary is less
than 2°C cooler or even warmer than the Columbia River, and/or are shallow and exposed.
Additionally, the extent of use by salmon and steelhead in these 11 non-primary CWR
tributaries is unknown and likely is limited due to one or more of the characteristics noted above
(Appendix 12.5).
18

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Tributary Name
River
Mile
August Mean
Mainstem
Temperature
(DART)
August Mean
Tributary
Temperature
(NorWeST)
August Mean
Temperature
Difference
August Mean
Tributary Flow
(NHD&USGS*)
Plume CWR
Volume
(> 2°C A)
Stream CWR
Volume
(> 2°C A)
Total CWR
Volume
(> 2°C A)


°C
°C
°C
cfs
m3
m3
m3
Skamokawa Creek (WA)
30.9
21.3
16.2
-5.1
23
450
1,033
1,483
Mill Creek (WA)
51.3
21.3
14.5
-6.8
10
110
446
556
Abernethy Creek (WA)
51.7
21.3
15.7
-5.6
10
81
806
887
Germany Creek (WA)
53.6
21.3
15.4
-5.9
8
72
446
518
Cowlitz River (WA)
65.2
21.3
16.0
-5.4
3634
870,000
684,230
1,554,230
Kalama River2 (WA)
70.5
21.3
16.3
-5.0
314*
14,000
27,820
41,820
Lewis River (WA)
84.4
21.3
16.6
-4.8
1291*
120,000
493,455
613,455
Sandy River (OR)
117.1
21.3
18.8
-2.5
469
9,900
22,015
31,915
Washougal River1 (WA)
117.6
21.3
19.2
-2.1
107*
740
32,563
33,303
Bridal Veil Creek (OR)
128.9
21.3
11.7
-9.6
7
120
0
120
Wahkeena Creek (OR)
131.7
21.3
13.6
-7.7
15
220
0
220
Oneonta Creek (OR)
134.3
21.3
13.1
-8.2
29
820
54
874
Tanner Creek (OR)
140.9
21.3
11.7
-9.6
38
1,300
413
1,713
Eagle Creek (OR)
142.7
21.2
15.1
-6.1
72
2,100
888
2,988
Rock Creek1 (WA)
146.6
21.2
17.4
-3.8
47
530
1,178
1,708
Herman Creek (OR)
147.5
21.2
12.0
-9.2
45
168,000
1,698
169,698
Wind River (WA)
151.1
21.2
14.5
-6.7
293
60,800
44,420
105,220
Little White Salmon River (WA)
158.7
21.2
13.3
-7.9
248*
1,097,000
11,661
1,108,661
White Salmon River (WA)
164.9
21.2
15.7
-5.5
715*
72,000
81,529
153,529
Hood River (OR)
165.7
21.4
15.5
-5.9
374
28,000
0
28,000
Klickitat River (WA)
176.8
21.4
16.4
-5.0
851*
73,000
149,029
222,029
Deschutes River (OR)
200.8
21.4
19.2
-2.2
4772*
300,000
580,124
880,124
Umatilla River1 (OR)
284.7
20.9
20.8
-0.1
*
1^
00
0
10,473
10,473
1 Only provides intermittent cold water refugia; CWR volume represents volume when river is greater than 2°C colder than Columbia River.
2 Tidally influenced and may be inaccessible during low tides.
Table 2-3 Twelve primary CWR tributaries (highlighted in bold and color)
19

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Tributaries Providing Co!d Water Refuge
in the Lower Columbia River
Skamdkawa^Gennany
r*	Creek ,'
5 £ hv@8kmui AkI
	¦ A JCretfc .)
M Y . 1
Bonnevilk
,0am


McNary Dam
John Day Dam
jjfejfB*
Dalles Dam

Dam Location
t Primary CWR tributary temperatures > 4*C cooSer than the Columbia
0 Primary CWR tributary temperatures between 2*C and 4*C cooler than the Columbia
£ Non-primary CWR tributary locations
	
Figure 2-8 Twelve primary cold water refuge tributaries (purple and green) to the Lower Columbia River as well as the 11 non-
primary cold water refuge tributaries (white)
20

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Four of the 12 primary CWR tributaries are below Bonneville Dam, seven are between the
Bonneville Dam and The Dalles Dam, and only one, the Deschutes River, is upstream of The
Dalles Dam. The two largest CWR are the Cowlitz River confluence area CWR and the Little
White Salmon River CWR, which drains into Drano Lake prior to entering the Columbia River.
The total volume of all 23 CWR is roughly 5 million cubic meters, which is equivalent to 2,000
Olympic-sized swimming pools. The 12 primary CWR constitute an estimated 98% of the total
CWR volume in the Lower Columbia River.
Each of the 12 primary CWR tributaries is shown in Figure 2-9 through Figure 2-20. On each
figure is a yellow pin showing the 'upstream extent,' which signifies how far upstream EPA
estimates salmon and steelhead will swim up the tributary when using it as a CWR (Appendix
12.4). Each figure includes the daily average temperature profile of both the Columbia River
(black) and the tributary (purple or green) to illustrate the difference in water temperatures over
time between the two (see Appendix 12.12 for location of temperature monitors). The bars
associated with the temperature profiles reflect the average diurnal range in temperature. Some
of the figures include a pink pin, which is the location of a PIT-tag antenna that records fish with
inserted PIT-tags if they swim past the receiver.
21

-------
Columbia River Cold Water Refuges Plan
Final January 2021
49 Cowlitz River
1.5 Million m3
CWR Volume
^—Cowlitz River {20015) - Average of 2001, 2002,2003, 2004, 2005,2006, 2007, 2008 - Bars Represent Average of Diurnal
Range
	Columbia River (5011) - Average of 1993,1995,1996,1997,1998,1999,2000, 2001, 2002, 2003 - Bars Represent
Average of Diurnal Range
Sample Date
Figure 2-9 Cowlitz River Cold Water Refuge and Associated Temperatures
*Yellow pin is estimated CWR upstream extent
24
22

-------
Columbia River Cold Water Refuges Plan
Final January 2021
*Yellow pin is estimated CWR upstream extent
Figure 2-10 Lewis River Cold Water Refuge
23

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Sandy River (19514) - Average of 2009, 2010 - Bars Represent Average of Diurnal Range
	Columbia River (9807) - Average of 1997,1998,1999, 2000, 2001, 2002, 2003, 2004, 2005, 2010, 2011 - Bars Represent
Average of Diurnal Range
Sample Date
Figure 2-11 Sandy River Cold Water Refuge and Associated Temperatures
*Yellow pin is estimated CWR upstream extent
24
24

-------
Columbia River Cold Water Refuges Plan
Final January 2021
91 Tanner Creek
1,700 m3CWR
Volume
Tanner Creek (21547) - Average of 2000, 2001, 2002, 2003, 2004, 2005 - Bars Represent Average of Diurnal Range
^—Columbia River (9794) - Average of 1995 through 2011 - Bars Represent Average of Diurnal Range
c c c c c c c c	aoMtaiwMMQOQDMMaaaaaciaaQa
777777777'iiii
HHHHNNtNrnrOlDOlN^COH'JNONWCOH^h.OfniDa)
H H H H (M (M N	H H H N N (N CO	H H H (N fNl (N (Nl
Sample Date
Figure 2-12 Tanner Creek Cold Water Refuge and Associated Temperatures
*Yellow pin is estimated CWR upstream extent
24
25

-------
Columbia River Cold Water Refuges Plan
Final January 2021
92_Eagle Creek
3,000 m3CWR
Volume
—Eagle Creek (20698) - Average of 2001, 2002, 2003, 2004, 2005, 2007, 2008 - Bars Represent Average of Dirunal Range
^—Columbia River (16436) - Average of 1996, 2006, 2007, 2008, 2009, 2010, 2011 - Bars Represent Average of Dirunal Range
Sample Date
*Yellow pin is estimated CWR upstream extent
Figure 2-13 Eagle Creek Cold Water Refuge and Associated Temperatures
26

-------
Columbia River Cold Water Refuges Plan
Final January 2021
96 Herman Creek
170,000 m3
CWR Volume
samsm
	Herman Creek (20672) - Average of 2000, 2001, 2002, 2003, 2004, 2005 - Bars Represent Average of Diurnal Range
Columbia River (19511) - Average of 2009, 2010, 2011 - Bars Represent Average of Dirunal Range
Sample Date
Figure 2-14 Herman Creek and Cove Cold Water Refuge and Associated Temperatures
*Yellow pin is estimated CWR upstream extent
24
27

-------
Columbia River Cold Water Refuges Plan
Final January 2021
100 Wind River
105,000m3
CWR Volume
Wind River (20290) - Average of 2000. 2001, 2002, 2004, 2005, 2006, 2007, 2009, 2010,2011 - Bars Represent Average of Dirunal Range
Columbia River (19511) - Average of 2009, 2010, 2011 - Bars Represent Average of Dirunal Range
Sample Date
Figure 2-15 Wind River Cold Water Refuge and Associated Temperatures
*Yellow pin is estimated CWR upstream extent
24
28

-------
Columbia River Cold Water Refuges Plan
Final January 2021
112_Little White
Salmon River
1.1 million m3
CWR Volume
*Yellow pin is estimated CWR upstream extent
24
4
2
	Little White Salmon (20395/22001/21021/21025/21026) - Average of 1996,1997, 1998,1999, 2000, 2001, 2006, 2007,
2008, 2011 - Bars Represent Average of Diurnal Range
Columbia River (19462) - Average of 2007, 2008, 2009, 2010, 2011 - Bars Represent Average of Diurnal Range
<3- r» o
—	— — — — — — — — — — QOQjOQDQJ3QjOQJDQDQjOQOQDQ.Q.Q.Q-Q.Q.
—	— — — — — — — — — — =3 Z3 =3 =3 Z3 D 3 =3 =3 Z3 <1J CL> 
-------
Columbia River Cold Water Refuges Plan
Final January 2021
aoogle earth
115 While Salmon River
150,000 m3
CWR Volume
*Yellow pin is estimated CWR upstream extent
24
6
4 	White Salmon (20293) - Average of 1995,1996,1997,1998,1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2009 - Bars
Represent Average of Diurnal Range
2	^—Columbia River (19462) - Average of 2007, 2008, 2009, 2010, 2011 - Bars Represent Average of Diurnal Range
— QDMOfiMQfiQflMQflcioMaaaaaaaaaa
_ Z3 Z3 Z3 Z3 Z3 Z3 Z3 Z3 Z3 Z3 (1) CL) (1) fi) fl) fl) <1J fl] fl) flJ
O m id m in in co
H H H H (N N (N
tH	O
crirMLncOT-i<3-r^orNLnoOT-i^r-«Oroioc?>
H *—I H fNJ Csl (N PO	H H H (N N N (N
Sample Date
Figure 2-17 White Salmon River Cold Water Refuge and Associated Temperatures
30

-------
Columbia River Cold Water Refuges Plan
Final January 2021
116 Hood River
28,000 m3
CWR Volume
'Yellow pin is estimated CWR upstream extent
Hood River (19466) - Average of 2006, 2008, 2009, 2010, 2011 - Bars Represent Average of Diurnal Range
Columbia River (19462) - Average of 2007, 2008, 2009, 2010, 2011 - Bars Represent Average of Diurnal Range
c c c c c £= c c c c ^	— — — — — QocioQcw)W)W)Qflciow)W)aaaaaaaaQ.a
77777777T>7ii'ii ¦ « ii«i<<<<<<<<<t/>(/)(/)(/)
TH^r^OMtDCT^rNJ^CO	HHHHtNNtNWMlDOllN^COH^rsO'NinCOH^lsOrOlDai
H H H rl (N (\| (N	H H H (N fM N m	H H H (N N N N
Sample Date
Figure 2-18 Hood River Cold Water Refuge and Associated Temperatures
31

-------
Columbia River Cold Water Refuges Plan
Final January 2021
125 Klickitat River
220,000 m3
CWR Volume
Klickitat River (19836) - Average of 2003, 2007, 2008, 2009, 2010 - Bars Represent Average of Diurnal Range
^—Columbia River (4936) - Average of 1998,1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011
- Bars Represent Average of Diurnal Range
Sample Date
Figure 2-19 Klickitat River Cold Water Refuge and Associated Temperatures
*Yellow pin is estimated CWR upstream extent
24
32

-------
Columbia River Cold Water Refuges Plan
Final January 2021
135 Deschutes River
880,000m3
CWR Volume
^—Deschutes River (USGSGauge 1410300) - Average of 2011, 2012, 2013, 2014, 2015, 2016- Bars Represent Average of
Diurnal Range
	Columbia River (16435) - Average of 1998,1999, 2000, 2001, 2002, 2003, 2004,2005, 2006, 2007, 2008, 2009, 2010,
2011 - Bars Represent Average of Diurnal Range
Sample Date
*Yellow pin is estimated CWR upstream extent
Figure 2-20 Deschutes River Coid Water Refuge and Associated Temperatures
33

-------
Columbia River Cold Water Refuges Plan
Final January 2021
3 SALMON AND STEELHEAD USE OF COLD WATER REFUGES
3.1 SALMON AND STEELHEAD MIGRATION TIMING AND COLUMBIA RIVER
TEMPERATURES
The date when fish migrate through the Lower Columbia River and the associated water
temperatures are significant factors in whether or not fish will use cold water refuges (CWR).
The migration timing of the salmon and steelhead species that migrate up the Columbia River
and pass Bonneville Dam each summer is displayed in Figure 3-1 along with the average
Columbia River temperature during that time. On average, temperatures in the Lower Columbia
River exceed 20°C from mid-July through mid-September and reach peak temperatures of
about 22°C in mid-August. The bulk and peak of the summer steelhead run (purple line) migrate
past Bonneville Dam during the two-month period when Columbia River temperatures exceed
20°C. The first half of the fall Chinook run (blue line) migrates past Bonneville Dam when
temperatures are above 20°C (fall Chinook are defined as Chinook passing Bonneville Dam
after August 1st). Accordingly, steelhead and fall Chinook are the species that most often
encounter warm Lower Columbia River temperatures and, as discussed later in this chapter, are
the species that use CWR the most to escape warm Columbia River temperatures.
Most of the sockeye (green line) and summer Chinook (yellow line) generally pass Bonneville
Dam and swim through the Lower Columbia River in June and early July, prior to the onset of
warm temperatures (summer Chinook are defined as Chinook passing Bonneville Dam between
June 1 and July 31). Accordingly, sockeye and summer Chinook are less likely to use CWR and
typically swim continuously through the Lower Columbia River. When the river does warm
earlier, coinciding with sockeye and summer Chinook fish runs, as it did in 2015, the use of
CWR is seen as an ineffective migration strategy for these fish. This appears to be because
delayed upstream migration by holding in CWR results in exposure to warmer mainstem
temperatures during their continued upstream migration as river temperatures continue to heat
up from early to mid-summer.
Due to their extensive use of CWR, this chapter focuses on characterizing summer steelhead
and fall Chinook use of CWR in the Lower Columbia River.
34

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Adult Salmon & Steelhead Passage at Bonneville Dam
June - September 2007-2016 Average
25000
20000
^	^	«s^
* v -i> -p "o & o v	rjF a ^
¦Fall Chi new
•Steelhead
Date
¦Sockeye
"Summer Chin
¦Temperature
Figure 3-1 Salmon and steelhead Bonneville Dam passage and temperature (DART)
3.2 COLUMBIA RIVER TEMPERATURES THAT TRIGGER COLD WATER REFUGE
USE
In the early 2000s, the University of Idaho's Department of Fish and Wildlife Sciences and
NMFS, under contract with the U.S. Army Corps of Engineers, conducted a series of salmon
and steelhead studies using radio-tagged fish to track movement and temperature during
migration up the Columbia River. These studies characterized salmon and steelhead use of
CWR in the Lower Columbia River. The study results have been summarized in several
scientific journals (Goniea et al. 2006, High et al. 2006, Keefer et al. 2009, Keefer et al. 2018)
and in the USAGE 2013 Report titled "Location and Use of Adult Salmon Thermal Refugia in the
Lower Columbia and Snake River" (USAGE 2013).
Figure 3-2 and Figure 3-3 show the relationship between Columbia River water temperature
and CWR use for steelhead and fall Chinook salmon (USAGE 2013). As shown in Figure 3-2,
migrating steelhead begin to use CWR when the Columbia River temperature reaches 19 C,
and when temperatures are 20 C or higher, approximately 60-80% of the steelhead use CWR.
As shown in Figure 3-3, fail Chinook initiate use of CWR at slightly warmer temperatures (20-
21°C), and about 40% use CWR when temperatures reach 21-22'C.
35

-------
Columbia River Cold Water Refuges Plan	Final January 2021
Mean temperature (°C) on reservoir entry date
Figure 3-2 Steeihead use of cold water refuge (black dots and 'Used tributaries' axis)
(Keefer et. al. 2009)
Mean weekly temperature (C)
Figure 3-3 Fall Chinook use of cold water refuge (Goniea et. al. 2006)
36

-------
Columbia River Cold Water Refuges Plan
Final January 2021
3.3 EXAMPLES OF SALMON AND STEELHEAD USE OF COLD WATER
REFUGES
It is enlightening to look at tracking study results for individual fish with internal temperature
sensors to illustrate how fish use CWR. Figure 3-4 shows the temperatures experienced by an
individual steelhead between the Bonneville Dam and The Dalles Dam. This steelhead quickly
swam from Bonneville Dam to the Little White Salmon River (Drano Lake) and stayed for
approximately two weeks, then rapidly swam up the Columbia River to the White Salmon River,
where it stayed for about five days before proceeding to pass The Dalles Dam. This figure
provides an example of how steelhead use CWR (in this case, for approximately three weeks)
to minimize their exposure to warm Columbia River temperatures as they wait for the river (gray
line) to cool before they continue their upward migration to spawn. Steelhead that use CWR
typically do so for 20 days in the Bonneville reservoir reach and for 2-6 days in the Dalles
reservoir reach based on research done in 2000 and 2002 (Keefer and Caudill, 2017).
Figure 3-4 Temperature profile of a steelhead using cold water refuges (Keefer & Caudill
2017)
Figure 3-5 shows another steelhead exhibiting a similar pattern of CWR use. This steelhead
used Herman Creek/Cove, the Little White Salmon River (Drano Lake), an unknown CWR
(potentially the mouth of the Klickitat River) between Bonneville Dam and The Dalles Dam. It
then took refuge in the Deschutes River CWR for a few days prior to proceeding up the
Columbia and Snake Rivers. Figure 3-5 shows how a steelhead can minimize its exposure to
elevated temperatures during its upstream migration in August by residing in a CWR and
continue migrating upstream in September when temperatures begin to cool.
37

-------
Columbia River Cold Water Refuges Plan
Final January 2021
24
Steelhead 24-73; tagged 8-8-2002 (DST 4603A)
22 -
20 -
18 -
ra 16 -
8.
I 14-1
(-
12 -
10 -
8
Unknown
V ' CWR
Deschutes
Thp
Dalles	Ice
Ualles McNary	ulc^
. '	Harbor Little
•	*4 nui lxji uiue
rNfM- L°^00056 - End
A*

Herman Creek
Lyons Ferry
		FiSh 30-minute
		BCN daily mean
	IHD daily mean
>1 >" >	MCN daily mean
! Aug
13 Aug
18 Aug
23 Aug
28 Aug
2 Sep
7 Sep
12 Sep
Figure 3-5 Temperature profile of a steelhead using cold water refuges (Keefer & Caudill
2017)
Figure 3-6 shows the temperature profile of a fall Chinook salmon. Fall Chinook salmon also
utilize CWR as part of their migration strategy, but for shorter periods than steelhead. Scientists
hypothesize that this is in part because fall Chinook spawn in the fall in upstream rivers and are
genetically driven to move to their spawning grounds in time to spawn (Goniea et al. 2006).
Conversely, steelhead spawn in the late winter and spring, so they have more time and flexibility
in their migration to reach their upstream spawning grounds (Keefer et al. 2009 and Keefer et al.
2018). The fall Chinook in Figure 3-3 used the Little White Salmon (Drano Lake), the White
Salmon and an unknown CWR area (potentially the Klickitat River) for a few days between
Bonneville Dam and The Dalles Dam, then found an unknown CWR area near McNary Dam.
Fall Chinook that use CWR typically do so for 1-2 days in the Bonneville reservoir reach and for
one day in the Dalles reservoir reach based on research done in 2000 and 2002, which were
relatively cool years during the Fall Chinook run. (Keefer and Caudill, 2017).
Figure 3-7 shows the temperature profile for a summer Chinook salmon. As reflected in Figure
3-1, summer Chinook salmon migrate past Bonneville Dam in June and July, typically prior to
the onset of warmer Columbia River temperatures. However, summer Chinook that pass
Bonneville Dam in late July, like the one shown in Figure 3-7, can be exposed to warm
Columbia River temperatures greater than 20°C. This summer Chinook used the Deschutes
River CWR for a brief time prior to proceeding upriver, which is typical for summer Chinook
CWR use (Keefer and Caudill, 2017). Summer Chinook salmon benefit less from using CWR,
since they migrate when Columbia River temperatures are rising. Thus, if a summer Chinook
held in a CWR, it would experience higher Columbia River temperatures during the rest of its
migration. It appears to be more advantageous for summer Chinook to quickly migrate through
the Lower Columbia River to avoid the warmest temperatures that generally occur in late July
and August. However, brief respites in CWR could provide some physiological benefit to
summer Chinook.
38

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Fall Chinook 25-429; tagged 8-14-2000 (DST 2650B)
Figure 3-6 Temperature profile of a fall Chinook salmon using cold water refuges (Keefer &
Caudill 2017)
Summer Chinook 10-145; tagged 7-22-2000 (DST 3547A)
Figure 3-7 Temperature profile of a summer Chinook salmon using cold water refuges
(Keefer & Caudill 2017)
39

-------
Columbia River Cold Water Refuges Plan
Final January 2021
3.4 NUMBER OF STEELHEAD IN COLD WATER REFUGES
In order to assess the sufficiency of CWR and to understand their importance to migrating
salmon and steelhead, it is important to estimate the number of fish using CWR. The research
conducted by the University of Idaho and NMFS demonstrates that salmon and steelhead move
in to CWR in the Lower Columbia River to avoid warm Columbia River temperatures. However,
there are no research studies estimating the number of salmon and steelhead that are in the
respective CWR areas.
EPA developed a method to estimate the number of steelhead in the CWR between Bonneville
Dam and The Dalles Dam by using daily passage counts of steelhead at these two dams from
DART. Figure 3-8 shows the average steelhead passage counts at each of the two dams and
the average Columbia River temperature at Bonneville Dam from 2007 to 2016. This figure
shows that as temperatures reach 20°C, many steelhead that pass Bonneville Dam in late July
and August (blue line) wait until September to pass The Dalles Dam (green line). Since more
steelhead are entering the Bonneville reach than leaving the reach during this time, it results in
an accumulation of steelhead within the Bonneville reach, which can be estimated. EPA
estimated the number of accumulated steelhead by summing the daily count of steelhead
passing Bonneville Dam minus the daily count passing The Dalles Dam and subtracting the
percentage of steelhead not expected to pass The Dalles Dam due to fishing harvest, straying,
and those returning to spawn in Bonneville reach tributaries. EPA estimated the percentage of
accumulated steelhead that is in the reservoir versus in CWR using scientific literature on the
relationship of temperature and the percentage of steelhead that enter CWR (Appendix 12.13).
Figure 3-9 shows the results of EPA's estimates of the number of steelhead in CWR within the
Bonneville reach in an average year (2007-2016). Up to approximately 80,000 steelhead
accumulate in the Bonneville reach in August. Of these, approximately 68,000 (85%) are
estimated to be in CWR at the same time during the peak period of use. The peak occurs in the
latter half of August since steelhead continue to accumulate within the reach until about the first
of September. At this time, temperatures cool to the point that more steelhead are exiting the
reach by passing The Dalles Dam than entering the reach by passing the Bonneville Dam as
shown in Figure 3-8 (Appendix 12.13).
To corroborate the EPA approach to estimating the number of steelhead in the Bonneville reach
CWR, empirical data from the University of Idaho was evaluated (M. Keefer, personal
communication, August 31, 2017). Figure 3-10 shows the daily location of 219 recorded
steelhead as they migrate through the Bonneville reach. As shown, on a given day when
Columbia River temperatures typically exceed 20°C, the vast majority of steelhead (80-90%) are
in CWR and only a portion are in the Columbia River. Further, the peak accumulation of
steelhead in CWR occurred in the latter half of August/early September. Thus, the EPA
estimation approach matches the pattern and percentage of radio-tagged steelhead in
Bonneville reach CWR very closely.
The volume of water in the Bonneville reach of the Columbia River is approximately 600,000
acre-feet, and the total volume of CWR in this reach is about 1,453 acre-feet, which means that
in late August and early September approximately 80-85% of the steelhead are in 0.2%
available water in this reach.
40

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Steelhead Passage Bonneville Dam vsThe Dalles Dam
10 Year Average (2007-2016)
9000
25
8000
ro 6000
"O
20
00
15 -6
O)
=j
rs
10
Date
¦ Bonneville Dam
¦The Dalles Dam
¦Temperature
Figure 3-8 Steelhead passage at Bonneville Dam and The Dalles Dam (Appendix 12.13)
41

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Average Number of Steelhead in Bonneville Reach
(2007-2016)
T3
fO

-Q
£
1-Jun	1-Jul	31-Jul
Accumulated in BON Reach
29-Sep
•In Reservoir
2 9-Oct
Figure 3-9 Estimated number of steelhead in Bonneville reach cold water refuges (Appendix
12.13)
80
70
60
50
40
0)
¦g 30
3
z 20
10
0
Daily estimates












_w-sn vvA/v/


-In reach
•In CWR
Not in CWR
14-May 3-Jun 23-Jun 13-Jul 2-Aug 22-Aug 11-Sep 1-Oct 21-Oct 10-Nov
Figure 3-10 Proportion of 219 radio-tagged steelhead in Bonneville cold water refuges (M.
Keefer, personal communication, August 31, 2017)
42

-------
Columbia River Cold Water Refuges Plan
Final January 2021
EPA applied a simplified approach to estimate the number of steelhead in the Bonneville reach
CWR for individual years from 1999 through 2016, which is shown in Table 3-1 (Appendix
12.13). The simplified approach estimates the peak number of steelhead that accumulate in the
Bonneville reach by taking the number of steelhead that would pass The Dalles Dam for the
July 15 - August 30 period if steelhead were not using CWR (expected to pass) and subtracting
the number of steelhead that actually pass The Dalles Dam during this period. Of the number of
accumulated steelhead in the Bonneville reach during the peak accumulation period (late
August), 85% were assumed to be in CWR (Appendix 12.13).
As shown in Table 3-1, the number of steelhead in CWR varies year to year and is primarily a
function of the size of the steelhead run (number passing Bonneville Dam) and the Columbia
River temperature. During a year with a large steelhead run and warm Columbia River
temperatures (2009), 155,000 steelhead are estimated to be in the Bonneville reach CWR.
During a year with a small steelhead run and cool Columbia River temperatures (2012), only
23,000 steelhead are estimated to be in CWR.
Table 3-1 Estimated number of steelhead in cold water refuges each year (1999-2016)
(Appendix 12.13)




Measured %
Expected



Avg
Passed
Passed
That Passed
to Passed



Temp
BON
Dalles
Dalles
Dalles
In BON Reach
In CWR (85%)
Year
July 15-Aug 31
July 15-Aug 31
July 15-Aug 31
June 1-Oct 31
July 15-Aug 31
Peak
Peak
2016
21.4
83,919
24,212
80%
66,868
42,656
36,258
2015
21.8
165,138
69,059
84%
137,893
68,834
58,509
2014
21.5
175,686
70,488
80%
140,923
70,435
59,869
2013
21.5
166,926
68,949
83%
138,059
69,110
58,743
2012
20.1
142,032
95,612
86%
122,797
27,185
23,107
2011
19.5
252,331
176,573
82%
207,452
30,879
26,248
2010
21.0
231,804
121,974
82%
189,445
67,471
57,350
2009
21.6
451,509
205,163
86%
388,094
182,931
155,492
2008
20.0
225,506
117,044
79%
177,048
60,004
51,004
2007
21.1
229,124
83,820
76%
173,420
89,600
76,160
2006
21.1
187,415
53,379
12%
134,561
81,182
69,005
2005
21.4
175,028
55,866
77%
135,090
79,224
67,340
2004
22.0
155,516
42,744
78%
120,905
78,161
66,437
2003
21.7
209,328
58,083
77%
160,904
102,821
87,398
2002
20.4
257,857
131,121
82%
210,238
79,117
67,250
2001
20.7
397,879
169,554
80%
319,544
149,990
127,491
2000
20.6
164,593
75,954
75%
124,114
48,160
40,936
1999
20.0
136,136
76,782
77%
104,458
27,676
23,524








Average
20.9
219,048
98,363

175,585
77,222
65,639
Table 3-2 includes the estimated number of steelhead in each of the eight CWR in the
Bonneville reach between Bonneville Dam and The Dalles Dam using the CWR volumes from
Table 2-2 and Table 2-3 as an approximate indicator of the distribution of steelhead in the eight
CWR. Over half of the steelhead (61%) are expected to be in the Little White Salmon (Drano
Lake) CWR with approximately 40,000 steelhead during the peak period for an average year,
43

-------
Columbia River Cold Water Refuges Plan
Final January 2021
with peaks ranging from 14,000 to 95,000 steelhead in low and high years. Other Bonneville
reach CWR tributaries with extensive steelhead CWR include Herman Creek, White Salmon
River, Wind River, and the Klickitat River.
Table 3-2 Estimated number of steelhead in each Bonneville reach cold water refuge
(Appendix 12.13)


Plume
Stream
Total

#Steelhead in
#Steelhead
#Steelhead


CWR
CWR
CWR
% of CWR
Each CWR
in Each CWR
in Each CWR

Tributary
Volume
Volume
Volume
in BON
(1999-2016
High Year
Low Year
Tributary Name
Temp
(>2°CA)
(> 2°C A)
(> 2°C A)
Reach
Avg)
(2009)
(2012)

C
m3
m3
m3




Eagle Creek
15.1
2,100
888
2,988
0.2%
109
259
39
Rock Creek
17.4
530
1,178
1,708
0.1%
63
148
22
Herman Creek
12.0
168,000
1,698
169,698
9.5%
6,216
14,726
2,188
Wind River
14.5
60,800
44,420
105,220
5.9%
3,854
9,131
1,357
Little White Salmon River
13.3
1,097,000
11,661
1,108,661
61.9%
40,613
96,208
14,297
White Salmon River
15.7
72,000
81,529
153,529
8.6%
5,624
13,323
1,980
Hood River
15.5
28,000
0
28,000
1.6%
1,026
2,430
361
Klickitat River
16.4
73,000
149,029
222,029
12.4%
8,133
19,267
2,863
Total

1.501.430
290,403
1.791.833
100%
65,639
155,492
23.107
To corroborate the EPA approach to estimate the number of steelhead in each CWR, empirical
data from the University of Idaho was evaluated (M. Keefer, personal communication,
September 11, 2017). Table 3-3 shows the distribution of 59 radio-tagged steelhead in the
Bonneville reach CWR on August 31, which represents the time of peak CWR use. The
distribution in Table 3-3 is generally consistent with predicting the number of steelhead in each
CWR based on volume shown in Table 3-2, with a large percentage (68%) of the steelhead in
the Little White Salmon River (Drano Lake) and a significant percentage (greater than 7%) in
Herman Creek, White Salmon River, and the Klickitat River CWR.
Table 3-3 Distribution of radio-tagged steelhead in the Bonneville reach cold water refuges
on August 31 (Combined 2000/2001 Data Set) (M. Keefer, personal communication, September
11, 2017)
CWR Locution
31-Aug
%
Predicted based
on CWR Volume
Herman Creek
6
10%
10%
Wind River
1
2%
6%
Little White Salmon/Drano Lake
40
68%
62%
White Salmon
4
7%
9%
Klickitat River
4
7%
12%
Unknown CWR
4
7%

Total
59 Steelhead


Table 3-4 shows the estimated density of steelhead in the Bonneville reach CWR under
different run size scenarios (average, high, low) and for two volume metrics of CWR (volume
44

-------
Columbia River Cold Water Refuges Plan
Final January 2021
that is 2°C cooler than the Columbia River and volume that is 18°C or cooler). The density is
estimated by dividing the estimated number of steelhead by the CWR volume. The density
associated with CWR volume that is 18°C or cooler may be a better indicator of density that fish
actually experience, because steelhead residing for an extended period are likely to seek
temperatures below 18°C. The maximum estimated density of steelhead is 0.16 steelhead per
cubic meter, which is 404 steelhead in an Olympic-sized swimming pool (Appendix 12.13). EPA
identified one fish per cubic meter as the maximum potential density from studies on adult
Chinook salmon and steelhead held in confined spaces (Berejikian et al. 2001, Hatch et al.
2013). Thus, using this comparison metric, the CWR volume in the Bonneville reach appears
sufficient for the number of steelhead using CWR in this portion of the Columbia River.
However, this comparison should be viewed with caution due to the different context and small
number of fish in the studies noted above and other unknown factors that may affect the
carrying capacity of CWR.
Table 3-4 Estimated steelhead density in cold water refuges (Appendix 12.13)

CWR Volume (> 2'C A

CWR Volume (< 18"C


Average
High
Low
Ave rage
High
Low

1999-2016
2009
2012
1999 -2016
2009
2012
#fish/m3
0.0366
0.0868
0.0129
0.0683
0.1617
0.0240
# fish/2500 m3
92
217
32
171
404
60
An analysis of PIT-tagged steelhead passing Bonneville and The Dalles Dams conducted by
Brian Maschhoff provides an additional corroborating line of evidence on the extent to which
steelhead use CWR in the Bonneville reach (see Appendix 12.13). This analysis shows
considerable delay and presumed CWR use by most steelhead during warm river temperatures.
This analysis also indicates that there is not a difference in the extent of steelhead delay and
presumed CWR use between hatchery and wild steelhead and steelhead that were transported
as juveniles versus those that were not (Appendix 12.13).
3.5 NUMBER OF FALL CHINOOK IN COLD WATER REFUGES
EPA used the methods described above for steelhead to estimate the number of fall Chinook
using CWR in the Bonneville reach. As shown in Figure 3-11, the estimated number of fall
Chinook in CWR (blue line) is estimated to be approximately 5,000 during the last week of
August and the first two weeks of September for an average year (2008-2017) (Appendix
12.13). This figure shows that, unlike steelhead, the majority of fall Chinook in the Bonneville
reach are estimated to be migrating in the reservoir. After mid-September, the number of fall
Chinook passing Bonneville Dam begins to decrease and the accumulated number of fall
Chinook in the reach begins to decrease as temperatures fall to 20°C and below.
In warmer years such as 2013, when temperatures remain above 21 °C into early September
during the peak of the fall Chinook run, EPA estimates a higher proportion of fall Chinook will
use CWR within the Bonneville reach to avoid mainstem temperatures. As shown in Figure
3-12, 20,000 to 40,000 fall Chinook are estimated to have been in the Bonneville reservoir CWR
in 2013 in the latter part of August through mid-September (blue line). This is four to eight times
45

-------
Columbia River Cold Water Refuges Plan
Final January 2021
the estimated number of 5,000 fall Chinook in CWR in an average year (see Figure 3-11). Late
August and early September temperatures were consistently around 22°C in 2013, which are
temperatures at which a significant number of fall Chinook seek CWR. 2013 also represents a
relatively high run year with 953,222 adult fall Chinook passing Bonneville Dam, which is about
twice the 10-year (2007-2016) annual average of 504,148 (FPC 2014 & 2016 Annual Report).
There is more uncertainty in the estimates of the number of fall Chinook in the Bonneville
reservoir CWR compared to the estimates of steelhead in CWR because fall Chinook use CWR
for a shorter duration (Appendix 12.13).
Average Number of Fall Chinook in Bonneville Reservoir
(2008-2017)
50,000
45,000
40,000
g 35,000
5 30,000
™ 25,000
o
5 20,000
-Q
E
3 15,000
¦Z.
10,000
5,000
0
1-Aug	31-Aug	30-Sep	30-Oct
Accumulated in BON Reach	In CWR	In Reservoir
Figure 3-11 Accumulation of fall Chinook in the Bonneville reach and the number of fall
Chinook in cold water refuges (2008-2017 average) (Appendix 12.13)
46

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Number of Fall Chinook in Bonneville Reach
(2013)
120,000
100,000
o
° 80,000
_£Z
u
* 60,000
M—
0

-------
Columbia River Cold Water Refuges Plan
Final January 2021
summer maximum temperatures. Conversely, as shown in Figure 3-13, there is not a significant
delay over The Dalles Dam in the decade after The Dalles Dam was built (1957-1966). Limited
temperature data collected in the 1950s depicted in Figure 3-14 shows summer peak
temperatures were lower compared to current day temperatures. Current daily average
temperatures exceed 20°C for about two months and exceed 21 °C for one month, but during
the 1950s daily average temperatures typically exceeded 20°C only for a short period (a week)
and did not exceed 21 °C. As described earlier, >20°C temperatures are associated with a high
level of CWR use by steelhead. These data suggest steelhead use of CWR in the Bonneville
reach was historically less than what we observe currently, and that steelhead are using CWR
more today in response to increased summer temperatures of the Lower Columbia River.
Steelhead Passage Bonneville Dam vs the Dalles Dam
10 Year Average (1957-1966)
4500
Bonneville Dam	The Dalles Dam
Figure 3-13 Steelhead passage at Bonneville Dam and The Dalles Dam, 1957-1966 (DART)
48

-------
Columbia River Cold Water Refuges Plan
Final January 2021
24
22
U
£ 20
5 18
2
I 16
a 14
I—
12
10
Date
^—1950-59 Daily Avg ^—2007-2016 Daily Avg
Figure 3-14 Current versus 1950s water temperatures in the Lower Columbia River (DART)
3.8 DESCHUTES RIVER COLD WATER REFUGE USE
The discussion above in Sections 3.4 - 3.7 characterizes the use of CWR by steelhead and fall
Chinook in the Bonneville reach between Bonneville Dam and The Dalles Dam. Upstream of
The Dalles Dam, the only other significant and primary CWR in the Lower Columbia River is the
Deschutes River. The Deschutes River is unique in that it has a PIT-tag detector, installed in
2013 near the mouth, which NMFS has used to analyze the extent that steelhead use the
Deschutes River for CWR (NMFS 2017a). Table 3-5 shows that an average of 873 PIT-tagged
steelhead were recorded in Deschutes River CWR comprised mostly of Snake River (61%) and
Middle Columbia steelhead (30%).
Table 3-5 Deschutes River mouth steelhead PIT-tag detections by calendar year and
Distinct Population Segment (DPS) (NMFS 2017a)
DPS
2013
2014
2015
Average
%
Lower Columbia
9
5
31
15
2%
Middle Columbia
174
214
385
258
30%
Snake River
541
506
540
529
61%
UpperColumbia
74
54
86
71
8%
Total
798
779
1042
873

Table 3-6 shows the number of Snake River PIT-tagged steelhead detected at The Dalles Dam
and the percentage of those steelhead detected at the Deschutes River mouth. Approximately
Bonneville Forebay Daily Average Temperatures


-------
Columbia River Cold Water Refuges Plan
Final January 2021
14% (12-18%) of the Snake River steelhead detected at The Dalles Dam were recorded in the
Deschutes River mouth. Extrapolating to all Snake River steelhead (including non-Pit-tagged
steelhead), Table 3-6 shows that NMFS' estimated total number of Snake River steelhead using
Deschutes River CWR in an average year is 27,659 (NMFS 2017a). Assuming 61% of all
steelhead in Deschutes River CWR are Snake River steelhead as presented in Table 3-5, the
total number of steelhead using the Deschutes River CWR in an average year is 45,343.
Table 3-6 Percent of Snake River (SR) steelhead using Deschutes cold water refuges and
number of steelhead using Deschutes cold water refuges (NMFS 2017a)


%of SRPIT



SR PIT tagged
tagged
Estimated
Estimated

Steelhead
Steelhead
Number of Total
Number of All

Detected @
Detected at
SR Steelhead in
Steelhead in

Dalles Dam
Deschutes
Deschutes CWR
Deschutes CWR
2013
2977
18%
26,162
42,889
2014
4201
12%
30,332
49,725
2015
3279
13%
26,483
43,415
Average
3486
14%
27,659
45,343
Figure 3-15 shows how many Snake River steelhead are estimated to be within Deschutes
River CWR for each month. As depicted in Figure 3-15, the peak period of use was September
in 2013 and 2014 and in August in 2015. During this peak period of use, approximately 10,000
to 16,000 Snake River steelhead were in the Deschutes River CWR. Assuming 61% of all
steelhead in Deschutes River CWR are Snake River steelhead, the total number of steelhead
using the Deschutes River CWR during the peak period of use is 16,000 to 26,000. 26,000
steelhead in the Deschutes River CWR would equate to a density of 0.087 steelhead per square
meter, which is the same upper range density estimated for Bonneville Reach CWR (based on
>2°C delta volume of CWR) reflected in Table 3-4.
As noted above, the overall percentage of Snake River steelhead that use the Deschutes River
as CWR is 12-18%. In August, during peak river temperatures, the percentage rises to near
25% (NMFS 2017a). This percentage is less than the percentage of steelhead that use
Bonneville Reach CWR, which is up to about 85% during peak temperatures. There are several
possible reasons for this lower percentage of use of the Deschutes River: 1) the percent of
steelhead using the Deschutes River reported here does not capture use of the Deschutes
plume only; 2) the Deschutes River is just one CWR on one side of the river and the Bonneville
Reach CWR consists of 7 primary CWR; and 3) steelhead are encountering the Deschutes
River after many have already spent time in CWR in the Bonneville Reach and later in the
summer as the Lower Columbia River begins to cool. Nonetheless, the Deschutes River is a
heavily used CWR and is the only primary CWR between The Dalles Dam and McNary Dam.
50

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Estimated numbers of PIT tagged Snake River Steelhead and total numbers present by
month in the Deschutes River based on detections at the Deschutes River mouth detector.
Population estimates are based on PIT tagging rates observed at tower Granite
350
18,000
k2Q0
: 150
SO
2013	tags
2014	tags
2015	tags
2013	pop est.
2014	pop. est
2015	pop est.
16,000
14,000
12,000 tjj
1
10,000
6,000
4,000
2,000
June	July
August September October November December January February
April
May
Figure 3-15 Estimated number of PIT-tagged Snake River steelhead and estimated total
number of Snake River steelhead (estimated by tag expansion) present in Deschutes River cold
water refuges by month 2013-2015 (NMFS 2017a)
3.9 USE OF CWR BY SPECIFIC POPULATIONS OF STEELHEAD AND FALL
CHINOOK
The specific populations of steelhead and fall Chinook that use CWR the most are those with
run timing that coincides with the warmest Columbia River temperatures. Figure 3-16 shows the
percent of specific steelhead populations that use CWR (solid circles and x-axis) and the
populations' median passage time (y-axis), which reflect how long individuals from each
population spend in CWR. Those steelhead populations in the upper right in Figure 3-16 use
CWR extensively while those populations in the lower left use CWR less. Figure 3-17 shows
the migration timing for the various steelhead populations, which shows that those steelhead
populations with high CWR use are those where a high proportion of the population migrates
through the Lower Columbia River when temperatures are warmest (i.e., late July through late
August as reflected in the shaded area). Steelhead populations from the John Day, Umatilla,
Grande Ronde, Imnaha, Yakima, Snake, Salmon, and Walla Walla Rivers all use CWR to a
51

