HomeMy WebLinkAboutMiddle Hoh Resiliency Plan December 2021
Photo: Raena Anderson, 10,000 Years Institute
Plan Information:
A product of Jefferson County through an agreement with Natural Systems Design, Cramer Fish Sciences and in collaboration with Hoh Tribe Natural Resources, Trout Unlimited, 10,000 Years
Institute and more than 50 resource agency representatives and valley landowners.
Authors include: Mike Ericsson, Tim Abbe, Shelby Burgess, Phil Roni, Tami Pokorny, Luke Kelly, Jill Silver, Kevin Fetherston and Paul Pittman
Project manager: Tami Pokorny, Jefferson County Public Health
Funded by the Washington Coast Restoration and Resiliency Initiative (WCRRI)
RCO #18-2005
JCPH WQ-20-195
JCPH WQ-19-177
Special thanks to the Hoh Tribe and participants in the Middle Hoh River Resiliency Steering and Leadership Committees.
December 2021
TABLE OF CONTENTS
Introduction 1
Problem Statement 1
Plan Assumptions and Context 2
Plan Goals and Objectives 3
Plan Overview 4
Plan Steering Committee 4
Resiliency 4
Existing Conditions 5
Landscape Setting 5
Geology & Geomorphology 5
Landslides 7
Sediment Sources 8
Channel Migration Zone 11
Riparian and Floodplain Native Forests 24
Methods & Study Area 24
Results & Discussion 27
Hydrology & Hydraulics 32
Hydrology and Boundary Conditions 33
Mesh Development and Roughness Categories 35
Modeled Infrastructure 35
Calibration 35
Results 36
Main Stem Aquatic Habitat 38
Methods 39
Results 43
Oxbow Canyon Reach 46
Willoughby Creek Reach 48
Morgan’s Crossing Reach 51
Spruce Canyon Reach 55
Huelsdonk-South Fork Reach 58
Future Fish Habitat Survey Needs 61
Anticipated Trends 61
Transportation 62
Trends & Anticipated Changes 65
Climate Change 65
Sediment Sources 66
Forests 67
Invasive Species Trends 69
Invasive Plants in the Hoh Watershed – Species, Sources, Impacts 73
Resiliency Corridor 82
Long-Term Desired Conditions (include overarching goals and specific objectives) 83
Intermediate-Term Desired Conditions 84
Short-Term Desired Conditions 84
Local Capacity toSupply Restoration Needs 91
Introduction 91
Recent Restoration – a Snapshot of Restoration Types and Funding Over Past Five Years. 93
Elements of Restoration Projects and Associated Capacity Needs 93
Survey Results from Restoration Practitioners 94
Inventory - Local Capacity to Supply Restoration Needs 96
Things to Consider 96
Discussion 97
Recommendations to Increase Local Restoration Capacity 100
Phase II Approach 100
References 103
Introduction 1
Problem Statement 1
Plan Assumptions and Context 2
Plan Goals and Objectives 3
Plan Overview 4
Plan Steering Committee 4
Resiliency 4
Existing Conditions 5
Landscape Setting 5
Geology & Geomorphology 5
Landslides 7
Sediment Sources 8
Channel Migration Zone 11
Riparian and Floodplain Native Forests 24
Methods & Study Area 24
Results & Discussion 27
Hydrology & Hydraulics 32
Hydrology and Boundary Conditions 33
Mesh Development and Roughness Categories 35
Modeled Infrastructure 35
Calibration 35
Results 36
Main Stem Aquatic Habitat 38
Methods 39
Results 43
Oxbow Canyon Reach 46
Willoughby Creek Reach 48
Morgan’s Crossing Reach 51
Spruce Canyon Reach 55
Huelsdonk-South Fork Reach 58
Future Fish Habitat Survey Needs 61
Anticipated Trends 61
Transportation 62
Trends & Anticipated Changes 65
Climate Change 65
Sediment Sources 66
Forests 67
Invasive Species Trends 69
Invasive Plants in the Hoh Watershed – Species, Sources, Impacts 73
Resiliency Corridor 82
Long-Term Desired Conditions (include overarching goals and specific objectives) 83
Intermediate-Term Desired Conditions 84
Short-Term Desired Conditions 84
Local Capacity toSupply Restoration Needs 91
Introduction 91
Recent Restoration – a Snapshot of Restoration Types and Funding Over Past Five Years. 93
Elements of Restoration Projects and Associated Capacity Needs 93
Survey Results from Restoration Practitioners 94
Inventory - Local Capacity to Supply Restoration Needs 96
Things to Consider 96
Discussion 97
Recommendations to Increase Local Restoration Capacity 100
Phase II Approach 100
References 103
LIST OF TABLES
Table 1. Inside the CMZ. Forest type and height class. 31
Table 2. Outside the CMZ. Forest classes and types. 32
Table 3. Estimated peak flows at each gage in the model domain. 33
Table 4. Modeled Discharge Values at Inflow Locations 34
Table 5. Calibrated Manning's n roughness values for each roughness category. 36
Table 6. Average depth by reach for the modeled peak floods. 36
Table 7. Average velocity by reach for the modeled peak floods. 36
Table 8. Pool frequency standards for functioning rivers developed by NMFS (1996). 41
Table 9. Identified road segments within the Middle Hoh CMZ, Resiliency Corridor and FEMA 100-yr floodplain. 63
Table 10. The magnitude of future peak flows in the Hoh River for 2070-2099, projected as result of the climate crisis under IPCC A1B scenario (broadly representing “business as usual”
through 2050, IPCC 2000). 66
Table 11. Objectives and strategies necessary to achieve the resiliency goal of restoring native vegetation. 71
Table 12. The Invasive plant species propagation, seed number, seed viability, and allelopathy (with sources) 72
Table 13. Middle Hoh River Resiliency Plan reaches relative to ISPC reaches 73
Table 14. The Risk and hazards: invasive plant spread via streamflow, roads, equipment, hillslope processes, and weather 73
Table 15. The Condensed ac of invasive plants inventoried and treated in 2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 76
Table 16. The Condensed ac of Scotch broom inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 77
Table 17. The Condensed ac of reed canarygrass inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 78
Table 18. The Condensed ac of knotweed inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 79
Table 19. The Condensed ac of herb Robert inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 80
Table 20. The Invasive plant survey and treatment timings based on plant life histories, treatment methods, and site and weather conditions 81
Table 1. Inside the CMZ. Forest type and height class. 31
Table 2. Outside the CMZ. Forest classes and types. 32
Table 3. Estimated peak flows at each gage in the model domain. 33
Table 4. Modeled Discharge Values at Inflow Locations 34
Table 5. Calibrated Manning's n roughness values for each roughness category. 36
Table 6. Average depth by reach for the modeled peak floods. 36
Table 7. Average velocity by reach for the modeled peak floods. 36
Table 8. Pool frequency standards for functioning rivers developed by NMFS (1996). 41
Table 9. Identified road segments within the Middle Hoh CMZ, Resiliency Corridor and FEMA 100-yr floodplain. 63
Table 10. The magnitude of future peak flows in the Hoh River for 2070-2099, projected as result of the climate crisis under IPCC A1B scenario (broadly representing “business as usual”
through 2050, IPCC 2000). 66
Table 11. Objectives and strategies necessary to achieve the resiliency goal of restoring native vegetation. 71
Table 12. The Invasive plant species propagation, seed number, seed viability, and allelopathy (with sources) 72
Table 13. Middle Hoh River