-------
Columbia River Cold Water Refuges Plan
Final January 2021
significant extent. The steelhead populations that use CWR the least are those that mostly
migrate through the Lower Columbia River before (Tucannon, Hanford, and Lyons Ferry) or
after (Clearwater) the warmest temperatures.
Figure 3-16 Percent of population-specific steelhead that used cold water refuges for >12
hours (solid circles) and associated median passage time from Bonneville Dam to the John Day
Dam for those that used and did not use (clear circles) CWR. TUC, Tucannon River; HAN,
Hanford Reach; LFH, Lyons Ferry Hatchery; UCR, Upper Columbia River; WWR, Walla Walla
River; CWR, Clearwater River; SAL, Salmon River; SNK, Snake River above Lower Granite
Dam; YAK, Yakima River; IMR, Imnaha River; GRR, Grande Ronde River; UMA, Umatilla River;
JDR, John Day River. (Keefer et al. 2009)
52

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Clearwater -
Salmon -
John Day -
Umatilla -
Snake > LGr -
c	Deschutes -
o
"(5 Grande Ronde -
q. Upper Columbia -
o
O- Walla Walla -
Imnaha -
Lyons Ferry -
Yakima -
Tucannon -
Hanford -
19 June 9 July 29 July 18 Aug. 7 Sept. 27 Sept. 17 Oct.
Date at Bonneville Dam
Figure 3-17 Median timing distributions (median, quartiles, and 10th and 90th percentiles) at
Bonneville Dam for steelhead that successfully returned to tributaries or hatcheries. Vertical
dotted lines show mean first and last dates that Columbia River water temperatures were 19°C;
the shaded area shows dates with mean temperatures >21 °C. (Keefer et al. 2009)
Similarly, those populations of fall Chinook that migrate through the Lower Columbia River in
August and early September use CWR the most. Figure 3-18 depicts the composition of the fall
Chinook run by date. Fall Chinook are classified as Chinook that pass Bonneville Dam after
August 1st. Radio-tag studies of fall Chinook use of CWR mirrors the composition of different fall
Chinook populations migrating past Bonneville Dam in August and early September. Hanford
reach fall Chinook and fall Chinook populations above Priest Rapids Dam were most
predominately in CWR, with lesser numbers of Snake River and Yakima fall Chinook (US Army
Corps, 2013). It should be noted, however, that the data in Figure 3-18 is from 1998 and the
early 2000s, and the composition of the fall Chinook populations may be different today. In
particular, the Snake River fall Chinook population has increased, so today we might expect a
higher proportion of Snake River fall Chinook using CWR.
	1	*
	1	•	h
	1	•	
—I	*	1—
	1	•	1—
	1	*	1	
	1	•	1	
	1	•	1	
	1	•	1	
	t	*	t	
	•	1	
	•	1	
•	1—
-•	1	
53

-------
Columbia River Cold Water Refuges Plan
Final January 2021
3
p
i—
o>
CL
Date at Bonneville Dam
Figure 3-18 Mean composition of upriver bright fall-run Chinook salmon at Bonneville Dam
using five-day intervals based on release dates of radio-tagged fish. 1998 and 2000-2004.
MCB-BPH = mid-Columbia River bright-Bonneville Pool hatchery stock. (Jepson et al. 2010)
> Priest Rapids Dam
60 -
:
40 -
Sriake
Deset>utes
54

-------
Columbia River Cold Water Refuges Plan
Final January 2021
4 TEMPERATURE AND FISH HARVEST IMPACTS ON MIGRATING
SALMON AND STEELHEAD
¦ I ADVERSE TEMPERATURE EFFECTS TO MIGRATING ADULT SALMON AND
STEELHEAD
Water temperatures significantly affect salmon and steelhead health and survival, since they are
ectothermic (cold-blooded) with their internal body temperature closely tracking river
temperatures. Salmon and steelhead experience harmful health effects when exposed to warm
water temperatures above their optimal range. Optimal temperatures for migrating adult salmon
and steelhead are in the 12-16°C range with minimal adverse effects below 18°C (EPA 2003).
Both the States of Oregon and Washington have a 20°C maximum water quality criteria for the
Lower Columbia River, which is consistent with EPA's recommended numeric criteria for large
mainstem rivers that naturally warm to this level and are used by salmon and steelhead for
migration (EPA 2003).
Table 4-1 summarizes the adverse effects to migrating adult salmon and steelhead in the Lower
Columbia River as temperatures rise above 18°C. The temperature ranges in Table 4-1
represent average river temperatures with multiple day exposure. In general, as temperatures
rise, disease risk, stress, energy loss, avoidance behavior, and mortality rates increase.
Sockeye are most susceptible to warm temperatures with limited mortality at 19-20°C and
significant mortality at 20-21 °C. Steelhead are also susceptible to these temperature ranges but
exhibit avoidance behavior by seeking cold water refuges (CWR) as is demonstrated in this
Plan. Chinook are more tolerant to warm temperatures, with avoidance behavior (seeking CWR)
and mortality occurring at higher temperatures (21-22°C and higher).
In other portions of this Plan, documented research on the effects summarized in Table 4-1 is
provided, specifically Chapter 2 related to avoidance behavior and CWR use and sections 4.2,
4.5, and 4.6 related to mortality, energy loss, and shifts in migration timing.
55

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Table 4-1 Summary of temperature effects to migrating adult salmon and steelhead in the
Lower Columbia River (EPA 2003; McCullough 1999, Richter and Kolmes 2005)
Temperature Range
Effects
Less than 18°C
~ Minimal effects to salmon and steelhead
18-20°C
~	Elevated disease risk
~	Low proportion of steelhead seek CWR
~	Slight increase in sockeye mortality
20-21 °C
~	Significant disease risk
~	Increased stress and energy loss
~	Majority of steelhead seek CWR
~	Significant sockeye mortality
~	Low proportion of Chinook seek CWR
21-22°C
~	High disease risk
~	High stress and energy loss
~	High percentage of steelhead move into CWR
~	High sockeye mortality
~	Moderate proportion of Chinook seek CWR
22-23°C
~	Very high disease risk
~	Very high stress and energy loss
~	Very high percentage of steelhead move into CWR
~	Very high sockeye mortality
~	Significant proportion of Chinook seek CWR
23-24°C
~	Very high disease risk
~	Very high stress and energy loss
~	High avoidance behavior for steelhead and all salmon
~	High mortality for steelhead and salmon species
4.2 RELATIONSHIP BETWEEN TEMPERATURE AND MIGRATION SURVIVAL OF
ADULT STEELHEAD AND FALL CHINOOK SALMON
The survival rates of migrating adult salmon and steelhead between Bonneville Dam and
McNary Dam can be estimated by comparing the passage counts at each of the dams. The Fish
Passage Center conducted an analysis of the survival rates between these two dams as a
function of Columbia River water temperature. Figure 4-1 shows that the survival rate for
steelhead (PIT-tagged 2003-2015) decreases at 18°C temperatures and higher, and that a 10%
reduction in survival occurs at 21-22°C temperatures compared to 18°C and below
temperatures. Figure 4-2 shows the survival rates for fall Chinook at three different temperature
ranges (below 20°C, 20-21 °C, and >21 °C) with a decline in survival with warmer temperatures.
There is approximately a 7-8% decrease in survival for temperature >21 °C versus below 20°C.
Figure 4-2 also shows that adults that were transported in barges down the Columbia River as
juveniles have less survival than those that migrated downstream in the Columbia River.
56

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Upper Columbia & Snake combined
Temperature
Figure 4-1 Estimated survival rate of adult steelhead between Bonneville Dam and McNary
Dam (FPC, October 31, 2016 Memo)
o 80-
0.75-
0.70
nJ
>
t 0.65-
W
060
3s
Range
$ Below 20C
$ Between 20-21C
^3 Over 21C
0.55-
0.50-
In-River	Transported
Juv.Transport
Figure 4-2 Estimated survival rate of adult fall Chinook between Bonneville Dam and
McNary Dam (FPC, May 8, 2018 Memo)
57

-------
Columbia River Cold Water Refuges Plan
Final January 2021
The results shown in Figure 4-1 and Figure 4-2 indicate that the migration survival of an
individual steelhead or a fall Chinook salmon between Bonneville Dam and McNary Dam
decreases by 7-10% as temperatures rise above 21 °C. It should be noted that other factors,
such as increased harvest of fish that moved into CWR due to the rise in temperature, could be
contributing to the decreased survival rates.
4.3 FISHING HARVEST OF SALMON AND STEELHEAD IN COLD WATER
REFUGES
As noted above in Section 4.2, the correlation between increased Columbia River temperature
and decreased migration survival of adult steelhead and fall Chinook in the Lower Columbia
River could also be associated with increased fishing harvest in CWR at warmer Columbia River
temperatures. Fishing harvest in CWR also makes it difficult to directly measure the benefits of
CWR to migrating adult salmon and steelhead.
Keefer et al. (2009) analyzed the migration success of steelhead that used CWR versus those
that did not use CWR. This study found that migration success to the spawning tributaries for
those steelhead (wild and hatchery) that used CWR was about 8% less than those steelhead
that did not use CWR, which initially suggests CWR use is not beneficial. However, the study
also indicated that fishing harvest in CWR likely explained the decreased survival. Wld
steelhead using CWR, which are required to be released when caught, experienced a 4.5%
decrease in survival during migration to their spawning tributaries compared to wild steelhead
that did not use CWR. This increased mortality, however, could be associated with catch and
release mortality and illegal catch of wild steelhead in CWR (Keefer et al. 2009).
Another confounding variable is salmon and steelhead that were transported (barged)
downstream as juveniles have a higher rate of straying and lower adult survival rates between
Bonneville and McNary dams (see Figure 4-2). Some of these strays could be in CWR, thus
juvenile transportation could explain some of the lower adult survival rates for fish that use
CWR. Further, fish that use CWR may be more susceptible to warm temperatures and may
have higher mortality in the mainstem than fish that don't use CWR (Keefer et al. 2009).
NMFS (2017a) also found that the survival rate for steelhead (wild and hatchery) from The
Dalles Dam to McNary Dam was about 9% less for those steelhead that used CWR (detected in
the Deschutes River) versus those that did not use CWR. NMFS assessment also provided data
on fish harvest in the Deschutes River that appears to explain the reduced survival for those
steelhead using CWR.
Due to fishing harvest in CWR, it is difficult to directly measure the extent to which steelhead
and fall Chinook CWR use may lead to higher migration survival rates due to avoidance and
minimization of exposure to warm Lower Columbia River temperatures. Similarly, it is difficult to
separate how much of the observed 7-10% decrease in steelhead and fall Chinook survival in
the Lower Columbia River when temperatures exceed 21 °C is due to temperature effects versus
fishing harvest. More sophisticated studies, perhaps during periods with no fishing, would likely
be needed to accurately answer these questions quantitatively.
58

-------
Columbia River Cold Water Refuges Plan
Final January 2021
4.4 SNAKE RIVER STEELHEAD AND FALL CHINOOK MIGRATION SURVIVAL
RATES IN THE LOWER COLUMBIA AND LOWER SNAKE RIVERS
Section 4.2 assessed the impact that river temperatures have on the survival rate of individual
steelhead and fall Chinook. This section looks at the survival rate in the Lower Columbia River
for ESA-listed Snake River steelhead and fall Chinook runs to ascertain if elevated
temperatures may be contributing to decreased survival rates. NMFS calculates the survival
rates of ESA-listed salmon and steelhead in the Lower Columbia River each year for the whole
run. As shown in Figure 3-17 and Figure 3-18 above, the Snake River steelhead run passes
Bonneville Dam from July through September, and the Snake River fall Chinook run passes
Bonneville Dam from August through early October, respectively. Thus, a portion of these runs
migrate through the Lower Columbia River when water temperatures exceed 20°C, while a
portion of the runs migrate through when temperatures are below 20°C.
Figure 4-3 shows the "adjusted" survival rate for Snake River steelhead between Bonneville
Dam and McNary Dam and between Bonneville Dam and Lower Granite Dam on the Snake
River for each year (2008-2017). "Adjusted" denotes the survival rate, factoring in the estimated
percentage of fish that are harvested or stray. Therefore, adjusted survival highlights the
percentage of fish that do not survive for unknown reasons. As shown in Figure 4-3, the ten-
year average adjusted survival rate from Bonneville Dam to McNary Dam is 94% (range of 90 to
100%) and from Bonneville Dam to Lower Granite Dam is 87% (range 81 to 94%). These data
indicate that there is an average of 6% unexplained mortality of adult Snake River steelhead
migrating between the Bonneville and McNary Dams and an additional 7% unexplained
mortality between McNary Dam and Lower Granite Dam. Part of this unexplained mortality is
likely attributable to mortality associated with prolonged exposure to Columbia River
temperatures above 20-21 °C during the upstream migration as has been observed to occur
(see Figure 4-1). Absent detailed studies, this 6% migration mortality rate appears to be equal
for hatchery and wild steelhead. For context, the estimated Snake River steelhead harvest
(primarily for hatchery steelhead) between Bonneville Dam and McNary Dam (Zone 6) is
approximately 15%, and the estimated stray rate is 5%.
It also should be noted that there is year-to-year variability in unexplained adult steelhead
mortality between the Bonneville and McNary Dams, with some years near 10% mortality (e.g.,
2009, 2011,2013, and 2017). Also, these data represent an average of all Snake River
steelhead populations, and some individual populations could have higher unexplained
mortality, especially if the majority of their migration occurs during peak summer temperatures
(see Figure 4-7).
59

-------
Columbia River Cold Water Refuges Plan
Final January 2021
100.0%
90.0%
80.0%
K 70.0%
<
cc
^ 60.0%
>
3 50.0%
i/i
2
| 40.0%
z
30.0%
20.0%
10.0%
0.0%


1
2008 2009 2010
^1
2011

!


2012 2013
YEAR
2014 2015 2016 2017
~ BON to MCN ¦ BON to LGR
Figure 4-3 Adjusted survival estimates of adult Snake River steelhead between Bonneville
Dam (BON) and McNary Dam (MCN) and between Bonneville Dam and Lower Granite Dam
(LGR) for the whole run (NMFS, 2019)
Figure 4-4 shows the "adjusted" survival rate for the Snake River fall Chinook run between
Bonneville Dam and Lower Granite Dam for each year (2008-2016). The average adjusted
survival for Snake River fall Chinook between Bonneville Dam and Lower Granite Dam is 90%,
which means there is 10% unexplained mortality of adult Snake River fall Chinook migrating
between the two dams. About half (5%) of this mortality occurs between Bonneville Dam and
McNary Dam, and half (5%) occurs between McNary Dam and Lower Granite Dam and likely is
the same rate for both hatchery and wild Snake River fall Chinook. In some years, the survival
rate is 80%, with 20% unexplained mortality (2011, 2013, 2016) between Bonneville Dam and
Lower Granite Dam. Part of this unexplained mortality is likely associated with prolonged
exposure to Columbia River temperatures above 21 °C during the upstream migration as has
been observed to occur per Figure 4-2 above. For context, the estimated Snake River fall
Chinook harvest rate (primarily for hatchery fall Chinook) between Bonneville Dam and McNary
Dam (zone 6) is approximately 23%, and the estimated stray rate is 3%.
60

-------
Columbia River Cold Water Refuges Plan
Final January 2021
110.0%
100.0%
90.0%
80.0%
£ 70.0%
c 60.0%
o
"to
a) 50.0%
>
c
® 40.0%
30.0%
20.0%
10.0%
0.0%
Adult Fall Chinook
•^—Average
2008 2009 2010 2011 2012 2013 2014 2015 2016
Year of Migration
Figure 4-4 Adjusted survival estimates of adult Snake River fall Chinook between Bonneville
Dam and Lower Granite Dam for the whole run (NMFS, 2019)
The information summarized above in this section and in Section 4.2 indicates exposure to
warm Lower Columbia (and Snake River) temperatures is likely contributing to mortality loss of
migrating adult steelhead and fall Chinook salmon. NMFS Biological Opinion (2020) on the
Columbia River System Operations (CRSO) recognized these adverse effects to adult
steelhead and fall Chinook from warm summer temperatures in the migration corridor. However,
NMFS concluded the overall adult survival rates for these species through the Lower Columbia
River were "relatively high" and the mortality losses were not at levels that would cause the
CRSO to appreciably reduce the survival and recovery of ESA-listed Snake River steelhead and
fall Chinook (NMFS 2020). As noted elsewhere, use of CWR by these species may be aiding
their migration survival rates through the Lower Columbia River during periods of warm
temperatures.
61

-------
Columbia River Cold Water Refuges Plan
Final January 2021
4.5 ENERGY LOSS AND PRE-SPAWNING MORTALITY OF FALL CHINOOK
SALMON FROM EXPOSURE TO WARM MIGRATION TEMPERATURES
As described in Section 4.1, prolonged exposure to warm river temperatures can have adverse
effects on migrating salmon. The rate of energy expenditure as a fish migrates directly depends
on swimming speed (fish speed plus water velocity) and temperature (Connor et al. 2018). For a
fish to successfully spawn at the end of its migration, it must have enough energy reserves for
gonad formation and to complete the spawning process. A recent study (Plumb 2018) used a
bioenergetics model to examine the effects of temperature on migration energy use and
spawning success. The study focused on Snake River fall-run Chinook migrating from
Bonneville Dam in the Columbia River to the confluence of the Snake and Salmon rivers in Hells
Canyon.
Based on previous studies (Bowerman et al. 2017), Plumb defined the energy threshold
criterion for successful spawning as 4 kJ/g3, where fish below this threshold typically die and do
not successfully spawn. Migrating salmon have finite energy reserves at the start of their
migration, and high river temperatures can hasten the rate at which fish reach this physiological
threshold, ultimately limiting spawning success (Plumb 2018).
Increases in time spent and distance traveled during migration lead to increases in pre-
spawning mortality, supporting a link between energy expenditure and spawning success
(Bowerman et al. 2017). Annual detections of PIT-tagged fish validate that slower travel rates
and greater exposure to higher temperatures affect arrival probabilities at spawning grounds.
The probability of fall Chinook having sufficient (>4 kJ/g) energy reserves to spawn depends in
part on two factors: (1) which day of the year a fish migrates from Bonneville Dam; and (2)
whether a fish uses CWR during migration. While early fall Chinook migrants are exposed to
warmer temperatures in comparison to later migrants, using CWR as a coping strategy can
influence the amount of energy reserves a fish has at time of spawning. Holding in CWR and
migrating later when Columbia and Snake River temperatures are lower can reduce thermal
exposure and energy loss.
Plumb (2018) modeled the thermal experience of simulated fall Chinook, which was a function
of the mainstem river temperatures during migration (Columbia and Snake Rivers), the
temperature difference between the mainstem river and a cold water tributary, and the
probability of a fish occupying a cold water tributary.
Figure 4-5 demonstrates that simulated fish using CWR experienced lower cumulative
temperatures and energy loss, which increased the proportion of early migrants surviving to
spawn. For instance, among fall Chinook migrating in August, those that used CWR (light grey
line) had a higher proportion with sufficient energy to complete spawning than those that did not
(dotted line).
3 kJ/g = kilojoules per gram of the fish and reflects the amount of energy per unit of mass or in
general terms the amount of stored fat relative to the size of the fish.
62

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Simulated without tributary use
Simulated with tributary use
Observed return of PIT-tagged fish
Figure 4-5 The proportion of simulated fish that had energy densities greater than the 4 kJ/g
threshold needed for sufficient energy to spawn (Plumb, 2018)
Supporting Plumb's findings, Figure 4-6 (Connor et al. 2018) shows that the early portion of the
spawning distribution of fall Chinook is predicted to drop below the energy threshold needed for
successful spawning and that these fish may experience pre-spawning or premature mortality.
63

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Pre-spawning
Premature
Successful
2011
2013
ilk
N
n
2015


ill
L
,1

L
-15-Oct	
14-DecJ
-15-Oct	
	14-Dec-l
Date span
Figure 4-6 Standardized, simulated spawning initiation date distributions for PIT-tagged,
hatchery-origin Snake River fall Chinook salmon adults, 2010-2015 (Conner et. al 2018)
Under scenarios to mimic future conditions with climate change, temperature increases of 1, 2,
and 3°C from baseline river temperatures showed a linear decline in the median energy
remaining at spawning and in the fraction of simulated fish having enough energy reserves to
spawn (Plumb 2018). As average temperatures increased, Chinook who did riot utilize CWR
were forced to migrate later in the year from Bonneville Dam to have enough energy reserves
left to spawn. However, for Chinook that did utilize CWR during migration under increasing river
temperatures, passage dates from Bonneville Dam were on average 18-27 days earlier than
fish that did not utilize CWR. This finding supports the conclusion that using CWR during upriver
migration may provide early migrants with an energetic advantage over fish that do not use
them. Further, the proportion of fish that seek and use thermal refuge is likely to increase as
temperature increases (Connor etal. 2018).
4.6 INCREASED MORTALITY AND SHIFT IN RUN TIMING OF SOCKEYE AND
SUMMER CHINOOK FROM WARM MIGRATION TEMPERATURES
As noted earlier, sockeye salmon do not appear to use CWR to avoid warm Lower Columbia
River temperatures, and it does not appear to be advantageous to do so. Sockeye salmon
migrate through the Lower Columbia River in June and July prior to the warmest summer river
temperatures that typically occur in August. If sockeye salmon were to delay their migration by
entering CWR, they would end up encountering warmer Columbia River temperatures during
their continued upstream migration.
64

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Warm Lower Columbia River temperatures, however, do have a significant impact on sockeye
salmon. The unusually warm June and July Lower Columbia River temperatures that occurred
in 2015 illustrate the relationship between warmer river temperatures and increased mortality of
sockeye salmon. As shown in Figure 4-7, in 2015 Lower Columbia River temperatures were
significantly warmer than average during the June-July sockeye run, reaching 20°C (68°F) at
the peak of the run, in late June. Typically, temperatures are about 16°C (61 °F) during the peak
of the sockeye run in late June.

75 1

35,000





70
			_/		
30,000

'erage Temperature
65
60
55
50
	~
rIf
Jr*
/ /
rl''
25,000
20,000
15,000
ckeye Adult Counts
<
45 •
2015 Counts
10,000
£

40
2015 Temp.
5,000


35
— ••Avg. Temp. {t»-'14) M
0





4/1 4/15 4/29 5/13 5/27 6/10 6/24 7/8 7/22 8/5 8/19


Figure 4-7 Sockeye passage and river temperature at Bonneville Dam (FPC, August 26,
2015 Memo)
Figure 4-8 shows how survival of sockeye from Bonneville Dam to McNary Dam dropped
significantly as temperature rose during the sockeye run in 2015. In early June when river
temperatures were below 19°C, survival between the two dams was high (90-100%). During
week 4 in Figure 4-8 (June 22-28), when river temperature climbed above 20 C, survival
dropped to 70% for Columbia River sockeye and 50% for Snake River sockeye (10% for Snake
River sockeye transported as juveniles). In weeks 5-8, when river temperatures exceeded 21 °C,
survival was very low (0-20%). Because most of the Snake River sockeye migrated in late June
and July, the overall survival for Snake River sockeye between Bonneville Dam and McNary
Dam was only 15% in 2015 (FPC 2015).
Although 2015's unusually warm June-July river temperatures had a dramatic effect on sockeye
salmon survival in the Lower Columbia River, warm Lower Columbia River temperatures result
in decreased sockeye survival in other years as well. Figure 4-9 shows the sockeye survival
rate between Bonneville and McNary dams as a function of river temperature across the
sockeye run for six different years (2010-2015). In 2010-2012 when the sockeye migrated
65

-------
Columbia River Cold Water Refuges Plan
Final January 2021
through the Lower Columbia River before river temperatures reached 64 F (18°C) survival rates
were relatively high (approximately 75%), In 2013 and 2014, for those sockeye migrating
through Lower Columbia River when temperatures exceeded 64 F (18 C) survival decreased,
most dramatically for Snake River sockeye.
Q Water Temperature at the Dalles ¦ Upper Columbia ¦ Snake River In-river Snake River Transported
1-Jun
4	5	6
Week {start date above bars)
Figure 4-8 Weekly survival estimates from Bonneville Dam to McNary Dam in 2015 for
Upper Columbia River Sockeye (blue bars), Snake River sockeye that migrated in-river as
juveniles (orange bars), and Snake River sockeye that were transported as juveniles (yellow-
orange bars) with water temperatures (red line) at The Dalles Dam (NMFS 2016)
66

-------
Columbia River Cold Water Refuges Plan
Final January 2021
100
0.75
0.50
O
| C.25
L
D
00 coo
^ 1 00
QC.75
m
0.50
0.25
COO J	1	1	1	1	1	1	1	1	1	1	1	1
60 64 63 72 60 64 68 72 60 64 68 72
Bonneville Dam Fore bay Temperature on Exit Day (F)
— Snake River — Upper Ccluirib a
Figure 4-9 Estimated relationship between Bonneville Dam forebay temperature and
Bonneville Dam to McNary Dam survival by return year for Snake and Upper Columbia adult
sockeye (FPC Memo 2015)
As described in Figure 4-8 and Figure 4-9, July Lower Columbia River temperatures have a
pronounced effect on sockeye salmon migration survival. Figure 4-10 shows how increasing
July river temperatures at Bonneville Dam (Panel B) over the past 60 years have resulted in
earlier migration of Columbia River sockeye salmon. The median passage date, which
historically was the first week of July, is now the last week of June (Figure 4-10, Panel A). Thus,
as July river temperatures have increased, the July sockeye migrant mortality has increased.
Over time, because the June sockeye migrants are more successful, the genetic traits of the
June migrants increase as a percentage of the population, contributing to the shift in migration
timing (Crozier et al. 2011).
2010
2011
2012
2013
r^i
2014
2015
1
67

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 4-10 Median sockeye salmon migration date (A), July mean temperature (B), and
June mean flow (C) at Bonneville Dam (Crozier et al. 2011)
Summer Chinook, like sockeye salmon, migrate through the Lower Columbia River in June and
July prior to the warmest summer temperatures (Figure 3-1). And, for the reasons described
above for sockeye salmon, summer Chinook likely do not use CWR, except for brief periods of
respite. Summer Chinook also have increased adult mortality with increased temperatures.
Figure 4-11 shows that 2013, 2014, and especially 2015 had above normal river temperatures
during the June-July migration period for Snake River summer Chinook passing Bonneville
Dam. Figure 4-12 shows the decreased survival rate of Snake River summer Chinook between
Bonneville and McNary dams for 2013, 2014, and 2015 relative to the average survival rate
(80%). The warmer-than-average temperatures in these years is likely a contributing factor to
the decreased survival.
68

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Return Yoor
2003
2004
2005
2006
2007
2OO0
2009
2010
	 2011
	 2012
	 2003-201 2
	 2013
	 201*
	1 20 1 5
0«' 15
07-01
Date
07-15
07-31
a> 70 -
I
1
¦ es -
~ CO -
06 Ol
Figure 4-11 Daily average temperature (°F) in the Bonneville Dam forebay from June 1 to
July 31 by return year (FPC 2016)
Bonneville-McNary
1.00-
iin
0.00 -
—i—1—1—i—
2003- 2013 2014 2015
2012
Figure 4-12 Hatchery Snake River summer Chinook adult reach survival with 95% confidence
intervals by return year (FPC 2016)
Much like the sockeye salmon run, the summer Chinook run has also shifted to earlier in the
year, likely in response to rising July temperatures. Figure 4-13 and Figure 4-14 show the
distribution of the summer Chinook run over Bonneville Dam from 1994 to 2018. Figure 4-14
shows that both the 50% passage date (yellow line) and the 90% passage date (blue line) have
shifted earlier by about 1 week over the past 25 years. Due to the increase in July temperatures
in the Lower Columbia River, only a small portion (10% or less) of the summer Chinook run
pass Bonneville Dam in the last two weeks of July.
69

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Historical Run Timing, 1994 • 2018
Adult Visual Counts Chinook
Bonneville Dam, 6/1 - 7/31
— First-Last — 5-95% — 10-90% — 25-75% — 50%
# Passage
www.cbr.washington.edu/dart	29 Mar 2019 11:29:39 PDT
Figure 4-13 Summer Chinook run timing past Bonneville Dam (1994-2018) (DART)
Historical Run Timing Regression, 1994 - 2018
Adult Visual Counts Chinook
Bonneville Dam, 6/1 - 7/31
Sep-1
Aug-1
JuH
Jun-1
First Passage. Mean DOY 152
5% Passage. Mean DOY 155
10% Passage. Mean DOY 157
50% Passage. Mean DOY 175
90% Passage. Mean DOY 199
95% Passage. Mean DOY 205
100% Passage. Mean DOY 212
	 First y = O.Ox+152, R2=nan
	 5% y = -O.lx+306, R^O.22
	 10% y — -O.lx+415. R2=0.22
50% y - -0.3X+737. R2-0.38
	 90% y - -0.3X+773. R2-0.42
	 95% y = -0.2X+632. R2=0.38
	 100% y = -O.Ox+212. R2=nan
3 in vo r* oo
g% cr* 0} en o>
O*	CPl O) G\
www.cbr.washington.edu/dart
8 8 8
8 |
O »-
s s
29 Mar 2019 11:29:39 PDT
Figure 4-14 Trends in summer Chinook run distribution past Bonneville Dam (1994-2018)
(DART)
70

-------
Columbia River Cold Water Refuges Plan
Final January 2021
5 HISTORIC AND FUTURE TRENDS IN COLUMBIA RIVER
TEMPERATURES
5.1 HISTORIC TEMPERATURE CONDITIONS OF THE LOWER COLUMBIA RIVER
Based on available literature and EPA analyses (Appendix 12.16), the estimated increase in
Columbia River temperatures from climate change since the 1960 baseline ranges from 0.2 CC
to 0.4°C per decade, for a total temperature increase to date of 1,5°C ± 0.5°C. EPA notes that
flow regulation, land use changes, natural variability, and other factors may have also influenced
the observed changes. Thus, increased water temperatures since 1960 may not be ascribed
solely to anthropogenic climate change influences.
Figure 5-1 Trend in Columbia River August temperatures at Bonneville Dam (National
Research Council 2004)
Historic measurement data shown in Figure 5-1 on the Columbia River at Bonneville Dam
indicate that the total warming of the river since the late 1930s in August (average) is
approximately 2.2'C (dashed line), rising from below 20 C to near 22 C. This increase
incorporates ail factors in river warming, including dam construction in the middle decades of
the century and climate change from 1960 to 2000. It is noted that monitoring data collected at
the dams and contained in the DART database prior to 1990 is uncertain due to a lack of data
quality procedures. Nevertheless, this is the best available information on historic temperatures,
and the increase in August temperatures appears to be generally consistent with current
71

-------
Columbia River Cold Water Refuges Plan
Final January 2021
estimates of anthropogenic impacts using EPA's RBM10 model (EPA 2020), combined with the
climate-related warming since 1960 noted above.
EPA's RBM10 model can predict past temperatures by using historic air temperatures and river
flow, and RBM10 model results were considered in the climate trend analysis in Appendix
12.16. Figure 5-2 is a simulation with the existing Columbia and Lower Snake River dams in
place (all dams were built prior to 1970 except Lower Granite, which was built in 1975). Figure
5-3 is a simulation without the U.S. Columbia and Lower Snake River dams (the simulation
retained Canadian dams on the Columbia River). A comparison of the two figures indicates that
August and September mean Columbia River temperatures at Bonneville Dam would have
warmed at a lower rate and to a lesser extent without the dams since 1970. The yellow-dashed
line representing the August warming rate in Figure 5-2 shows 0.4 C increase per decade,
while the yellow-dashed line in Figure 5-3 shows a 0.26 C increase per decade. For July (red-
dashed lines), however, the rate of warming is approximately the same in the two simulations,
indicating that the increase in warming since 1970 is primarily attributable to air temperature
increases from climate change, and that the dams have not exacerbated the warming trend in
July. Therefore, the dams appear to have exacerbated the rate of climate change induced
warming in the Columbia River in the late summer (August-September).

• •
• *

			 i
• 	
••fr
• • _
•
• • •
•
• • •


•


* • ~ * * , 	
—jCii			
• •
• . •

• •
• •


• •
24
22
v 20
b
3
no
£18
!
c 16

-------
Columbia River Cold Water Refuges Plan
Final January 2021
24
22
10
1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
Year
• July	August	• September	• October
	Linear (July) Linear (August) Linear (September) Linear (October)
Figure 5-3 Simulated monthly mean temperatures at Bonneville Dam (free flowing) (EPA
2020)
As discussed above in Sections 3.7 and 4.6, the increase in summer river temperature has
increased the use of cold water refuges (CWR) by steelhead and fall Chinook in the Lower
Columbia River, has contributed to increased mortality of migrating adult sockeye and summer
Chinook, and is contributing to earlier sockeye salmon and summer Chinook runs.
5.2 FUTURE TEMPERATURE CONDITIONS OF THE LOWER COLUMBIA RIVER
AND ITS TRIBUTARIES
Climate change has already influenced and is projected to continue to influence river
temperatures across the Northwest, including the temperatures of the Columbia and Snake
Rivers. Climate change will also influence multiple aspects of river hydrographs, including timing
and magnitude of river flow. As noted above, climate change is estimated to have increased
temperatures in the Columbia and Snake River mainstems by 1.5 C ± 0.5°C since 1960 (0.3 C
per decade). From this new baseline, the warming trend is expected to continue in the coming
decades.
Figure 5-4, Figure 5-5, and Figure 5-6 display Lower Columbia River August mean
temperatures under current conditions, in 2040, and in 2080, respectively, assuming a
continuation of the 0.3 C degree per decade warming trend. A continued 0.3 C degree per
decade warming trend is very similar to Lower Columbia River reported model predictions using
the AB1 scenario of future greenhouse emissions and global warming (Isaak et ai. 2018,
Yearsley 2009, Appendix 12.19), which represents a mid-range reduction in annual global
greenhouse gas emissions over the 21st century.
73

-------
Columbia River Cold Water Refuges Plan
Final January 2021
As shown in Figure 5-5 and Figure 5-6, August mean temperatures in the Lower Columbia
River are projected to increase from near 22 C currently to near 23 C in 2040 and near 24 C in
2080. August mean temperatures in the 23-24 C range would likely result in a significant
amount of lethality to migrating adult salmon and steelhead (Table 4-1). It is therefore likely that
fewer salmon and steelhead will migrate in the Lower Columbia River during mid-July through
August in the future under these warming trends, resulting in a change in the timing of salmon
and steelhead runs. Adult sockeye salmon and summer Chinook will likely continue to migrate
earlier as already observed, with very few migrants in July. Adult fall Chinook are likely to
migrate later with minimal migrants in August, and those that do migrate then will likely need to
use CWR to have sufficient energy to successfully spawn. Steelhead may use CWR for a longer
duration to avoid peak temperatures, or they may not be able to use CWR over the mid-summer
like they currently do because mainstem temperatures are too warm in late July/early August for
steelhead to reach the CWR in the Bonneville reach . If the latter proves true, this may result in a
bi-modal migration pattern for steelhead with early summer and late summer runs. However,
whether these species can shift their migration timing to adapt to the rate of warming, and
whether such shifts can be done successfully without disruption to their full freshwater life cycle,
is uncertain (Crazier et al. 2011 and Keefer & Caudill 2017).
Figure 5-4 Current August mean water temperature in the Columbia River and tributaries
(2011-2016) (Appendix 12.14)
74

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 5-5 Estimated 2040 August mean water temperature in the Columbia River and
tributaries (Appendix 12.14)
Figure 5-6 Estimated 2080 August mean water temperature in the Columbia River and
tributaries (Appendix 12.14)
75

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Temperatures in the tributaries to the Lower Columbia River, including the 23 tributaries that
currently provide CWR, are also predicted to increase due to climate change. Table 5-1
displays the predicted increase in August mean temperatures for the 23 CWR tributaries (12
primary CWR highlighted in blue) using the NorWeST SSN model (Appendix 12.17). August
mean temperatures for the CWR tributaries are predicted to increase by 1.2-1,5°C by 2040 and
by 2.1-2.7°C by 2080 relative to current baseline (1995-2011).
Of significant concern are those primary CWR tributaries that are predicted to have August
mean temperatures that exceed 18°C. Tributary temperatures exceeding 18°C, although still
serving as CWR if more than 2°C cooler than the Columbia River, are at levels associated with
increased risk of disease and energy loss. For instance, by 2040, the Deschutes, Lewis, and
Sandy Rivers are predicted to exceed 18°C, temperatures that will diminish their CWR function.
By 2080, the Cowlitz, White Salmon, and Klickitat Rivers are predicted to have August mean
temperatures exceeding 18°C, diminishing their CWR function.
Table 5-1 Future temperature conditions of the Lower Columbia River tributaries (Appendix
12.17)
Tributary Name
Current (°C)
(1995-2011)
2040 (°C)
Change
between 2040
and current (°C)
2080 (°C)
Change
between 2080
and current (°C)
Skamokawa Creek
16.2
17.6
1.4
18.6
2.4
Mill Creek
14.5
15.9
1.4
16.8
2.3
Abernethy Creek
15.7
17.1
1.4
18.1
2.4
Germany Creek
15.4
16.8
1.4
17.8
2.4
Cowlitz River
16.0
17.4
1.4
18.4
2.4
Kalama River
16.3
17.7
1.4
18.8
2.5
Lewis River
16.6
18.0
1.4
19.0
2.5
Sandy River
18.8
20.3
1.5
21.4
2.6
Washougal River
19.2
20.7
1.5
21.8
2.7
Bridal Veil Creek
11.7
12.9
1.2
13.8
2.1
Wahkeena Creek
13.6
15.0
1.3
15.9
2.3
Oneonta Creek
13.1
14.4
1.3
15.4
2.2
Tanner Creek
11.7
12.9
1.2
13.8
2.1
Eagle Creek
15.1
16.5
1.4
17.5
2.4
Rock Creek
17.4
18.9
1.5
19.9
2.5
Herman Creek
12.0
13.4
1.4
14.3
2.3
Wind River
14.5
15.9
1.4
16.8
2.4
Little White Salmon
River
13.3
14.8
1.4
15.7
2.3
White Salmon River
15.7
17.2
1.5
18.2
2.4
Hood River
15.5
17.0
1.4
17.9
2.4
Klickitat River
16.4
17.8
1.5
18.8
2.4
Deschutes River
19.2
20.7
1.5
21.7
2.5
Umatilla River
20.8
22.4
1.5
23.4
2.6
76

-------
Columbia River Cold Water Refuges Plan
Final January 2021
6 SUFFICIENCY OF COLD WATER REFUGES IN THE LOWER
COLUMBIA RIVER
6.1 CWR SUFFICIENCY ASSESSMENT FRAMEWORK
Assessing whether there is a sufficient amount of cold water refuge (CWR) in the Lower
Columbia River to attain the Oregon water quality standard is complex. Oregon's CWR narrative
standard stipulates the Lower Columbia River must have CWR that is sufficiently distributed to
allow salmon and steelhead migration without significant adverse effects from higher water
temperatures elsewhere in the water body (i.e., Columbia River). One of the purposes of this
Plan is to provide a framework to make this CWR sufficiency assessment given the current state
of information available.
Through the scientific assessment and development of this Plan, EPA identified important
context issues for the evaluation of CWR sufficiency. The first issue is the assumption that CWR
are beneficial to migrating salmon and steelhead in the Lower Columbia River. There are two
exceptions to this assumption in the Lower Columbia River. The first exception is fish mortality
from fishing in CWR. As presented in Section 4.3, fish that enter into CWR have a lower adult
migration survival rate through the Lower Columbia River compared to fish that do not use
CWR. This appears to be explained mostly by fish harvest in CWR and potentially mortality of
caught and released fish, although the higher tendency of fish that were barged downstream as
juveniles to stray into CWR may also be a causal factor. However, the role of water quality
standards under the Clean Water Act (CWA) is to ensure the water is of sufficient quality (in this
case, water temperature) to protect designated uses of the water body (in this case, salmon and
steelhead). Therefore, EPA did not consider fishing mortality in the assessment of CWR
sufficiency, recognizing that the amount of fish mortality in CWR can change through fish
management decisions. Thus, EPA evaluated the sufficiency of CWR in the Lower Columbia
River as if there was no fishing to focus our assessment on water quality conditions to support
migrating salmon and steelhead.
The second exception to the assumption that CWR are beneficial to migrating salmon and
steelhead is that using CWR may cause harm due to the delay in their migration. As discussed
in this Plan, sockeye salmon and summer Chinook migrate through the Lower Columbia River
prior to the onset of the warmest summer temperatures, and extended CWR use would likely be
harmful due to exposure to warmer conditions during their continued migration. Wth these two
exceptions explained, the evidence presented in this Plan suggests that CWR use appears to
be physiologically beneficial for those species that use CWR the most, which are summer
steelhead and fall Chinook.
The second context issue is the temperature of the Columbia River itself. As described in this
Plan, the degree to which salmon and steelhead use CWR depends on the Columbia River
mainstem temperature. The warmer the river, the more fish use CWR. Thus, assessing CWR
sufficiency can be viewed as a function of the Columbia River temperature. However, although
CWR can help mitigate adverse effects to migrating salmon and steelhead when Columbia
River temperatures exceed 20°C, the CWR narrative standard should not be interpreted
77

-------
Columbia River Cold Water Refuges Plan
Final January 2021
to "allow for" or to "fully compensate for" Columbia River water temperatures higher than the
20°C numeric criterion.
EPA assessed whether CWR is sufficient to attain Oregon's CWR narrative criteria based on
current Columbia River conditions because such water quality data are available, and because
water quality standard assessments are generally based on current conditions. However, to
address the dynamic of different temperatures in the Lower Columbia River, EPA evaluated
sufficiency at three different temperature regimes: August mean temperature of 20°C, which
reflects historical conditions; 21.5°C, which reflects current conditions; and 22.5°C, which
reflects a predicted 2040 condition. This analytical framework to address sufficiency is helpful to
understand the use of CWR in the past, present, and future. Some of the recommendations in
this Plan consider predicted future temperature conditions in the Lower Columbia River and the
CWR tributaries as practical considerations to improve water quality for migrating salmon and
steelhead.
To evaluate sufficiency of CWR at different Lower Columbia River temperatures, EPA
considered several factors based on information presented in previous chapters, as well as in
the HexSim model discussion below: (1) the extent of CWR use in terms of number of salmon
and steelhead in CWR and the proportion of the run using the CWR; (2) a qualitative
assessment of the potential for the current volume of CWR to have capacity limitations; (3) the
distribution of CWR in the Lower Columbia River; (4) observed and modeled indicators offish
health and risk, including mortality rates, energy loss, and cumulative exposure to stressful
temperatures for migrating salmon and steelhead in the Lower Columbia River; and (5) the
overall importance of adult migration risk factors in the recovery of salmon and steelhead from
review of ESA recovery plans and NMFS' Columbia River Systems Operations Biological
Opinion.
6.2 HEXSIM MODEL
To aid in examining sufficiency of CWR in the Lower Columbia River, EPA developed a fish
behavior simulation model using the HexSim modeling platform (Schumaker and Brookes,
2018) that simulates behavior, movement, and tracks thermal exposure of individual fish
migrating through the Lower Columbia River. The model description and the initial application of
the model through the Bonneville reach of the Columbia River between Bonneville Dam and
The Dalles Dam is summarized in Snyder et al. 2019. The model has been expanded to include
the 178-mile portion of the Columbia River from Bonneville Dam to the Snake River confluence
(Snyder et al., 2020).
The HexSim model provides the opportunity to simulate different scenarios and evaluate how
they affect CWR use and important indicators related to fish health. For the initial model runs for
this Plan, EPA selected the following scenarios: (1) existing CWR; and (2) no CWR. Both
scenarios were run under different Columbia River temperatures representing past, current, and
predicted future average conditions. These model scenarios help examine how the current
amount of CWR affects fish health indicators at different Columbia River temperatures to assess
CWR benefits. Health indicators assessed include cumulative energy expenditure, cumulative
degree days above warm temperature thresholds (e.g., 21 °C and 22°C), and predicted acute
mortality between Bonneville Dam and the confluence with the Snake River. EPA evaluated
78

-------
Columbia River Cold Water Refuges Plan
Final January 2021
these scenarios and resultant indicators for two populations of summer steelhead, Grande
Ronde summer steelhead and Tucannon summer steelhead; and two populations of fall
Chinook salmon, Snake River fall Chinook and Hanford reach fall Chinook. An additional model
run examined the change in these health indicators if five extra evenly-spaced CWR were
theoretically added between the John Day Dam and the Snake River. The size of the extra
CWR were 6,000 cubic meters each, which is about twice the size of the Eagle Creek CWR.
The extra CWR model run analyzed the Grande Ronde summer steelhead and Snake River fall
Chinook population for 2017 temperatures. The results of these model runs are presented in
Appendix 12.21.
The following is a summary of the HexSim model assessment. The summary below highlights
model results for Grand Ronde summer steelhead because that population represents a
steelhead population that uses CWR extensively, as shown in Section 3.9.
Cumulative Number of Hours in CWR as a Function of Columbia River Temperature
The number of hours individuals spend in CWR increases with increased Columbia River
temperatures for all four populations evaluated, which is consistent with the CWR use estimates
in Chapter 3. For Grande Ronde summer steelhead, the number of hours per fish in CWR is
modeled to be 124 hours at past/historical temperatures, 389 hours at current temperatures,
and 497 hours at predicted 2040 temperatures (Appendix 12.21).
Energy Loss Under Different Scenarios
The energy loss (fat loss) within the model reach (Bonneville Dam to Snake River confluence)
increased for all four populations with increased Columbia River temperatures. Figure 6-1
summarizes the energy loss for Grande Ronde summer steelhead for the different scenarios. If
too much energy is lost during migration and pre-spawning, a fish may not have enough energy
to complete spawning as discussed in Section 4.5. Because use of CWR increases the amount
of time in the model reach, CWR use somewhat increases the population's median amount of
energy loss in the model reach relative to no CWR use as shown in Figure 6-1. However, to
evaluate the implications of energy use on spawning success, energy loss needs to be
evaluated within the context of the entire migration, including holding and spawning. For
example, Grande Ronde summer steelhead migrate another 170 miles upstream in the Snake
River before traveling up the Grande Ronde River to their spawning grounds. Under scenarios
of no CWR use, there is a much earlier average arrival at the end of the modeled reach (Snake
River confluence) (Figure 6-2), when Snake River temperatures are warmer. The use of CWRs
extends the range of arrival dates at the Snake River confluence, which may decrease energy
loss for those late arriving individuals who will then migrate through the Snake River when it is
cooler. Therefore, while the entire population does not see an energy benefit in the modeled
reach of the migration corridor, CWRs potentially increase the diversity of energy conserving
migration strategies.
In summary, it is necessary to model the full migration to the spawning grounds to fully assess
energy loss and the potential for pre-spawning mortality, as was done in the Plumb (2018) and
Conner et al. (2018) papers, which concluded CWR in the Lower Columbia River were
beneficial to reduce pre-spawning mortality for early migrating Snake River fall Chinook (Section
79

-------
Columbia River Cold Water Refuges Plan
Final January 2021
4.5). These papers indicate that most of the energy loss for Snake River fall Chinook occurs
upstream of the Lower Columbia River. Thus, the river temperature during the latter part of the
fall Chinook migration, when the fish are preparing to spawn, is an important factor in spawning
success, and CWR in the Lower Columbia River can serve to allow the fish to arrive at the
spawning grounds when river temperatures are cooler.
Grande Ronde River Summer Steelhead
50-
40"
30"
Q
20
10-


rc^	OF^° G°

or.

CP^C
CP
CP"
Figure 6-1 Simulated energy loss for Grande Ronde summer steelhead from Bonneville
Dam to the Snake River under various scenarios (Appendix 12.21)
80

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Grande Ronde River Summer Steelhead
600-
Columbia Current, CWR Current
400-
	1	1	1	1	
Aug	Sep	Oct	Nov
Arrival date at end of modeled reach
Figure 6-2 Simulated arrival date at the Snake River for Grande Ronde summer steelhead
with and without CWR use under current conditions (Appendix 12.21)
Acute Mortality
The model runs with and without CWR at past, current, and future (2040) Columbia River
average temperatures did not show any significant acute mortality for the four populations in the
model reach (Appendix 12.21). This was not unexpected because acute temperature stress
mortality was based on a study that indicates acute stress mortality begins to occur at 24°C
(less than 1% chance), climbing to a 10% chance at 27°C with 24-hour exposure (Railsback et
al., 2009). Columbia River maximum daily average temperatures currently reach 23°C and are
not predicted to reach 24°C until 20404 (Appendix 12.1).
Because current and predicted future daily average temperatures are at the threshold of acute
temperature stress mortality, an uncertainty analysis was conducted using three different
temperature-acute mortality relationships based on multiple studies (Jager2011, Sullivan etal.
2000, Railsback et al. 2009). Under the more conservative relationship when acute stress
mortality starts at 23°C, the HexSim model predicted 18% acute stress mortality in 2040 for
Grande Ronde summer steelhead with the current available CWR and 28% acute stress
mortality absent CWR (Appendix 12.21). This indicates that use of CWR may serve an
4 Columbia River temperatures used in the HexSim model, as well as reflected in the figures and tables of this Plan,
are from the main channel dam site monitors that are about 30-35 feet deep. As presented in Appendix 12.1, the
upper surface layer of the John Day and McNary reservoirs can reach 25-26°C, but exposure to these temperatures
was not included in the HexSim model.
81

-------
Columbia River Cold Water Refuges Plan
Final January 2021
important role to reduce acute stress mortality for migrating adult salmon and steelhead in the
future when Lower Columbia River temperatures are predicted to reach 24°C.
Cumulative Degree Days under Different Scenarios
The model runs show large differences in cumulative degree days above warm temperature
thresholds of 21 °C and 22°C with and without CWR for Grande Ronde steelhead. As shown in
Figure 6-3, under current Columbia River temperatures the cumulative number of degree days
above 21 °C is much higher if there were no CWR compared to the current amount of CWR. The
average number of cumulative degree days above 21 °C is 139 days for the Grande Ronde
summer steelhead population using CWR. If no CWR were available, the population would have
272 degree days above 21 °C.
Figure 6-4 shows the cumulative degree days above 22°C for Grande Ronde steelhead. Under
current Columbia River temperatures, the 10-year mean of daily average temperatures
(reflected in Figure 6-4) rarely exceeds 22°C in the Columbia River, so cumulative degree days
above 22°C are near zero with and without CWR. However, under predicted 2040 average
conditions, the cumulative degree days above 22°C for the Grande Ronde steelhead population
will be higher (286) if no CWR were available compared to the current amount of CWR (118). It
is also notable that for current warm years (e.g. 2017 and other recent warm years when
Columbia River temperatures were warmer than the 10-year average with numerous days
exceeding 22°C), CWR use reduced the cumulative exposure for steelhead above 22°C, similar
to what is displayed in Figure 6-4 for 2040 average temperatures (Appendix 12.21).
Figure 6-5 shows the modeled difference in cumulative degree days above 21 °C under 2017
Columbia River temperatures for Grande Ronde steelhead with current CWR, with added CWR,
and with current CWR if the Columbia River was 1°C cooler than 2017 temperatures. For the
added CWR scenario, CWR was theoretically added at five evenly spaced locations between
the John Day Dam and the Snake River for a total of 30,000 cubic meters of added CWR. This
modest addition of CWR (less than 1% of the total CWR volume in the Lower Columbia River)
to this reach with limited current CWR, shows a slight decrease in the cumulative degree days
above 21 °C for this population. However, if the Columbia River 2017 temperatures were
hypothetical^ 1°C cooler, the reduction in cumulative degree days would be greater, indicating
the importance of the river temperature itself and that CWR does not fully offset exposure to
warm river temperatures (Appendix 12.21 and Snyder etal., 2020)
The difference in cumulative degree days above the 21 °C and 22°C thresholds in the above
scenarios illustrate the benefits of CWR use for migrating steelhead by avoiding peak warm
temperatures and is consistent with the information and discussion presented in Chapter 3.
Prolonged exposure to temperatures greater than these thresholds is stressful for migrating
salmon and steelhead and likely increases disease risk associated with mortality as discussed
in Chapter 4.
82

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Grande Ronde River Summer Steelhead
800"
600-
o
' 400"
o
200-
o-
¦J***



G°






G<**V&
^5^
Figure 6-3 Simulated cumulative degree days above 21 °C for Grande Ronde summer
steelhead between Bonneville Dam and the Snake River under different scenarios (Appendix
12.21)
83

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Grande Ronde River Summer Steelhead
G<>
G°^C^
G°


oW

G°

,600-
ro
Q
0
0
o) 400 -\
0
Q
0
¦J; 200 h
ro
O
o-
Current CWRs
Added CWRs
Cooler Columbia
River (-1°C)
Current CWRs
Figure 6-5 Simulated cumulative degree days above 21 °C under 2017 Columbia River
temperatures for Grande Ronde summer steelhead between Bonneville Dam and the Snake
River under different scenarios (Appendix 12.21)
84

-------
Columbia River Cold Water Refuges Plan
Final January 2021
6.3 ASSESSMENT OF SUFFICIENTLY DISTRIBUTED CWR
As noted above, EPA assessed whether CWR is sufficient to attain Oregon's CWR water quality
standard under current Lower Columbia River August average temperatures (21.5°C),
considering the factors listed in Section 6.1. For context, EPA also evaluated CWR sufficiency
under past (20°C) and future (22.5°C) conditions.
Current Conditions: Fish Use and CWR Capacity
As shown in Chapters 2 and 3, current Lower Columbia River temperatures typically exceed
20°C for two months and exceed 21 °C for one month, and use of available CWR by steelhead
and fall Chinook is well documented and extensive. Based on information in Chapters 3 and 4,
and HexSim model results, current steelhead and fall Chinook use of CWR appears to provide
some individuals physiological and energetic benefits by allowing them to avoid warm mid-
summer Columbia River temperatures and continue migrating upstream when temperatures
have cooled. As described in Chapter 3 and displayed in Figure 3-1 and Figure 3-17, the
majority of the overall summer steelhead run, as well as most individual summer steelhead
populations, migrate though the Lower Columbia River during the peak summer temperatures.
This indicates that CWR use in the Lower Columbia River is an important aspect of the
contemporary migration strategy for most steelhead populations (Keefer et al. 2009). As
described in Chapter 3 and displayed in Figure 3-1, about half the fall Chinook run occurs in
August and the first half of September when some fall Chinook salmon seek CWR to avoid
warm Columbia River temperatures. Thus, CWR use is an important migration strategy for part,
but not all, of the fall Chinook run.
From the density estimates in Chapter 3 and HexSim modeling, it does not appear the capacity
in CWR is exceeded, except for Eagle Creek and Rock Creek. The HexSim model showed
these small CWR reaching capacity (Snyder et al. 2019). EPA reviewed literature on the density
of adult salmon and steelhead held in confined spaces to define a maximum fish density of 1
fish per cubic meter, but it is uncertain whether this is representative of maximum density in
CWR (Berejikian et al. 2001, Hatch et al., 2013). Disease risk from high density of fish in CWR
is also a concern that could factor into CWR capacity consideration, especially for marginal
CWR that are at temperatures (18°C or higher) associated with elevated disease risk. However,
EPA is unaware of anecdotal evidence or studies that indicate incidents of disease for adult
steelhead or fall Chinook in CWR. Additional research on factors regulating capacities of CWR
and disease risk in CWR is needed.
Current Conditions: CWR Distribution
Regarding the distribution of CWR in the Lower Columbia River, migrating salmon and
steelhead have several CWR opportunities below Bonneville Dam and extensive CWR
opportunities in the Bonneville Dam reservoir reach and the Deschutes River above The Dalles
Dam (see Figure 2-8). The cluster of CWR in the Bonneville Dam reservoir reach and the
Deschutes River is approximately midway from the ocean to the confluence of the Snake River.
It takes approximately one week for salmon and steelhead to travel from the ocean to this
cluster of CWR and another week to pass the McNary Dam and get to the Snake River
confluence area. Thus, the CWR distribution is advantageous in that the CWR provide the
opportunity to escape the warm Columbia River midway through their upstream migration of the
85

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Lower Columbia River and avoid approximately two weeks of continuous exposure to warm
temperatures over this 325-mile reach.
However, the lack of CWR in the nearly 100 miles between the Deschutes River and McNary
Dam, including the John Day reservoir which has the highest temperatures in the Lower
Columbia River, is of concern. This nearly 100-mile reach poses the greatest risk from warm
temperatures for migrating salmon and steelhead. Thus, it is difficult to conclude that CWR is
sufficiently distributed due to the absence of CWR in this reach. Opportunities to restore CWR in
this reach are limited. Under natural conditions there were likely only a few small tributaries (e.g.
Willow Creek, Rock Creek) and the Umatilla River that may have provided CWR. As noted in
Chapter 2, the Umatilla River is currently warmer than the Columbia River in July and most of
August and only provides marginal and intermittent CWR in late August and September after
the Umatilla River has cooled relative to the Columbia River. Cooling Lower Umatilla River
temperatures in August and September consistent with the Oregon and Confederated Tribes of
the Umatilla Indian Reservation (CTUIR) Temperature TMDLs (ODEQ 2001 and EPA 2005) to
provide increased CWR volume would make the Umatilla River a more consistent and viable
CWR and would help address the overall distribution of CWR in the Lower Columbia River.
HexSim model runs indicate small additions of CWR in this reach may be beneficial to reducing
cumulative exposure to warm Columbia River temperatures (see Figure 6-5).
Current Conditions: Adult Survival
The strongest line of evidence that the current amount of CWR is sufficient under current
Columbia temperatures is the adult survival rates from Bonneville Dam to McNary Dam. As
discussed in Section 4.4, the adult survival rate after accounting for harvest and straying for
Snake River steelhead and fall Chinook is over 90%. Table 2-1 shows the estimates of adult
survival after accounting for harvest and straying for Snake River species from Bonneville Dam
to McNary Dam from 2012-2016 (NMFS 2017b). Snake River fall Chinook adult survival is near
96% and Snake River steelhead is 93%. While NMFS recognizes that warm Lower Columbia
River temperatures are a concern and cause adverse effects to ESA-listed species, NMFS
views the adult migration survival rates for these species as "relatively high" and the losses are
not at levels that would cause the Columbia River System Operations to appreciably reduce the
survival and recovery of ESA-listed Snake River steelhead and fall Chinook (NMFS 2020).
NMFS also noted the importance of CWR to these summer migrating species.
Table 6-1 Adult salmon and steelhead survival estimates after correction for harvest and
straying based on PIT-tag conversion rate analysis from Bonneville (BON) to McNary (MCN)
dams, McNary to Lower Granite (LGR) dams, and Bonneville to Lower Granite dams (NMFS
2017b)
Species
Years
BON to MCN
MCN to LGR
BOX to I.GR
SR Fall Chinook
2012-2016 Avg
95.8%
94.9%
91.0%
SR Spr/Sum Chinook
2012-2016 Avg
93.1%
94.0%
87.3%
SR Sockcve
2012-2016 Avg
59.9%
74.2%
49.7%
SR Steelhead
2012-2016 Avg
93.2%
94.3%
87.9%
86

-------
Columbia River Cold Water Refuges Plan
Final January 2021
The current amount of CWR may be helping to maintain the average survival rates (after
adjusting for harvest and straying) above 90% shown in Table 6-1 by minimizing salmon and
steelhead exposure to peak summer temperatures in the Lower Columbia River. As illustrated in
Figure 6-3 for Grand Ronde summer steelhead, CWR use, relative to no CWR use, reduces the
cumulative exposure to temperatures above 21 °C, which is associated with increased stress
and disease mortality. Moreover, CWR use in the Lower Columbia River also reduces
cumulative exposure to warm temperatures for fish migrating up the Snake River due to
migrating later in the summer/fall, which likely aids in the survival rates up the Snake River to
Lower Granite Dam (LGR). Notably, Snake River sockeye, which do not use CWR due to their
early summer run timing, have a much lower adult survival rate due to mortality from warm
Columbia River temperatures as discussed in Chapter 4.
Snake River summer steelhead and Snake River fall Chinook adult survival rates (NMFS
2017b) from Bonneville Dam to McNary are generally representative of survival rates of other
steelhead species (Upper Columbia River and Middle Columbia River) and other fall Chinook
species (Hanford reach) that use CWR. As presented in Section 3.9, Upper Columbia River
steelhead migrate earlier in the year compared to Snake River steelhead and therefore have
less overall exposure to warm Lower Columbia River temperatures and use CWR less.
Likewise, most Hanford reach fall Chinook migrate later than Snake River fall Chinook and
therefore have less overall exposure to warm Lower Columbia River temperatures and use
CWR less.
However, as discussed in Section 4.4, there is year-to-year variability in unexplained mortality
for adult steelhead and Fall Chinook between the Bonneville and McNary Dams. Some years
with more than 10% unexplained mortality (i.e., less than 90% adjusted survival) could be
associated with exposure to warm migration temperatures. Further, these data for steelhead
represent an average of all Snake River steelhead populations, and some individual populations
could have higher unexplained mortality, especially if a high percentage of the population's
migration occurs during peak summer temperatures. Thus, the variation and uncertainty in the
adjusted survival rates are important to recognize.
Current Conditions: Summary
EPA's assessment is that CWR is sufficient under current Columbia River temperatures if the
volume of the 12 primary CWR is maintained and the Umatilla River is cooled to provide
increased CWR volume in August and September. EPA reached this assessment primarily
because there do not appear to be significant capacity limitations on the use of currently
available CWR, adult steelhead and fall Chinook migration adjusted survival rates generally
exceed 90% between Bonneville Dam and McNary Dam, and increasing CWR in the Umatilla
River is important for the overall distribution of CWR in the Lower Columbia River.
Past Conditions
When the Lower Columbia River is 20°C (August mean), which represents historical Columbia
River temperatures, EPA's assessment is that the current amount of CWR appears to be
sufficient to support migrating salmon and steelhead. Under the scenario of 20°C, CWR use is
modest by steelhead and very limited for fall Chinook, as first described in Chapter 3. The level
of CWR use when August mean temperature is 20°C is far less than what is observed under
current conditions. Because the current CWR volume appears to be sufficient with the exception
87

-------
Columbia River Cold Water Refuges Plan
Final January 2021
of the Umatilla CWR under current Columbia River temperatures, as discussed above, the
current CWR volume would likely be sufficient when the Columbia River is cooler. Although an
August mean temperature of 20°C during migration is above optimal and presents risks in terms
of elevated disease occurrence and sub-lethal effects, observed mortality to migrating adults is
low under these conditions.
Future Conditions
When the Lower Columbia River is 22.5°C (August mean), which reflects predicted future
(2040) conditions, EPA's assessment is that there is significant risk that the current amount of
CWR will not be sufficient to minimize the risk to migrating salmon and steelhead. As presented
in this Plan, a warmer Lower Columbia River at these temperatures (22.5°C August mean with
daily average temperatures frequently reaching 23-24°C) will significantly increase the stress,
energy loss, and mortality risk to salmon and steelhead migrating in the Lower Columbia River
in the summer. Under these temperatures, the extent of CWR use, as discussed in Chapter 3
and presented in HexSim model results, is expected to be higher. Steelhead may be less apt to
leave the CWR at these peak summer temperatures. Further, these temperatures will trigger fall
Chinook to use CWR at a higher rate. As a result, the density of fish in CWR will be higher,
calling into question the capacity of the currently available CWR. Additionally, the CWR
tributaries are predicted to warm. This is of particular concern for marginal CWR (Table 7-1).
For example, the Deschutes River, which although cooler than the Columbia River, currently
has an August mean temperature of 19°C, which is above optimal for migrating salmon. These
factors suggest there is significant risk that the Lower Columbia River adult migration survival
rates for steelhead and fall Chinook will decrease in the future. However, as noted earlier, CWR
cannot be expected to fully compensate for warm Lower Columbia River temperatures. As such,
the causal factor to the increased risks for salmon and steelhead noted above is the warm
Columbia River temperatures, not the lack of adequate CWR to minimize those risks. That said,
increasing CWR may serve to mitigate some of the risks of warmer Columbia River
temperatures.
Conclusion
EPA's assessment is that CWR is sufficient to attain Oregon's CWR narrative criteria in the
Lower Columbia River if the volume of the 12 primary CWR is maintained and the Umatilla River
is cooled to provide increased CWR volume in August and September consistent with the
Oregon and CTUIR Temperature TMDLs. Therefore, maintaining the current temperatures and
flows of the 12 primary CWR tributaries and cooling the Umatilla River is needed to limit
significant adverse effects to migrating adult salmon and steelhead from higher water
temperatures in the Columbia River. Further, predicted continual future warming of the Lower
Columbia River is expected to increase salmon and steelhead use of CWR and diminish the
extent to which the current amount of CWR reduces the risks to migrating adult salmon and
steelhead. Therefore, increasing the amount of CWR in the future through restoration and
enhancement is recommended to help offset the predicted increased future adverse effects
associated with a warmer Lower Columbia River.
It is important to note that EPA's assessment of CWR needed to attain Oregon's CWR narrative
criteria does not imply that current Columbia River temperatures are at levels to protect salmon
and steelhead migration. Current river temperatures exceed the 20°C numeric criterion and
cause adverse effects to salmon and steelhead which are not fully mitigated by CWR.
88

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7 ACTIONS TO PROTECT & RESTORE COLD WATER REFUGES
As summarized in Chapter 6, EPA's assessment is that to provide sufficient cold water refuges
(CWR) in the Lower Columbia River for migrating adult salmon and steelhead, it will be
necessary to maintain the existing amount of cold water that is provided by the 12 primary CWR
tributaries and to provide increased CWR in the Umatilla River. This chapter summarizes
actions to protect and restore the 12 primary CWR tributaries to both avoid human actions that
could increase temperatures and to cool temperatures to partially or fully counteract predicted
warming from climate change (Appendix 12.15). In addition, this chapter summarizes actions to
restore CWR in the Umatilla River. EPA also included
Fifteenmile Creek to highlight a tributary with potential to be
restored into a CWR based on the temperature TMDL
suggesting substantial cooling potential and the fact that
Fifteenmile Creek has been prioritized for restoration for ESA-
listed steelhead recovery. These 14 tributaries are illustrated in
Figure 7-1.
The other ten non-primary CWR tributaries identified in Chapter
2 may be able to increase the amount of CWR near their confluence areas, if restored. Due to
time limitations, EPA did not address those tributaries (Appendix 12.20).
A temperature TMDL is a
waterbody plan that sets the
maximum amount of heat
allowed to enter a waterbody
so that the waterbody will
meet temperature water
quality standards.
iiSwisiRfcgj;
McNary
Dam
Bonneville
Dam fl,
Dam Location
O Primary CWR tributary locations
£ Restore CWR tributary locations
Figure 7-112 primary and 2 "restore" cold water refuge tributary locations
89

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.1	COLD WATER REFUGE WATERSHED SNAPSHOTS
EPA developed "cold water refuge watershed snapshots" of the 12 "primary" CWR tributaries
and the two "restore" CWR tributaries to highlight information about the CWR and their
respective watersheds. The snapshots describe the quality and characteristics of each refuge,
background on the watershed, features of the watershed that can affect CWR quality, and
actions in the watershed that can protect and restore the CWR.
One focus of the snapshots is to identify watershed features that help to maintain CWR quality.
These are used as the basis for actions to protect those watershed features. A second focus is
to identify features that degrade CWR quality. These are used as a basis for restoration actions
to reduce temperatures and potentially offset future warming from climate change. These
protection and restoration actions are regulatory - related to management actions already
established - and voluntary in nature. Whenever possible, an effort is made to identify agencies
and organizations that have jurisdictional authority over the actions.
The actions are also intended for local stakeholders and regional planning groups to use in
focusing their work and leveraging resources for projects that protect and restore CWR. Most of
the restoration actions are actions identified in salmon recovery and watershed restoration plans
to benefit species within the watersheds. To this end, the snapshots emphasize ongoing work in
the watersheds that provide multiple local benefits in addition to enhancing CWR and put a
spotlight on the important regional benefits provided by these restoration actions.
To develop these snapshots, EPA relied on work described in the previous chapters regarding
CWR plume volume, upstream extent of fish use, and documented fish use by migrating
salmonids. EPA also developed maps for the land cover and land ownership in each CWR
tributary and conducted other analyses for riparian cover and water allocation. For background
on different activities in each watershed, EPA conducted a literature search relying heavily on
subbasin plans, regional salmon recovery plans, and local watershed priority plans. See
Appendix 12.20. Chapter 11 includes a bibliography of the sources for each snapshot.
EPA shared drafts of these documents with interested parties in the basin including Tribal
Governments, Lower Columbia Estuary Partnership, counties, Washington Department of Fish
and Wldlife (WDFW), Oregon Department of Fish and Wildlife (ODFW), Washington
Department of Ecology (Ecology), Oregon Department of Environmental Quality (ODEQ), U.S.
Forest Service (USFS), NMFS, watershed councils, and other groups. The snapshots are
relatively concise, providing a brief overview of the watersheds, distilling meaningful information
for stakeholders, and including actions to protect and restore the CWR.
More detail on the development of the snapshots is included in Appendix 12.20.
7.2	CHARACTERISTICS OF PRIMARY COLD WATER REFUGE TRIBUTARIES
Each of the 12 primary CWR tributaries have characteristics that help to create and maintain
cold temperatures during the summer. Figure 7-1 shows that all of the 12 primary CWR
tributaries originate from volcanic mountains or forested areas in the Cascade Mountain Range
in Washington and Oregon. Many of the tributaries have a large percentage of forest land within
their basins, much of which is federal forest land.
The Cowlitz River, Lewis River, and Sandy River share similar features. They are the three most
downstream CWR in the Columbia River below Bonneville Dam, whose headwaters include
90

-------
Columbia River Cold Water Refuges Plan
Final January 2021
volcanic mountains that provide snowmelt runoff in the summer. Along with forested
headwaters, each of these rivers have development (urban, rural, and/or agricultural) in the
lower part of their basins and have dams that deliver cool regulated summer flow to the lower
section of these rivers.
Tanner Creek, Eagle Creek, and Herman Creek are small, well-forested watersheds that are
part of the Columbia River Gorge with cool river temperatures. Wind River, Little White Salmon
River, White Salmon River, and Hood River are moderate sized basins, include a significant
percentage of forested land, and also drain into the Columbia River Gorge. The White Salmon
River and Hood River also have a significant amount of farmland.
The Klickitat River and Deschutes River are located east of the Cascade Mountain range, where
the climate is significantly drier and warmer and the percentage of forested land drops
significantly. However, both tributaries have volcanic geology which creates opportunity for
groundwater infiltration, important for providing a reliably steady source of cold water in the
summer which enhances CWR quality. The lower Deschutes River's flow and temperature are
influenced by the Pelton-Round Butte Dam 100 miles upstream of the mouth.
Herman Creek and the Little White Salmon River are unique because they drain into artificial
cove areas created by infilling (Herman Cove) and by a highway (Drano Lake). These
embayments pool inflowing cool tributary flows, creating coves that provide CWR.
Table 7-1 includes a temperature-based classification of CWR quality based on optimal and
sub-optimal water temperatures for fish from EPA's Region 10 Temperature Guidance
(Appendix 12.20):
~	"Excellent" cold water refuge - Average August tributary temperatures cooler than 16°C.
~	"Good" cold water refuge - Average August tributary temperatures 16-18°C.
~	"Marginal" cold water refuge - Average August tributary temperatures greater than 18°C.
91

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Table 7-1 Location and characteristics of primary cold water refuges

Watershed Characteristics
River Name and
Location/

Watershed
River
Percent
Dam
CWR Quality
River Mile
Headwaters
Size
(square
miles)
Length
(miles)
Forested
Influenced
Cowlitz River
Below Bonneville
Mt. Rainier




(good)
Dam
(RM 65.2)
Mt. St. Helens
Mt. Adams
2,586
105
62%
X
Lewis River
Below Bonneville





(good)
Dam
(RM 84.4)
Mt. Adams
Mt. St. Helens
1,046
95
66%
X
Sandy River
Below Bonneville





(marginal)
Dam
(RM 117.1)
Mt. Hood
508
56
77%
X
Tanner Creek
Below Bonneville
Mt. Hood




(excellent)
Dam
(RM 140.9)
National
Forest
46
6
87%

Eagle Creek
Bonneville Dam
Mt. Hood




(excellent)
Reservoir
(RM 142.7)
National
Forest
90
15
90%

Herman Creek
Bonneville Dam
Mt. Hood




(excellent)
Reservoir
(RM 147.5)
National
Forest
50
8
98%

Wind River
Bonneville Dam
Gifford




(excellent)
Reservoir
Pinchot





(RM 151.1)
National
Forest
225
30
84%

Little White
Bonneville Dam
Gifford




Salmon River
Reservoir
Pinchot




(excellent)
(RM 158.7)
National
Forest
136
19
70%

White Salmon
Bonneville Dam





River
Reservoir
Mt. Adams
400
44
66%

(excellent)
(RM 164.9)





Hood River
Bonneville Dam





(excellent)
Reservoir
(RM 165.7)
Mt. Hood
279
25
62%

Klickitat River
Bonneville Dam





(good)
Reservoir
(RM 176.8)
Mt. Adams
1,350
96
48%

Deschutes
The Dalles Dam
Mt. Hood




River
Reservoir
Mt. Jefferson
10,500
252
32%
X
(marginal)
(RM 200.8)
Three Sisters




92

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.3 COWLITZ RIVER (RIVER MILE 65) - PROTECT AND ENHANCE
Refuge Volume: 1,554,230 m3 (largest)
Average August Temperature: 16°C
Distance to Downstream Refuge: N/A
Distance to Upstream Refuge: 19 mi. (Lewis
River)
Cold Water Refuge Rating: Good (16-18 C)
Photo 7-1 Cowlitz River
What features make the Cowlitz River an
important cold water refuge to protect
and enhance?
The Cowlitz River enters the Columbia River
at river mile 65, about 3.5 miles south of
Longview, Washington. Cowlitz River
temperatures in August average 16 C,
almost 5 C cooler than the Columbia River's
average August temperature of 20.75 C.
This makes the Cowlitz River a good CWR
(<\ 6_i 8°C)
^	''	Photo 7-2 Aerial view of the Cowlitz River; yellow piri denotes
The lower portion of the Cowlitz River is ups,ream extent ofrefuge
designated for salmonid spawning, rearing, and migration by the Washington Department of
Ecology, which assigns a water quality criterion of 17.5°C for maximum water temperatures.
The maximum modeled temperature for the Cowlitz River is 21 °C (1993-2011) (Appendix
12.18). Based on measured maximum temperature readings, the lower Cowlitz River is on the
303(d) list for temperature impaired
waters. The Cowlitz River is the first
major tributary upstream of the mouth of
the Columbia where migrating salmonids
can seek refuge during their migration,
likely using both the mouth and lower
portion of the refuge, estimated to be
1.75 miles upstream (yellow pin, Photo
7-2). Of the tributaries along the lower
Columbia River, the Cowlitz River has
the largest volume of cold water at the
confluence in summer months. In
August, the Cowlitz River has an
average flow of 3,634 cfs, which
Photo 7-3 Map of the Cowlitz River Basin	produces a CWR estimated to be
93

-------
Columbia River Cold Water Refuges Plan
Final January 2021
1,554,230 cubic meters,
or approximately 622
Olympic-sized swimming
pools. The next available
cold water refuge for
migrating saimonids
leaving the Cowlitz River
is 19 miles upstream in
the Lewis River.
Introduction to the
Cowlitz River
Watershed
The Cowlitz watershed
drains heavily-timbered
mountainous slopes
surrounding Mount
Rainier, Mount Adams,
Mount St. Helens, and
I	J Barren ¦ z 4%
¦ Forest -61.6%
[	j Kianretvgumvasea - 3.4%
[ ) Wetland-2.5%
0	5 10 Mifes
	1	I	I
Sources Esn. USGS. NOAA
Figure 7-2 Cowlitz River land cover
the Goat Rocks Wilderness. Flowing for 105 miles in a west-southwest direction, the mainstem
Cowlitz passes through the cities of Kelso and Longview near its confluence with the Columbia
River. Mayfield Dam at River Mile 42 divides the Cowlitz River watershed into an Upper and
Lower Basin.
National Park Service
own and manage most of
the upper basin; in total,
public agencies own
approximately half of the
watershed. Forest covers
Figure 7-2 and Figure 7-3 show land cover and ownership in the Cowlitz watershed. A large
extent of the upper basin is in the Mount Rainier National Park and the Gifford Pinchot National
Forest. Together the U.S.
Forest Service and
STATE LAND-5 1%
TRIBAL LAND - <1%
nearly two-thirds of the
watershed - particularly
in the upper basin where
high levels of riparian
canopy cover shade
headwater streams,
helping to maintain cool
water temperatures.
Shrubland (18%) grows in
fragmented patches
throughout the watershed. Nearly the entire Lower Basin is privately owned. Cultivated crops
(-3%) and developed areas (-5%) are concentrated along the mainstem and lower tributary
valleys below Mayfield Dam and near the river mouth, respectively.
5 10 Miles
_J	I
Sources Esn. USGS. NOAA
Figure 7-3 Cowlitz River land ownership
94

-------
Columbia River Cold Water Refuges Plan
Final January 2021
The Toutle River, which enters the Cowlitz at river mile 20, is a major tributary that drains Mount
St. Helens. In 1980, the volcano's eruption filled the Toutle Valley with billions of tons of erodible
debris. Increased sediment loads can lead to the widening and shallowing of rivers and, as a
result, can increase water temperature. The U .S. Army Corps of Engineers constructed
sediment retaining dams on the Toutle and continuously dredge the channels of both the Toutle
and Cowlitz Rivers.
Factors that Influence Temperature in the Cowlitz River Watershed
Riparian
Vegetation: The
Cowlitz River
watershed has
well-forested areas
in the tributaries of
the upper
watershed. Figure
7-4 shows the
difference between
the maximum
potential and
current shade,
demonstrating
which areas have
the highest
restoration
potential. The lower
mainstem Cowlitz
River and
associated tributaries are not as well shaded as the upper basin. The riparian forests along the
lower 20 miles of the Cowlitz River have been severely degraded through industrial and
commercial development, and
between river miles 20 and 52 the
river lacks mature forests and
adequate buffer widths. Riparian
shade has been degraded on
private commercial forest lands,
which cover much of the lower
Cowlitz basin, but shade is
expected to improve through time
and implementation of
Washington's State Forest Practice
Rules. Loss of riparian shade is
likely a primary cause of several
tributaries to the Lower Cowlitz
River, including Coweeman River,
Ostrander Creek, and
Figure 7-4 Cowlitz River shade difference between potential maximum and current shade
122° 30' W	122° 00' W
Figure 7-5 Map of Cowlitz River Dams
95

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Photo 7-4 Cowlitz River as seen from above
Arkansas/Monahan/Delameter Creeks,
being listed on the state's 303(d) list for
temperature impairment. These warm
creeks have daily maximum temperatures
that exceed 20°C.
Hydromodification: The Cowlitz River is
currently modified by three hydroelectric
dams in the upper basin (Figure 7-5).
Tacoma Power operates the Mossyrock and
Mayfield Dams; Bonneville Power
Administration (BPA) operates the Cowlitz
Fails Dam. The Mossyrock Dam is the
tallest dam in Washington State and forms
23.5-mile-long Riffe Lake. At river mile 52,
Mayfield Dam, built in 1956, blocks natural passage of anadromous fish. Tacoma Power's
FERC license for Mayfield Dam requires 2,000 cfs of minimum flow below the dam in August,
which approximates August flow prior to the building of the dam. Typically, however, August
flows below the dam are higher than this level, which provide most of the flow in the Lower
Cowlitz River from the dam to the mouth. The lower 20 miles of the Cowlitz River was
channelized to facilitate industrial,
agricultural, and urban
development, resulting in a
significant loss in floodplain
function.
Water Use: Tacoma Power has
senior water rights in the region for
power production for the two dams,
but as noted above has minimum
flow requirements below Mayfield
Dam which provide significant flow
to the Lower Cowlitz River that
Photo 7-5 Sediment retaining structure on the north fork of the Toutle River,
which eventuaiiy flows to the Cowlitz
helps maintain cool summer river temperatures. Currently there are no instream flow rules
(water rights to protect fish) for the Lower Cowlitz River with current flow viewed as sufficient to
meet future anticipated demands. The Cowlitz River watershed is intensely farmed based on
Washington Department of Ecology's Water Availability Summary (2012). Irrigation withdrawals
have contributed to low flows and high water temperatures in several of the tributaries to the
Lower Cowlitz River. WDFW has issued surface water source limitations (SWSLs) for minimum
instream flow on the Lower Cowlitz River and Salmon Creek and for closure to new water rights
for Arkansas Creek, Olequa Creek, and Hazel Dell Creek to protect fish from low flows. SWSLs
serve to advise Ecology on the issuance of new water rights. Limiting additional water
withdrawals can help maintain cool river temperatures and the CWR volume of the Lower
Cowlitz River.
Climate Change: In 2040, average August temperatures in the Cowlitz River are predicted to
rise to 17 C compared to 23°C in the Columbia River. In 2080, August temperatures in the
Cowlitz River are expected to rise further to 18 C compared to 24 C in the Columbia River.
Therefore, the Cowlitz River could still be considered a marginal CWR by 2080. However, as
96

-------
Columbia River Cold Water Refuges Plan
Final January 2021
temperatures rise, mountain glaciers which help the Cowlitz River stay cool, will recede. Studies
at the University of Washington have shown that climate change will likely exacerbate low
summer flows in the mainstem Cowlitz River, because of lower snowpack melt in the summer.
Ongoing Activities in the Cowlitz River Watershed and Recommended Actions to Protect
and Enhance the Cold Water Refuge
In 2010, the Washington Lower Columbia Salmon Recovery and Fish and Wildlife Subbasin
Plan, which includes the Lower Cowlitz and Coweeman subbasins, was adopted by the Lower
Columbia Fish Recovery Board as an integrated plan for salmon recovery, the Northwest Power
and Conservation Council (NPCC) fish and wildlife program, and Washington State watershed
management. This plan was adopted by NMFS in 2013 as the salmon recovery plan under the
ESA. The management plans detail key priorities contributing to recovery and mitigation in the
basin, such as managing regulated stream flows through the hydropower system and restoring
floodplain and riparian function. Specific restoration projects for the Lower Cowlitz River have
been identified in the Lower Cowlitz River and Floodplain Habitat Restoration Project Siting and
Design Report (2007).
The Watershed Plan for WRIA 26 (2005) and associated updated WRiA 26 Water Supply and
Streamflow Flow Review (2014) adopted by Cowlitz and Lewis Counties provide
recommendations to Ecology for water resources in the Lower Cowlitz River. The
recommendations in the 2014 update include reservations for future use along with closure and
instream flow rules for most of the Lower Cowlitz River tributaries, including the Coweeman,
Ostrander, Arkansas/Delameter/Monahan, Olequa, Lacamas, Mill, and Salmon Creeks.
Although the 2005 plan called for closure of new water rights for the Lower Cowlitz River, the
2014 update recommended it remain open to future appropriations due to adequate flows with
reservations for the counties and cities.
Cowlitz County and the U.S. Army Corps of Engineers maintain levees and flood control in the
river to regulate legacy sediment contributions caused by the Mount St. Helens eruption. In
2013, USACE initiated a $4.5 million project to construct a sediment retention structure on the
Toutle River to prevent further sediment seepage into mainstem Cowlitz River. This action helps
reduce sediment deposition into the Cowlitz River CWR.
The Capitol Land Trust manages a 17-acre land parcel along the Lower Cowlitz River, including
1,500 feet of streambank which protects and maintains critical habitat for salmonids and other
wildlife species.
As the largest CWR used by migrating salmonids, the Cowlitz River is an important refuge to
protect and enhance. Actions to protect and enhance the Cowlitz River CWR include:
•	On National Forest Lands, continue to implement the
(1990) and its amendments, which include
to protect and restore riparian shade and
stream functions to maintain cool river temperatures. (USFS)
•	Continue to implement	on private forest lands
and the on
state lands to protect and restore riparian shade and stream functions to maintain
cool river temperatures. (WDNR)
97

-------
Columbia River Cold Water Refuges Plan
Final January 2021
•	On private and county lands, continue to implement the riparian protections in the
Cowlitz and Lewis Shoreline Master Programs ( ; ) and critical areas
ordinances to regulate development in the Lower Cowlitz River and its tributary
shoreline areas to protect and restore riparian shade and stream functions to maintain
cool river temperatures. (Cowlitz County and Lewis County)
•	Implement actions and projects in the
(2010) and the
(2007) to restore riparian shade, floodplain
functions, and channel complexity that improve salmon habitat and maintain cool
temperatures in the Lower Cowlitz River and tributaries. (Multiple parties)
•	Address temperature impairments in the Lower Cowlitz River basin by supporting
riparian restoration and other projects, many of which are identified in existing plans,
and establish a water clean-up plan/TMDL alternative or temperature TMDL, as
warranted. (Ecology)
•	Continue to provide cool summer flows from Mayfield Dam per the	to
maintain the CWR volume and temperatures. (Tacoma Power)
•	Consider adopting a watershed management rule with the reservations, closures, and
minimum instream flows as recommended in the
(2014) to balance future water uses with maintaining future
flow and Cowlitz River CWR volume. Consider a revised SWSL or an instream flow rule
for the Lower Cowlitz River to help maintain cool river temperatures and protect CWR
volume.(Ecology, WDFW)
•	Continue sediment retention on the Toutle River to prevent excess sedimentation at the
confluence of the Cowlitz River. (Army Corps)
•	Continue to develop state and local partnerships with local land trusts, like the Capitol
Land Trust, to obtain and preserve pieces of land to keep riparian areas intact. (Multiple
parties)
98

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.4 LEWIS RIVER (RIVER MILE 84) - PROTECT AND ENHANCE
Photo 7-6 Lewis River looking upstream towards railroad
bridge
What features make Lewis River an
important cold water refuge to protect
and enhance?
Refuge Volume: 613,455 m3 (4th largest)
Average August Temperature: 16.6 C
Distance to Downstream Refuge: 19 mi.
(Cowlitz River)
Distance to Upstream Refuge: 33 mi. (Sandy
River)
Cold Water Refuge Rating: Good (16-18 C)
The Lewis River, located at river mile 84.4
of the Columbia River, provides a
significant CWR below Bonneville Dam.
Average August water temperatures in the
Lewis River are estimated to be 16.6 C,
approximately 5 C colder than the
Columbia River. This classifies the Lewis
River as a qood CWR (16-18°C) The Lewis
y	v " ' ,	P/iofo 7-7 Aerial View of Lewis River at the Confluence with
River GWR is 19 miles upstream of the	Columbia River; yellow pin denotes upstream extent; Photo: Google
Cowlitz River CWR. The Lewis River CWR Earth
includes the confluence area and an estimated 1.7 miles upstream (yellow pin, Photo 7-7).
The Washington Department of Ecology designates the lower portion of the Lewis River for
salmonid spawning, rearing, and migration and
assigns a water quality criterion of 17.5C for
maximum water temperatures. The maximum
modeled temperature for the Lewis River is 20.8°C
(1993-2011) (Appendix 12.18). Based on
measured maximum temperature readings, the
Lower Lewis River is on the 303(d) list for
temperature impaired waters. The Lewis River's
relatively high discharge averages 1,291 cfs in
August. The Lewis River CWR, including the lower
portion of the river and the plume, is estimated to
be 613,455 cubic meters, the fourth largest refuge
in the Columbia River and the size of
photo 7-8 Lower Lewis River Falls	approximately 245 Olympic-sized swimming pools.
Fall Chinook salmon, and steelhead trout leaving
the Lewis River will swim 33 miles before reaching the next refuge in the Sandy River.
99

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 7-6 Lewis River land ownership
The Lewis River watershed
drains the southern slopes
of Mount St. Helens and the
western flank of Mount
Adams. For most of its
journey, the Lewis River is
synonymous with the North
Fork Lewis River. The
smaller East Fork joins the
North Fork to form the
mainstem Lewis River 3.5
miles above the confluence
with the Columbia River.
0	5 10	20 Miles
	1	J	 I 	i I
Sources Esn, USGS. NOAA
Introduction to the Lewis
River Watershed
Figure 7-7 Lewis River land cover
watershed is forested.
Shrubland (15%) and
grassland (5%) are found in fragments throughout the basin. In its last 12 miles, the Lewis River
flows through a broad valley predominated by cultivated crops (4%) and urban development,
including the City of Woodland and the rapidly growing community of Battle Ground (Figure
7-7). The East Fork Lewis River is impaired for temperature with exceedances of maximum
water temperatures of 16°C, the water quality criteria for core salmonid habitat.
Both forks of the Lewis River
have steep, heavily forested
headwaters in the Gifford
Pinchot National Forest
managed by the U.S. Forest
Service (Figure 7-6). The
North Fork begins on the
western slope of Mount
Adams, while the East Fork
Lewis originates near Green
Lookout Mountain in the
southern portion of the
watershed. Approximately
two-thirds of the entire
HQ Water-2.2%	Q_) Shrubiand -14.6%
Developed - 5.1%	| ) Grassland - 4.5%
( | Barren-1,5%	( ) Planted/Cultivated - 3.9%
Forest-65.8%	( ) Wetland -2.4%
0	5 10	20 Miles
	1	I	I	I	I
Sources: Esri, USGS, NOAA
A series of dikes along the lower 7 miles of the Lewis River protect farmland and urban
development. The dikes and associated channel modifications are estimated to have
disconnected the river from more than half of its historic floodplain.
Factors that Influence Temperature in the Lewis River Watershed
Protecting and Enhancing Riparian Vegetation: Shade levels are high on most of the upper
tributaries of the North Fork Lewis River, but shade levels are significantly lower in its middle
reaches (Figure 7-8). The lowest levels of shade are found on the impounded sections of the
100

-------
Columbia River Cold Water Refuges Plan
Final January 2021
mainstem Lewis River
(Swift Reservoir, see
Figure 7-8), where the
reservoir is much wider
than the stream would
be, inhibiting the ability
of riparian vegetation to
shade the water surface.
Figure 7-8 shows that
overall stream shade is
close to its potential or in
reasonable shape, with
portions of the lower
reaches having the
greatest potential for
stream shading. The
2010 Washington Lower
Columbia Salmon
Recovery and Fish and
Wildlife Subbasin Plan noted poor riparian conditions on the mainstem between the mouth and
river mile 15. Further, the East Fork Lewis is currently listed as impaired for temperature.
Washington Department of Ecology completed the East Fork Lewis River Watershed Bacteria
and Temperature Source Assessment Report (2018) for the East Fork Lewis in 2018 that
specifies reaches lacking riparian shade that contribute to temperature exceedances.
Dams and Hydromodifications: PacifiCorp operates three dams on the North Fork Lewis that
have substantial impact on anadromous salmon: Merwin (1931), Yale (1953), and Swift (1958)
(iPhoto 7-9). Merwin Dam, the most downstream structure, is at river mile 19.5. The hydropower
operations have altered the natural hydrology by decreasing peak flows that historically flooded
the lower valley. Decreased peak flows coupled with extensive channelization of the Lower
Lewis River through dikes and bank stabilization have reduced floodplain functions that help
maintain cool river temperatures. PacifiCorp received a new 50-year FERC license in 2008 that
includes minimum flows downstream of Merwin Dam of 2,300 -1,500 cfs from July 1 to July 30
and 1,200 cfs from July 31- October 15, which approximates pre-dam flows. Water releases
from Merwin Dam are taken from Merwin Reservoir at a fixed depth of about 178 feet below the
surface when the reservoir is at full pool. Because the reservoir is stratified in the summer with
cool water at depth, the dam
delivers relatively cool, stable
flows in August. The cool flows
from Merwin Dam are important
for cool river temperatures at the
mouth that provide the CWR.
The lower part of the East Fork
Lewis flows through a broad,
alluvial valley that historically was
an active floodplain with diverse
riparian forests. Channel
101
Figure 7-8 Lewis River shade difference between potential maximum and current shade

MERWIN DAM
Figure 7-9 Map of Lewis River dams

-------
Columbia River Cold Water Refuges Plan
Final January 2021
modifications over the years have dramatically
altered natural channel migration, floodplain
processes, and riparian shading, contributing to
warm river temperatures and degraded fish
habitat.
Water Use: Senior water rights for PacifiCorp to
maintain reservoir levels in Lake Merwin and Yale
Lake limit the water available for new sources in
the Lewis River upstream of the dams. In addition,
farms on the Lower Lewis River hold surface water
rights for irrigation. Since snowpack is depleted in
the summer, the demands for water are greatest
when the supply is lowest, the same time that
migrating salmon use the Lewis River mouth as a
refuge.
Washington Department of Ecology has assigned instream flow rules (water rights to protect
fish) at several locations in the basin. Minimum instream flows at river mile 19 of the Lewis River
range from 1,200-2,700 cfs between June and August. For the East Fork Lewis River, minimum
instream flows at river mile 10.1 in the summer
range from 122-420 cfs. There are also areas within
each basin where additional flow withdrawals are not
allowed, including the Lower Lewis River upstream
of river mile 7.1 and the East Fork Lewis River. The
Salmon-Washougal & Lewis Watershed
Management Plan, WRIAS 27-28 adopted in 2006
provided the analysis and recommendations for
Ecology's 2008 instream flow and closures rules
noted above for Lewis River basin.
Climate Change: in 2040, August temperatures in
the Lewis River are projected to rise to 18°C,
compared to 23°C in the Columbia River. In 2080,
August temperatures are expected to further rise to
19°C compared to 24°C in the Columbia River.
Therefore, increases in Lewis River temperatures are expected to shift the refuge from a good
quality refuge (16-18°C) to a marginal quality refuge (>18 C). Still, the Lewis River is expected to
be 5°C cooler than temperatures in the Columbia River in the summer, even under climate
change projections.
Ongoing Activities in the Lewis River Watershed and Recommended Actions to Protect
and Enhance the Cold Water Refuge
Groups such as Clark County, Clark County Conservation District, Cowlitz Indian Tribe, non-
profit organizations, private citizens, and state and federal agencies have identified and
prioritized projects in the Lewis River. Recent plans include the Washington Lower Columbia
Salmon Recovery and Fish and Wildlife Subbasin Plan (2010), which includes North Fork Lewis
and East Fork Lewis subbasin plans, and the Lower East Fork Lewis River Habitat Restoration
Plan (2009). The 2010 plan, adopted by the Lower Columbia Fish Recovery Board as an
102

-------
Columbia River Cold Water Refuges Plan
Final January 2021
integrated plan for salmon recovery, the Northwest Power and Conservation Council Fish and
Wildlife Program, and Washington State watershed management was adopted by NMFS in
2013 as the salmon recovery plan. These plans provide a comprehensive assessment of
restoration needs and project priorities in the basins. Recommended actions include increasing
floodplain function, restoring riparian habitat, and increasing channel complexity, which can help
maintain cool river temperatures as well as improve salmon habitat. In addition, the Washington
Department of Ecology is currently developing a water clean-up plan/TMDL alternative to
address warm temperatures in the East Fork Lewis River based on the 2018 temperature
source assessment.
Actions to protect and enhance the Lewis River CWR include:
•	On National Forest Lands, continue to implement the
(1990) and its amendments, which include
to protect and restore riparian shade and
stream functions to maintain cool river temperatures. (USFS)
•	Continue to implement	on private forest lands
and the on
state lands to protect and restore riparian shade and stream functions to maintain
cool river temperatures. (WDNR)
•	On private and county lands, continue to implement the riparian protections in the
Cowlitz and Clark County Shoreline Master Plans ( ; ) and critical areas
ordinances to regulate development in the Lewis River and East Fork Lewis River
shoreline areas to protect and restore riparian shade and stream functions to
maintain cool river temperatures. (Cowlitz County and Clark County)
•	Continue to provide cool summer flows from Merwin Dam per the	to
maintain the CWR volume and temperatures. (PacifiCorp)
•	Continue to implement instream flow and new water consumptive use closures per
Ecology's water management rule for the Lewis River basin (WRIA 27). (Ecology)
•	Continue to implement actions and projects in the
(2010) and the
(2009) to restore riparian shade, floodplain functions, and
channel complexity that improve salmon habitat and help maintain cool temperatures
in the Lower Lewis River and East Fork Lewis River. (Multiple parties)
•	Complete the East Fork Lewis River water clean-up plan and implement the plan's
actions to reduce river temperatures. (Ecology)
103

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.5 SANDY RIVER (RIVER MILE 117) - PROTECT AND ENHANCE
Refuge Volume: 31,915 m3(11th largest)
Average August Temperature: 18.8 C
Distance to Downstream Refuge: 33 mi.
(Lewis River)
Distance Upstream Refuge: 24 mi. (Tanner
Creek)
Cold Water Refuge Rating: Marginal (>18°C)
Photo 7-12 Upper Sandy River
What features make the Sandy
River an important cold water
refuge to protect and enhance?
The Sandy River is located at river
mile 117 of the Columbia River,
downstream of the Bonneville Dam.
Sandy River temperatures in August
are 2.5 C cooler than the Columbia Photo 7_13 Aeria| view of Sandy Rlver de!ta at the confluence with Columbia
River, averaging 18.8 C. This makes River; yellow pin denotes upstream extent
the Sandy River a marginal CWR (>18°C) for migrating salmonids. The Sandy CWR is 33 miles
upstream of the Lewis River CWR. ODEQ assigns a water quality criterion of 18°C for maximum
temperatures to protect salmonid rearing and migration in the lower portion of the Sandy River.
The maximum modeled temperature for the Sandy River is 23.6C (1993-2011) (Appendix
12.18). Based on measured maximum temperature readings, the Lower Sandy River is on the
303(d) list for temperature impaired waters. Migrating salmon are thought to use the confluence
of the rivers and an estimated 1.10 miles up the Sandy Riveras a CWR (yellow pin, Photo 7-
13). Bull Run River, a major tributary to the Sandy River, supplies the drinking water for the City
of Portland, and withdrawals from Bull Run River affect the amount of water that reaches the
		- Sandy River. The Sandy River mainstem is
currently undammed from the headwaters
to the confluence, helping temperatures
stay cooler with a more natural flow
regime. Historical lahars (fast-moving
mudflows) formed a large debris fan with a
braided channel in the lower reaches and
mouth of the Sandy River, and the glacier
that feeds the Sandy River is heavily laden
with sediment. Sediment build-up at the
mouth can make the refuge shallower and
Photo 7-11 sandy River at Dodge Park, upstream of confluence subsequently warmer over time. The Sandy
104

-------
Columbia River Cold Water Refuges Plan
Final January 2021
River is the tenth largest CWR in the Lower Columbia River with an estimated volume of 31,915
m3, the size of approximately 13 Olympic-sized swimming pools, and a mean flow of 469 cfs.
The next upstream CWR is 24 miles away in Tanner Creek.
Introduction to the Sandy River
Watershed
Glaciers on the western slopes of Mount
Hood feed the Sandy River. Much of the
upper basin is protected as part of the
Mount Hood National Forest and remains
heavily forested. The Sandy River
watershed includes the Bull Run River
subbasin, Portland's drinking water
source. Given its proximity to the Portland
metropolitan area and its high quality
natural areas, the Sandy River watershed
is a popular recreation area.
Approximately 25 miles of the Upper and
Lower Sandy River is designated as a
federal Wild and Scenic River and state
Scenic Waterway. The Upper Sandy River
has a wild designation for 4.5 miles and a
recreational designation for 16.6 miles.
The Lower Sandy River has a scenic
designation from Dodge Park (river mile
18.75) to Dabney Park (river mile 6). The Wild and Scenic designations and the Bull Run River
watershed's status as an important
drinking water source provide
protections by limiting development in
the middle and upper watersheds.
Approximately three-quarters of the
watershed is forested, predominately in
the Mount Hood National Forest which
makes up about 2/3 of the watershed
(,Figure 7-10 and Figure 7-11). The
lower watershed is mostly privately
owned. The lower watershed is also in
the Columbia River Gorge National
Scenic Area, which includes Forest
Service, state, and private lands. The flat
topography of the lower watershed
supports a mix of cultivated crops (4%)
and the cities of Gresham and Troutdale,
the only significant areas of developed
land other than State Highway 26, which
winds through the watershed before
passing south of Mount Hood.
Figure 7-10 Sandy River land ownership
Figure 7-11 Sandy River land cover
105

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 7-13 Sandy River Delta Dam pre-removal - white line indicates location
of former dam (USACE, 2015)
Protecting and Enhancing
Riparian Vegetation: The
Sandy River watershed has
high levels of riparian shade
throughout the upper and
middle forested tributaries.
These are federal, state, and
private lands that are
governed by the USFS
Mount Hood Forest Land
and Resource Management
Plan (1990) and its
amendments, which include
the Aquatic Conservation
Strategy, Oregon's State
Forest Management Plan,
and Oregon's Forest Practices Act. The Upper Sandy River Basin is in designated wilderness
and is subject to limited management. This shade serves to block solar radiation and maintain
cool stream temperatures. However, there are reaches that have been degraded and have
potential for increased shade in the Lower Sandy River. Shade from riparian vegetation reduces
solar exposure to the stream channel and helps maintain cool water temperature. Figure 7-12
shows the difference between
maximum and current shade levels
highlighting the reaches that could
benefit the most from riparian
revegetation . Beaver and Keily
Creeks, tributaries to the Lower
Sandy River, have the greatest
potential for more riparian shade.
Figure 7-12 Sandy River shade difference between potential maximum and current shade
Factors that Influence
Temperature in the Sandy
River Watershed
Water quality modeling in ODEQ's
Sandy River Basin TMDL (2005)
predicted a temperature decrease
of approximately 0. 5°C with
maximum potential vegetation
under low flow conditions.
Increased riparian shade can help
to reduce sedimentation and
maintain CWR volumes and
temperatures.
Dams and Hydromodifications: The mainstem Sandy River is currently undammed for 56
river miles from the headwaters to the confluence. The removal of several dams, including
106

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Marmot Dam (2007), the Little Sandy Diversion Dam
(2008), and the Sandy River Delta Dam (2013) has
restored a more natural flow regime, increased floodplain
connectivity, and added channel complexity. The Sandy
River Delta Dam (Figure 7-13) had blocked the east
channel of the delta, impeding fish passage and access.
The U.S. Army Corps of Engineers identified habitat
improvements from removal of the Sandy River Delta Dam
as including year-round access for salmon to the east
channel, cooler waters in the east channel during the
summer, and additional shallow water. The State of Oregon
owns the land under the East Channel, and the Forest
Service owns most of the rest of the Sandy River Delta, all
of which is part of the Columbia River Gorge National
Scenic area.
The Bull Run River is a significant tributary to the Sandy
River, which includes two reservoirs that provide drinking
water for the City of Portland. Historically, the unused water
from the top of the thermally-stratified Bull Run reservoirs
was released to the Bull Run River and warmed
temperatures in the Sandy River. In the past few years,
however, the Portland Water Bureau has used a selective withdrawal system to release higher
volumes of colder water in the summer,
which has resulted in colder waters
reaching the Sandy River. This along
with other measures in the Bull Run
Water Supply Habitat Conservation
Plan (2008) have helped to reduce
harmful effects to salmon from the Bull
Run River reservoirs.
The State of the Sandy (2017) report by
the Sandy River Watershed Council
indicates that a dam on Kelly Creek on
the Mount Hood Community College
campus creates an artificial pond which
raises temperatures as much as 4 C in
the summer. The community college is
considering removing this dam, which
could cool the water temperatures in the
lower Sandy watershed. Other dams
continue to operate on many tributaries
to the Sandy River.
Water Use: Water availability is
overallocated in the Sandy River
primarily due to Portland's diversion of
the Bull Run River for its drinking water.
107
Photo 7-12 East Channel post-Sandy Delta
Diversion Dam removal (USAGE)
Table 7-2 Water Availability Analysis, 5/20/20, Sandy River at mouth,
Oregon Water Resources Department
SAN DY RIVER>COLUMBIA R - AT MOUTH
(@ 80% exceedance)
Month
Mont
nly Streamflow (cfs)
Natural
Streamflow
Water
Allocated or
Reserved
% Allocated*
JUNE
1,190
1,932
162%
JULY
726
1,067
147%
AUGUST
539
583
108%
SEPTEMBER
503
730
145%
Top Users: Municipal (97%), Domestic (2%)
*% Allocated: [Water Allocated or Reserved]/[Natural
Streamflow]. This is the percentage of water either allocated
or reserved for in-stream or other uses compared with the
natural streamflow. Percentages over 100% indicate the water
is overallocated atthe mouth of the river.
Reference:
https://apps.wrd.state.or.us/apps/wars/wars_display_wa_tab
ies/display_wa_details.aspx?ws _id=71480&exlevel=80&scen
ario id=l

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Table 7-2 shows that the Sandy River is overallocated June through September, and that
municipal uses account for 97% of the water use, leaving little water for other uses. As
discussed above, the Portland Water Bureau implements the Bull Run Water Supply Habitat
Conservation Plan to manage water flows. In addition, each year, the City of Portland prepares
a seasonal water supply augmentation and contingency plan called the Seasonal Supply Plan
(SSP) for water releases. The releases must be consistent with the Bull Run Water Supply
Habitat Conservation Plan and final Temperature Management Plan (2009) requirements.
Oregon Department of Fish and Wildlife applied for and was granted instream water rights
(ISWRs) to protect fish at several locations in the basin in 1991 and 1992. ISWRs function like
all water rights, and are junior to any earlier water rights. ISWRs provide targets for the flows
needed to support fish, wildlife, their habitats and recreation. For the Sandy River, ISWRs at
river mile 18.5 (Bull Run River) to the mouth in the summer range from 400 cfs (in August) to
1400 cfs, and from river mile 42.8 (Zigzag River) to river mile 37.5 (Salmon River) from 100-250
cfs. There are 17 ISWRs on tributaries to the Sandy River, including on the Salmon River
(summer range: 60-250 cfs) and on the Zigzag River (summer range: 75-150 cfs). These
ISWRs serve to help maintain existing flows, although senior water holders primarily for
municipal uses can still diminish flows below these levels in low flow years.
Climate Change: In 2040, average August temperatures in the Sandy River are predicted to
rise to 20°C compared to 23°C in the Columbia River. In 2080, August temperatures in the
Sandy River are expected to rise further to 21°C compared to 24°C in the Columbia River.
Therefore, although the Sandy River will still be cooler than the Columbia River by 3°C in 2040
and 2080, the absolute temperature of the Sandy River will be higher, which decreases its
benefit to salmon.
Ongoing Activities in the Sandy River Watershed and Recommended Actions to Protect
and Enhance the Cold Water Refuge
Groups such as the Portland Water Bureau, Sandy River Watershed Council, USFS, Bureau of
Land Management, East Multnomah Soil and Water Conservation District, Lower Columbia
Estuary Partnership and others have identified, prioritized, and implemented projects in the
Sandy River Basin. Oregon adopted the Sandy River Basin Total Maximum Daily Load and
Water Quality Management Plan (2005) to address warm river temperatures. The Portland
Water Bureau's Temperature Management Plan (2009), included in the Bull Run Water Supply
Habitat Conservation Plan (HCP), set up riparian forest protections, set reservoir flow releases
to meet temperature TMDL targets, and called for construction of the selective withdrawal
structure in the Bull Run Reservoir currently in use. The Portland Water Bureau is implementing
the Bull Run Water Supply HCP, a 50-year plan with 49 habitat, temperature, and flow
mitigation measures such as conservation easements on 240 acres of private land, engineered
logjams, and releases of cold water withdrawals. Implementation includes annual compliance
reports to implement the HCP.
The lower part of the Sandy River basin is within the Columbia River Gorge National Scenic
Area and covered by the Management Plan for the Columbia River Gorge National Scenic Area
(2016). National Scenic Area land use designations, policies, and guidelines in the Sandy River
basin area include buffer requirements and limitations on development to help protect water
quality and the Sandy River CWR.
108

-------
Columbia River Cold Water Refuges Plan
Final January 2021
The Lower Sandy River from river miles 6 to 18.5 was designated a State Scenic Waterway in
1972 and a part of the National Wild and Scenic River System in 1988. River miles 16.5 to 18.5
are in Clackamas County, and the remaining portions are in Multnomah County. The Sandy
Wild and Scenic River and State Scenic Waterway Management Plan (1993) provides limits on
development and timber harvest to help protect water quality and the Sandy CWR.
The Sandy River Watershed Council's State of the Sandy report highlights restoration work in
the basin, including improving and planting riparian vegetation, conducting large wood
placement and channel alteration, and improving fish passage. Ongoing and planned activities,
particularly increasing riparian vegetation near the confluence and the removal of the Kelly
Creek Dam, could benefit the CWR.
Actions to protect and enhance the Sandy River CWR include:
•	On national forest lands, continue to implement	and actions in the
(1990) and
its amendments, and the
(2016) to protect and restore riparian shade and stream functions to
maintain cool river temperatures. (USFS)
•	Continue to implement Oregon's	on private forest lands throughout
the watershed to protect and restore riparian shade and stream functions to maintain
cool river temperatures. (ODF)
•	On private and county lands, continue to implement the riparian protections in the
and	land use regulations, and the Multnomah Columbia
River Gorge National Scenic Area to regulate development in the Sandy River
watershed to protect riparian shade and stream functions to maintain cool river
temperatures. (Clackamas County and Multnomah County)
•	Continue to implement higher flows, colder temperatures, riparian restoration,
floodplain reconnection, and stream habitat restoration actions in the mainstem Sandy
River, Bull Run Reservoir Basin, and other tributaries noted in the
(2008),	(2009), and
(2005). (Portland Water Bureau)
•	Continue to implement ongoing protections from the
(1993) scenic designation in the Lower
Sandy River that limit development and maintain riparian habitat. (BLM, Oregon State
Parks and Recreation Department, Multnomah and Clackamas counties)
•	Continue to implement instream water rights for fish protection in the Sandy River
basin, particularly the Lower Sandy River, to protect existing flow and CWR volume.
(OWRD)
•	Continue collaboration in the watershed among multiple interested parties for
restoration, increased large woody debris, and other watershed restoration activities.
(Multiple parties)
•	Cool river temperatures by considering the removal of a small dam on Kelly Creek as
noted in the	(2017). (Mount Hood Community College)
109

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.6 TANNER CREEK (RIVER MIILE 141) - PROTECT AND ENHANCE
Refuge Volume: 1,713 m3(15th largest)
Average August Temperature: 11,7°C
Distance to Downstream Refuge: 24 mi. (Sandy
River)
Distance to Upstream Refuge: 2 mi. (Eagle
Creek)
Cold Water Refuge Rating: Excellent (<16°C)
Photo 7-13 Tanner Creek drainage from Hamilton Island
(2005)
What features make Tanner Creek an
important cold water refuge to protect
and enhance?
Tanner Creek provides a small CWR
located immediately below Bonneville
Dam at river mile 141, 24 miles upstream
of the refuge in the Sandy River. With an
estimated average temperature of 11,7°C
in August, Tanner Creek is approximately
10°C colder than the Columbia River,
classifying the creek as an excellent quality
refuge (<16°C).
ODEQ has designated the lower portion of Tanner Creek for core cold water habitat and salmon
and steelhead spawning and has assigned water quality criteria of 16°C and 13°C for maximum
water temperatures during spawning (August 15 - May 15), respectively. The maximum
modeled temperature for Tanner Creek is 14.5°C (1993-2011) (Appendix 12.18). However,
based on measured maximum temperature readings, the lower portion of Tanner Creek is not
on the 303(d) list for temperature impaired waters. Migrating salmonids use both the mouth and
the stream channel below Tanner Creek Bridge and an estimated 0.08 miles upstream as a
refuge (yellow pin, Photo 7-14). While the creek is very cold relative to the Columbia River, the
August flow is modest at only 38 cfs. However, the Bonneville Hatchery uses groundwater,
which is discharged to Tanner Creek and increases flows below the hatchery. As a result, the
CWR is estimated to be 1,713 m3 in size, or approximately % of an Olympic-sized swimming
pool, making it the smallest of the 12 primary refuges on the Lower Columbia River. Returning
adults must pass over Bonneville Dam and swim two miles before encountering Eagle Creek,
the next primary CWR.
Photo 7-14 Aerial view of Tanner Creek at the confluence with
Columbia River; yellow pin denotes upstream extent
110

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Tanner Creek originates from a groundwater spring below
Tanner Butte on the southern bank of the Columbia River
Gorge. The heavily forested watershed combined with the
creek's steep gradient and short length (6.5 miles) produce
reliably cold water. Cascading downhill in a nearly due north
, 7 . , J. c „	direction, Tanner Creek collects lateral tributaries from the
Photo 7-15 Wahclella Falls	'
east and west hillslopes. The upper portion of Tanner Creek
is protected as part of the Mark O. Hatfield Wilderness Area, and no urban development or
agricultural land exists in the watershed. Forest (87%) predominates in the basin; shrubland
(12%) grows on portions of the upper and middle watershed. Bonneville Fish Hatchery, the only
developed site, is located north of Highway 84 adjacent to the creek's confluence with the
Columbia River (Figure 7-14). The USFS owns and manages the entire watershed except for
the State of Oregon's Bonneville Fish Hatchery (Figure 7-15).
Introduction to the Tanner Creek Watershed
The watershed lies in the Mount Hood National Forest and
Columbia River Gorge National Scenic Area. Famous for its
picturesque Wahclella Falls (Photo 7-18), the Gorge
attracts many visitors who hike along the creek's lower
reaches. The Bull Run watershed, which supplies water to
the City of Portland, borders the basin to the southwest; the
Eagle Creek watershed abuts Tanner Creek to the east.
Tanner Creek is a priority watershed in the USFS
Watershed Condition Framework (2011) for the Columbia
River Gorge National Scenic Area.
In 2017, the Eagle Creek Fire burned a significant portion of the watershed. Potential post-fire
impacts to the refuge include increased water temperatures due to reduced riparian canopy
Figure 7-15 Tanner Creek land ownership
m Developed - <1%
CD Bwricn • <1%
H Forest-87.1%
^"1 Shruenaftd-12.1%
B j Gr3s*l*r»d - <1%
¦<1%
Figure 7-14 Tanner Creek land cover
111

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Shade Difference
£	Less than 10
O	11-20
G	21-30
O	31-40
O	41-50
O	51-60
©	61-70
#	71 - 80
#	81-90
#	91-100
Bonneville
Fish Hatchery
Figure 7-16 Tanner Creek shade difference between potential
maximum and current shade
(not shown in Figure 7-16). The USFS has
identified National Forest Service Road 8400777
Road, a mid-slope road on the east side of Tanner
Creek, as having the largest risk of sediment
delivery to Tanner Creek.
Post-fire analysis conducted by the USFS indicated
large extents of the mid-basin hillslopes were
moderately (yellow) or severely burned (red),
meaning the fire consumed at least 80% of the
ground cover and surface organic matter (Figure
7-17). Fortunately, most of the severe burn areas
occurred outside the riparian zone. A GIS analysis
of the Burn Severity Assessment data indicated that
14% of the riparian zone suffered low severity fire
disturbance, 31% experienced moderate severity
disturbance, and 12% experienced high severity
cover and sedimentation of the creek
mouth resulting from rainfall on bare,
steep slopes.
Factors that Influence Temperature in
the Tanner Creek Watershed
Protecting and Enhancing Riparian
Vegetation; Prior to the Eagle Creek Fire,
high levels of canopy cover shaded
Tanner Creek and its tributaries, except
for the lowermost portion of the mainstem
channel that has less than 50% cover due
to the Bonneville Fish Hatchery and
associated development.
Areas in the watershed with the highest
potential for canopy cover restoration
include the mouth of the creek in and
around Bonneville Fish Hatchery and
along the riparian areas affected by
moderate-to-severe fire severity
disturbance levels, predominately along
the upper portions of lateral tributaries
Eagle Creek Fire Burn Severity
uv>d«»dable Qtatuf&ance
1 low Severity Dtshiibance
I I Mc iterate Sfvwty Distuibane*
Ibgh S«i «rty Disturbance
IZOIt BuRei A aurvd Sfrturns
WZZy
Figure 7-17 Eagle Creek Fire Burn Severity map in the
Tanner Creek Watershed. (Peter Leinenbach and
USFS)
112

-------
Columbia River Cold Water Refuges Plan
Final January 2021
disturbance. The Middle and Upper Tanner
Creek are designated wilderness, which
relies on passive rather than active
restoration.
Dams and Hydromodifications: Except for
two small dams on the creek's last mile, the
basin's landcover and stream channel retain
natural characteristics. There is a diversion
dam at 0.8 miles above the creek mouth,
which withdraws water and blocks fish
passage and may cause some stream
warming.
Water Use: Water use is not limited in
Tanner Creek. Bonneville Fish Hatchery is Photo 7-19 Tanner creek
the only water user in the small watershed.
There is no instream water right for fish protection in Tanner Creek. To support fish cultivation,
the ODFW owns two year-round water rights: a surface water right that allows for the diversion
of up to 50 cfs and a groundwater right that allows for the pumping of an additional 2.2 cfs of
water. The diversion and point of use for both water
rights is in and around the creek mouth, and the
majority of pumped or diverted water returns to the
stream after being used in the Hatchery. In addition, up
to 39 cfs of cold water is pumped from the aquifer
below the Columbia River into the hatchery and is
discharged into Lower Tanner Creek resulting in a total
flow that exceed 50 cfs at the mouth. Further, the
basin's steep topography and designation as a
Wilderness Area limit the potential for new water uses
in the future.
Photo 7-17 Tanner Creek Drainage post Eagle
Creek Fire (USFS)
under climate change projections.
Climate Change: In 2040, Tanner Creek's average
August water temperature is projected to increase to
13 C while the mainstem Columbia River is projected
to average 23 C. In 2080, average August water
temperature in Tanner Creek is expected to rise by an
additional degree to 14 C compared to 24 C in the
Columbia River. Therefore, while water temperatures
are projected to increase in future decades, Tanner
Creek is predicted to provide a small plume of excellent
quality refuge (<16 C) for migrating salmonids, even
It is important to note that temperature modeling of Tanner Creek occurred prior to the Eagle
Creek Fire.
113

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Ongoing Activities in the Tanner Creek Watershed and Recommended Actions to Protect
and Enhance the Cold Water Refuge
Tanner Creek's small size and absence of residents make it one of the few watersheds in
Oregon without an established watershed council to coordinate restoration and outreach
activities. Since almost the entire watershed falls within USFS lands, USFS plans (i.e., the
USFS Water Condition Framework Transition Watershed Action Plan for Tanner Creek and
Hamilton Creek - Columbia River (2011; updated 2016)) provide recommended actions to
protect and enhance water quality in the watershed. USFS's highest ranked essential project in
the USFS Water Condition Framework is to eliminate the road accessing the diversion that
confines the stream channel to restore fish passage at the diversion dam and habitat conditions
near the mouth.
The lower part of Tanner Creek is part of the Columbia River Gorge National Scenic Area and
covered under the Management Plan for the Columbia River Gorge National Scenic Area
(2016). Most of Tanner Creek is in the Special Management Area of the National Scenic Area
under the authority of the USFS, which provides a very high level of protection within the
watershed.
Since nearly the entire Tanner Creek watershed is protected as part of the Mark O. Hatfield
Wilderness Area and the Columbia River Gorge National Scenic Area, the basin is not at risk of
new development and, as a result, is in a good position to maintain cold water temperatures in
the future. Actions to protect and enhance the Tanner River CWR include:
•	On National Forest lands, continue to implement	and actions in the
(1990)
and its amendments, and the
(2016) to protect and restore riparian shade and stream
functions to maintain cool river temperatures. Protect existing riparian vegetation
corridors in the watershed in accordance with federal forest protections under the
Mark O. Hatfield Wilderness Area.(USFS)
•	Apply the protection of cold water quality standard (OAR 430-0410-0028 (11)) to limit
new sources and activities to a cumulative warming of no more than 0.3°C above the
current ambient summer maximum temperature. (ODEQ)
•	Consider revising the designated use in Tanner Creek from 'Salmon and Trout
Rearing and Migration Use' to 'Core Cold Water Habitat Use' because current
temperatures attain the 16°C criteria associated with Core Cold Water Habitat use
(ODEQ).
•	Consider applying for instream water rights for fish protection to help maintain existing
flows and Tanner Creek CWR volume. (ODFW)
•	Implement Tanner Creek Essential Projects in the USFS Watershed Condition
Framework Tanner Creek Action Plan (2011) that include eliminating or relocating the
surface water diversion from Tanner Creek, eliminating the fish passage barrier on
Tanner Creek, removing the access road along 0.6 miles of stream to restore
floodplain connectivity, and replanting riparian habitat along the 0.6 miles of stream.
(USFS, ODFW)
114

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.7 EAGLE CREEK (RIVER MILE 143) - PROTECT AND ENHANCE
Refuge Volume: 2,988 m3(14th largest)
Average August Temperature: 15.1 C
Distance to Downstream Refuge: 2 mi
(Tanner Creek)
Distance to Upstream Refuge: 4.5 mi.
(Herman Creek)
Cold Water Refuge Rating: Excellent
(<16°C)
Photo 7-18 Eagle Creek confluence facing Columbia
River (Courtesy photo: Jonnel Deacon)
What features make the Eagle Creek an
important cold water refuge to protect
and enhance?
Located at river mile 143 in Oregon, Eagle
Creek is the first CWR tributary salmon
encounter upstream of the Bonneville
Dam. The confluence of Eagle Creek
emerges from a narrow channel, becomes
shallow and broad, flows south past
Interstate 84, and enters the Columbia
River. Eagle Creek temperatures in August
are 6 C cooler than the Columbia River,
with average temperatures of 15.T C. This
classifies Eagle Creek as an excellent CWR (<16°C). ODEQ designates the lower portion of
Eagle Creek for salmonid rearing and migration and has assigned a water quality criterion of
18 C for maximum water temperatures. The maximum modeled temperature for Eagle Creek is
18.8°C (1993-2011) (Appendix 12.18). The lower portion of Eagle Creek is not on the 303(d) list
for temperature impaired waters. However, there have been measured exceedances of 18°C in
Lower Eagle Creek. Eagle Creek is the first among a cluster of eight CWR between Bonneville
Dam and The Dalles Dam. Migrating fish use the confluence and an estimated 0.15 miles
upstream of the confluence as CWR (yellow pin, Photo 7-19).
Eagle Creek has a mean flow of 72 cfs in August, and the twelfth largest CWR in the Columbia
River, estimated at 2,988 m3, slightly larger than one Olympic-sized swimming pool. Though
Eagle Creek provides a smaller CWR compared to others, it presents a reliably colder stream of
water on average compared to the Columbia River. The next available CWR is 4.5 miles
upstream in Herman Creek.
Photo 7-19 Aerial view of Eagle Creek confluence with Columbia
River; yellow pin denotes upstream extent
115

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Introduction to the Eagle Creek Watershed
The Eagle Creek watershed drains north-
facing slopes of the Columbia River's southern
bank, immediately upstream of Bonneville
Dam. Prior to the 2017 Eagle Creek Fire that
originated in the watershed, the Eagle Creek
Trail was the most popular hiking trail in the
Columbia Gorge. Many visitors have hiked to
Metlako and Punch Bowl Falls and beyond
into the Mark O. Hatfield Wilderness Area
within the Mount Hood National Forest, which
covers most of the watershed except for a
portion of Lower Eagle Creek.
USFS manages nearly the entire watershed
except for the State of Oregon's control of the Cascade Hatchery near the creek mouth (Figure
7-18). The watershed retains natural vegetation - a mix of forest (89%) and shrubland (9%)
cover the steep slopes (Figure 7-19). The Eagle Creek Recreation Area and trailhead, fish
hatchery, and Eagle Creek Overlook Group campground at the creek mouth are the only
developed areas in the basin. Development at the mouth of Eagle Creek impacts floodplain
connectivity.
In September 2017, the Eagle Creek Fire spread from the watershed and burned tens of
thousands of acres in the Columbia Gorge. In the context of CWR, it is crucial to collect more
information on the impacts of the fire on riparian vegetation, channel banks, erosion, and
corresponding effects on water temperature and quality.
Figure 7-18 Eagle Creek land ownership	_ . , ,
3	a	r	Figure 7-19 Eagle Creek land cover
pravATE • *1%
116

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Factors that Influence Temperature in the
Eagle Creek Watershed
Protecting and Enhancing Riparian
Vegetation: Prior to the Eagle Creek Fire,
large amounts of riparian vegetation cover
shaded Eagle Creek and its tributaries except
for portions of Middle and Lower Eagle Creek.
Figure 7-19 compares the shade differences
between the potential maximum and shade
prior to the 2017 Eagle Creek Fire.
Post-fire analysis conducted by the USFS
indicated large extents of Eagle Creek were
moderately (yellow) or severely burned (red) in
tributaries to Eagle Creek and Middle and
Upper Eagle Creek, meaning the fire
consumed at least 80% of the ground cover
and surface organic matter (Figure 7-20).
Much of the riparian zone corridor along Lower
Eagle Creek, however, experienced
Eagle Creek Fire Burn Severity
	Undetectable Disturbance
| Low Severity Disturbance
| Moderate Severity Disturbance
High Severity Disturbance
| 120ft Buffer Around Streams
SourcesifisiT: (JSGS. NOAA
shade 'Pirr«ren<:«
•
Lmi tfetn 10
o
11-20
0
*1-30
o
31-40
o
41' SO
o
M 60
o
61 - 7D
o
71-80
•
&1 -90
Tigure 7-20 Eagle Creek shade difference between potential
maximum and pre-2017 fire shade
"undetectable disturbance" in terms of loss of
vegetation. A GIS analysis of the Burn Severity
Assessment data indicated that 23% of the
riparian zone suffered low severity fire
disturbance, 24% experienced moderate
severity disturbance, and 5% experienced high
severity disturbance.
Dams and Hydromodifications. Cascade
Hatchery operates a diversion dam at River
Mile 2 for approximately 2800 feet that impacts
temperatures and flows in that reach. It is also
an aquatic organism passage barrier identified
by the USFS as a priority for restoration.
Figure 7-21 Eagle Creek Fire Burn Severity map in the Eagle
Creek Watershed (Peter Leinenbach and USFS)
117

-------
Columbia River Cold Water Refuges Plan
Final January 2021
EAGLE CR > COLUMBIA R - AT MOUTH
(@ 80% exceedance)
Month
Mont
nly Streamflow (cfs)
Natural
Streamflow
Water
Allocated or
Reserved
% Allocated*
JUNE
93
0
0%
JULY
69
0
0%
AUGUST
42
0
0%
SEPTEMBER
44
0
0%
Top Users: None
*% Allocated: [Water Allocated or Reserved]/[Natural
Streamflow]. This is the percentage of water either allocated
or reserved for in-stream or other uses compared with the
natural streamflow. Percentages over 100% indicate the water
is overal located atthe mouth of the river.
Reference:
https://apps.wrd.state.or.us/apps/wars/wars_display_wa_tab
1 es/di s pi ay_wa_deta i 1 s .as px?ws_i d=304105 lO&exl evel =80&s
cenario_id=l
Table 7-3 Water Availability Analysis, Eagle Creek at mouth, 5/20/20,
Oregon Water Resources Department
Climate Change, In 2040, average August
temperatures in Eagle Creek are predicted to
be 17°C compared to 22 C in the Columbia
River. In 2080, August temperatures in Eagle
Creek are expected to rise further to 18 C
compared to 23°C in the Columbia River.
Therefore, Eagle Creek is expected to shift
from an excellent CWR (<16 C) to a good
CWR (16-18 C), unless restoration actions
such as increased riparian vegetation offset
increasing water temperatures. Eagle Creek is
still expected to be more than 5°C cooler than
temperatures in the Columbia River in the
summer, even under climate change
projections.
Water Use: There are no consumptive
or instream uses at the mouth of Eagle
Creek. Thus, the net stream availability
is the same as the natural streamflow
as shown in Table 7-3. The water
availability analysis from the Oregon
Water Resources Department (OWRD)
indicates water is available in Eagle
Creek. At river mile 2, the ODFW has a
surface water right to divert up to 45 cfs
for the Cascade Hatchery and return the
water just downstream of the hatchery
at the mouth of Eagle Creek. This has
resulted in significantly lower flows in
this reach during late summer and early
fall. There are no instream water rights
to protect fish.
Preserving flows in Eagle Creek can
help keep temperatures cold. No
modeling has been done to determine
minimum stream flows that would
preserve current cold temperatures.
Photo 7-21 Eagle Creek looking out to Columbia River, August
2016
Ongoing Activities in the Eagle Creek Watershed and Recommended Actions to Protect
and Enhance the Cold Water Refuge
Eagle Creek is well protected from future development activities. The Mark O, Hatfield
Wilderness protects the middle and upper part of the watershed. The lower part of Eagle Creek
is part of the Columbia River Gorge National Scenic Area and covered under the Management
Plan for the Columbia River Gorge National Scenic Area (2016). Most of Eagle Creek is in the
118

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Special Management Area of the National Scenic Area under the authority of the USFS, which
provides a very high level of protection within the watershed. The September 2017 fire,
however, burned a significant amount of the watershed.
Actions to protect and enhance the Eagle Creek CWR include:
•	On national forest lands, continue to implement	and actions in the
(1990) and
its amendments, and the
(2016) to protect and restore riparian shade and stream functions to
maintain cool river temperatures. Protect existing riparian vegetation corridors in the
watershed in accordance with federal forest protections under the Mark O. Hatfield
Wilderness Area.(USFS)
•	Review data and consider listing Eagle Creek for temperature impairments on the
303(d) List below the Cascade Hatchery diversion at river mile 2. (ODEQ)
•	Consider applying for instream water rights for fish protection to help maintain existing
flows and Eagle Creek CWR volume. (ODFW)
•	Identify impacts from the 2017 Eagle Creek Fire that have reduced riparian vegetation
and hillslope and stream bank stability in the lower watershed. Revegetate or stabilize
bare areas to cool water temperatures and reduce sedimentation.
•	Evaluate and take appropriate actions to address impacts from the Cascade Hatchery
diversion dam flow withdrawal to increase flows in the diversion reach, increase
floodplain connectivity, and help maintain cool river temperatures. Evaluate, and if
feasible, consider groundwater sources to offset surface withdrawals from Eagle Creek.
(ODFW)
119

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.8 HERMAN CREEK (RIVER MILE 147.5) - PROTECT AND ENHANCE
Refuge Volume: 169,698 m3 (6th largest)
Average August Temperature: 12°C
Distance to Downstream Refuge: 4.5 mi. (Eagle
Creek)
Distance to Upstream Refuge: 3.5 mi (Wind
River)
Cold Water Refuge Rating: Excellent (<16'C)
Photo 7-22 Herman Creek near the confluence with the
Columbia River, August 2017
What features make Herman Creek an
important cold water refuge?
Located at river mile 147.5, Herman
Creek is one of eight primary CWR
between Bonneville Dam and the Dalles
Dam that fish use as they migrate
upstream. Herman Creek is 4.5 miles
upstream of the next closest refuge at
Eagle Creek. Herman Creek temperatures
in August average 12°C, 9°C cooler than
the Columbia River. This temperature
makes Herman Creek an excellent quality
CWR (<16°C). The lower portion of
Herman Creek is designated by ODEQ for salmon and trout rearing and migration, with a water
quality criterion of 18 C for maximum water temperatures. The maximum modeled temperature
for Herman Creek is 13.7 C (1993-2011) (Appendix
12.18). Based on measured maximum temperature
readings, the lower portion of Herman Creek is not
on the 303(d) list for temperature impaired waters.
Herman Creek and Herman Creek Cove provide
169,698 m3 of cold water, the size of approximately
68 Olympic-sized swimming pools, and the sixth
largest CWR in the Lower Columbia River. In August,
the creek has an average flow of 45 cfs.
Constructed levees protect Herman Creek Cove from
inflow of warmer Columbia River waters. Thermal
stratification of the water in the cove provides a cool
layer of water. The CWR is estimated to be primarily
limited to the cove, the hatchery discharge channel,
Photo 7-23 Aerial view of Herman Creek and Herman Cove at
confluence with Columbia River; yellow pin denotes upper extent of
refuge
120

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 7-22 Herman Creek land cover
and mouth in the Columbia River Gorge
National Scenic Area. Nearly the entire basin
(98.5%) is forested; the small amount of
developed and cultivated land is concentrated
at the lower reaches of Herman Creek and
along Herman Creek Cove. ODFW operates
Oxbow Hatchery on Herman Creek. Waterfront
property on the eastern side of Herman Cove
has been pursued for light commercial and
industrial development. Over the last decade,
Nestle Corporation proposed a plan to bottle
water from Oxbow Springs, reflecting the high
quality of water from Oxbow Springs that feeds
Little Herman Creek. In August 2017, the
Eagle Creek fire affected areas near the
Herman Creek watershed, but initial post-fire
burn severity analysis conducted by the USFS
indicated the watershed experienced only
minor impacts from the fire.	F'9ure 7~23 Herman Creek land ownership
and an estimated 0.3 miles upstream on the Herman Creek mainstem. The Port of Cascade
Locks has noted high levels of sediment at the mouth of Herman Creek, causing water levels to
be shallower. The next available CWR is 3.5 miles upstream in the Wind River.
introduction to the Herman Creek
Watershed
The Herman Creek watershed is relatively
small, covering 50 square miles. Herman
Creek originates at Hicks Lake and flows
steeply downhill in a due north direction for 8.5
miles before emptying into the Columbia River.
Herman Creek Cove at the mouth of the
tributary is an area where fish are known to
congregate. Herman Creek Cove is fed by
Herman Creek and the hatchery discharge
channel. Waterfalls are a natural barrier to fish
passage at river mile 2.8 for coho and at river
mile 3.5 for steelhead. Oxbow Fish Hatchery
also operates a diversion dam at river mile 0.8.
The watershed consists almost entirely of
protected USFS land (Figure 7-23), with most
of the watershed protected as part of the Mark
O. Hatfield Wilderness Area and lower reaches
STATE LAMO <«*
121

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Factors that Influence Temperature in
the Herman Creek Watershed
Protecting and Enhancing Riparian
Vegetation: The Herman Creek
watershed has high levels of riparian
shade throughout the well-forested
watershed. This shade serves to block
solar radiation and maintain cool
temperatures. Riparian shade also
maintains channel complexity and
groundwater, which keeps water
temperatures cold. Figure 7-24 compares
the shade difference between the
potential maximum and current shade.
Lower Herman Creek (from the
confluence of two small tributaries with
the creek to the mouth of the cove) offers
potential for restoration of riparian
vegetation to help improve stream cover
and contribute to maintaining cool stream
temperatures. This is the only area along
the creek that has been developed.
Dams and Hydromodifications:
Hydromodifications are minimal in the
upper parts of the watershed. The Oxbow Figure 7-24 Herman Creek shade difference between potential
11*. ,	. .	, maximum and current shade
Hatchery operates two diversion dams that
divert water into the hatchery before the
water is returned to the creek.
Herman Creek Cove itself is the result of
levees constructed in the mid-20th century to
produce a harbor for milling operations on the
shore. The levees now serve to protect the
cove from warmer Columbia River waters.
The cove is located within the impoundment
Photo 7-25 Oxbow Hatchery on Herman Creek, August 2017
Forest surveys conducted by USFS found little to no large woody debris in the lower and middle
reaches due to culverts and channelization. The amount of large woody debris in the watershed
did not meet the Aquatic Conservation
Strategy goals of the Northwest Forest Plan.
Placement of large woody debris in Herman
Creek could help trap sediment, create pools
of cold water, and improve habitat conditions
for fish.
122

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Table 7-4 Water Availability Analysis, Herman Creek at mouth, 5/20/20,
Oregon Water Resources Department
area of the downstream Bonneville
Dam, and the water surface level
can vary by as much as two feet in
response to reservoir operations,
potentially affecting fish access to
CWR in the impoundment area if
certain points become too shallow.
Water Use: Table 7-4 shows the
water availability in Herman Creek.
There is minimal water use, and
water availability in the summer
months is close to the natural
stream flow. The minimal
consumptive uses of Herman Creek
consist of domestic water supply by
the City of Cascade Locks and for
fish cultivation at Oxbow Fish
Hatchery. Established in 1913, the
hatchery holds water rights to
withdraw 19 cfs from Oxbow
Springs to the hatchery, which is
discharged into Herman Creek. The
hatchery has two ponds withdrawing
water from Herman Creek. The
upper pond withdraws water from
Herman Creek and discharges back into the creek. The lower pond withdraws water from
Herman Creek as well but discharges into the hatchery discharge channel. The added cold
water from Oxbow Springs supplements flows in Herman Creek and Herman Creek Cove.
There are no instream water rights for fish protection in Herman Creek.
HERMAN CR> COLUMBIA R-AT MOUTH
(@80% exceedance)
Month
Mont
ily Streamflow (cfs)
Natural
Streamflow
Water
Allocated or
Reserved
% Allocated*
JUNE
44
0
0%
JULY
28
0
1%
AUGUST
15
0
1%
SEPTEMBER
15
0
1%
Top Users: Domestic (71%), Irrigation (29%)
*% Allocated: [Water Allocated or Reserved]/[Natural
Streamflow]. This is the percentage of water either al located
or reserved for in-stream or other uses compared with the
natural streamflow. Percentages over 100% indicate the water
is overal located atthe mouth of the river.
Reference:
https://apps.wrd.state.or.us/apps/wars/wars_display_wa_tab
I es/di s pi ay_wa_deta i I s .a s px?ws _i d=30410515&exl evel =80&s
cenario id=l
Climate Change: In 2040, average August temperatures in Herman Creek are expected to be
13°C compared to 22 C in the Columbia River, in 2080, August temperatures in Herman Creek
are expected to rise further to 14°C compared to 23°C in the Columbia River. Therefore,
Herman Creek will remain an excellent CWR
(<16°C), even under future climate change
projections. This contrasts with many other CWR
in the Lower Columbia River where climate
change will warm refuges to sub-optimal
temperatures for salmon.
Photo 7-26 Herman Creek side channel, August 2017
Ongoing Activities in the Herman Creek
Watershed and Recommended Actions to
Protect and Enhance the Cold Water Refuge
Herman Creek is protected as part of the Mark
O. Hatfield Wilderness. In the early 2000s, the
Hood River Soil and Water Conservation District
123

-------
Columbia River Cold Water Refuges Plan
Final January 2021
worked with USFS, Confederated Tribes of Warm Springs, the Columbia River Inter-Tribal Fish
Commission and various state agencies in Oregon to develop the Hood River Subbasin Plan
(2004). This plan was submitted to the Northwest Power and Conservation Council to meet
Endangered Species Act requirements for salmon recovery and adopted by NMFS in 2013. The
plan identifies several projects to improve riparian and habitat conditions in Herman Creek that
align with the goals for maintaining cold water temperatures and protecting Herman Creek as a
CWR. To protect steelhead and rainbow trout, the plan also identifies protecting and restoring
Herman Creek from the Hatchery Diversion Dam to the falls between river miles 0.8 and 2.8. It
also recommends increasing riparian vegetation and large woody debris to increase stream
complexity in the middle and lower reaches.
The lower part of Herman Creek is part of the Columbia River Gorge National Scenic Area and
covered under the Management Plan for the Columbia River Gorge National Scenic Area
(2016). Most of Herman Creek is in the Special Management Area of the National Scenic Area
under the authority of the USFS, which provides a very high level of protection within the
watershed. A small segment of Herman Creek near the mouth is designated urban use.
Actions to protect and enhance Herman Creek and Herman Creek Cove include:
•	On national forest lands, continue to implement	and actions in the
(1990) and
its amendments, and the
(2016) to protect and restore riparian shade and stream functions to
maintain cool river temperatures. Protect existing riparian vegetation corridors in the
watershed in accordance with federal forest protections under the Mark O. Hatfield
Wilderness Area.(USFS)
•	Apply the protection of cold water quality standard (OAR 430-0410-0028 (11)) to limit
new sources and activities to a cumulative warming of no more than 0.3°C above the
current ambient summer maximum temperature. (ODEQ)
•	Consider revising the designated use in Herman Creek from 'Salmon and Trout Rearing
and Migration Use' to 'Core Cold Water Habitat Use' because current temperatures
attain the 16°C criteria associated with Core Cold Water Habitat use. (ODEQ)
•	Consider applying for instream water rights for fish protection to help maintain existing
flows and Herman Creek CWR volume. (ODFW)
•	Implement projects in the	(2004) including increasing large
woody debris in Herman Creek to decrease excess sedimentation at the mouth and
increase riparian vegetation in Lower Herman Creek from the confluence of two small
tributaries of the creek to the mouth of Herman Creek Cove. (Multiple parties)
•	Conduct a sediment removal feasibility study in the cove to maintain CWR volumes and
fish access.
124

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.9 WIND RIVER (RIVER MILE 151)-PROTECT AND ENHANCE
Refuge Volume: 105,220 m3 (8th largest)
Average August Temperature: 14.5 C
Distance to Downstream Refuge: 3.5 mi.
(Herman Creek)
Distance to Upstream Refuge: 7.7 mi. (Little
White Salmon River)
Cold Water Refuge Rating: Excellent (<16 C)
Photo 7-27 Wind River looking out to Columbia River, August
2016
What features make the Wind River an
important cold water refuge to protect and
enhance?
Located at river mile 151, the Wind River is
one of eight primary CWR between
Bonneville Dam and The Dalles Dam that fish
use as they migrate upstream. The Wind
River is 3.5 miles upstream of the next closest
refuge in Herman Creek. Wind River
temperatures in August are estimated to be Photo 7-28 Aerial view of Wind River confluence with Columbia
7°C cooler than the Columbia River with	River; yellow pin denotes upstream extent
average temperatures of 14.5°C, making the Wind River an excellent quality CWR (<16°C).
Washington Department of Ecology has designated the lower portion of the Wind River as core
summer salmonid habitat with a water quality criterion of 16 C for maximum water
temperatures. The maximum modeled water temperature for the Wind River is 18.3 C (1993-
2011) (Appendix 12.18). Based on measured maximum temperature readings, the Lower Wind
			River is on the 303(d) list for temperature
impaired waters.
Photo 7-29 Wind River, August 2016
The confluence of the Wind River has a large
amount of sediment which has made the river
mouth broader and shallower, increasing water
temperatures and reducing the volume and
quality of CWR habitat. This is due to a
combination of anthropogenic causes, such as
historical logging and natural processes. It is
estimated that migrating fish use the lower 0.8
miles of the Wind River, below Shipherd Falls,
as CWR (yellow pin, Photo 7-28). The Wind
River has the eighth largest CWR in the
125

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Columbia River estimated at 105,220 m3, the size of approximately 42 Olympic-sized swimming
pools, with mean flows of 293 cfs. The next available CWR is 7.7 miles upstream in the Little
White Salmon River
Figure 7-25 Wind River land cover
Figure 7-26 Wind River land ownership
Introduction to the Wind River Watershed
The Wind River originates in the Gifford Pinchot
National Forest, Snowmelt runoff and high
levels of canopy shading produce cold water
temperatures. In addition, large groundwater
spring inputs in Upper Trout Creek, the
mainstem near Carson Hatchery, and Panther
Creek contribute to the river's cold
temperatures. Panther Creek, the Wind's
largest tributary, joins the mainstem at river mile
4.3. Panther Creek is particularly important in
keeping the lower portion of the mainstem cool
during the summer due to its current cool
conditions, flow, and proximity to the mouth of
the Wind River. The Wind River meanders and
broadens at the mouth, where it passes under
State Highway 14 near Home Valley, WA,
before entering the Columbia River.
The Wind River watershed is mostly forested
with 90% of the land owned by the USFS, with
private ownership concentrated from the
Middle Wind River to its confluence with the
Columbia River (Figure 7-26). The land cover
near the mouth of the Wind River is primarily
developed and de-forested (Figure 7-25) and
has the greatest impact upon temperature and
complexity of the CWR at the mouth of the
Wind River.
Factors that Influence Temperature in the
Wind River Watershed
Protecting and Enhancing Riparian
Vegetation: The Wind River watershed has
high levels of riparian shade throughout most of
the watershed, especially in the upper well-
forested tributaries. These are on federal, state,
and private lands that are governed by the
USFS Gifford Pinchot National Forest Land and
Resource Management Plan, the Washington
Department of Natural Resource's Habitat
Conservation Plan, and Washington's Forest
126

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Dry Creek
Upper and
Lower Trout
Creek
SHade Difference
0	LflssThan *0
O	11-2D
0	21 -30
0	31-10
0	41-50
0	51-60
O	61 - 70
O	T1-80
O	8'-SO
•	91-100
Panther Creek
Lower and Middle
Wind River
Figure 7-27 Wind River shade difference between potential maximum
shade and current shade
Watershed Temperature TMDL (2002) predicted that
maximum potential vegetation could decrease water
temperatures at the mouth from 18°C to 14°C under
low flow conditions.
Dams and Hydromodifications: There are no dams
in the Wind River watershed. Hemlock Dam on Trout
Creek, located two miles upstream from the
tributary's confluence with the Wind River, was
removed in 2009. Since then, there have been
significant improvements in habitat complexity in the
former reach. Fish population data to date suggest a
trend in increased adult and juvenile steelhead
populations in Trout Creek relative to the rest of the
watershed.
Water Use: Figure 7-28 shows the water rights and
availability in the Wind River watershed (WRIA 29).
Water rights are heavily allocated for agricultural uses
Low flows exist in the Upper and Lower Trout Creek
Practice Rules, respectively. This shade
serves to block solar radiation and
maintain cool stream temperatures.
However, there are several reaches that
have been degraded and have potential
for increased shade. Figure 7-27
compares the shade difference between
the potential maximum and current
shade. Most of the watershed is at or
near the maximum vegetation for
shading (dark and medium green). The
areas with greatest potential to increase
riparian shade are the Wnd River
mainstem, Upper and Lower Trout
Creek, and Dry Creek (yellow and light
green areas). Increasing riparian
vegetation above the confluence is
important because cooling water
temperatures upstream will transfer
downstream.
Water quality modeling in Washington
Department of Ecology's Wind River
WRIA 29 Wind/White Salmon
Water Availability
A/ WRIA Boundary
(SWSL)
Low Flow
Highways
JS/ County I
f~*~l Counryw
Figure 7-28 Wind River Basin - Water rights and
availability, Washington Department of Ecology
127

-------
Columbia River Cold Water Refuges Plan
Final January 2021
and Lower and Middle Wind River. Trout Creek is designated by WDFW as a surface water
source limitation area that advises Ecology to protect instream flows and restrict issuance of
new water uses. Because water use is high and supply is limited, more water use may reduce
the CWR plume volume and increase temperatures in the CWR.
Climate Change: In 2040, average August temperatures in the Wind River are predicted to be
16 C compared to 22 C in the Columbia River.
In 2080, August temperatures in the Wind River
are expected to rise further to 17°C compared
to 23°C in the Columbia River. Therefore, the
Wind River will change from being an excellent
CWR (<16 C) to a good CWR (16-18°C),
unless restoration actions such as increased
riparian vegetation offset increasing water
temperatures. The Wnd River is still expected
to be more than 6 C cooler than temperatures
in the Columbia River in the summer, even
under climate change projections.
Photo 7-30 Wind River looking downstream to confluence,
Ongoing Activities in the Wind River	August 2017
Watershed and Recommended Actions to Protect and Enhance the Cold Water Refuge
Ecology adopted the Wind River Watershed Temperature TMDL and associated implementation
plan (2004) to address warm river temperatures, and in 2005 the Watershed Management Plan
for WRIA29 (including the Wind River) was adopted to guide water resource management. The
2004 TMDL implementation plan includes
restoration benchmarks to measure progress.
In 2010, the Washington Lower Columbia
Salmon Recovery and Fish and Wildlife
Subbasin Plan, which includes the Wnd
subbasin, was adopted by the Lower Columbia
Fish Recovery Board as an integrated plan for
salmon recovery, the Northwest Power and
Conservation Council fish and wildlife program,
and Washington State watershed
management. This plan was adopted by NMFS
in 2013 as the salmon recovery plan under the
ESA. More recently, detailed implementation
plans have been developed that have identified
and prioritized reach-scale watershed projects
and specific water resource actions, which
include the Wind River Habitat Restoration Strategy (2017) and the WRIA 29a Watershed
Planning Detailed Implementation Plan (2015).
The lower part of the Wnd River basin is part of the Columbia River Gorge National Scenic
Area and covered under the Management Plan for the Columbia River Gorge National Scenic
Area (2016). This plan includes "open space" land use designation and associated limits on new
Photo 7-31 Wind River at confluence, August 2017
128

-------
Columbia River Cold Water Refuges Plan
Final January 2021
development and buffer restrictions for a significant portion of the Lower Wind River, which
serves to help protect water quality and the Wind River CWR.
Actions in these plans align directly with actions that would benefit CWR. These include moving
the boat ramp and parking area to the southeast corner of the mouth, and converting the current
boat ramp and parking area to multi-threaded side channels and vegetated islands to increase
complexity. Other projects include bank stabilization projects and revegetation, which would
reduce erosion and sediment at the Wind River confluence and cool waters.
Actions to protect and enhance the Wind River CWR include:
•	On national forest lands, continue to implement	and actions in the
(1990) and
its amendments, and the
(2016) to protect and restore riparian shade and stream functions to
maintain cool river temperatures. (USFS)
•	Continue to implement	on private forest lands and
the on state
lands to protect and restore riparian shade and stream functions to maintain cool river
temperatures. (WDNR)
•	On private and county lands, continue to implement the riparian protections in the
¦¦¦., ¦¦¦¦¦:.;	and :
to regulate development in the Wnd River
shoreline areas to protect riparian shade and stream functions to maintain cool river
temperatures. (Skamania County)
•	Continue to implement riparian restoration, floodplain reconnection, and stream habitat
restoration actions in the mainstem Wind River, Little Wnd River, and Upper and Lower
Trout Creek noted in the	(2017),
(2010) and
(2004) to cool river temperatures
and reduce sedimentation into the Wind River CWR. (Multiple parties)
•	Conduct a sediment removal feasibility study at the mouth to enhance CWR volume and
fish access.
•	Consider establishing surface water source limitation areas and/or adopting instream
flow rules for the Lower Wind River and Panther Creek as recommended in the
(2015) to help protect stream
flows for fish and Wnd River CWR volume. (WDFW, Ecology)
•	Conduct a temperature TMDL implementation review to assess progress in meeting
established restoration benchmarks along with recommendations for further actions.
(Ecology)
129

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.10 LITTLE WHITE SALMON RIVER (RIVER MILE 158.7) - PROTECT AND
ENHANCE
Refuge Volume: 1,108,661 m3 (2nd largest)
Average August Temperature: 13.3 C
Distance to Downstream Refuge: 7.7 mi.
(Wind River)
Distance to Upstream Refuge: 6.3 mi. (White
Salmon River)
Cold Water Refuge Rating: Excellent (<16 C)
Photo 7-32 Little White Salmon upstream view of lower
hatchery intake
What features make the Little White
Salmon River an important cold water
refuge to protect and enhance?
Photo 7-33 Aerial view of the Little White Salmon cold water refuge;
yellow pin denotes the upper boundary of the refuge
The Little White Salmon River is located at
river mile 159 and is one of eight primary
CWR between Bonneville Dam and The
Dalles Dam that fish use to migrate
upstream. The Little White Salmon River
flows into Drano Lake before entering the
Columbia River and is 7.7 miles upstream
of the next closest refuge in Wind River.
The mean August temperature of the Little
White Salmon River where it enters Drano Lake is 13 C, almost 8 C cooler than the mainstem
Columbia River in August, making the Little White Salmon River an excellent quality refuge
(<16°C). The lower portion of the Little White Salmon is designated for core summer salmonid
habitat by the Washington Department of
Ecology with a water quality criterion of 16 C
for maximum water temperatures. The
maximum modeled temperature for the Little
White Salmon is 15.6 C (1993-2011)
(Appendix 12.18). Based on measured
maximum temperature readings, there are
reaches of the Middle and Upper Little White
Salmon River upstream of Moss Creek that
are on the 303(d) list for temperature
impaired waters. Moss Creek near river mile
7, a particularly cold tributary, cools the Little
White Salmon River by roughly 4 C in
August from 12 C upstream to 8°C
Photo 7-34 The confluence of the Little White Salmon River via downstream (ADDendix 20212 22)
Drano Lake flowing into the Columbia River
130

-------
Columbia River Cold Water Refuges Plan
Final January 2021
The cooler water in the thermal refuge is
primarily near the inlet of the Little White
Salmon River into Drano Lake (-10 C-18 C),
and at the bottom of Drano Lake (16 C-2T C),
and migrating salmon are estimated to use up
to 1.3 miles upstream as a refuge. Drano Lake
makes the Little White Salmon River
confluence the second largest CWR along the
Columbia River, with a total volume of
1,108,661 m3, approximately 443 Olympic-
sized swimming pools. The Little White Salmon
River has an August mean flow of 248 cfs near
its confluence with Drano Lake (Appendix
12.23). Fish leaving the Little White Salmon
will travel 6.3 miles upriver before encountering
the White Salmon River, the next CWR.
Introduction to the Little White Salmon
River Watershed
The Little White Salmon River provides snow-
fed water from its headwaters east of the
Cascade crest to the confluence. The Gifford
Pinchot National Forest makes up roughly 79%
of the Little White Salmon River basin (Figure
7-29). The National Forest protects the
watershed from urban and industrial
development. The riparian forest buffers shade
the snow- and groundwater-fed streams,
keeping them cool as they flow toward the
Columbia River. However, a legacy of timber
harvesting has left lasting habitat impacts on
the subbasin in the form of stream-side clear
cuts and roads.
State and private lands in the Little White
Salmon River subbasin are generally
undeveloped. Less than 1% of the subbasin is
used for traditional agriculture (Figure 7-30)
Only 4% of the subbasin is developed land and
is concentrated near the confluence, where
most private lands are found. Timber
management in Gifford Pinchot National Forest
is the dominant land use (Figure 7-29, Figure
7-30). The Gifford Pinchot National Forest
prevents major urban development from
Figure 7-29 Little White Salmon River Basin land ownership
Figure 7-30 Little White Salmon River Basin land cover
131

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Shade Difference
© Less thai 10
Eastern mairistem Little
White Salmon River
occurring throughout the subbasin.
The lower part of the Little White
Salmon River basin is part of the
Columbia River Gorge National
Scenic Area. Current land uses and
associated protections will likely
continue in the Little White River
subbasin. The quality refuge habitat
of Drano Lake makes it a popular
fishing destination.
Factors that Influence
Temperature in the Little White
Salmon River Watershed
Protecting and Enhancing
Riparian Vegetation: The Little
White Salmon River watershed has
high levels of riparian shade to
maintain cool river temperatures,
except for a few areas. Federal,
state, and private lands are governed
by the USFS Gifford Pinchot National
Forest Land and Resource
Management Plan (1990), the
Figure 7-31 Difference between potential stream shade conditions and	Washinaton Department Of Natural
current stream shade
Resource s Washington Habitat
Conservation Plan, and Washington's Forest Practice Rules, respectively. Figure 7-31
compares the shade difference between the potential maximum and current shade. Note the
figure displays the greatest potential shade difference is located within a lava bed, where the
river is subsurface, so it does not represent actual riparian shading potential. The eastern
mainstem of the river has the greatest potential for restoration. Although stream shade potential
difference is small, restoring riparian shade in this reach could still have a positive impact on
mainstream temperatures. Overall, the Little
White Salmon River is well shaded with riparian
buffers. The Gifford Pinchot Forest Land and
Resource Management Plan requires wide
buffers which protect water quality from timber
harvest practices by reducing the effects of
erosion and sedimentation. The Management
Plan for the Columbia River Gorge National
Scenic Area (2016) includes "open space" land
use designation and associated limits on new
development and buffer restrictions for the
Lower Little White Salmon River.
Photo 7-35 Drano Lake
132

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Hydromodifications: The natural hydrology of the
Little White Salmon River confluence was altered
by the construction of Bonneville Dam. Backwater
from Bonneville Dam and the dike that supports
Highway 7 spurred the formation of Drano Lake.
Drano Lake backwater inundated roughly one mile
of spawning habitat at the Lower Little White
Salmon River and Columbia River confluence.
Historically, the Little White Salmon River provided
primary spawning habitat for salmonids up to river
mile 3 where Spirit Falls serves as a natural fish
barrier. Although inundation led to significant
spawning habitat loss, Chinook and steelhead can
use the cool water of Drano Lake and the lower
reach of the Little White Salmon River as CWR during their migration up the Columbia River
The Little White Salmon River has a unique geological feature, Big Lava Bed, that covers
16,000 acres in the upper western subbasin.
Lava Creek descends into the lava bed, then
reappears downstream, cooling the river as
the stream flows underground. This
geological feature is one of the reasons the
Little White Salmon River provides such cold
water to the confluence at Drano Lake.
Water Use: The Little White Salmon River is
located within the Wnd-White Salmon Water
Resource Inventory Area 29. The
Washington Department of Ecology's Water
Resource Explorer indicates that many water
diversions exist along the river, although
water is considered available for out of
stream uses. The largest certified use of
water is to the U.S. Fish and Wildlife Service for the Willard and Little White Salmon National
Fish Hatchery. Each of these hatcheries withdraw about 55 cfs for use, but the water is returned
to the river. Maintaining water flows is important to keeping high CWR volume and cold water
temperatures in the summer.
Climate Change: In 2040, average August temperatures in the Little White Salmon River are
predicted to be 15 C compared to 22 C in the Columbia River. In 2080, August temperatures in
the Little White Salmon River are expected to rise further to 16°C compared to 23CC in the
Columbia River. Therefore, the Little White Salmon River will change from being an excellent
CWR (<16 C) to a good CWR (16-18 'C), unless restoration actions such as increased riparian
vegetation offset increasing water temperatures. The Little White Salmon River is still expected
to be more than 7 C cooler than temperatures in the Columbia River in the summer, even under
climate change projections.
Photo 7-37 Spirit Falls on the Little White Salmon River
Source: https://curiousgorgeblog.wordpress.com/44~
spirit-falls/
Photo 7-36 View of the Lower Little White Salmon River above
Drano Lake
133

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Ongoing Activities in the Little White Salmon River Watershed and Recommended
Actions to Protect and Enhance the Cold Water Refuge
In 2010, the Washington Lower Columbia Salmon Recovery and Fish and Wildlife Subbasin
Plan, which includes the Little White Salmon subbasin, was adopted by the Lower Columbia
Fish Recovery Board as an integrated plan for salmon recovery, the Northwest Power and
Conservation Council Fish and Wildlife Program, and Washington State watershed
management. This plan was adopted by NMFS in 2013 as the salmon recovery plan under the
ESA. The subbasin plan was developed in a partnership between the Lower Columbia Fish
Recovery Board, NPCC, federal agencies, state agencies, tribal nations, local governments,
and others.
Historically, due to natural barriers at Spirit Falls, there was limited use of the Upper Little White
Salmon River Basin by salmonids. Therefore, the Little White Salmon River serves a small role
in contributing to salmon recovery objectives due to the very limited available spawning habitat
in the lower river. However, due to its cold water, the Little White Salmon River is used for
anadromous salmon production in the hatcheries.
The subbasin plan provides for broader watershed recovery. The Little White River subbasin
plan identified the lower and middle mainstem as priority areas to improve habitat connectivity,
forest practices related to sediment, riparian vegetation, and floodplain function. These
restoration efforts will benefit habitat in these areas and contribute to maintaining cool river
temperatures that provide CWR in the lower river and Drano Lake. However, the current
implementation status of the subbasin restoration activities is unknown.
The Watershed Management Plan for WRIA 29 (2005) and associated WRiA 29 Watershed
Planning Detailed Implementation Plan (2015) adopted by Skamania County provides
recommendations to Ecology for water resources in the Lower Cowlitz River. The
recommendations include reservations for future use and adoption of an instream flow rule for
the Little White Salmon River for long-term protection of fish uses.
Ongoing protection through current plans and restoring riparian and watershed conditions in the
basin will maintain and enhance its importance as refuge habitat for migrating salmonid species.
Actions to protect and enhance the Little White Salmon River CWR include:
•	On national forest lands, continue to implement	and actions in the
(1990) and its
amendments, and the
(2016) to protect and restore riparian shade and stream functions to maintain cool
river temperatures. (USFS)
•	Continue to implement	on private forest lands and
the on state
lands to protect and restore riparian shade and stream functions to maintain cool river
temperatures. (WDNR)
•	On private and county lands, continue to implement the riparian protections in the
, and
to regulate development in the Little White
River shoreline areas to protect riparian shade and stream functions to maintain cool
river temperatures. (Skamania County)
134

-------
Columbia River Cold Water Refuges Plan
Final January 2021
•	In addition to riparian restoration actions in the forest plans noted above, implement
riparian restoration on private lands in the middle mainstem of the Little White
Salmon River as identified in the
(2010) to cool river temperatures. (Multiple parties)
•	Consider establishing surface water source limitation areas and/or adopting instream
flow rules for the Lower Little White Salmon River near Cook as recommended in the
(2015) to help protect
stream flows for fish and Wind River CWR volume. (WDFW, Ecology)
•	Apply antidegradation requirements to limit temperature increases associated with
any proposed thermal discharges into the Little White Salmon River. (Ecology)
135

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.11 WHITE SALMON RIVER (RIVER MILE 165) - PROTECT AND ENHANCE
Refuge Volume: 153,529 m3 (7th largest)
Average August Temperature: 15.7 C
Distance to Downstream Refuge: 6.3 mi. (Little
White Salmon River)
Distance to Upstream Refuge: 1 mi. (Hood River)
Cold Water Refuge Rating: Excellent (<16 C)
Photo 7-38 Upstream view of the White Salmon River
What features make the White Salmon
River an important Cold Water Refuge
to protect and enhance?
Located at river mile 165, the White
Salmon River is one of eight primary
CWR between Bonneville Dam and The
Dalles Dam that fish use to migrate
upstream. The White Salmon River is 6.3
miles upstream of the next closest refuge
at the Little White Salmon River.
Average water temperatures in the White photo 7-39 Aerial view of the White Salmon River cold water refuge;
Salmon River in August are roughly	vellow P'n denotes upstream extent
15.7 C, 5.5°C cooler than the Columbia River. This feature makes the White Salmon River an
excellent CWR (<16°C). The Washington Department of Ecology designates the lower portion
of the White Salmon River for core summer salmonid habitat and has assigned a water quality
	 criterion of 16°C for maximum water temperatures.
The maximum modeled temperature for the White
Salmon River is 19.6 C (1993-2011) (Appendix
12.18). However, based on measured maximum
temperature readings, the lower White Salmon River
is not on the 303(d) list for temperature impaired
waters except for a segment at the confluence which
is currently listed as impaired. This impairment
appears to be influenced by the Columbia River.
Photo 7-40 Upstream of the White Salmon River
confluence with the Columbia River
Migrating Chinook and steelhead are estimated to use
the lower 1.3 miles of the White Salmon River as a
CWR (yellow pin, Photo 7-39). The cold water refuge
has a volume of roughly 153,529 m3, the equivalent of
39 Olympic-sized swimming pools, and mean flows of
136

-------
Columbia River Cold Water Refuges Plan
Final January 2021
715 cfs, making the White Salmon River confluence the seventh largest CWR identified on the
Lower Columbia River. The next available CWR is one mile upstream in the Hood River.
Introduction to the White Salmon River Watershed
With headwaters in the Gifford Pinchot National Forest, the White Salmon River watershed
drains glaciers on the southwest flank of Mount Adams. The mainstem flows south for 44 miles
before emptying into the Columbia River directly across from the City of Hood River, Oregon.
Portions of the mainstem are designated as Wild and Scenic and managed by the USFS and
Klickitat County, and the river is a popular destination for commercial and recreational activities
including fishing, kayaking, and rafting. Major tributaries include Trout Lake, Buck Creek, Mill
Creek, Dry Creek, Gilmer Creek, and Rattlesnake Creek. The river remains cool throughout the
year due to snowmelt runoff and contributions from groundwater. Groundwater recharge
provides an estimated 200 cfs or more of baseflow to the river throughout the year, with the
largest contribution occurring between June and September when precipitation averages below
2 inches per month.
The Gifford Pinchot National Forest, managed by the USFS, protects the slopes of Mount
Adams in the upper watershed and composes nearly half of the basin's land area (48%). The
lower portion of the basin is a mix of private and state-owned land (Figure 7-32). The White
Salmon River basin is largely forested (66%), with developed (5%) and cultivated lands (3%)
along riparian areas south of Trout Lake to the Columbia River confluence. The lower three
miles of the river are part of the Columbia River Gorge National Scenic Area. Road networks
exist throughout the watershed, but the most heavily developed areas surround the
Figure 7-33 White Salmon River Basin land ownership	Figure 7-32 White Salmon River Basin land cover
137

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Photo 7-41 White Salmon River confluence before and after
and Resource Management Plan (1990) in the
upper watershed. State and private forest lands
in the middle and lower watershed are governed
by the Washington Department of Natural
the removal of the condit Dam; usgs, u.s. Department of Resource's Washington Habitat Conservation
interior, 2015	Plan and the Washington's Forest Practice Rules,
respectively. Figure 7-34 highlights the difference between current and potential maximum
shade. The yellow, orange, and red river segments reflect the areas with the most potential for
enhancing riparian cover (Figure 7-34).
There is some potential for enhancing
riparian vegetation along the mainstem
segments and tributaries around and
south of the Trout Lake Creek confluence,
and in segments of the Rattlesnake Creek
tributary in the southeastern area of the
subbasin. The largest potential for
restoration is in the eastern portion of the
mid-basin where there is a high proportion
of agricultural or pastureland (circled,
Figure 7-34).
unincorporated community of Underwood near
the river's confluence with the mainstem
Columbia River.
Factors that Influence Temperature in the
White Salmon River Watershed
Protecting and Enhancing Riparian
Vegetation: The White Salmon River watershed
has high levels of riparian shade throughout most
of the watershed, except for some areas mostly
on private land. Federal lands are governed by
the USFS Gifford Pinchot National Forest Land
Hydromodifications: Currently, there are
no dams in the White Salmon River. The
most significant hydromodifications on the
White Salmon River relate to the removal
(2012) of Condit Dam at river mile 3.4,
which reestablished salmon and
steelhead access to historical habitat in
the basin. The initial breaching of the dam
was rapid, resulting in short-term damage
to salmonid and aquatic life, as large
amounts of sediment were flushed
downstream. Conditions have since
Settled and improved Much of the built- Figure 7-34 White Salmon River shade difference potential maximum
and current shade
up sediment previously trapped behind
the dam settled downstream near the
138

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Columbia River confluence. This resulted in the
formation of a new beach line at the confluence,
reducing the average depth and total volume of the
CWR used by salmon at the confluence plume.
Confluence conditions are dynamic; gravel banks
continue to shift and expand in the lower stem during
high flow events.
Water Use: Water rights for the White Salmon River
basin are managed under Washington WRIA 29,
which includes the Wind River and Little White
Salmon River to the west. There are no existing
instream flow rules (water rights to protect fish).
There is a need for more water use data to determine
the risk and protection needs in the subbasin.
Maintaining water flows is important to keeping high
CWR volume and cold water temperatures in the
summer.
rnoio ™¦ ngn®,n u,e vvma-vvrme o.mon, climate Change: In 2040, average August
December 2016 (Washington Department of Ecology)	3	>	a a
temperatures in the White Salmon River are predicted
to be 17 C compared to 23 C in the Columbia River. In 2080, August temperatures in the White
Salmon River are expected to rise further to 18°C compared to 24 C in the Columbia River.
Therefore, the White Salmon River is expected to be a good CWR (16-18C), even under
climate change projections. The White Salmon River is still expected to be more than 6°C cooler
than temperatures in the Columbia River in the summer.
Ongoing Activities in the White Salmon River Watershed and Recommended Actions to
Protect and Enhance the Cold Water Refuge
The removal of Condit Dam resulted in an increase in restoration projects and initiatives to
protect returning salmonid populations and their spawning and rearing habitats. Along with the
Wild and Scenic River land designation protections, these initiatives align with many of the
same best practices to protect and enhance the confluence as a CWR. Goals for Wild and
Scenic Rivers include keeping rivers "largely primitive and [their] shorelines undisturbed," which
aligns with CWR goals of reduced
sedimentation and the preservation of riparian
vegetation.
The Yakama Nation, Klickitat County, and
Washington Department of Fish and Wildlife
were the lead entities in the development of the
White Salmon Subbasin Plan (2004) adopted by
the Northwest Power and Conservation Council.
Building on this effort, NMFS finalized the ESA
Recovery Plan for the White Salmon River
Subbasin (2013). These plans identify
Rattlesnake Creek and Indian Creek, which are
on the state's 303(d) list as impaired for water
temperature, as priority areas to Improve riparian
Skamania
LEGEND
CoBoundary
1 Ecology Regwn Boundary
C?
Applications
Permits
CefUfteates
Oaim*

139

-------
Columbia River Cold Water Refuges Plan
Final January 2021
conditions and stream complexity to reduce water temperatures and improve habitat. Restoring
riparian habitat and shade along the previous reservoir behind Condit Dam (Northwestern Lake)
and along the agricultural land near Trout Lake are other opportunities to cool the river.
The site of the Underwood Indian Village was inundated by sediments after the removal of the
Condit Dam, limiting fishery access for Columbia River Treaty Tribes. Yakama Nation Fisheries
conducted a restoration project in 2018 to manage the sediment delta that formed at the White
Salmon/Columbia River confluence. This project included dredging the navigation channel and
using the dredge material to build islands to minimize shallow nearshore habitats near the
confluence and restore habitat for juvenile salmonids.
The lower part of the White Salmon River basin is part of the Columbia River Gorge National
Scenic Area and covered under the Management Plan for the Columbia River Gorge National
Scenic Area (2016). This plan includes "open space" land use designation and associated limits
on new development and buffer restrictions for a significant portion of the lower White Salmon
River, which serves to help protect water quality and the White Salmon River CWR.
White Salmon River from River Mile 12.7 at Gilmer Creek to River Mile 5 at the head of the
former Northwestern Lake is designated a wild and scenic area. The Lower White Salmon Wild
and Scenic River Management Plan (1991) calls for many actions including maintaining or
enhancing riparian habitat within and outside of a 200-foot buffer, preventing development that
would have a serious adverse effect on water quality, and establishing instream flows.
Actions to protect and enhance the White Salmon River CWR include:
•	On national forest lands, continue to implement	and actions in the
(1990) and
its amendments, and the
(2016) to protect and restore riparian shade and stream functions to
maintain cool river temperatures. (USFS)
•	Continue to implement	on private forest lands and
the on state
lands to protect and restore riparian shade and stream functions to maintain cool river
temperatures. (WDNR)
•	Continue to implement the
(2016) open space land use designation and riparian protections along the
White Salmon River shoreline areas to protect riparian shade and stream functions to
maintain cool river temperatures. (Columbia River Gorge Commission and Skamania
County)
•	On private and county lands, continue to implement the riparian protections in the
(1998) to protect riparian shade and stream
functions to maintain cool river temperatures, and update the plan to meet state
requirements. (Klickitat County)
•	Continue to implement actions in the
(1991) to protect riparian shade and stream functions and land uses
to maintain cool river flows and temperatures. (USFS, Klickitat County, and others)
•	Restore riparian vegetation to reduce water temperatures in Rattlesnake Creek and
Indian Creek and along the previous Northwest Lake location on the White Salmon
	River	
140

-------
Columbia River Cold Water Refuges Plan
Final January 2021
as identified in the	(2013) to
help maintain cool temperatures in the White Salmon CWR. The White Salmon River
around Trout Lake may also have potential for riparian restoration for increased shade
and cooler river temperatures. (Multiple parties)
•	Consider establishing surface water source limitation areas and/or adopting instream
flow rules for the White Salmon River as recommended in the
(1991) to help protect stream flows for fish,
recreation, and Wind River CWR volume. (WDFW, Ecology)
•	Assess residual sediment impacts to CWR from the 2012 Condit Dam removal and to
CWR volume and temperature. Continue conducting excess sediment removal
feasibility studies at the mouth of the White Salmon River to preserve CWR volume and
temperatures.
•	Apply antidegradation requirements to limit temperature increases associated with any
proposed thermal discharges into the Little White Salmon River. (Ecology)
141

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.12 HOOD RIVER (RIVER MILE 166) - PROTECT AND ENHANCE
Refuge Volume: 28,000 m3 (12th largest)
Average August Temperature: 15.5 C
Distance to Downstream Refuge: 1 mi. (White
Salmon River)
Distance to Upstream Refuge: 11 mi. (Klickitat
River)
Cold Water Refuge Rating: Excellent (<16 C)
Photo 7-44 Hood River
What features make the Hood River an
important cold water refuge to protect
and enhance?
Located at river mile 166 of the Columbia
River, the Hood River is approximately
halfway between the Bonneville Dam and
Dalles Dam. It is located one mile upstream
from the White Salmon River, the next
downstream refuge. Hood River
temperatures in August average 15.5°C, 6°C	, fU ,D. ... . ...
r	a	a	'	Photo 7-45 Aerial view of Hood River at the confluence with
cooler than the Columbia River. This	Columbia River; yellow pin denotes upstream extent
classifies the Hood River an excellent CWR
(<16 C). However, the large sand bar at the confluence, channelization in the lower Hood River,
and relatively low depth (-0.8 meters) in the summer may present barriers to salmon using the
Hood River as a refuge. Additionally, a fish monitoring station near the mouth of the Hood River
detected few out-of-basin steelhead (10-15 annually) migrating upstream of the station between
2010-2015. For that reason, only the
mouth of the Hood River is included as a
CWR (Photo 7-45).
The lower portion of the Hood River is
designated by ODEQ as core cold water
habitat with an assigned water quality
criterion of 16 C for maximum water
temperatures. The maximum modeled
temperature for the Hood River is 19.1 C
(1993-2011) (Appendix 12.18). Based on
measured maximum temperature readings,
the lower Hood River is on the 303(d) list
for temperature impaired waters. The Hood River is the eleventh largest CWR in the Lower
Columbia River with a cold water plume volume of 28,000 m3, or 11 Olympic-sized swimming
142

-------
Columbia River Cold Water Refuges Plan
Final January 2021
pools, and mean flows of 374 cfs. The next available CWR is 11 miles upstream in the Klickitat
River.
Introduction to the Hood River Watershed
Figure 7-35 Hood River land cover
Figure 7-36 Hood River land ownership
The Hood River watershed drains the snow-
laden eastern flank of Mount Hood and the
land to the north of the volcano. Three major
tributaries, the East, West, and Middle Forks,
cascade down from the mountainous
headwaters. The longest tributary, East Fork,
drains Mount Hood Meadows ski and
snowboard resort and flows east and then
north, collecting Dog River and the Middle
Fork before meeting the West Fork near the
small unincorporated community of Dee,
Oregon, approximately 11 miles south of the
City of Hood River, the only significant urban
development in the basin. Above this
confluence, the East Fork is considered the
mainstem Hood River.
Protected as part of the Mount Hood
National Forest, much of the upper basin
retains natural land cover, contributing to
high levels of riparian shading.
Approximately 60% of the basin is forested;
shrubland (16%) is found in fragments
throughout the watershed, and cultivated
crops (11%) predominate on flat topography
south of Hood River and surrounding Dee.
USFS owns and manages 56% of the
watershed, with the remaining 44% privately
owned (Figure 7-36). The City of Hood
River, located at the confluence of the Hood
and Columbia Rivers, has the largest
population in the watershed. In the past, the
Hood River delta and lowlands were flooded
during the construction of Bonneville Dam.
Currently, the mouth of Hood River is
channelized. The mouth of the Hood River is
in the Hood River Urban Area of the
Columbia River Gorge Scenic Area and is
managed by the City of Hood River and the
Port of Hood River.
143

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Factors that Influence Temperature in the Hood River Watershed
Protecting and Enhancing Riparian Vegetation
Although much of the Hood River watershed is well-shaded to maintain cool river temperatures,
there are several developed river
reaches that have lost much of their
riparian shade. Figure 7-37 displays
the difference between potential
maximum and current shade
conditions, helping to identify
reaches in the Middle and Lower
Hood River that could be restored to
provide more riparian shade where
high levels of development and
agriculture occur. On average,
shading from riparian conditions
could be improved by 37% to cool
temperatures at the confluence.
Areas with the most potential for
riparian shade include Indian Creek,
Odell Creek, Neal Creek, and the
East Fork Hood River Creek. Water
quality modeling in ODEQ's Western
Hood Subbasin TMDL (2001)
predicted maximum potential
vegetation and a minimum instream
flow of 250 cfs from Powerdale Dam
could decrease maximum water
u	v	temperatures at the mouth from 18 C
Figure 7-37 Hood River shade difference between potential maximum and	0
current shade	to 15 C.
Dams and Hydromodifications: In the
past, Powerdale Dam, located on river mile
4.5 of the Hood River, withdrew a significant
amount of water that affected the water
quality and quantity downstream in a 3-mile
bypass reach. In 2010, the Powerdale Dam
was decommissioned. Although there are no
permanent flow and temperature gauges
since Powerdale Dam was removed, the
updated 2018 Western Hood Subbasin
TMDL projected that temperatures would Photo 7*47 Hood River at the site of the former Powerdale Dam
decrease with increased flows in the lower
4.5 miles of the Hood River. A small hydroelectric dam on Odell Creek was removed in 2016,
which has expanded the time for resident salmonid spawning. The dam on Clear Branch, a
tributary to the Middle Fork Hood River, raises temperatures downstream of the reservoir during
144

-------
Columbia River Cold Water Refuges Plan
Final January 2021
most of the summer. The Confederated Tribes
of the Warm Springs Reservation also
operates and manages a fish hatchery on the
Middle Fork Hood River.
Water Use: Irrigation is the dominant water
use, and there are past and ongoing efforts to
improve the efficiency of irrigating crops to
reduce water demand, decrease agricultural
runoff, and increase flow in streams. The three
primary irrigation districts are: Farmer's
Irrigation District (FID), Middle Fork Irrigation
District (MFID), and East Fork Irrigation District
(EFID). MFID operates the Clear Branch Dam
for irrigation. EFID has the largest water
withdrawals for irrigation. Figure 7-38, from
the 2006 USFS Mount Hood National Forest
Aquatic Habitat Restoration Strategy, shows
the large amount of diversions throughout the
basin, especially the lower Hood River. Photo
7-47 Hood River at the site of the former Powerdale
DamPhoto 7-47 also shows the now-
decommissioned Powerdale Dam. In 2016, the Figure 7-38 Estimated flow diversions in the Hood River
Basin in 2006
Hood River Soil and Water Conservation
District published the Hood River Water
Conservation Strategy, a report
developed with the agricultural
community to evaluate different
alternatives to reduce water usage.
Table 7-5 shows that the Hood River is
overallocated during the summer
months at river mile 0.75. ODFW
applied for and was granted instream
water rights (ISWRs) to protect fish at
several locations in the basin in different
years. ISWRs function like all water
rights, and are junior to any earlier
water rights. ISWRs provide targets for
the flows needed to support fish,
wildlife, their habitats and recreation. In
1966, 1983, and 1998, ODWR
approved three ISWRs on Hood River
at river mile 4.5 (former Powerdale
Dam) to the mouth at 45, 100, and 250
cfs, respectively, in August. There were
18 ISWRs on tributaries to the Hood
River granted from 1966 to 2016,
145
» 0X4.
Table 7-5 Water Availability Analysis, 5/20/20 Hood River at river mile
0.75, 5/23/18, Oregon Water Resources Department
HOOD R> COLUMBIA R-ATRM0.75
(@ 80% exceedance)
Month
Mont
ily Streamflow (cfs)
Natural
Streamflow
Water
Allocated or
Reserved
% Allocated*
JUNE
745
1,069
144%
JULY
588
1,031
175%
AUGUST
457
989
216%
SEPTEMBER
438
918
210%
Top Users: Other (68%), Irrigation (21%)
*% Allocated: [Water Allocated or Reserved]/[Natural
Streamflow]. This is the percentage of water either allocated
or reserved for in-stream or other uses compared with the
natural streamflow. Percentages over 100% indicate the water
is overallocated atthe mouth of the river.
Reference:
https://apps.wrd.state.or.us/apps/wars/wars_display_wa_tab
les/display_wa_details.aspx?ws _id=30410575&exlevel=80&s
cenario id=l

-------
Columbia River Cold Water Refuges Plan
Final January 2021
including on the West Fork Hood River (summer range: 100-255 cfs), East Fork Hood River
(summer range: 75-210 cfs), and Middle Fork Hood River (summer range: 10-233 cfs). These
ISWRs serve to help maintain existing flows, although senior water holders primarily for
irrigation can still diminish flows below these levels in low flow years. Therefore, improving
irrigation water efficiency will increase the water quality and quantity for resident and migratory
fish in the tributaries and mouth of the Hood River.
Climate Change: In 2040, August temperatures in the Hood River are projected to rise to 16°C,
compared to 23°C in the Columbia River. In 2080, August temperatures in the Hood River are
expected to rise to 17°C compared to 24°C in the Columbia River. Therefore, increases in Hood
River temperatures are expected to keep the Hood River as a good CWR (16-18°C). Still, the
Hood River is expected to be more than 7°C cooler than temperatures in the Columbia River in
the summer, even under climate change projections.
Ongoing Activities in the Hood River Watershed and Recommended Actions to Protect
and Enhance the Cold Water Refuge
The existing watershed plans with targeted actions and partnerships provide a solid foundation
for protecting and improving conditions in the basin and at the confluence. In 2004, the Hood
River Soil and Water Conservation District completed the Hood River Subbasin Plan, a
comprehensive review of the watershed with prioritized actions identified by many stakeholders
in the basin, which was adopted by the Northwest Power and Conservation Council. In 2014,
the Hood River Watershed Group updated the subbasin plan and published the Hood River
Watershed Action Plan (2014), which provides a list of new projects to be implemented over
several years. In 2006, the USFS completed the Hood River Aquatic Habitat Restoration
Strategy, which targets the lower watershed for greater riparian cover and increased flows. In
2016, the Soil and Water Conservation District released a study on water conservation and
efficiency, Hood River Water Conservation Strategy. ODEQ updated its Western Hood Basin
TMDL in 2018, retaining the riparian shade targets from the 2001 TMDL. Numerous other plans
have been developed targeting efforts on USFS lands, more efficient water use, reduction of
pesticide use and runoff, improvement offish passage and habitat, among other plans. The
Confederated Tribes of Warms Springs has worked extensively in the basin conducting
monitoring and restoration projects. Many recommendations in these plans will benefit the
downstream CWR area. Increased riparian vegetation on agricultural land will reduce pesticide
runoff and shade streams, helping improve water quality.
The lower part of the Hood River basin is part of the Columbia River Gorge National Scenic
Area and covered under the Management Plan for the Columbia River Gorge National Scenic
Area (2016). This plan includes "open space" land use designation and associated limits on new
development and buffer restrictions for a significant portion of the lower Hood River, which
serves to help protect water quality and the Hood River CWR.
Actions to protect and enhance the Hood River CWR include:
• On national forest lands, continue to implement	and actions in the
(1990) and
its amendments, and the
(2016) to protect and restore riparian shade and stream functions to
maintain cool river temperatures. (USFS)
146

-------
Columbia River Cold Water Refuges Plan
Final January 2021
•	Continue to implement Oregon's	on private forest lands in the
watershed to protect and restore riparian shade and stream functions to maintain cool
river temperatures. (ODF)
•	On private and county lands, continue to implement the riparian protections in the
(2016) through
the	to regulate development in the lower Hood River
watershed to protect riparian shade and stream functions to maintain cool river
temperatures. (Hood River County)
•	Restore riparian vegetation in the Hood River basin including Indian Creek, Neal
Creek, Odell Creek, and the area of the decommissioned Powerdale Dam (Photo
7-47) as identified in the Western Hood Basin TMDL ( , ),
(2006), and the	(2014).
•	Continue implementing water efficiency projects to maintain and increase flows in the
Hood River basin noted in the	(2016).
(Multiple parties)
•	Increase the amount of instream large woody debris to create pools of cold water and
trap sediment that would otherwise reach the river mouth. (Multiple parties)
•	Support education and outreach opportunities for habitat and riparian restoration on
privately-owned properties in Hood River watershed plans. (Multiple parties)
147

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.13 KLICKITAT RIVER (RIVER MILE 177) - PROTECT AND ENHANCE
Refuge Volume: 222,029 m3 (5th largest)
Average August Temperature: 16.4 C
Distance to Downstream Refuge: 11 mi.
(Hood River)
Distance to Upstream Refuge: 24 mi.
(Deschutes River)
Cold Water Refuge Rating: Good (16-18 C)
Photo 7-48 Klickitat River near the confluence with the
Columbia River
What features make the Klickitat River an
important cold water refuge to protect and
enhance?
The Klickitat River is located at river mile 177
of the Columbia River. It is one of the first
tributaries migrating salmon encounter east of
the Cascades. The Klickitat River is eleven
miles upstream of the CWR in the Hood River.
Average August temperatures in the Klickitat
River are estimated to be 16.4°C,
approximately 5 C cooler than the Columbia Photo 7.49 Aerjai view of Klickitat River confluence with
River. This classifies the Klickitat River as a Columbia River; yellow pin denotes upstream extent,
good CWR (16-18 C). With mean flows of 851
cfs and lower temperatures relative to the Columbia River, migrating fish to use the confluence
and approximately 1.8 miles of stream in the
Klickitat River as a CWR (yellow pin, Photo 7-49).
Photo 7-50 Klickitat River, upstream of confluence
The lower portion of the Klickitat River is designated
as core summer salmonid habitat by Washington
Department of Ecology, which assigns a water
quality criterion of 16 C for maximum water
temperatures. The maximum modeled temperature
for the Klickitat River is 20.5°C (1993-2011)
(Appendix 12.18). Based on measured maximum
temperature readings, the lower Klickitat River is on
the 303(d) list for temperature impaired waters. The
Klickitat River has the fifth largest CWR in the
148

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 7-39 Klickitat River land cover
~
J	SllflJblaniJ 27.B%
[ l	Hanted/cultrvatsc - ? A
f 1	Grassland - 6.'%
im	Fotosl 4a 3K
^	Developed 4.7%
f |	Barren -<1%
Columbia River with a flow of 851 cfs and
volume estimated at 222,029 m3, the size of
approximately 89 Olympic-sized swimming
pools. The next available CWR is 24 miles
upstream in the Deschutes River.
Introduction to the Klickitat River
Watershed
The Klickitat River originates from snowmelt
off Gilbert Peak on the Yakama Indian
Reservation. The river flows south, collecting
water from the eastern slopes of Mount
Adams and drains the Lincoln Plateau before
cutting through steep canyons on its way to
the Columbia River near Lyle, WA. Snowmelt
runoff and the underlying volcanic basalt
rock that create groundwater pools recharge
the Klickitat River and provide cool water to
the river throughout the summer.
Figure 7-40 Klickitat ownership
The Klickitat River watershed is semi-arid
with a mix of land uses. Forested lands
cover nearly half the basin (48%), primarily
in the upper watershed (Figure 7-39).
Shrubland (28%) is found in fragments
throughout the basin and along the lower
mainstem Klickitat River. Grasslands are
interspersed throughout the upper basin
(8%), and planted/cultivated lands (7%)
surround the small community of Centerville,
WA, the patch of developed land (5%) in the
southeast of the basin.
The Yakama Nation owns and manages
most of the upper watershed (42%),
including the largest extent of forested areas.
The lower half of the watershed is mostly
privately owned (47%) with a mix of forested,
shrubland, planted/cultivated land, and
developed areas. State lands make up 9% of
the watershed; the Bureau of Land
Management, USFS, and U.S. Fish and
Wildlife Service each manage small (<1%)
portions of the basin (Figure 7-40). The
lower 10 miles of the Klickitat River have
federal Wild and Scenic designations. The
mouth and lower Klickitat River are located
within the Columbia River Gorge National Scenic Area.
149

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Shade Difference
© Less than 10
Little Klickitat
Creek

Swale
Creek
Factors that Influence Temperature in
the Klickitat River Watershed
Protecting and Enhancing Riparian
Vegetation:
Tributaries to the Klickitat River have
relatively higher shade levels than the
mainstem Klickitat River. The lower and
mid-mainstem are shaded because of
canyons along the Klickitat River. Figure
7-41 compares the riparian shade
differences between the potential
maximum and current shade. Swale Creek
is impacted by floodplain filling, grading,
and bank armoring associated with railroad
construction, which has increased erosion
and decreased the amount of vegetation.
Little Klickitat Creek has the most potential
for increased shading in the Klickitat
Watershed. Water quality modeling in
Washington Department of Ecology's Little
Klickitat River Watershed Temperature
TMDL (2002) concluded that potential
maximum vegetation and reduced width-to-
Figure 7-47 Klickitat River shade difference between potential	depth ratios COUld decrease temperatures
maximum and current shade	at the mouth from 23°C to 21,5°C under
average flow conditions.
Dams and Hydromodifications: There are no
dams in the mainstem Klickitat River. Lyle Falls is
a series of five cascades at river mile 2.2. The
creation of the Bonneville Pool altered the
conditions at the mouth. Before the construction
of the Bonneville Dam, historic aerial photos of
the confluence show a multi-thread channel with
expansive cottonwood. Today, the Klickitat River
is confined to a straight, simplified channel that
lacks the complexity of the natural confluence.
Water Use: Water availability is limited in the
watershed, both in the Upper Klickitat River,
within the Yakama Nation tribal boundaries, and
in the lower portions. WDFW has recommended a
surface water source limitation for Swale Creek
and in certain areas of the Little Klickitat
watershed, where Washington Department of
Ecology can condition or deny new water rights
permits. Figure 7-42 shows that Little Klickitat, Mill
Figure 7-42 Water Availability in WRIA 30 (Washington
Department of Ecology, Revised 2012)
WRIA 30 Klickitat River Basin
Water Availability
OREGON
150

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Creek, and Blockhouse Creek Basins are
"adjudicated basins," which means that water right
disputes may be resolved in courts. Basins with
past adjudications typically indicate that little water
is available for new permits. Because water use is
high and supply is limited, more water use may
reduce the CWR plume volume and increase
temperatures in the CWR.
Climate Change: In 2040, average August
temperatures in the Klickitat River are predicted to
be 18°C compared to 23 C in the Columbia River.
In 2080, August temperatures in the Klickitat River
are expected to rise further to 19°C compared to
24°C in the Columbia River. Therefore, the
Klickitat River will change from being a good CWR (16-18 C) to a marginal CWR (>18°C),
unless restoration actions such as riparian vegetation and increased water flows offset
increasing water temperatures. The Klickitat River is still expected to be more than 5 C cooler
than temperatures in the Columbia River in the summer, even under climate change projections.
Ongoing Activities in the Klickitat River
Watershed and Recommended Actions to
Protect and Enhance the Cold Water Refuge
The Klickitat River watershed has been studied
by many entities in the watershed. The Yakama
Nation, Klickitat County, and Washington
Department of Fish and Wildlife were the lead
entities in the development of the Klickitat
Subbasin Plan (2004) adopted by the Northwest
Power and Conservation Council. Building from
this Plan, NMFS adopted the Middle Columbia
River Steelhead Recovery Plan (2009), which
includes a salmon recovery plan for the Klickitat
River. Klickitat County, City of Goldendale, and
the Klickitat County Public Utility District
completed the Watershed Management Plan (2005), which addresses water quantity, quality,
and fish habitat outside Yakama Indian Reservation boundaries. Yakama Nation's Klickitat
Watershed Enhancement Project (KWEP) includes past and ongoing projects to restore,
enhance, and protect aquatic habitats in the Klickitat Basin. The Klickitat Lead Entity Salmon
Recovery Strategy (2013) is a non-regulatory document describing the vision for salmonid
habitat recovery and protection and was led by the Klickitat County Natural Resources
Department with involvement from Eastern and Central Klickitat Conservation Districts,
Underwood Conservation District, Yakama Nation, environmental, sport fishing, timber interests,
USGS, and NMFS. Ecology's Little Klickitat Watershed Temperature TMDL (2002), Little
Klickitat River Watershed Temperature TMDL Detailed Implementation Plan (2005), and the
Riparian Vegetation Assessment, Little Klickitat River and Swale Creek (2009) highlight the
need for increased riparian protections to cool river temperatures. The focus of these projects is
to restore stream processes and improve habitat conditions and water quality. Completed
151

-------
Columbia River Cold Water Refuges Plan
Final January 2021
projects include restoration of fish passage, meadows restoration, forest road management,
floodplain reconnection, wood replenishment, and side channel reconnection. These actions in
the lower watershed directly align with and benefit CWR. Studies in the Little Klickitat River also
identified locations and actions to reduce river temperatures and restore thermal complexity that
align with the goal of reducing temperatures in the lower Klickitat River.
The lower part of the Klickitat River basin is part of the Columbia River Gorge National Scenic
Area and covered under the Management Plan for the Columbia River Gorge National Scenic
Area (2016). This plan includes "open space" land use designation and associated limits on new
development and buffer restrictions for a significant portion of the lower Klickitat River, which
serves to help protect water quality and the Klickitat River CWR.
The USFS Lower Klickitat River Wild and Scenic Management Plan Final Environmental Impact
Statement (1991) addresses the lower 10.8 miles of the Klickitat River and calls for many
actions including maintaining or enhancing riparian habitat within and outside of a 200-foot
buffer, preventing development that would have a serious adverse effect on water quality, and
establishing instream flows. The Klickitat County Shoreline Master Plan (1998) requires riparian
protections from development activities in the basin, and the county is currently amending the
plan.
Actions to protect and enhance the Klickitat River CWR include:
•	Continue to implement projects on and off Yakama Indian Reservation boundaries in
the Klickitat Water Enhancement Project. (Yakama Nation)
•	Continue to implement	on private forest lands and
the on state
lands to protect and restore riparian shade and stream functions to maintain cool river
temperatures. (WDNR)
•	Continue to implement the
(2016) to regulate development in the Klickitat River shoreline areas to
protect riparian shade and stream functions to maintain cool river temperatures.
(Columbia River Gorge Commission, USFS)
•	On private and county lands, continue to implement the riparian protections in the
(1998) to protect riparian shade and stream
functions to maintain cool river temperatures, and update the plan to meet state
requirements. (Klickitat County)
•	Consider adopting an instream flow rule or surface water source limitation the Lower
Klickitat River as recommended in the
(1991) to help protect stream flows for fish, recreation, and Wind
River CWR volume. Consider establishing a surface water source limitation and/or adopt
instream flow rules for Swale Creek to help protect stream flows for fish and Klickitat
River CWR volume. (Ecology, WDFW)
•	Continue to implement projects identified in the
(2018) and through the KWEP that restore stream processes, including
increasing large woody debris, channel complexity, and floodplain reconnection on the
mainstem Klickitat River, Little Klickitat River, and Swale Creek to maintain riparian
shade and stream functions to maintain cool river temperatures. (Multiple parties)
152

-------
Columbia River Cold Water Refuges Plan
Final January 2021
•	Continue to implement projects in the
(2005) and
(2009), including increasing riparian shade and
implementing restoration projects to improve stream functions and floodplain
reconnection and to maintain cool water temperatures. (Multiple parties)
•	Support education and outreach about grant and tax benefits for habitat and riparian
restoration on privately-owned properties to maintain cool water temperatures.
(Multiple parties)
•	Continue to maintain or increase flows in the Klickitat River through flow conservation,
water quantity trading, and minimum instream flows in the summer to maintain CWR
volumes. (Multiple parties)
153

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.14 FIFTEENMILE CREEK (RIVER MILE 188.9) - RESTORE
Refuge Volume: N/A
Average August Temperature: 19.15 C
Distance to Downstream Refuge: 11.9 mi. (Klickitat
River)
Distance to Upstream Refuge: 12.1 mi (Deschutes
River)
Cold Water Refuge Rating: Marginal (>18°C)
Photo 7-53 Looking downstream from the confluence
with The Dalles Dam in the background
What features make Fifteenmile Creek a
potential cold water refuge to restore?
Entering the Columbia River at river mile
188.9 immediately downstream of The Dalles
dam, Fifteenmile Creek is in the drier, eastern
end of the Columbia River Gorge. It is located
twelve miles upstream of the CWR in the
Klickitat River. Average August water
temperatures in Fifteenmile Creek are
estimated to be 19°C, approximately 2°C
colder than the Columbia River. Currently, an
annual August stream flow of 4 cfs and
relatively high stream temperatures prevent
Fifteenmile Creek from serving as a CWR for
migrating salmonids. If restored, Fifteenmile
Creek could serve as an additional refuge for migrating salmonids.
The lower portion of Fifteenmile Creek is designated for salmon and trout rearing and migration
with an assigned water quality criterion of 18 C for maximum water temperatures. The
maximum modeled temperature for Fifteenmile Creek is 26 C (1993-2011) (Appendix 12.18).
Based on measured maximum temperature readings, the lower portion of Fifteenmile Creek is
on the 303(d) list for temperature impaired waters. Migrating salmonids will need to travel twelve
miles upstream before reaching the next CWR in the Deschutes River.
Introduction to the Fifteenmile Creek Watershed
Fifteenmile Creek originates from Senecal Spring in the eastern foothills of Mount Hood. The
creek flows in a northeast direction before making a large bend to the west prior to joining the
mainstem Columbia River. Its primary tributaries include Eightmile Creek, Dry Creek, Fivemile
Creek, Ramsey Creek, and Larch Creek.
Photo 7-54 Flow of Fifteenmile Creek into the Columbia River in
August, 2017; the water pooled below is backwater from the
Columbia River
154

-------
Columbia River Cold Water Refuges Plan
Final January 2021
The Fifteenmile Creek basin is dominated by
private landownership (>85%). A portion of the
Mount Hood National Forest managed by the
USFS, the only federally-owned land in the
watershed, covers the forested slopes of the
upper basin, and composes 15% of the basin
(JFigure 7-43). Although USFS land is
harvested for timber, land management
practices are designed to minimize impacts on
streams by conserving headwaters and
associated riparian buffers.
Forested lands (18%) are confined to the
higher elevation slopes and narrow riparian
corridors bordering tributaries in the upper
watershed (Figure 7-44). Fragmented patches
of grasslands (6%) can be found in the upper
basin as well. In the lower, flatter, and more
arid portions of the basin, shrubland (47%) and
cultivated crops (27%) predominate. The
watershed's only developed land (3%) is
concentrated near the creek mouth in the
eastern end of The Dalles, and the small
community of Dufur in the middle of the
watershed.
Fed by snowmelt runoff and groundwater
contributions, Fifteenmile Creek could
potentially deliver cold water down to the
confluence, providing additional CWR for
migrating salmonids with continued water
quantity and riparian habitat
restoration. However, agriculture is vital to the
local economy, valued at roughly $22 million
per year. Agricultural land types here include
orchards, vineyards, and pasture. Primary
agricultural products include wheat, cattle, and
cherries.
Figure 7-43 Fifteenmile Creek land ownership
Factors that Influence Temperature in the
Fifteenmile Creek Watershed
Riparian Vegetation: There is a substantial
area for additional riparian
vegetation restoration in the lower watershed
along the tributary streams and creeks on the
mainstem (Figure 7-45) The lower watershed
was widely denuded for use as agricultural land.
Figure 7-44 Fifteenmile Creek land cover
Fifleenmite Creek
155

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Figure 7-45 highlights areas
with potential for substantial
restoration on the finger
tributaries that contribute to
the mainstem. These areas
include the Lower Fifteenmile
subbasin, Eightmile Creek,
and small tributaries of Dry
Creek. There is
also potential for restoration
on the southeast portion of
the subbasin. The conversion
of riparian areas to
agricultural lands has resulted
in the removal of tali grasses
and small trees. Water quality
modeling in ODEQ's Middle
Columbia-Hood (Miles Creek)
Subbasin TMDL (2008)
predicted that maximum
potential vegetation and
increased flows could
decrease water temperatures at the mouth from 25 C to 18 C under low flow conditions, a
significant decrease.
Hydromodifications: Stream channels have been modified via road crossings, diversions,
dikes, ditches, etc. to develop farmland, accommodate roads, and protect infrastructure. There
are significant surface water alterations to accommodate agricultural irrigation in the subbasin.
These modifications alter the hydrologic connectivity to the floodplain and intensify streambank
erosion. Historical modeling indicates that flows were likely naturally low in the basin, so
additional water withdrawals and diversions during the critical summer period can have an
exacerbated effect. There are several aquifers in the Fifteenmile Creek drainage
basin. Groundwater levels are declining. Despite the unknowns regarding groundwater-surface
water connections, it is clear that these decline rates can be reduced by improving well
construction and reducing pumping through cooperative agreements.
Water Use: Consumptive water right use is highest in July. Watermasters are limited in their
regulatory authority, as they can only regulate based on priority date of the water right and not
on protection of water quality or species. Of the ten 6th order watersheds within the basin, three
- Middle Eightmile, Lower Fifteenmile, and Upper Eightmile - have 75% or more of the instream
flow diverted. Information to better understand the connective hydrodynamics between
authorized underground pumping and Fifteenmile Creek will inform the sustainability of pumping
and may impact the Watermaster's decision making.
Climate Change: Like the other cold water tributaries, average August temperatures in
Fifteenmile Creek are predicted to increase approximately 1.5 C in 2040 for a temperature of
20.7 C, compared to 23 C in the Columbia River. In 2080, August temperatures in Fifteenmile
Creek are expected to rise further to 21,7°C, compared to almost 24°C in the Columbia River.
Figure 7-45 Fifteenmile Creek shade difference between potential maximum and
current shade
156

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Ongoing Activities in the Fifteenmile Creek Watershed and Recommended Actions to
Restore the Cold Water Refuge
The 2004 Fifteenmile Subbasin Plan developed for Northwest Power and Conservation Council
by the Fifteenmile Coordinating Group (including, but not limited to, Confederated Tribes of
Warm Springs, Wasco County Soil and Water Conservation District, NMFS, ODEQ, ODFW,
OWRD, and USFS) highlights the need for continued collaboration and the importance of cross-
leveraging funds to implement best management practices and priority restoration projects. The
plan promotes a restoration philosophy to protect the remaining high quality, productive aquatic
habitats in the basin, which is typically the most effective and least costly approach long-term.
Other plans include USFS's Fifteen Mile Creek Basin
Aquatic Habitat Restoration Strategy (2010), Middle
Columbia-Hood (Miles Creek) TMDL, and Wasco County
Soil and Water Conservation District's Fifteenmile
Watershed Assessment (2003). ODFWs Conservation
and Recovery Plan for Oregon Steelhead Populations in
the Middle Columbia River Steelhead Distinct Population
Segment (2010) as part of NMFS' Middle Columbia
Steelhead ESA Recovery Plan (2009) identified
Fifteenmile Creek as important for steelhead populations.
As a result, many agencies have focused restoration
actions in Fifteenmile Creek. Because of these efforts and
the potential to reduce temperatures, EPA included
Fifteenmile Creek as a CWR to be restored.
Restoring habitat along riparian areas and restoring flow
are both important to reestablish Fifteenmile Creek as a
CWR. Groundwater decline can be reduced through
improved well construction and reduction of pumping
through cooperative agreements. The Wasco County Soil
and Water Conservation District manages a program,
Fifteenmile Action to Stabilize Temperature (FAST),
based on predictive modeling that alerts local irrigators to alter their practices when
temperatures are lethal for salmon and steelhead at two or more sites for two or more days. It
also provides financial compensation to irrigators for their participation in the program. The
Fifteenmile Watershed Council spurred work to install new gauges to improve the understanding
of flow throughout the basin and increase the ability to regulate water withdrawals.
Actions to further restore Fifteenmile Creek include:
•	On national forest lands, continue to implement the aquatic strategies and actions in the
USFS Mount Hood National Forest Land and Resource Management Plan (1990) and
its amendments to protect and restore riparian shade and stream functions to maintain
cool river temperatures. (USFS)
•	Continue to implement Oregon's Forest Practices Act on private forest lands in the
watershed to protect and restore riparian shade and stream functions to maintain cool
river temperatures. (ODF)
Photo 7-55 Looking upstream from the
confluence toward the Fifteenmile Creek flow
157

-------
Columbia River Cold Water Refuges Plan
Final January 2021
•	On private and county lands, continue to implement the riparian protections in the
(2016) through
the	to regulate development in the lower Fifteenmile
Creek watershed to protect riparian shade and stream functions to maintain cool river
temperatures. (Wasco County)
•	Continue partnerships to purchase or lease in-stream water rights during critical
periods for salmonids. (Multiple parties)
•	Promote and fund irrigation efficiency activities and equipment to adaptively manage
practices when temperatures rise. (Multiple parties)
•	Improve channel connectivity with floodplains and side-channels as noted in salmon
recovery plans and the
(2010). (Multiple parties)
•	Restore riparian buffers and maintain the riparian restoration work done in previous
years as noted in the
(2010),
(2010), and
(2008). (Multiple parties)
•	Encourage private landowners to enter riparian buffer programs. Fund fencing projects
for pasture lands near riparian areas to minimize the impacts of grazing. (Multiple
parties)
•	Refer to the	(2004) to focus restoration efforts on priority
areas identified by the locally-vetted prioritization method. (Multiple parties)
158

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.15 DESCHUTES RIVER (RIVER MILE 201) - PROTECT AND ENHANCE
Photo 7-56 Deschutes River, directly upstream of its
confluence with the Columbia River
What features make the Deschutes River
an important cold water refuge to protect
and enhance?
Refuge Volume: 880,124 m3 (3rd largest)
Average August Temperature: 19.2°C
Distance to Downstream Refuge: 24 mi.
(Klickitat River)
Distance to Upstream Refuge: No Upstream
Refuge before Snake River
Cold Water Refuge Rating: Marginal (>18 C)
The Deschutes River joins the Columbia
River at river mile 201, approximately 24
miles upstream of Klickitat River, the closest
downstream refuge. In August, the mouth of
the Deschutes River averages 19°C, typically
about 2 C colder than the Columbia River in
August Because migrating salmon and
Steelhead are more vulnerable in	Photo 7-57 Aerial view of the Deschutes River; the upstream
temperatures above 18°C, the Deschutes	boundary of the cold water refuge is demarcated by the yellow pin
confluence is a marginal quality CWR (>18 C) (See Figure 2-20). The lower portion of the
Deschutes River is designated for salmon and trout rearing and migration by ODEQ, which
assigns a water quality criterion of 18 C for maximum water temperatures. The maximum
modeled temperature for the Deschutes River is 26.9 C (1993-2011) (Appendix 12.18). Based
on measured maximum temperature readings, the lower Deschutes River, as well as a number
of tributaries, is on Oregon's 303(d) list as impaired for temperature.
The average August volume of the CWR at the mouth of the Deschutes River is 880,124 m3,
and the average flow is 4,772 cfs. This makes the Deschutes River one of the largest CWR in
the Lower Columbia River system, with a plume approximately the size of 352 Olympic-sized
swimming pools. A PIT-tag receiver at the mouth of the Deschutes River and radio-tag studies
Photo 7-58 Lower Deschutes River, viewed from the west bank
159

-------
Columbia River Cold Water Refuges Plan
Final January 2021
have documented extensive use of the lower 3.2 miles of the river for cold water use by salmon
and steelhead (yellow pin, Photo 7-57). The Deschutes River is the last significant CWR before
the confluence with the Snake River.
Introduction to the Deschutes River
Watershed
The Deschutes River watershed is the second
largest river drainage system in Oregon, flowing
through the eastern, more arid, side of the
Cascades. The Deschutes River and its
tributaries are fed by large amounts of
precipitation, mostly snow, coming from the
Cascade Mountains. This amounts to more than
100 inches annually, while additional sources of
precipitation come from the Ochoco Mountains
(40 inches), and lower central areas (10 inches).
The Deschutes River's large flow and relatively	, . kl J r
Photo 7-59 Moody Rapids, approximately 1 km upstream of
cooler water results in an observable plume of the confluence
cold water at the confluence with the Columbia
River. The Deschutes River has one major hydroelectric complex, the Pelton Round Butte
Hydroelectric Project, which forms Lake Billy Chinook approximately 100 miles upstream of its
confluence with the Columbia River. The Upper Deschutes, Crooked, and Metolius Rivers each
flow into Lake Billy Chinook. The Metolius River is heavily groundwater fed and provides cool
summer flows into Lake Billy Chinook.
Grassland • 3 0%
Wfew-<1%	Forest-31.7%
Slwuoiaofi - 57.4%	mi Developed - 2.0%
Planted.'CulllvalOd ¦ 3 5% f [ Barren t 1*
Figure 7-46 Land cover in the Deschutes Basin
Figure 7-47 Land ownership in the Deschutes Basin
160

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Buck
Hollow
Creek
White
River
Wapinita
Creek
Shade Difference
© Less than 10
O 11-20
Bakeoven
Creek
Warm
Springs
River
Trout
Creek
Just over half of the Deschutes River drainage area consists of shrubland (57%), in addition to
moderate amounts of forested area located mostly near the headwaters (32%) (Figure 7-46).
The top two landowners/managers in the Deschutes River drainage area are private landowners
(42%) and the USFS (32%). Tribal land comprises 7% of land ownership. The Bureau of Land
Management manages about 18% of the land in the watershed, some adjacent to the lower
Deschutes River, and the majority of which is in the Crooked River watershed above the Pelton
Round Butte Project (Figure 7-47). In the Deschutes River watershed, degradation has
occurred through livestock use,
forestry and agricultural practices,
invasion by western juniper, and
water storage and diversions.
Degradation from urbanization in the
Bend, Prineville, Redmond, and
Sisters areas has also occurred.
Figure 7-48 Deschutes River shade difference between potential maximum and
current shade
In 1970, Oregon designated the
lower 100 miles of the Deschutes
River as a state scenic waterway,
and in 1988 the U.S. Congress
designated this same reach as a
National Wild and Scenic River. In
1993, the Lower Deschutes River
Management Plan was adopted by
the BLM in collaboration with the
State, Confederated Tribes of Warm
Springs, and others to implement
both the federal and state
requirements. Most of the land
adjacent to the river in this reach is
public land administered by the BLM
or the State. There is also tribal land
and private land adjacent to the
river. The plan helps to protect and
enhance the river's outstandingly
remarkable and related values,
including the riparian conditions.
Near the confluence, lands adjacent to the Deschutes River are also part of the Columbia River
Gorge National Scenic Area and covered under the Management Plan for the Columbia River
Gorge National Scenic Area (2016). These lands are designated as open space under the plan.
Factors that Influence Temperature in the Deschutes River Watershed
Riparian Vegetation; The riparian vegetation analysis has focused on the lower part of the
watershed below Pelton-Round Butte Project. Although the headwaters in the cascades on
forest lands is currently well-shaded, a large portion of the lower basins is not well-shaded. The
mainstem of the Deschutes River does not have a high potential for shade, due to its large
width. Figure 7-48 compares the shade differences between the system potential and current
shade. Efforts to restore riparian vegetation would likely make the largest difference in areas
161

-------
Columbia River Cold Water Refuges Plan
Final January 2021
with the largest shade difference. Large portions of the lower Deschutes River watershed have
a semi-arid climate, and habitat restoration in these areas is likely to be slow. Most of the land in
areas with the highest potential for improvement is located on privately owned or tribal lands.
Thus, restoration activities will need cooperation from landowners as well as the Confederated
Tribes of Warm Springs. Revegetation in the tributaries will improve their overall health and may
also have a cumulative cooling effect on the Deschutes River itself. It should be noted that these
maps were developed prior to the summer 2018 fire, which burned much of the riparian
vegetation in the lower 38 miles of the Deschutes River.
Dams and Hydromodifications; The Deschutes River, particularly the lower portion below
Lake Billy Chinook to its confluence with the Columbia River, is influenced by the Pelton Round
Butte Hydroelectric Project. Pelton Round Butte is composed of three dams, beginning
downstream of Lake Billy Chinook: The Round Butte Dam, the Pelton Dam, and the Re-
regulating Dam. Pelton Round Butte is owned jointly by Portland General Electric (PGE) and the
Confederated Tribes of Warm Springs/Warm Springs Power Enterprises. A new FERC license
was issued in 2005 to operate the Project for 50 years.
In 2010, the building of a Selective Water Wthdrawal (SWW) tower at the Round Butte Dam
was completed. Prior to the installation of the SWW tower, water was released from the bottom
gate of Round Butte Dam. The SWW facilities were built to provide surface withdrawal for
downstream juvenile fish passage and
to allow the temperature of
downstream water releases to be
regulated to more closely match
temperatures that would occur absent
the dams. This is achieved by
releasing water downstream of the
dam from different depths and
targeting temperatures to match the
average temperature of the three rivers
inflowing Lake Billy Chinook. The
SWW operations have increased
temperatures in spring and early
summer and cooled temperatures in
August and September in the lower
Deschutes River. Thus, the SWW
operations appear to have a somewhat
beneficial effect by providing cooler
water during the CWR use period.
Although the cooler released water
attenuates due to the long distance
between the dam and the confluence
of the Deschutes River, modeling as
part of a PGE Water Quality Study
(2019) by Max Depth Aquatics
indicated that in 2017, late August and September temperatures at river mile one of the
Deschutes River would have been 1-2°C warmer under the pre-SWW operations.
Table 7-6 Water Availability Analysis, 5/20/20 for the Deschutes River
confluence with the Columbia River
DESCHUTES R > COLUMBIA R-AB MOUTH
(@ 80% exceedance)
Month
Mont
lly Streamflow (cfs)
Natural
Streamflow
Water
Allocated or
Reserved
% Allocated*
JUNE
5,560
5,670
102%
JULY
4,610
5,407
117%
AUGUST
4,320
4,812
111%
SEPTEMBER
4,410
4,997
113%
Top Users: Irrigation (87%), Municipal (8%)
*% Allocated: [Water Allocated or Reserved]/[Natural
Streamflow]. This is the percentage of water either a I located
or reserved for in-stream or other uses compared with the
natural streamflow. Percentages over 100% indicate the water
is overal located at the mouth of the river.
Reference:
https://apps.wrd.state.or.us/apps/wars/wars_display_wa_tab
I es/di s pi ay_wa_deta i I s .a s px?ws _i d=70087&exl evel =80&s cen
ario id=l
162

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Water Use: Table 7-6 displays Oregon Water Resources Department data on water usage in
the Deschutes River watershed. Water availability is overallocated in the Deschutes River
primarily due to irrigation, municipal use, and storage uses. Oregon Department of Fish and
Wildlife applied for instream water rights (ISWRs) to protect fish at several locations in the basin
in different years. ISWRs function like all water rights, and are junior to any earlier water rights.
ISWRs provide targets for the flows needed to support fish, wildlife, their habitats, and
recreation. From 1983 to 1991, for the lower Deschutes River, the OWRD approved several
ISWRs at river mile 100 (Pelton Round Butte Dam) to the mouth that range from 3000 and 3500
cfs in August to 4000 cfs in the rest of the summer. There are additional ISWRs in the Upper
Deschutes River and 84 ISWRs on tributaries to the Upper and Lower Deschutes River,
including on the White River (summer range: 60-341), the Metolius River (summer range: 110-
335 cfs) and the Crooked River (summer range: 20-150 cfs). These ISWRs serve to help
maintain existing flows, although senior water holders primarily for irrigation can still diminish
flows below these levels in low flow years.
Efforts to reduce irrigation diversions and maintain higher flows in the lower Deschutes River
and in tributaries to the lower Deschutes River, like Trout Creek, can serve to maintain and
potentially enhance the CWR at the confluence.
Climate Change; Currently, the Deschutes River averages 19.2°C in August. Modeled stream
temperature data from NorWeST shows that by 2040, this is predicted to increase to 20.5°C,
and by 2080 to 21,6°C. Comparatively, the mainstem of the Columbia River at river mile 201
where the Deschutes River enters currently averages 21.5°C in August. At this location the
Columbia River is predicted to rise to 23.0°C and 24.0°C by 2040 and 2080, respectively. While
the Deschutes River is predicted to remain relatively cooler than the Columbia River by about
2.5°C, by 2040, it is likely to be above accepted temperature thresholds for migration. By 2080,
it is likely to reach lethal levels for steelhead and salmon.
Ongoing Activities in the Deschutes River Watershed and Recommended Actions to
Protect and Enhance the Cold Water Refuge
The Deschutes Subbasin Plan (2004) adopted by the Northwest Power and Conservation
Council, provides a comprehensive assessment and management plan to protect and restore
the basin to support fish and wildlife resources. Building on this work, the State of Oregon, with
many partners, developed the Conservation and Recovery Plan for Oregon Steelhead
Populations in the Middle Columbia River Steelhead Distinct Population Segment (2010), which
is a component of NMFS' Middle Columbia River Steelhead Recovery Plan (2009). In addition
to protection programs, these plans identify a variety of habitat restoration actions across the
Deschutes basin. Specific implementation actions of the Steelhead Recovery Plan have been
developed and are being implemented, which are summarized by ODFW in the 2010-2016
Implementation Progress Report (2019). In the Lower Deschutes basin, nearly all the major
tributaries have degraded habitat and warm summer water temperatures from grazing,
agricultural practices, roads, and irrigation withdrawals (e.g., Bakeoven Creek, Buck Hollow
Creek, Warm Springs River, Trout Creek, and Shitike Creek). Prioritized actions to restore
habitat and reduce temperatures include restoring riparian vegetation, decreasing channel
width, increasing channel complexity and floodplain connection, and restoring flows. Portions of
the lower Deschutes River also have been identified as needing improved riparian conditions,
floodplain connection, and reduced stream width. Implementing these actions may contribute to
cooling the Deschutes CWR.
163

-------
Columbia River Cold Water Refuges Plan
Final January 2021
As discussed above, the water temperature of the releases from the Pelton Round Butte Project
in accordance with the FERC license conditions influences the water temperature in the lower
100 miles of the Deschutes River. Under current operations, the released water from the dam
contains a mix of warmer water near the surface and cooler water at depth, with a maximum of
60% percent cooler water. By mid-August and September, typically 50-60% of the releases are
from the cooler water at depth. PGE recently
developed a Water Quality Study to assess the
effects of different mixes of surface and sub-
surfaces releases from the dam. There appears to
be some potential to consistently provide 60% of
cooler water from early August through September
to help cool the Deschutes CWR.
In 2019, a Draft Deschutes River Habitat
Conservation Plan (HCP) was submitted to the
USFWS and NMFS by eight irrigation districts and
the City of Prineville and released by the agencies
for public comment. The HCP focuses on changes
in surface flows and irrigation conservation
measures in the upper Deschutes basin to improve
habitat conditions for ESA-listed Oregon spotted frog, steelhead, and bull trout as well as non-
listed Chinook salmon and sockeye, by restoring more natural river flows, including higher
winter flows and in some reaches lower summer flows. It is unclear if these changes will
significantly affect the SVWV operations and temperature downstream of Pelton Round Butte
Project.
The Deschutes River both above and below the Pelton Round Butte Project exceeds
temperature water quality standards and is listed on the State of Oregon's 303(d) list of impaired
waters along with many tributaries to the Deschutes River. Although ODEQ has initiated work
on the temperature TMDL for the upper Deschutes basin and to a lesser extent the lower
Deschutes basin, no temperature TMDLs have been completed in the Deschutes Basin.
The Deschutes River has many active watershed groups looking to restore more favorable
habitat for cold water fish. One group, the Deschutes River Conservancy, is engaged in
restoring stream flow to the river. Most of their work is focused upstream of the Pelton Round
Butte Hydroelectric Project where more of the water is diverted for irrigation. Their activities
include water rights transfers, water rights leasing, and promotion of water conservation. The
Crooked River and Middle and Upper Deschutes Watershed Councils have been actively
working on riparian restoration in their respective watersheds. The Lower Deschutes Weed
Control Project is an ongoing partnership with several agencies and organizations, focusing on
invasive species removal in the lower 40 miles of the Deschutes River. While this may not
directly impact temperatures, it is important for improving the overall health of the riparian
corridor.
Actions to protect and enhance the Deschutes River CWR include:
• As part of the Pelton Round Butte Project water quality management and monitoring
plan, consider the temperature effects of the selective water withdrawal operations on
the Deschutes River CWR. Specifically, consider maximum sub-surface cool water
blend
164

-------
Columbia River Cold Water Refuges Plan
Final January 2021
(60% percent) in August and September to help maintain temperatures below 18°C
when CWR use is highest. (ODEQ/PGE/Warm Springs Tribes)
•	Continue to implement projects to restore riparian vegetation, reduce channel width,
increase channel complexity, and restore flow in the White River basin, Bakeoven
Creek, Wapinita Creek, Buck Hollow Creek, Warm Springs River, Trout Creek, Shitike
Creek, and the Lower Deschutes as identified in the
(2010) and the	(2004) to help cool river
temperatures in the Deschutes CWR. (Multiple parties)
•	Develop temperature TMDLs and associated implementation plans for the upper and
lower Deschutes River basin. (ODEQ)
•	In the review and/or implementation of the
, fully consider the effects on summer temperatures downstream of the Pelton
Round Butte Project and the Deschutes River CWR to ensure August-September
temperatures are not warmed and preferably cooled. (NMFS, USFWS, PGE)
•	Protect sources of groundwater from degradation in quality and quantity. Specifically,
continue the existing protections and mitigation requirements in place for new
groundwater withdrawals above Pelton Round Butte Project. (Multiple parties)
•	Support partnerships to purchase or lease in-stream water rights during critical periods
to benefit salmonids. (Multiple parties)
•	On national forest lands, continue to implement the	and actions in the
(1990) and
(1989) and
associated amendments to protect and restore riparian shade and stream functions to
maintain cool river temperatures. (USFS)
165

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.16 UMATILLA RIVER (RIVER MILE 284.7) - RESTORE
Refuge Volume: 10,473 m3 (13th largest)
Average August Temperature: 20.8 C
Distance to Downstream Refuge: 83.7 mi.
(Deschutes River)
Distance to Upstream Refuge: N/A
Cold Water Refuge Rating: Marginal (>18 C)
Photo 7-67 Photo of the Umatilla River confluence with
the Columbia River
What features make the Umatilla River
a potential cold water refuge to
restore?
The Umatilla River confluence with the
Columbia River is located at river mile
284.7, just downstream of McNary Dam.
The Deschutes River is the nearest
downstream refuge, 84 river miles
downstream. The Umatilla River is only
considered a CWR in late August and
September when it is cooler than the
Columbia River. The average temperature of the Umatilla River is warmer than the Columbia
River in June and July, and the two rivers have the same average temperature of 20.8 C in
August. In September, the Umatilla River is on average 1.9 C cooler than the Columbia River
but has portions of the day that are more than 2°C cooler than the Columbia River, thereby
providing intermittent CWR (Figure 7-50). This qualifies the Umatilla River as a marginal CWR
(>18 C) for late August and September. ODEQ has designated the lower portion of the Umatilla
River for salmon and trout rearing and migration and has assigned a water quality criterion of
18°C for maximum water temperatures. The maximum modeled temperature for the Umatilla
River is 27 C (1993-2011) (Appendix 12.18). Based on measured maximum temperature
readings, the lower Umatilla River is on the 303(d) list for temperature impaired waters.
With a mean August flow of 87 cfs, the Umatilla River CWR is estimated to have a volume of
10,473 m3, the size of four Olympic-sized swimming pools during the time the river is 2°C cooler
than the Columbia River. The refuge is estimated to consists of cool water within the lower
tributary up to one mile upstream (Figure 7-49). The confluence is shallow and sandy.








« *










¦ ~


¦^r* " , . ,




~ Vl —• j" ¦
- *"* •



* -*¦ »»




-C
" i




* W»_UPS.

m


v r-
-
*

•<»s. ' * '. ***
v-

A

. - H, - -


N
Google earth s
1

v C_
Figure 7-49 Aerial view of the confluence of the Umatilla and Columbia
Rivers; yellow pin denotes upstream extent of refuge
166

-------
Columbia River Cold Water Refuges Plan
Final January 2021
-Umatilla R*vef (1$12) - Average of 200S, 200$, 2010, 2011 - Bars Represent Average of Diumat Range
¦Columbia Riven5092) - Average of 2QQ9, 2010, 2011 ¦ Bars Represent Average of Diurnal Range
t-	t- c. i- l- t- l. i_ _ _ _ _ _ _	__ B)Qouswi«uB«»»wtiuaiAU.uaatici
H rt H rt N (N IN	»H »-l 
-------
Columbia River Cold Water Refuges Plan
Final January 2021
Factors that Influence Temperature in the Umatilla River Watershed
Riparian Vegetation:
The loss of riparian
vegetation in the
Umatilla Basin -
primarily due to
agricultural
development - has
played a role in
increasing stream
temperatures. Figure
7-53 shows the
difference between
existing and system
potential shade,
highlighting the riparian
areas that should be
targeted for
revegetation. The areas
with potential to
Figure 7-53 Umatilla River shade difference between potential maximum and current shade
a Wetland!) - <1 %
(2$ Water- <1%
Snrubland -43.7%
[ J Planted.'Cultivated-27.9%
[	J Grassland -6.0%
Forest - 18.3%
Developed - 3.4%
[ I Barren-<1%
Figure 7-52 Land cover in the Umatilla Basin
Ranching and agriculture predominate in the
basin. Forest (18%) covers the higher elevation
upper portions of the basin. In the gullies and
hills of the southern portion of the watershed,
shrubland (43%) grows extensively. Cultivated
crops (28%) cover the flat lands north of the
mainstem river and south of the Cities of
Pendleton, Umatilla, and Hermiston, located on
the middle and lower mainstem. Other than the
road networks, these cities and small towns
throughout the subbasin are the only developed
(3%) land in the watershed (Figure 7-52).
Extensive water withdrawals in the 20th century
to irrigate farmlands resulted in very low flows
and occasionally no flow in the Lower Umatilla
River. In the 1980s and 1990s, flow restoration
and fish passage projects were developed,
leading to improved conditions in the
confluence for salmonids. The most notable
recent restoration projects were the
construction of "water exchanges" in the 1990s
that pump Columbia River water into the basin
for irrigation in exchange for leaving water in the
Umatilla river that was previously diverted for agriculture. The preserved Umatilla River water
flows back to the Columbia River for an intended no net depletion of Columbia River water.
interbasin
water
transfer
*not in natural drainage
basin
Wildhorse Creek
Upper Umatilla
River mainstem
168

-------
Columbia River Cold Water Refuges Plan	Final January 2021
increase riparian shade include
Wildhorse Creek and the upper
mainstem of the Umatilla River. The
restoration of associated riparian
wetlands would also contribute to
increased water temperature buffering
in the mainstem Umatilla River. Land in
these sub-watersheds is primarily made
up of private agricultural land and
private shrubland (Figure 7-52),
rendering it highly important that there
be funding and institutional capacity in
the basin to develop revegetation
opportunities with private landowners.
Water quality modeling in ODEQ's
Umatilla River Total Maximum Daily
Load (TMDL) and Water Quality
Management Plan (WQMP) (2001)
predicted that maximum potential
vegetation and restored flows could
decrease maximum water temperatures
at the mouth from 24°C to 21 °C under
low flow conditions. The CTUIR TMDL
for Temperature and Turbidity (2005)
indicates that there is potential for temperature reduction between river miles 56-82 on tribal
land.
Hydromodification: There is one main storage reservoir in the Umatilla Basin, McKay
Reservoir on McKay Creek, which captures winter flows to be delivered to farms in the summer
through an extensive network of irrigation canals. A second storage reservoir, Cold Springs, is
not within the natural drainage basin but is diverted into the lower watershed, impacting
temperature at the confluence.
In the 1990s, two water exchange projects were built, which collectively pump 380 cfs of water
up from the Columbia River into irrigation canals in exchange for an equal amount of Umatilla
River water - that otherwise would have been diverted - left instream to benefit fish. These
water exchanges were authorized under the Umatilla Basin Project Act of 1988 and
implemented by the Bureau of Reclamation. Phase 1 pumped up to 140 cfs of Columbia River
water into the West Extension Irrigation District system and helped retain flow in the Umatilla
River below the Three Mile Dam diversion. Phase 2 pumped up to 240 cfs into the Stanfield
Irrigation District and helped retain flow in the Umatilla River below Stanfield Dam at river mile
32. These exchanges have improved flow in the Lower Umatilla River, but low flow conditions
remain. Target flows associated with the project in the Lower Umatilla River to the mouth are 75
cfs (July 15-August 15) and 250 cfs (August 16-September 30), but current flows do not achieve
the 250 cfs target.
Water Use: The surface water in the Umatilla Basin - much of which is stored in two main
storage reservoirs, McKay and Cold Springs - is over-appropriated, meaning that there are
Table 7-7 Water Availability Analysis, 5/20/20 for the Umatilla River
confluence with the Columbia River
UMATILLA R > COLUMBIA R-AT MOUTH
(@80% exceedance)
Month
Mont
lly Streamflow (cfs)
Natural
Streamflow
Water
Allocated or
Reserved
% Allocated*
JUNE
187
1,043
558%
JULY
83
541
654%
AUGUST
48
399
830%
SEPTEMBER
57
488
862%
Top Users: Irrigation (89%), Municipal (11%)
*% Allocated: [Water Allocated or Reserved]/[Natural
Streamflow]. This is the percentage of water either allocated
or reserved for in-stream or other uses compared with the
natural streamflow. Percentages over 100% indicate the water
is overal located atthe mouth of the river.
Reference:
https://apps.wrd.state.or.us/apps/wars/wars_display_wa_tab
les/display_wa_details.aspx?ws_id=221&exlevel=80&scenari
o id=l
169

-------
Columbia River Cold Water Refuges Plan
Final January 2021
more water rights allocated in the basin than the river can satisfy during normal years. In the
peak summer months, over 600% of the natural flow of the river is allocated for out-of-stream
uses, over 88% of which is for irrigation and 11% of which is for municipal use (Table 7-7). Prior
to full implementation of the water exchanges, water withdrawals primarily for irrigation led to
very minimal to no Umatilla River flows reaching the Columbia River confluence during the
summer irrigation season. Since implementation of the water exchanges and a 2006 agreement
to provide for lamprey passage, Umatilla River flows are maintained throughout the summer.
However, groundwater aquifers in the basin have been tapped for irrigation, resulting in
significant declines in water tables in parts of the basin by more than 500 feet. Because of
groundwater decline, the Umatilla Basin has four of Oregon's six Critical Groundwater Areas,
leading the OWRD to withhold the groundwater irrigation rights of over 120,000 acres of
farmland in the basin, with the goal of steadying the declining groundwater table. The CTUIR
have also expressed their concern over unmet claims to tribal reserved water rights, some of
which they would likely put towards restoring river flows. Much of the river is diked or flanked by
agriculture, which reduces floodplain connection and hyporheic flows. Efforts to conserve and
increase water flows will help to cool water temperatures and increase CWR volume.
ODFW applied for and was granted instream water rights (ISWRs) to protect fish at several
locations in the basin. ISWRs function like all water rights and are junior to any earlier water
rights. ISWRs provide targets for the flows needed to support fish, wildlife, their habitats, and
recreation. ISWRs granted in 1983 at river mile 51 (McKay River) range from 85 cfs (in August)
to 250 cfs and at river mile 79 (Meacham Creek) range from 60 to 200 cfs in the summer. There
were 24 ISWRs granted from 1983 to 1990 on tributaries to the Umatilla River. These ISWRs
serve to help maintain existing flows, although senior water holders primarily for irrigation can
still diminish flows below these levels in low flow years.
Climate Change: In 2040, average August temperatures in the Umatilla River are predicted to
be 21 °C compared to 22°C in the Columbia River. In 2080, August temperatures in the Umatilla
River are expected to rise further to 22°C compared to 23°C in the Columbia River. If the
Umatilla River is restored, there could be a greater difference between Umatilla and Columbia
River water temperatures to make the Umatilla River a more consistent CWR.
Ongoing Activities in the Umatilla River Watershed and Recommended Actions to
Restore the Cold Water Refuge
Restoration of the Umatilla CWR will involve a multifaceted effort to restore river flows, riparian
vegetation, and floodplain function in the basin to balance human and ecological demands.
Established plans include: Northwest Power and Conservation Council's Umatilla/Willow
Subbasin Plan (2004); ODFWs Conservation and Recovery Plan for Oregon Steelhead
Populations in the Middle Columbia River Steelhead Distinct Population Segment (2010), which
is part of NMFS' Middle Columbia Steelhead ESA Recovery Plan (2009); Umatilla River Basin
TMDL and WQMP; and the CTUIR TMDL for Temperature and Turbidity (2005). Implementing
actions identified in these plans can contribute to cooler Lower Umatilla River water
temperatures and increase CWR volume. Decreasing temperatures by 2°C in late August and
early September would result in average temperature near 16-17°C and maximum temperatures
near 18°C, which would provide suitable continuous CWR temperatures when salmon and
steelhead migrate through this part of the Columbia River when its temperatures commonly
exceed 20°C.
170

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Both the NPCC UmatillaAA/illow Subbasin Plan and the NMFS ESA Recovery Plan identify
implementation of a Phase 3 Umatilla Basin Project water exchange as a top priority to provide
critical increased summer flows in the Lower Umatilla River. The Bureau of Reclamation
(USBR) completed the Umatilla Basin Water Supply Study in 2012 that examined options for
increased flow in the Lower Umatilla River including water exchange options to pump additional
Columbia River water that could be used for irrigation in exchange for retaining additional flow in
the Lower Umatilla River. Related to this effort, the Confederated Tribes of the Umatilla Indian
Reservation (CTUIR) are currently in negotiations with federal and Oregon state officials and
basin stakeholders to settle CTUIR's Umatilla Basin water right claims, which include instream
flows to support fisheries. The settlement is predicated on a series of water rights trades
whereby Umatilla River Basin water users would trade their water rights to the CTUIR, and the
stakeholders would obtain contemporary water rights and supply from the Columbia River. The
settlement, which would result in retaining more summer flow in the Lower Umatilla River,
requires federal legislation and subsequent funding and agreement among various parties.
Another ongoing effort in eastern Oregon is to find long-term, sustainable solutions to aging
flood control levees, which involve the CTUIR and the Governor's Greater Eastern Regional
Solutions Team. This initiative provides the opportunity to include enhancing floodplain function
into decision making around levees.
Due to the current low summer flow levels, increasing the flow is an important action to cool the
Lower Umatilla River as illustrated in the Oregon Umatilla River Basin Temperature TMDL. With
many projects completed and local champions throughout the basin, there is momentum for
ongoing progress to increase summer flow in the Lower Umatilla River through collaboration
and partnership as noted above. Actions to further restore the Umatilla Basin include:
•	As identified in the NPCC	and the
, seek agreement on and implementation of an additional Umatilla Basin Project
water exchange to increase summer flow in the Lower Umatilla River, thereby
decreasing summer river temperatures and increasing CWR volume. (USBR, Oregon,
CTIUR, irrigation districts, and others)
•	Where feasible, set back levees to reduce channelization, restore natural channel
complexity, reconnect the river with its floodplain, and restore groundwater
interactions., as identified in the	CTUIR
Temperature TMDL (2005), the
(2010). (Multiple parties)
•	Restore vegetation of riparian areas across the basin's streams as identified in the
(2001) CTUIR TMDL for Temperature and
Turbidity (2005), and the
(2010).(Multiple parties)
•	Continue to implement on-farm efficiency projects to restore flow to the Umatilla River,
particularly in August and September, which will help to cool river temperatures and
expand CWR volume. (Multiple parties)
•	On national forest lands, continue to implement the	and actions in the
(1990) and its
amendments to protect and restore riparian shade and stream functions to maintain
cool river temperatures. (USFS)
171

-------
Columbia River Cold Water Refuges Plan
Final January 2021
7.17 SUMMARY OF ACTIONS TO PROTECT AND RESTORE COLD WATER
REFUGES
The following is a summary of the actions in the 12 primary CWR and two "restore" tributaries
highlighted in this chapter to protect and restore CWR in the Lower Columbia River.
Additionally, a brief discussion of other opportunities to expand CWR, including the non-primary
CWR tributaries identified in Table 2-1, is presented.
Regulatory Protection Programs56
All 14 tributary watersheds include existing regulatory programs and land use provisions that
serve to protect watershed conditions and help keep waters cool. Since the 14 tributary
watersheds include forest lands for significant portions of their watersheds, important protective
actions include continued implementation of: 1) USFS plans on federal forest land (e.g., aquatic
strategies in the USFS Gifford Pinchot National Forest Land and Resource Management Plan
(1990) and the USFS Mount Hood National Forest Land and Resource Management
Plan(1990)), 2) the state of Oregon and Washington's forest management plans on state forest
land, and 3) the states' forest practice regulations on private forest lands.
Protecting existing riparian buffer areas of the CWR tributaries from development activities on
non-forest lands (e.g., agricultural, rural, and urban lands) is critical to maintain cool river
temperatures. The Management Plan for the Columbia River Gorge National Scenic Area
(2016) applies to the lower portion of 10 of the 12 primary CWR in the Columbia Gorge National
Scenic Area and helps provide this protection on federal, state, county, and private lands. Four
tributaries have Wld and Scenic River designations and associated management plans (Sandy
River, White Salmon River, Klickitat River, and Deschutes River), which help protect the riparian
areas in the designated reaches. County land use regulations also serve an important role in
protecting the existing riparian buffers (e.g., Cowlitz, Lewis, Clark, Skamania, and Klickitat
Shoreline Master Plans).
Since additional water withdrawal during the summer can diminish the size and function of the
CWR tributaries, minimizing additional water withdrawals will help maintain CWR quality and
function. The Cowlitz, Lewis, and Deschutes Rivers have upstream dams with FERC license
conditions. The Sandy River has an upstream dam on the Bull Run River with an HCP for
minimum summer flows. State instream flow rules for Lewis, Sandy, Hood, and Deschutes
Rivers help serve to maintain existing summer flows in the lower portion of these tributaries.
Existing plans for other CWR tributaries include recommendations to establish minimum
instream flows to help maintain current flows (Wnd, Little White Salmon, White Salmon, and
Klickitat Rivers). Hood River, Fifteenmile Creek, Deschutes River, and the Umatilla River are
5	Many of the programs and plans noted in this section are intended to meet various state or federal requirements,
and some are updated on occasion based on new information. By citing a program or plan in this CWR Plan as
important to prevent degradation of water quality and maintain cool river temperatures, EPA is not stating the
program or plan meets applicable state or federal requirements. EPA recognizes some of these plans will be updated
as warranted.
6	EPA recommends use of WDFW's Management Recommendations for Washington's Priority Habitats (Riparian
Ecosystems) for riparian buffer protection programs and regulations on non-federal lands in cold water refuge
watersheds identified in this Plan (see https://wdfw.wa.gov/species-habitats/at-risk/phs/recommendations).
172

-------
Columbia River Cold Water Refuges Plan
Final January 2021
overallocated, so there are limits on additional surface water withdrawals, and efforts in these
basins focus on irrigation conservation measures to increase depleted summer stream flows.
Tanner Creek, Herman Creek, Little White Salmon River, and the White Salmon River have
river temperatures below the temperature water quality standards, and it is important to maintain
these cool temperatures to protect the CWR associated with these rivers. The states'
antidegradation water quality standard provisions, and Oregon's protecting cold water standard
(OAR 430-0410-0028 (11)), serve to help protect these cool conditions from proposed actions
and discharges that may warm these rivers. Additionally, revised use designations in Tanner
Creek and Herman Creek based on the existing cool river temperatures to establish colder
temperature standards could provide added protection.
Restoration Actions Identified in Existing Plans
Restoring degraded portions of the 12 primary CWR watersheds would enhance the quality of
the CWR and help counteract future increases in tributary river temperature from climate
change. In addition, restoration of the two "restore" watersheds, consistent with current plans,
would improve habitat and thermal conditions within the watershed, as well as increase the
availability of CWR in the Lower Columbia River. All but three (Tanner, Eagle, Herman Creeks)
of the 14 watersheds have subbasin plans adopted by the Northwest Power and Conservation
Council in 2004 as part of the NPCC's Columbia River Basin Fish and Wldlife Program. These
subbasin plans help prioritize BPA funding for projects that protect, mitigate, and enhance fish
and wildlife that have been adversely impacted by the development and operation of the
Columbia River hydropower system.
All of the 14 watersheds are covered under NMFS approved salmon recovery plans that identify
actions to recover ESA-listed species. Six of the watersheds (Sandy River, Wnd River, Hood
River, Klickitat River, Fifteenmile Creek, and the Umatilla River) have temperature TMDLs with
associated implementation plans.
These plans, along with other habitat restoration project plans and water use plans noted in the
watershed 'snapshots' in this chapter, provide a thorough list of actions to improve habitat and
water quality conditions in the 14 watersheds. Many of these actions can serve to reduce river
temperatures and increase river flows to provide cooler and expanded volumes of CWR at the
river confluences with the Columbia River. These actions include: 1) restoring riparian
vegetation to provide river shading; 2) restoring stream morphology and floodplain connectivity
to reduce channel widths and create pools and groundwater connectivity; and 3) restoring
summer river flows that are more resistant to warming and increase CWR volume.
Each of the 14 watershed 'snapshots' in this chapter identify priority actions in existing plans
that can serve to cool river temperatures and maintain or increase the volume of CWR. These
actions target restoration for river reaches and tributaries that improve salmon habitat, which
also serve to improve CWR at the tributary's confluence with the Columbia River. Examples
include riparian and habitat restoration in Coweeman River (Cowlitz), East Fork of the Lewis
(Lewis), Middle Wnd River (Wnd), Rattlesnake Creek (White Salmon), and Trout Creek
(Deschutes). Implementing many of these projects typically involves grant funds available from
a variety of sources (e.g., BPA fish and wildlife funds, salmon recovery funds, Clean Water Act
funds, agricultural conservation funds) and local partnerships.
173

-------
Columbia River Cold Water Refuges Plan
Final January 2021
Existing plans prioritize the water conservation, irrigation efficiency, and water
trading/exchanges in the Hood River, Fifteenmile Creek, Little Klickitat River, Deschutes River,
and Umatilla River to address overallocation and low summer flows. The Umatilla River has
very low summer flows at the confluence, resulting in warm temperatures and limited CWR.
Water exchanges that pump water from the Columbia River to serve irrigation districts in
exchange for reduced irrigation withdrawals from the Umatilla River as part of the Umatilla Basin
Project have helped restore flows to the Lower Umatilla River to some extent. An additional
water exchange (Phase 3) is identified as a high priority in existing plans to provide additional
summer flow to the Lower Umatilla River. Additional summer flow could improve fish passage in
the Lower Umatilla River, and help cool and increase the volume of CWR at the confluence with
the Columbia River.
To supplement existing plans in the Deschutes River, completing a temperature TMDL(s) for the
Deschutes River basin is highlighted to aid in the implementation of actions to help cool the
lower Deschutes River and its CWR. The Deschutes CWR is heavily used by salmon and
steelhead, is a relatively warm CWR, and is vulnerable to future warming.
Cool Water Releases from Dams
Upstream dams on the Cowlitz, Lewis, Sandy, and Deschutes Rivers currently serve important
roles in providing cool river flows in the lower segments of these rivers that provide CWR.
Mayfield Dam on the Cowlitz River and Merwin Dam on the Lewis River release cool water from
deep within their respective reservoirs. Both the Bull Run Reservoir Dam in the Sandy River
Basin and the Pelton Round Butte Dam on the Deschutes River have the selective ability to
release water from different depths, which helps provide cool summer flows. Due to the
Deschutes River's high CWR use by migrating salmonids, marginally cool current temperatures,
and predicted temperature increases due to climate change, the potential to release cooler
water from the Pelton Round Butte Dam in August to provide cooler water at the mouth should
be assessed as part of the hydroelectric project's water quality management plan. Under current
operations, the maximum amount of cool deeper water is generally released by mid-to-late
August, but there may be potential to release more cool water starting at the beginning of
August that may influence temperatures at the mouth.
Sediment Management in CWR
Sediment deposition may be a concern for fish access to CWR at the mouth of several CWR
tributaries, including Herman Creek Cove, Wnd River, White Salmon River, and the Klickitat
River. Feasibility studies for habitat restoration and sediment removal at the confluence areas is
recommended to assess the potential for increased fish access, increased depth, and reduced
warming.
Opportunities for Additional CWR
The protection measures discussed above also generally apply to the other 10 non-primary
CWR tributary watersheds, and implementing restoration actions in these watersheds could
potentially increase the availability of CWR. As discussed in Chapter 2, most of these tributaries
are relatively small with limited availability of CWR.
174

-------
Columbia River Cold Water Refuges Plan
Final January 2021
The Lower Columbia Estuary Partnership is analyzing the feasibility of augmenting CWR for fish
by building a log structure at the mouth of Oneonta Creek to deflect mainstem flow and create a
pool of cold water at the mouth. Building this structure will help create a larger volume of CWR
at the mouth at Oneonta Creek and potentially serve as a model to expand CWR at other small
cold streams in the Lower Columbia River.
It may be possible to augment the existing CWR with pumped groundwater to provide increased
cool flows in the lower reaches of the tributaries. Both Tanner Creek and Herman Creek are
supplemented with cold groundwater that supplies fish hatcheries, which is then discharged into
the creeks. If the hatchery on Eagle Creek supplemented its water supply with groundwater and
decreased its reliance on surface flows, it could cool the river and increase the CWR volume.
Due to the limited availability of CWR in the John Day Reservoir, opportunities to add CWR in
this reach could be explored if fish use of CWR continues to increase in the future as predicted
due to warmer Columbia River temperatures.
7,18 ACTION TO ADDRESS FISHING IN COLD WATER REFUGES
As discussed in Chapter 4, fishing in CWR appears to reduce the survival of steelhead that use
CWR compared to those that do not, offsetting the benefits to fish using CWR. This plan may
inform future updates to fishing regulations in the primary CWR, especially for the CWR with the
highest amount of CWR use during periods of warm Columbia River temperatures (e.g., Cowlitz
River, Lewis River, Herman Creek Cove, Wnd River, White Salmon River, Little White Salmon
River (Drano Lake), Klickitat River, and Deschutes River).
175

-------
Columbia River Cold Water Refuges Plan
Final January 2021
8 UNCERTAINTIES AND ADDITIONAL RESEARCH NEEDS
This Plan relies upon the most recent scientific studies, field observations, expert input, and
analyses to characterize the amount of cold water refuges (CWR) in the Lower Columbia River
and salmonid use of the CWR. However, the study of CWR use is an area with a large degree
of uncertainty because of the complex behaviors exhibited by salmonids. This section highlights
some of the main uncertainties in this plan and recommends future studies to address them.
Adult Salmon and Steelhead Use of Cold Water Refuges below Bonneville Dam
There have not been any scientific studies characterizing fish use of CWR below Bonneville
Dam. The extent different species of salmon and steelhead use the CWR areas below
Bonneville Dam is unknown. In this plan, EPA relied on fishing boat presence in the confluence
area of tributaries cooler than the Columbia River as the primary basis for determining use as a
CWR in tributaries downstream of Bonneville Dam. EPA did, however, visually (from shore and
snorkel) document presence of likely out-of-basin salmon and steelhead in the Tanner Creek
CWR.
Study Recommendations: Fund a radio-tagging study to characterize salmon and steelhead use
of CWR below Bonneville Dam. Install PIT-tag detectors near the mouth of the Cowlitz and
Lewis Rivers.
Adult Salmon and Steelhead Use of Cold Water Refuges above Bonneville Dam
Extensive studies characterizing CWR use above Bonneville Dam have been conducted by the
University of Idaho. EPA relied upon those studies in this Plan. However, those studies were
conducted in the late 1990s and early 2000s. Since then, there have been changes that may
have altered CWR use. Those changes include an increased number of returning adult fall
Chinook and steelhead, decreased percentage of returning adults that were transported as
juveniles, increased sedimentation at the entrance of some CWR areas (e.g., White Salmon
River), changes in thermal regimes of CWR (e.g., Deschutes River), and increased mainstem
Columbia River temperatures. Additionally, there has been very limited study of CWR use by
sockeye and summer Chinook. This Plan concludes CWR use by sockeye and summer
Chinook is very limited, but studies would be beneficial to confirm the extent these species use
CWR.
The installation of a PIT-tag detector at the mouth of the Deschutes River in 2013 is an
investment that has provided valuable information on CWR use in the Deschutes River CWR.
Installation of PIT-tag detectors at the mouths of other CWR would benefit future analysis.
Study Recommendations: Fund a radio-tagging study to provide updated characterization of
CWR use above Bonneville Dam under current conditions for Chinook, steelhead, and sockeye.
Install PIT-tag detectors at the entrance to Drano Lake, Herman Creek Cove, White Salmon
River, Klickitat River, Wnd River, and Eagle Creek. Conduct a radio-tagging study after PIT-tag
detectors are installed to calculate the detection efficiency of the detectors.
Benefits of Cold Water Refuge Use for Migrating Adult Salmon and Steelhead
As discussed in this Plan, measuring the extent to which CWR use provides physiological
benefits to migrating adult salmon and steelhead in terms of decreased mortality and other end
176

-------
Columbia River Cold Water Refuges Plan
Final January 2021
points is confounded by fish harvest within CWR. Comparing survival rates of fish that use CWR
to those that do not shows higher survival rates for fish that do not, but the reduced survival
appears to be explained by increased harvest levels in CWR. As noted in this plan, modeling
predicts that CWR use can reduce energy loss and increase spawning success. CWR use is
also predicted to provide reduced exposure to warm Columbia River mainstem temperatures
that is likely to reduce disease risk and stress responses and decrease adult migration mortality,
but this has not been documented.
Study Recommendations: Design and fund research studies to document and evaluate the
benefits of CWR use to migrating adult salmon and steelhead.
Effects to Migrating Adult Salmon and Steelhead from Exposure to Elevated Columbia River
Temperatures
This Plan highlights analysis that shows a correlation between increased mainstem Columbia
River temperatures and decreased adult migration survival through the Lower Columbia River. It
also notes that some of the decreased survival could be attributed to fish moving into CWR as
temperatures rise and being harvested. There are numerous studies documenting various
adverse effects (mortality, disease, increased energy loss, decreased swimming speed,
avoidance behavior) at temperature in excess of 18-20°C, but there are more studies on
juveniles than adults due to challenges of conducting temperature effect studies on adult fish.
Better quantification of mortality and adverse effects is needed for adult salmon and steelhead
exposed to temperature increments in the 20-25°C range for different durations in the Lower
Columbia River.
Study Recommendations: Design and fund research studies to isolate the temperature-mortality
relationship for migrating salmon and steelhead in the Lower Columbia River. Studies should
also include assessment of the cumulative effects of elevated temperature for the entire return
migration to spawning grounds.
Volumes of Cold Water Refuges and Tracking Temperature and Flow Trends
EPA relied upon modeling and, in some cases, measurement techniques to estimate the
volume of CWR (steam and plume portion) in each of 23 CWR areas identified in this plan as
described in the technical memoranda listed in this plan's appendices. There is significant
variability around EPA's CWR volume estimates that EPA did not attempt to quantify. In addition
to the uncertainty with the modeling and volume measurements, the actual amount of CWR
varies throughout the day and season, depending on variable tributary and Columbia River
temperatures, flow, and Columbia River water levels. EPA generalized CWR volume based on
August mean tributary and Columbia River temperatures and flows. Further, EPA relied on
modeled August mean stream temperatures (NorWeST) and flow (USGS) for some tributaries.
USGS continuous flow gauges currently operate near the mouth of the Cowlitz, Lewis, Sandy,
White Salmon, Hood, Klickitat, Deschutes, and Umatilla Rivers. USGS gauges near the mouth
of the Wnd and Little White Salmon River have operated in the past but do not currently
(Appendix 12.23).
Of the eight currently operating USGS flow gauges noted above, only the Deschutes River
gauge includes a continuous temperature gauge. State agencies and other organizations have
operated continuous temperature gauges during the summer near the mouth of most the CWR
177

-------
Columbia River Cold Water Refuges Plan
Final January 2021
tributaries in the past, but these gauges are not currently operational on an annual basis.
(Appendix 12.22)
Study Recommendations: All of the 12 primary CWR tributaries and the Umatilla River should
have monitors in the lower portion of the tributaries to track both temperatures and flow over
time and to provide input data for more detailed and variable estimates of CWR volume for
future analysis. EPA recommends continued use of the currently operating USGS flow gauges
noted above, USGS flow gauges be re-installed near the mouth of the Wind and the Little White
Salmon Rivers, and long-term continuous flow gauges be installed near the mouth of Tanner,
Eagle, and Herman Creeks below the hatchery discharges. EPA recommends continued use of
the USGS Deschutes temperature gauge noted above and that long-term temperature gauges
be established and operated on the Umatilla River and the rest of the primary CWR tributaries
at or near USGS flow gauge sites.
In addition, monitoring and research is needed to better understand and estimate CWR volumes
available to fish and how those vary through time and in response to management actions
(Columbia River pool levels, dredging, flow management in tributaries, etc.).
Upstream Extent of Tributary Cold Water Refuge Use
Most of the 12 primary CWR do not have a barrier limiting how far upstream out-of-basin
salmon and steelhead may travel. As described in Appendix 12.4, EPA relied on a variety of
scientific lines of evidence to estimate the upstream extent of salmon and steelhead use of a
tributary as a CWR, which included a radio-tagging study on the Deschutes River documenting
that approximately 85% of out-of-basin steelhead used the lower five kilometers as CWR.
Study Recommendations: Install PIT-tag receivers approximately 3-5 kilometers upstream on
the White Salmon, Klickitat, and Deschutes Rivers, or devise other research and monitoring
approaches to document and track the extent out-of-basin salmon and steelhead use these
tributaries as CWR.
Density Effects and Carrying Capacity of Cold Water Refuges
There is no research on the carrying capacity of CWR for adult salmon or steelhead. The
closest research EPA could draw upon was adult fish held in confinement. It is fairly speculative
as to what densities cause fish to avoid or leave CWR. Additionally, research is needed to
understand how CWR characteristics (e.g., bathymetry, dissolved oxygen levels, submersed
aquatic vegetation, presence of other fish species, human disturbance including angling, etc.)
may influence CWR use and capacity. Also, high densities of adult fish are known to contribute
to the spread of disease. This could be a concern for CWR that are colder than the Columbia
River but are in the 18-20°C range, which are temperatures at which disease risk is elevated
(e.g., Deschutes River). The extent to which CWR use at varying densities contributes to
increased disease (and associated mortality) is unknown.
Study Recommendations: Design and fund a study to define the carrying capacity of CWR for
salmon and steelhead, with particular focus on Drano Lake and Herman Creek which have fixed
amounts of CWR that is available for use due to upstream barriers.
Effects of Sediment Deposition on Cold Water Refuge Use
178

-------
Columbia River Cold Water Refuges Plan
Final January 2021
As discussed in this Plan, sediment has deposited near the confluence areas of most the 12
primary CWR. This may affect the extent to which salmon and steelhead use the CWR. As
noted in Chapter 7, EPA recommends feasibility studies and implementation of projects to
remove sediment in several CWR.
Study Recommendations: As part of any project to remove sediment from the CWR, a study
should be designed to estimate the amount of CWR use before and after the sediment removal.
179

-------
9 SUMMARY AND RECOMMENDATIONS
The following is a summary of EPA's Columbia River Cold Water Refuge Plan. These findings
and recommendations are grounded in the technical and planning information presented in
previous chapters, the plan's technical appendices, and referenced scientific studies.
Lower Columbia River Temperatures
1.	The numeric temperature water quality standard for the Lower Columbia River is 20°C,
which is intended to minimize the risk of adverse effects to migrating salmon and
steelhead from exposure to river temperatures warmer than 20°C.
2.	Current daily average water temperatures in the Lower Columbia River (mouth to
McNary Dam) exceed 20°C for approximately two months, from mid-July to mid-
September, and exceed 21 °C for approximately one month. River temperatures are
typically the warmest in August with peak daily temperatures in the 22-23°C range.
3.	Historically, pre-1940 Lower Columbia River summer temperatures were cooler, with
August mean temperatures approximately 2-2.5°C cooler than the current August mean
temperature of near 22°C. Both regional anthropogenic sources (e.g., dams/reservoirs)
and global climate change have contributed to this warming.
4.	Lower Columbia River summer temperatures are predicted to continue to rise. August
mean temperatures are predicted to be near 23°C by 2040 and approximately 24°C by
2080.
Cold Water Refuges in the Lower Columbia River
5.	There are 12 primary CWR tributaries in the Lower Columbia River. The CWR for each
tributary are in and/or near the confluence with the Columbia River. These 12 CWR are
known or presumed to be used by steelhead and fall Chinook and constitute 98% of
CWR volume in the Lower Columbia River. In addition, there are 11 other tributaries that
collectively provide a limited amount of CWR, are smaller in scale, and have limited
information on fish use.
6.	Four primary CWR are below Bonneville Dam (Cowlitz River, Lewis River, Sandy River,
and Tanner Creek); seven primary CWR are between Bonneville Dam and The Dalles
Dam (Eagle Creek, Wnd River, Herman Creek, White Salmon River, Little White
Salmon River, Hood River, and Klickitat River); and one primary CWR (Deschutes River)
is between The Dalles Dam and the John Day Dam. There are no primary CWR
between John Day Dam and McNary Dam.
7.	The Cowlitz River, Lewis River, Little White Salmon River (Drano Lake), and the
Deschutes River are the largest CWR.
Salmon and Steelhead Use of Cold Water Refuges
8.	Summer steelhead and fall Chinook are the primary species that use CWR in the Lower
Columbia River. Summer steelhead use CWR for extended periods (multiple weeks),
180

-------
while fall Chinook use CWR for shorter periods (days to a week). Use of CWR is
generally considered to be a successful migration strategy for these fish that allows them
to both escape peak Columbia River temperatures and delay migration until
temperatures are cooler.
9.	Duration of CWR use is very limited (hours) for summer Chinook, which may provide a
brief respite from warm temperatures. Sockeye salmon do not appear to use CWR as a
migration strategy, although tracking studies of sockeye CWR use has not been done.
Extended use of CWR in the Lower Columbia River is generally considered to be an
ineffective and ultimately unsuccessful migration strategy for these fish due to their run
timing; extended CWR use would likely expose them to warmer Columbia and Snake
River temperatures during the remaining part of their migration later in the summer.
10.	Steelhead begin to use CWR when mainstem temperatures reach 19°C. Fall Chinook
begin to use CWR when mainstem temperatures reach 21 °C. Both species use CWR
extensively when temperatures exceed 21 °C.
11.	CWR use by summer steelhead and fall Chinook likely provides physiological benefits by
reducing the adverse effects associated with prolonged exposure to warm Columbia
River temperatures. Prolonged exposure to warm temperatures increases disease risk,
stress, loss of energy reserves, and mortality risk, and ultimately decreases the
probability to successfully spawn.
12.	Simulation modeling (HexSim) indicates that existing CWR allows steelhead populations
to reduce the cumulative exposure to warm Columbia River temperatures above 21 °C
and 22°C thereby reducing risk of disease and stress-related mortality.
13.	Peak use of Bonneville reservoir CWR by steelhead occurs mid-August through early
September, and peak use by fall Chinook occurs in late August through mid-September.
During an average year (river temperatures and run size), approximately 65,000
steelhead and 5,000 fall Chinook are in Bonneville reservoir CWR. During years with
warm August-September Columbia River temperatures and high run size, as many as
155,000 steelhead and 40,000 fall Chinook are estimated to be in Bonneville reservoir
CWR during the period of peak refuge use, although these peak numbers for steelhead
and fall Chinook may not occur in the same years.
14.	The number of salmon and steelhead in CWR each year is a function of summer
Columbia River temperatures and run size - the larger the run size, the greater number
of fish in CWR; and the warmer the Columbia River temperature, the greater proportion
of the run using CWR.
15.	CWR use appears to be a behavioral adaptation in response to increased summer
Lower Columbia River temperatures. Under colder historical Columbia River
temperatures, which exceeded 20°C for only a short period (a few days) and rarely
exceeded 21 °C, CWR use was likely to be significantly less than what occurs today. This
hypothesis is supported by observations in recent years that show significantly less
CWR use during years when Columbia River water temperatures were relatively cool.
Adverse Effects to Migrating Adult Salmon and Steelhead from Warm Columbia River
Temperatures
181

-------
16.	Optimal Columbia River temperatures for migrating adult salmon and steelhead is below
18°C. Increased stress, disease, mortality, and stored (fat) energy loss that can
ultimately reduce spawning success occur with increasing severity as river temperatures
rise above 20°C. At average river temperatures of 22-23°C, all adverse effects become
significant.
17.	Increased river temperature is correlated with decreased survival for migrating adult
summer steelhead and fall Chinook between Bonneville Dam and McNary Dam. Survival
rates decrease by about 7-10% at >21 °C temperatures relative to temperatures below
20°C. Current CWR use by steelhead and fall Chinook may be minimizing survival loss
by reducing exposure to >21 °C temperatures. However, CWR use may also be
contributing to survival loss from harvest in CWR.
18.	River temperatures above 18°C reduce adult sockeye survival between Bonneville Dam
and McNary Dam. Sockeye mortality rates are moderate at river temperatures of 18-
20°C and are significant at 20-22°C.
19.	The migration timing of sockeye and summer Chinook has shifted to earlier in the year
by approximately a week due to warming of the Lower Columbia River in July. Peak
migration past Bonneville Dam for these fish is now in late June, with very few migrants
in mid- to late July.
20.	Absent use of CWR, a portion of the early fall Chinook exposed to warm Lower
Columbia River temperatures in August are predicted to experience total cumulative
migration energy loss such that they cannot successfully spawn in the fall in the Snake
River.
Sufficiency of Cold Water Refuges to Support Migrating Adult Salmon and Steelhead
21.	EPA's assessment is that CWR is sufficient to attain Oregon's CWR narrative criteria in
the Lower Columbia River if the volume of the 12 primary CWR is maintained and the
Umatilla River is cooled to provide increased CWR volume in August and September
consistent with the Oregon and CTUIR Temperatures TMDLs. Therefore, maintaining
the current temperatures and flows of the 12 primary CWR tributaries and cooling the
Umatilla River is needed to limit significant adverse effects to migrating adult salmon and
steelhead from higher water temperatures in the Columbia River7.
22.	Predicted continual future warming of the Lower Columbia River is expected to increase
salmon and steelhead use of CWR and diminish the extent to which the current amount
of CWR minimizes the risks to migrating adult salmon and steelhead. Therefore,
increasing the amount of CWR in the future through restoration and enhancement is
recommended to help offset the predicted increased future adverse effects associated
with a warmer Lower Columbia River
7 EPA's assessment of CWR needed to attain Oregon's CWR narrative criteria does not imply that current Columbia
River temperatures are at levels to protect salmon and steelhead migration. Current river temperatures exceed the
20°C numeric criterion and cause adverse effects to salmon and steelhead which are not fully mitigated by CWR.
182

-------
Watershed Characteristics of 12 Primary Cold Water Refuges
23.	The 12 primary CWR tributaries are in watersheds with important characteristics and
geographic features that serve to keep the tributaries relatively cool during the summer
period. Some drain from the glaciers of Mount Rainier, Mount Adams, or Mount Hood,
providing cold headwater source water (Cowlitz, Lewis, Sandy, Hood, White Salmon,
Little White Salmon, Klickitat, Deschutes). Some have significant groundwater inflows
that serve to keep the tributary cool (Tanner, Eagle, White Salmon, Little White Salmon,
Klickitat, Deschutes). Ten of the tributary watersheds are in the central or western
Cascades with high percentages of forested areas that minimize solar heating and help
keep waters cool.
24.	Four of the primary tributaries (Cowlitz, Lewis, Sandy, Deschutes Rivers) have upstream
storage dams that play an important role in providing cool summer river flows by
releasing cool water that exists deep within the storage reservoir.
25.	Although the 12 primary CWR tributaries are relatively cool, there are impacts within the
watershed that can warm the tributary, including floodplain degradation, water
withdrawals and reduced summer flow, sedimentation, and loss of riparian
shade. Climate change has already warmed all tributaries to some extent and is
predicted to continue to warm these tributaries in the future. Restoration activities to
address the anthropogenic impacts within the watershed can help offset predicted
warming.
26.	Most of the 12 primary tributaries have sediment build-up at the confluence with the
Columbia River that may impede salmon and steelhead access to the CWR, fill deep
pools preferred by fish, and create shallow areas more susceptible to solar warming.
Recommended Actions to Protect and Restore Cold Water Refuges
27.	Protect existing CWR tributaries through the implementation of existing programs and
regulatory actions8 that help keep waters cool.
a.	Since extensive portions of the priority CWR tributaries include forest lands,
important protective programs include continued implementation of USFS plans
and aquatic strategies on national forest land, State of Oregon and Washington
management plans on state forest land, and the states' forest practice
regulations on private forest land.
b.	Protect existing riparian buffers along the CWR tributaries on non-forest lands
through ongoing implementation of the Columbia River Gorge National Scenic
Area Management Plan in the Columbia River Gorge National Scenic area, Wld
and Scenic River managements plans for the Sandy, White Salmon, Klickitat,
8 Many of the programs and plans referenced in this recommendation are intended to meet various state or federal
requirements, and some are updated on occasion based on new information. By citing a program or plan in this CWR
Plan as important to prevent degradation of water quality and maintain cool river temperatures, EPA is not stating the
program or plan meets applicable state or federal requirements. EPA recognizes some of these plans will be updated
as warranted.
183

-------
and Deschutes rivers, and county land use regulations and plans to protect
shoreline areas.
c.	Maintain existing stream flows, which are important for the size and function of
the primary CWR tributaries, by continued implementation of the minimum
instream flow requirements below Mayfield (Cowlitz), Merwin (Lewis), Bull Run
(Sandy), and Pelton Round Butte (Deschutes) dams and the state minimum
instream flow rules for the Lewis, Sandy, Hood, and Deschutes Rivers. In
accordance with recommendations in existing plans, consider establishing
minimum instream flows for the Wind, Little White Salmon, White Salmon, and
Klickitat Rivers.
d.	Apply state water quality antidegradation requirements and Oregon's protecting
cold water standard (OAR 430-0410-0028 (11)) to help maintain the current
summer river temperatures in Tanner Creek, Herman Creek, Little White Salmon
River, and the White Salmon River, which are currently colder than the
temperature standards. Consider use designation revisions for Tanner Creek and
Herman Creek to reflect the current cold water habitat use.
28.	Implement projects identified in existing plans (e.g., NPCC subbasin plans, Salmon
Recovery Plans, TMDL implementation plans) to restore degraded portions of the 12
primary CWR and the Umatilla River watersheds to enhance the quality of the CWR and
to counteract predicted future increases in tributary river temperature. Projects include:
1) restoring riparian vegetation to provide river shading; 2) restoring stream morphology
and floodplain connectivity to reduce channel widths and create pools and groundwater
connectivity; and 3) restoring summer river flows that are more resistant to warming and
increase CWR volume.
29.	Develop a temperature TMDL(s) and associated implementation plan(s) for the
Deschutes River basin to aid in restoration actions to cool the lower Deschutes River
temperatures. Due to the Deschutes River's high CWR use by migrating salmonids,
marginally cool current temperatures, and predicted temperature increases due to
climate change, efforts to target 18°C temperatures or less during the August-September
high CWR use period is a high priority.
30.	To increase the CWR in the Umatilla River and help attain Oregon's CWR narrative
standard, implement actions in existing plans (e.g., NPCC Umatilla subbasin plan, Mid-
Columbia River Steelhead Recovery Plan, and the Umatilla River TMDL implementation
plan) that help increase shade, increase floodplain connectivity, and restore rivers flows.
A priority project is agreement on and implementation of an additional Umatilla Basin
Project water exchange or alternative measure to increase summer flows in the Lower
Umatilla River, thereby decreasing summer river temperatures and increasing CWR
volume.
31.	Maintain or enhance cool water releases during late July through mid-September from
upstream dams on the Cowlitz, Lewis, Sandy, and Deschutes Rivers to maintain or
increase CWR. Due to the importance and vulnerability of the Deschutes River CWR,
assess the potential to release cooler water from Pelton Round Butte Dam in August to
potentially cool the river at the mouth.
184

-------
32.	Consider feasibility studies for restoration and sediment removal at the confluence areas
of the following watersheds to increase fish access to CWR and increase depth: Herman
Creek Cove, Wind River, White Salmon River, and Klickitat River.
33.	In addition to protecting and restoring 12 primary CWR tributaries and the Umatilla River,
restoring the other non-primary CWR tributaries, Fifteenmile Creek, and potentially other
tributaries to the Lower Columbia River is recommended to provide additional CWR in
the future to help address the expected increased use of CWR due to a warmer
Columbia River. Construct the proposed log structure at the mouth of Oneonta Creek as
a demonstration project for CWR augmentation at small stream confluences. Examine
the feasibility of groundwater supply supplementation at the Cascade Hatchery to cool
and augment the Eagle Creek CWR.
Fishing in Cold Water Refuges
34.	Fishing in CWR appears to reduce the survival of steelhead that use CWR compared to
those that do not, offsetting the benefits to fish using CWR. This Plan may inform future
updates to fishing regulations in the primary CWR, especially related to periods of warm
Columbia River temperatures for the CWR with the highest use (Cowlitz River, Lewis
River, Herman Creek Cove, Wnd River, White Salmon River, Little White Salmon River
(Drano Lake), Klickitat River, and Deschutes River).
Recommended Studies and Monitoring to Address Uncertainties and Trends
35.	In Chapter 8, several scientific uncertainties associated with this Plan were highlighted
with recommended future studies to address them, which include: radio-tag studies to
track fish use of CWR below Bonneville Dam, repeated radio-tag studies to track fish
use of CWR above Bonneville Dam under current conditions, installation of PIT-tag
detectors at the mouth of CWR tributaries, installation of temperature and flow gages
near the mouth of CWR tributaries where there are none currently, and studies designed
to better characterize the adverse effects to fish from exposure to elevated temperatures
in the Lower Columbia River and the associated benefits of CWR use to reduce the
adverse effects.
36.	Immediate monitoring priorities include: install PIT-tag detectors in Little White Salmon
River/Drano Lake and Herman Creek Cove; re-establish USGS flow gauges, including
temperature gauges, near the mouth of Little White Salmon River and Wnd River; and
install and operate long-term annual summer temperature monitors at the USGS flow
gauge sites near the mouth of the Cowlitz, Lewis, Sandy, White Salmon, Hood, Klickitat,
and Umatilla Rivers.
185

-------
10 REFERENCES
Berejikian, B. A., Tezak, E. P., & Schroder, S. L. 2001. Reproductive behavior and breeding
success of captively reared Chinook salmon. North American Journal of Fisheries
Management, 21(1), 255-260.
Bowerman, T. E., Pinson-Dumm, A., Peery, C. A., & Caudill, C. C. 2017. Reproductive energy
expenditure and changes in body morphology for a population of Chinook salmon
Oncorhynchus tshawytscha with a long distance migration. Journal of Fish Biology, 90(5), 1960-
1979.
Connor, W. P., Tiffan, K. F., Chandler, J. A., Rondorf, D. W., Arnsberg, B. D., & Anderson, K. C.
2018. Upstream migration and spawning success of Chinook salmon in a highly developed,
seasonally warm river system. Reviews in Fisheries Science & Aquaculture, 1-50.
Crozier, L. G., Scheuerell, M. D., & Zabel, R. W. 2011. Using time series analysis to
characterize evolutionary and plastic responses to environmental change: a case study of a shift
toward earlier migration date in sockeye salmon. The American Naturalist, 178(6), 755-773.
DART (Columbia River Data Access in Real Time). Columbia Basin Research, School of
Aquatic and Fishery Sciences, University of Washington, Seattle, WA 98101.
Fish Passage Center. 2014. Annual Report. BPA Contract #67312, BPA Project #1994-033-00.
Fish Passage Center, Portland, Oregon.
Fish Passage Center. 2015. Memorandum from Fish Passage Center Staff, to Mary Lou Soscia,
EPA; Christine Psyk, EPA; and John Palmer, EPA, re: Information regarding 2015 conditions at
Bonneville Dam. Fish Passage Center, Portland, Oregon. August 26, 2015.
Fish Passage Center. 2015. Memorandum from Michele DeHart, Fish Passage Center, to
Charles Morrill, WDFW; Erick VanDyke, ODFW, and Steven Hawley, citizen, re: Requested
data summaries and actions regarding sockeye adult fish passage and water temperature
issues in the Columbia and Snake Rivers. Fish Passage Center, Portland, Oregon. October 28,
2015.
Fish Passage Center. 2016. Annual Report. BPA Contract #7075, BPA Project #1994-033-00.
Fish Passage Center, Portland, Oregon.
Fish Passage Center. 2016. Memorandum from Michele DeHart, Fish Passage Center, to John
Palmer, U.S. Environmental Protection Agency, re: The effect of water temperature on
steelhead upstream passage. Fish Passage Center, Portland, Oregon. October 31, 2016.
Fish Passage Center. 2016. Memorandum from Michele DeHart, Fish Passage Center, to
Charles Morrill, WDFW, and Erick VanDyke, ODFW, re: Requested data summaries regarding
summer Chinook adult fish passage and water temperature in the Columbia and Snake River.
Fish Passage Center, Portland, Oregon. January 29, 2016.
186

-------
Fish Passage Center. 2018. Memorandum from Michele DeHart, Fish Passage Center, to John
Palmer, U.S. Environmental Protection Agency, re: Fall Chinook survival between Bonneville
and McNary Dams, and the relationship between water temperature, travel velocity, and arrival
timing. Fish Passage Center, Portland, Oregon. May 8, 2018.
Goniea, T. M., Keefer, M. L., Bjornn, T. C., Peery, C. A., Bennett, D. H., & Stuehrenberg, L. C.
2006. Behavioral thermoregulation and slowed migration by adult fall Chinook salmon in
response to high Columbia River water temperatures. Transactions of the American Fisheries
Society, 135(2), 408-419.
Hatch, D. R., Fast, D. E., Bosch. W. J., Blodgett, J. W., Whiteaker, J.M., Branstetter, R., &
Pierce, A. L. 2013. Survival and Traits of Reconditioned Kelt Steelhead Oncorhynchus mykiss in
the Yakima River, Washington. North American Journal of Fisheries Management 33:615-625.
High, B., Peery, C. A., & Bennett, D. H. 2006. Temporary staging of Columbia River summer
steelhead in coolwater areas and its effect on migration rates. Transactions of the American
Fisheries Society, 135(2), 519-528.
Isaak, D., Wenger, S., Peterson, E., Ver Hoef, J., Nagel, D., Luce, C & Parkes-Payne, S. 2017.
The NorWeST summer stream temperature model and scenarios for the western U.S.: A crowd-
sourced database and new geospatial tools foster a user community and predict broad climate
warming of rivers and streams. Water Resources Research, 53, 9181-9205.
Isaak, D., Wenger, S., Peterson, E., Ver Hoef, J., Nagel, D., Luce, C., Hostetler, S., Dunham, J.,
Roper, B., Wollrab, S., Chandler, G., Horan, D., & Parkes-Payne, S. 2018. Global Warming of
Salmon and Trout Rivers in the Northwestern U.S.: Road to Ruin or Path Through Purgatory?
Transactions of the American Fisheries Society. 147(3): Pages 566-587.
Jager, H. I. 2011. Quantifying temperature effects on fall Chinook salmon. Oak Ridge, TN: US
Department of Energy, Oak Ridge National Laboratory, 10, 1047614.
Jepson, M. A., Keefer, M. L., Naughton, G. P., Peery, C. A., & Burke, B. J. 2010. Population
composition, migration timing, and harvest of Columbia River Chinook salmon in late summer
and fall. North American Journal of Fisheries Management. 30: 72-88.
Keefer, M. L., Peery, C. A., Bjornn, T. C., Jepson, M. A., & Stuehrenberg, L. C. 2004.
Hydrosystem, dam, and reservoir passage rates of adult Chinook salmon and steelhead in the
Columbia and Snake rivers. Transactions of the American Fisheries Society, 133(6), 1413-1439.
Keefer, M. L., Caudill, C. C., Peery, C. A., & Bjornn, T. C. 2006. Route selection in a large river
during the homing migration of Chinook salmon (Oncorhynchus tshawytscha). Canadian Journal
of Fisheries and Aquatic Sciences, 63(8), 1752-1762.
Keefer, M. L., Peery, C. A., & High, B. 2009. Behavioral thermoregulation and associated
mortality trade-offs in migrating adult steelhead (Oncorhynchus mykiss): variability among
sympatric populations. Canadian Journal of Fisheries and Aquatic Sciences, 66(10), 1734-1747.
Keefer, M. L., Jepson, M. A., Clabough, T. S., & Caudill, C. C. 2017. Migration of adult
salmonids in the federal Columbia River hydrosystem: a summary of radiotelemetry studies,
187

-------
1996-2014. Prepared for U.S. Army Corps of Engineers, Portland District. Technical Report,
2017-2.
Keefer, M. L., & Caudill, C. C. 2017. Assembly and analysis of radiotelemetry and temperature
logger data from adult Chinook salmon and steelhead migrating through the Columbia River
Basin. Prepared forTetra Tech and the U.S. Environmental Protection Agency. Technical
Report 2017-1.
Keefer, M. L., Jepson, M. A., Clabough, T. S., Johnson, E. L., & Caudill, C. C. 2015. Reach
conversion rates of radio-tagged Chinook and sockeye salmon and steelhead in the Lower
Columbia River (A Report for Study Code ADS-P-13-2). Prepared for U.S. Army Corps of
Engineers, Portland District. Technical Report 2015-8.
Keefer, M. L., Clabough, T. S., Jepson, M. A., Johnson, E. L., Peery, C. A., & Caudill, C. C.
2018. Thermal exposure of adult Chinook salmon and steelhead: Diverse behavioral strategies
in a large and warming river system. PloS one, 13(9), e0204274.
Lower Columbia River Estuary Partnership, (n.d.) Lower Columbia River Thermal Refugia
Study: Results Summary for 2014-2015 Columbia River Gorge Stream Temperature Monitoring.
McCullough, D. 1999. A Review and Synthesis of Effects of Alterations to the Water
Temperature Regime on Freshwater Life Stages of Salmonids, with Special Reference to
Chinook Salmon. Prepared for the U.S. Environmental Protection Agency. EPA 910-R-99-010.
Columbia River Inter-Tribal Fish Commission, Portland, OR.
National Research Council. 2004. Managing the Columbia River: instream flows, water
withdrawals, and salmon survival. National Academies Press.
Naughton, G. P., Jepson, M. A., Peery, C. A., Brun, C. V., & Graham, J. C. 2009. Effects of
temporary tributary use on escapement estimates of adult fall Chinook salmon in the Deschutes
River, Oregon. North American Journal of Fisheries Management, 29(6), 1511-1518.
NMFS. 2014. Endangered Species Act Section 7(a)(2) Supplemental Biological Opinion,
Consultation on Remand for Operation of the Federal Columbia River Power System. NWR-
2013-9562. National Marine Fisheries Service, Northwest Region.
NMFS. 2016. 2015 Adult Sockeye Salmon Passage Report. National Marine Fisheries Service.
NMFS. 2017a. Use of the Deschutes River by Migrating Adult Snake River Steelhead. National
Marine Fisheries Service. Draft.
NMFS. 2017b. Supplemental Recovery Plan Module for Snake River Salmon and Steelhead -
Mainstem Columbia River Hydropower Projects. National Marine Fisheries Service.
NMFS. 2019. Endangered Species Act Section 7(a)(2) Biological Opinion and Magnuson-
Stevens Fishery Conservation and Management Act Essential Fish Habitat Consultation for the
Continued Operation and Maintenance of the Columbia River System. WCRO-2018-00152.
National Marine Fisheries Service.
NMFS. 2020. Endangered Species Act Section 7(a)(2) Biological Opinion and Magnuson-
Stevens Fishery Conservation and Management Act Essential Fish Habitat Response for the
188

-------
Continued Operation and Maintenance of the Columbia River System. WCRO 2020-00113.
National Marine Fisheries Service.
Oregon Department of Environmental Quality, 2001. Umatilla River Basin Total Maximum Daily
Load (TMDL) and Water Quality Management Plan (VVQMP),
Plumb, J. M. 2018. A bioenergetics evaluation of temperature-dependent selection for the
spawning phenology by Snake River fall Chinook salmon. Ecology and Evolution, 8(19), 9633-
9645.
Railsback, S. F., Harvey, B. C., Jackson, S. K., & Lamberson, R. H. 2009. InSTREAM: the
individual-based stream trout research and environmental assessment model. Gen. Tech. Rep.
PSW-GTR-218. Albany, CA: US Department of Agriculture, Forest Service, Pacific Southwest
Research Station. 254 p, 218.
Richter, A., & Kolmes, S. A. 2005. Maximum temperature limits for Chinook, coho, and chum
salmon, and steelhead trout in the Pacific Northwest. Reviews in Fisheries Science, 13(1), 23-
49.
Robards, M. D., & Quinn, T. P. 2002. The migratory timing of adult summer-run steelhead in the
Columbia River over six decades of environmental change. Transactions of the American
Fisheries Society, 131(3), 523-536.
Snyder, M. N., Schumaker, N. H., Ebersole, J. L., Dunham, J. B., Comeleo, R. L., Keefer, M. L.,
... & Palmer, J. 2019. Individual based modeling offish migration in a2-D river system: model
description and case study. Landscape Ecology, 34(4), 737-754.
Snyder, M.N., N.H. Schumaker, J.B. Dunham, M L. Keefer, P. Leinenbach, A. Brookes, J.
Palmer, J. Wu, D. Keenan, J.L. Ebersole. 2020. Assessing contributions of cold-water refuges to
reproductive migration corridor conditions for adult Chinook Salmon and steelhead trout in the
Columbia River, USA. Journal of Ecohydraulics.
Sullivan, K., Martin, D. J., Cardwell, R. D., Toll, J. E., & Duke, S. 2000. An analysis of the effects
of temperature on salmonids of the Pacific Northwest with implications for selecting temperature
criteria. Sustainable Ecosystems Institute, Portland, OR.
Torgersen, C., Ebersole, J. & Keenan, D. 2012. Primer for identifying cold-water refuges to
protect and restore thermal diversity in riverine landscapes - EPA 910-C-12-001. U.S.
Environmental Protection Agency, Washington, DC, EPA/600/R-12/038.
USACE (U.S. Army Corps of Engineers). 2013. Location and use of adult salmon thermal
refugia in the Lower Columbia and Lower Snake Rivers. Federal Columbia River Power System,
Amendment 1 of the Supplemental FCRPS BiOp.
USACE (U.S. Army Corps of Engineers). 2017. Endangered Species Act. Federal Columbia
River Power System, 2016 Comprehensive Evaluation. U.S. Army Corps of Engineers
Northwestern Division, Bureau of Reclamation Pacific Northwest Region, and Bonneville Power
Administration.
U.S. Environmental Protection Agency. 2003. EPA Region 10 Guidance for Pacific Northwest
State and Tribal Temperature Water Quality Standards. EPA 910-B-03-002. Region 10 Office
189

-------
of Water, Seattle, WA.
U.S. Environmental Protection Agency. 2005. Confederated Tribes of the Umatilla Indian
Reservation Total Maximum Daily Load For Temperature and Turbidity. July 2005. Prepared by
the Confederated Tribes of the Umatilla Indian Reservation, Department of Natural Resources.
Region 10, Seattle, WA.
U.S. Environmental Protection Agency. 2020a. Columbia and Lower Snake Rivers Temperature
Total Daily Maximum Load. Region 10, Seattle, WA.
U.S. Environmental Protection Agency. 2020b. Assessment of Impacts to Columbia and Snake
River Temperatures using the RB10 Model. Scenario Report. March 2020. Region 10, Seattle,
WA.
Yearsley, J. R. 2009. A semi-Lagrangian water temperature model for advection-dominated
river systems. Water Resources Research, 45 (12).
190

-------
11 CHAPTER 7 BIBLIOGRAPHY
Applicable to multiple snapshots:
2017. USFS and EPA. NorWeST model runs and analysis from Chapter 1.
2017. EPA. Draft Shade Images North.
2017.	EPA. Land Use and Ownership Maps.
2004, as amended through 2016. Columbia River Gorge Commission and U.S. Forest Service.
Management Plan for the Columbia River Gorge National Scenic Area.
1990. U.S. Forest Service.
.
https://ecology.wa.govA/Yater-ShorelinesA/Yater-supply/Sireamflow-resiorationA/Yatershed-plan-
COWLITZ RIVER
2018.	Cowlitz County. Cowlitz County Shoreline Master Program 2018.
2017. Cowlitz County. Cowlitz County Comprehensive Plan: 2017 Update.
2017. Lewis County. Lewis County Shoreline Master Program: Environment Designations,
Policies, & Regulations.
2015. Climate Impacts Group, University of Washington. Climate Change Impacts on Tacoma
Power Watersheds.
2014. Lower Columbia Fish Recovery Board. WRiA 26 Water Supply and Stream Flow Review:
Findings and Recommendations.
2012. WA Department of Ecology. Focus on Water Availability: Cowlitz Watershed, WRIA 26.
2010. Lower Columbia Fish Recovery Board. WA Lower Columbia Salmon Recovery and Fish
and Wildlife Subbasin Plan - Lower Cowlitz Subbasin.
2010. Lower Columbia Fish Recovery Board. WA Lower Columbia Salmon Recovery and Fish
and Wildlife Subbasin Plan - Upper Cowlitz, Cispus, and Tilton Subbasin.
2002. Federal Energy Regulatory Commission. Cowlitz River Hydroelectric Project No. 2016
License. FERC Order Project No. 2016-044.
2002. Washington State Department of Ecology. Water Quality Certification for the Cowlitz River
Hydroelectric Project (FERC Number 2016).
Websites:
Capitol Land Trust. Cowlitz River Preserve:
Tacoma Public Utilities. Cowlitz River Project:
191

-------
TDN.com. "Cowlitz river deepens slightly for first time since Mount St. Helens eruption:"
(April 7, 2014)
The Daily Chronicle. "Commentary: Climate, Carbon and an Uncertain Future for the Cowlitz
and other Waterways:"
(January
22, 2018)
WA Department of Fish and Wildlife. Cowlitz Wildlife Area:
LEWIS RIVER
2018. Cowlitz County. Cowlitz County Shoreline Master Program 2018.
2018. WA Department of Ecology. East Fork Lewis River Watershed Bacteria and Temperature
Source Assessment Report
2016.	WA Department of Ecology. Focus on Water Availability: Lewis River Watershed, WRIA
27.
2012. Clark County. Clark County Shoreline Master Program 2012 Update.
2010. Lower Columbia Fish Recovery Board. North Fork Lewis Washington Lower Columbia
Salmon Recovery and Fish & Wildlife Subbasin Plan.
2010. Lower Columbia Fish Recovery Board. East Fork Lewis Washington Lower Columbia
Salmon Recovery and Fish & Wildlife Subbasin Plan.
2008. Washington Administrative Code. Water Resources Management Program for the Lewis
Basin, WRIA 27.
2008. Federal Energy Regulatory Commission. Merwin Hydroelectric License
w/Amendment. Piroiiei	35. FERC Order.
2006. Lower Columbia Fish Recovery Board, Lead Agency. Counties of Clark, Cowlitz, and
Skamania. Salmon-Washougal & Lewis Watershed Management Plan, WRIAs 27-28
Websites:
Columbia River Images. Lewis River, Washington:
PacifiCorp. Lewis River:
Lewis River:
USGS. Lakes and Drainages Around Mount St. Helens:
SANDY RIVER
2017.	Sandy River Basin Watershed Council. State of the Sandy.
192

-------
2015. U.S. Army Corps of Engineers. Sandy River Delta Dam Removal Project. Powerpoint
presentation.
2013. U.S. Army Corps of Engineers. Environmental Assessment Sandy River Delta Section
536 Ecosystem Restoration project, Multnomah County.
2012. USGS. Geomorphic Response of the Sandy River, Oregon, to Removal of Marmot Dam.
2010. Oregon Fish and Wildlife. Oregon Lower Columbia Plan.
2008. Portland Water Bureau. Bull Run Water Supply Habitat Conservation Plan.
2005. Oregon Department of Environmental Quality. Sandy River Basin Total Maximum Daily
Load.
1994. USFS. Upper Sandy National Wild and Scenic River Management Plan.
1993. Bureau of Land Management. Sandy River Wild and Scenic River and State Scenic
Waterway Management Plan.
Websites:
Columbia River Gorge National Scenic Area, Sandy River Delta. USFS.
Friends of the Sandy River Delta.
National Wld and Scenic River System, Sandy River, Oregon. USFS.
Partnerships in the Sandy River. USFS.
Sandy River Basin Partnership: Case Study by Freshwater Trust.
Sandy River Basin Watershed Council.
Sandy River Delta dam removal. U.S. Army Corps of Engineers.
Sandy River Water Trail Paddle Guide. City of Sandy.
TANNER CREEK
2018. EPA. EPA Region 10 Internal Memo re: Estimated the extent of riparian disturbance
resulting from the 2017 Eagle Creek Fire. To: Jenny Wu. Author: Peter Leinenbach. Date:
12/10/2018.
1978. Oregon Water Resources Department. Water Right Permit No. G 5646 (Certificate No.
45793). Retrieved from:
on July 17, 2019.
193

-------
1962. Oregon Water Resources Department. Water Right Permit No. S 1310 (Certificate No.
30039). Retrieved from:
on July 17, 2019.
EAGLE CREEK
2017. OWRD. Water Availability Analysis Eagle Creek > Columbia River-At Mouth, Hood
Basin.
2011. USFS. Watershed Condition Framework: A Framework for Assessing and Tracking
Changes to Watershed Condition.
1990. USFS. Mount Hood Forest Land Resources Management Plan.
Websites:
USFS Mount Hood National Forest Land and Resources Management website.
ODEQ. Figure 160A: Fish Use Designations Map, Hood Basin, Oregon.
HERMAN CREEK
2017. Conversations with Port of Cascade Locks. (July, August 2017)
2010. Cramer Fish Sciences. Temperature Characteristics of Herman Creek Cove and its
Function as a Cool-Water Refuge for Adult Salmon and Steel head in the Columbia River.
2008. ODFW. Oxbow Hatchery Program Management Plan.
2004. Hood River Soil and Conservation District. Hood River Subbasin Plan Including Lower
Oregon Columbia Gorge Tributaries.
USFS. Herman Creek Trail #406.
https://www.fs.usda.gov/lntemet/FSE_DOCUMENTS/stelprdb5397Q1Q.pdf
WIND RIVER
2017. Underwood Conservation District. Landowner Handbook: A Guide to Best Management
Practices in Skamania and Klickitat Counties.
2017. Underwood Conservation District, Lower Columbia Fish Recovery Board, Wnd River
Work Group. Wind River Habitat Restoration Strategy. Inter-fluve.
2017. Washington Department Fish and Wldlife. Abundance and Productivity of Wind River
Steel head and Preliminary Assessment of their Response to Hemlock Dam Removal, 2016,
BP A Project#: 1998-019-00.
194

-------
2015. Washington Department Fish and Wildlife. Abundance and Productivity of Wind River
Steeihead and Preliminary Assessment of their Response to Hemlock Dam Removal, 2015,
BP A Project#: 1998-019-00.
2015. Western WRIA 29 Planning Unit. WRIA 29A Watershed Planning Detailed
Implementation Plan.
2014.	R.A. Lovett. Dam removals: Rivers on the run. Nature.
2013. USGS. Wnd River subbasin restoration: U.S. Geological Survey annual report November
2012 through December 2013.
2012. Washington Department of Ecology. Focus on Water Availability: Wind-White Salmon
Watershed, WRIA 29. Publication No. 11-11-033.
2012. USGS. Wind River Watershed Restoration Annual Report, November 2011.
2010. Lower Columbia Fish Recovery Board. Lower Columbia Salmon Recovery and Fish and
Wildlife Subbasin Plan - Wind Subbasin.
2008. Washington Department of Ecology. Table 3: Goals and Accomplishments Associated
with Wind River Temperature Total Maximum Daily Load Daily Implementation Plan.
2005. Western WRIA 29 Planning Unit. Watershed Management Plan for Western Water
Resource Inventory Area 29.
2004.	Washington Department of Ecology. Wind River Watershed Temperature Total Maximum
Daily Load: Detailed Implementation Plan. Publication No. 04-10-037.
2002. Washington Department of Ecology. Wind River Watershed Temperature Total Maximum
Daily Load. Publication No. 02-10-029.
Lower Columbia Fish Recovery Board. Wnd River Restoration Workgroup.
USFS. Wnd River Stream Restoration Cost Estimates.
USFS. Wnd River Watershed Restoration Projects Annual Reports 2000-2002.
LITTLE WHITE SALMON RIVER
2015.	Western WRIA 29 Planning Unit. WRIA 29A Watershed Planning Detailed
Implementation Plan.
2010. Lower Columbia Fish Recovery Board. Lower Columbia Salmon Recovery and Fish and
Wildlife Subbasin Plan - Little White Salmon Subbasin.
2005.	Western WRIA 29 Planning Unit. Watershed Management Plan for Western Water
Resource Inventory Area 29.
2004b. Columbia River Inter-Tribal Fish Commission (CRITFC). Little White Salmon River
Subbasin Plan Summary.
2004c. Northwest Power and Conservation Council. Volume II, Chapter 17 Little White Salmon
Sub-basin.
195

-------
2000. Prepared for the Northwest Power Planning Council by Dan Rawding at the WA
Department of Fish and Wildlife. NWCP - Little White Salmon River Subbasin Summary.
Undated. Northwest Power and Conservation Council. Little White Salmon Subbasin
Dashboard.
Undated. WQS. Chapter 173-201A WAC Water Quality Standards for Surface Waters of the
State of Washington.	. Accessed
8/1/2016.
WHITE SALMON RIVER
2016. Salmon and Steelhead in the White Salmon River after the Removal of Condit Dam—
Planning Efforts and Recolonization Results. Fisheries, Vol. 41, No. 4.
2015.	USGS & U.S. Department of Interior. Salmon Habitat Assessment for Conservation
Planning in the Lower White Salmon River, Washington.
2013. National Marine Fisheries Service. ESA Recovery Plan for the White Salmon River
Subbasin.
2013. Klickitat County Natural Resource Department. Klickitat Lead Entity Region - Salmon
Recovery Strategy. John Foltz.
2013.	Mid-Columbia Fisheries Enhancement Group. Literature Review of Mainstem Columbia
River Anadromous Salmonid Habitat and Restoration Potential Between Bonneville Dam and
the Yakima River Confluence.
2012. Washington Department of Ecology. Wnd-White Salmon Watershed, WRiA 29: Focus on
Water A vail ability, 11 -11 -033.
2009. National Marine Fisheries Service. Appendix B: Recovery Plan for the Klickitat Population
of Middle Columbia River Steelhead Distinct Population Segment.
1991. The Lower White Salmon Wild and Scenic River Management Plan.
Yakama Nation Fisheries. Underwood In-Lieu Navigation and Habitat Restoration Project.
HOOD RIVER
2018. ODEQ. Western Hood Subbasin Temperature TMDL, revised.
2016.	Water Professionals Network for Hood River Basin SWCD. Hood River Basin Water
Conservation Strategy.
2014.	Hood River Watershed Group. Hood River Watershed Action Plan.
2014. CRITFC. Spirit of the Salmon: Wy-Kan-Ush-Mi Wa-Kish-Wit, 2014 update.
2006. USDA/USFS Mount Hood National Forest. Hood River Aquatic Habitat Restoration
Strategy.
2004. Hood River SWCD. Hood River Subbasin Plan.
196

-------
2001.	ODEQ. Western Hood Subbasin TMDL.
1999. Hood River Soil and Water Conservation District. Hood River Watershed Assessment.
1990. USFS. Mount Hood Forest Land Resources Management Plan.
Websites:
Hood River SWCD:
Columbia Land Trust, Powerdale Dam:
KLICKITAT RIVER
2013. WRIA 30 Watershed Planning and Advisory Committee, Klickitat County, Central Klickitat
Conservation District. Thermal Refuges and Fish Habitat Assessment, WRIA 30.
2013. Klickitat County Natural Resources Department. Klickitat Lead Entity Region-Salmon
Recovery Strategy.
2012. Washington Department of Ecology. Klickitat Watershed, WRIA 30: Focus on Water
Availability. Publication 11-11-034.
2009. National Marine Fisheries Service. Middle Columbia River Steelhead Recovery Plan.
2009. WRIA 30 Water Resource Planning and Advisory Committee. Riparian Vegetation
Assessment, Little Klickitat River and Swale Creek: Water Resource Inventory Area 30.
2008. Narum, S.R. et al. Influence of landscape on resident and anadromous life history O.
mykiss. Canadian Journal of Fisheries and Aquatic Sciences 65:1013-1023.
2008. Yakama Nation in cooperation with WDFW. Klickitat River Anadromous Fisheries Master
Plan.
2007. WRIA 30 Water Resource Planning and Advisory Committee. Framework for Water
Management in WRIA 30 Klickitat River Watershed.
2005. WRIA 30 Watershed Planning Unit. Klickitat Basin (WRIA 30) Watershed Management
Plan. 2005. Klickitat County Planning Unit. WRIA 30 Phase II Watershed Assessment.
2004. NPCC. Klickitat Subbasin Plan.
2004. BPA. Yakama/Klickitat Fisheries Project-Klickitat Only Monitoring and Evaluation, 2003-
2004 Annual Report. Publication DOE/BP-00014033-1.
2003. Sampson, M., & Evenson, R. Yakima/Klickitat Fisheries Project: Klickitat Only Monitoring
and Evaluation, 2002-2003 Annual Report (No. DOE/BP-00005934-1). Bonneville Power
Administration (BPA), Portland, OR.
2003. Aspect Consulting for Klickitat County. Multipurpose Water Storage Screening
Assessment Report WRIA 30.
2002.	Washington Department of Ecology. Little Klickitat River Watershed Temperature Total
Maximum Daily Load. Publication 02-03-031.
1998. Klickitat County. Klickitat County Shoreline Master Plan.
197

-------
Undated. National Marine Fisheries Service. Northwest Salmon & Steelhead Recovery Planning
& Implementation: Middle Columbia River Steelhead DPS Recovery Plan Summary.
ftp://ftp.library.noaa.gov/noaa_documents.lib/NMFS/NWFSC/Middle_Columbia_steelhead_DPS
_plan_summary.pdf
Undated. Klickitat County. Links to Klickitat River Planning Documents.
Websites:
Columbia Land Trust.
FIFTEENMILE CREEK
2010. United States Department of Agriculture. Fifteenmile Creek Basin Aquatic Habitat
Restoration Strategy. Mount Hood National Forest, Barlow Ranger District, Dufur, Oregon.
2010. Oregon Department of Fish and Wildlife. Conservation and Recovery Plan for Oregon
Steelhead Populations in the Middle Columbia River Steelhead Distinct Population Segment.
2009. National Marine Fisheries Service. Middle Columbia River Steelhead Recovery Plan.
2008. Oregon Department of Environmental Quality. Middle Columbia-Hood (Miles Creeks)
Total Maximum Daily Load.
2004. Graves, R. Fifteenmile Creek Riparian Buffers Project, 2002-2003 Annual Report, Project
No. 200102100, 10 electronic pages (BPA Report DOE/BP-00004935-2).
2003. Wasco County Soil and Water Conservation District, for Fifteenmile Watershed Council.
Fifteenmile Watershed Assessment.
Oregon Water Resources Department. 5/29/2013. Presentation on "Groundwater Supplies in
the Fifteenmile Creek Drainage, The Dalles, Oregon."
Fifteenmile Watershed Council. 1/25/2012. Endangered Species Act Workshop Meeting
Minutes.
Websites:
The Freshwater Trust. 6/14/2018. "Grant lays foundation for alert system to protect fish."
National Wild and Scenic Rivers System. Fifteenmile Creek, Oregon.
DESCHUTES RIVER
2016. McMillan, G., Hafele, R., Bond, J., Faux, R., & Diabat, M. Airborne Thermal Infrared
Remote Sensing of the Lower Deschutes River. Deschutes River Alliance.
2013. French, R., Stuart, A., Hooton, B., & Dale, C. White Paper-2013 Lower Deschutes River
Temperatures and Fisheries. Oregon Fish and Wldlife.
198
Draft

-------
2009.	National Marine Fisheries Service. Middle Columbia River Steel head Recovery Plan.
2006. High, B., Peery, C. A., and Bennett, D. H. Temporary staging of Columbia River summer
steelhead in coolwater areas and its effect on migration rates. Transactions of the American
Fisheries Society 135(2):519-528.
2005. Northwest Power and Conservation Council [NPCC]. Deschutes Subbasin Plan.
2003.	O'Connor, J. and Grant, G. A Peculiar River: Geology, Geomorphology, and Hydrology of
the Deschutes River, Oregon. American Geophysical Union, Washington, D.C.
Websites:
Deschutes River Conservancy.
NorWeST Regional Database and Modeled Stream Temperatures.
Portland General Electric. "Deschutes Fish Runs."
Oregon Department of Fish Wldlife Mid-Columbia River Conservation and Recovery Plan
Upper Deschutes Watershed Council.
UMATILLA RIVER
2016. Pagel, Martha. AMP Insights. Oregon's Umatilla Basin Aquifer Recharge and Basalt bank.
Case Study for: Political Economy of Water Markets.
2015. Confederated Tribes of the Umatilla Indian Reservation. Draft Agricultural Management
Environmental Assessment.
2012. U.S. Bureau of Reclamation. Umatilla Basin Water Supply Study.
2010.	Oregon Department of Fish and Wldlife. Conservation and Recovery Plan for Oregon
Steelhead Populations in the Middle Columbia River Steelhead Distinct Population Segment.
2009. National Marine Fisheries Service. Middle Columbia River Steelhead Recovery Plan.
2008. Umatilla County Critical Groundwater Task Force. Umatilla Subbasin 2050 Water
Management Plan.
2005. Confederated Tribes of the Umatilla Indian Reservation Total Maximum Daily Load for
Temperature and Turbidity. Confederated Tribes of the Umatilla Indian Reservation, Department
of Natural Resources. Pendleton (Mission), OR.
2004.	Northwest Power and Conservation Council. Umatilla/Willow Subbasin Plan.
2001 Oregon Department of Environmental Quality, Umatilla River Basin Total Maximum Daily
Load (Tfy'DL) and Wafer Quality fy'anagemeni Plan (WQMP).
199

-------
1990. U.S. Forest Service. Umatilla National Forest Land and Resource Management Plan.
1988. U S Bureau of Reclamation. Umatilla Basin project, Oregon: Planning report final
environmental impact statement.
Websites:
Bureau of Reclamation. Umatilla Basin Project.
200

-------
12 APPENDICES
12.1
12.2
BY TRIBUTARIES WITHIN THE LOWER/MIDDLE COLUMBIA RIVER
BASED ON NORWEST TEMPERATURE MODEL
12.3	• ¦¦¦ " ¦ ¦ : :¦	^ ¦ : - ¦ ' . ¦	- /. - ¦ ¦ /
CURRENTLY PROVIDE CWR IN THE LOWER COLUMBIA RIVER
12.4	¦¦ ¦/ ¦ : ¦¦ .¦ ' • ¦ ' V ;¦ ¦ ¦ : ¦' .¦ ¦	/, ¦¦ ¦ ¦ .¦ ¦ ¦¦
MIGRATING SALMON AND STEELHEAD
12.5
TRIBUTARIES PROVIDING CWR IN THE LOWER COLUMBIA RIVER AND
SELECTION OF THE 12 PRIMARY CWR
12.8 '• ¦ - - /, " ¦¦¦¦¦ ' " '• A/. ¦ " " ¦ ¦ ' ¦ V. ¦ ¦ ¦ ¦ :
MODELING REPORT
12.7
WITHIN TRIBUTARIES THAT DISCHARGE INTO THE COLUMBIA RIVER
12.8	' ¦ :	' ' ¦¦¦ ¦¦¦ ' ¦¦¦¦¦¦ " /. " A
RIVER CONFLUENCE SITES USING USEPA FIELD DATA COLLECTED
IN 2016
12.9
SALMON RIVER/DRANO LAKE
12.10
12.11	• ¦ ¦ - ¦ : .¦ ¦ ¦ ¦.	¦ - ¦ : . ¦ - - T : : - ¦ ¦
COVE
12.12
SUMMARY
12.13	¦ ' : 'T ¦ ' '¦ ¦ " ¦¦ , ' • ' ¦ ¦ • ¦ A ¦¦¦, : ^
CWR IN THE BONNEVILLE RESERVOIR REACH
12.14
TRIBUTARIES IN 2040 AND 2080
12.15
CLIMATE SCENARIOS IN THE COLUMBIA RIVER BASIN
201

-------
12.18
OF THE COLUMBIA AND SNAKE RIVERS
12.17	¦:¦//. ' " ' ' ' " '• " ' ' ¦ : '¦¦¦, ' . ¦ ¦¦ .¦ ' . ' ¦¦
COLUMBIA RIVER AND TRIBUTARIES IN 2040 AND 2080 BASED ON
THE NORWEST MODEL
12.18	" ' '• '	- - ' - ' " ¦ " ' ¦ ¦ :¦ • ' ¦ "
MODEL IN 12 PRIMARY COLD WATER TRIBUTARIES AND 2
"RESTORE" TRIBUTARIES
12.19
A CONTINUATION OF HISTORICAL WARMING TRENDS IN THE LOWER
COLUMBIA RIVER
12.20
12.21
RESULTS
12.22	:	: ¦¦ .¦ ¦ ¦ - 'H ¦ ¦ ^ ¦ ¦¦ : ¦¦ ¦ ¦ ¦ , ^¦ . ¦ ,
MONITORING DATA IN THE TWELVE PRIMARY CWR
12.23
CONDITIONS AND AVAILABLE FLOW DATA COLLECTED AT THE
PRIMARY COLD WATER REFUGIA (CWR) STREAMS
202

-------
Page intentionally left blank for double-sided printing

-------
SEPA
U.S. EPA
Region 10
1200 Sixth Avenue, Suite 155
Seattle, Washington 98101

-------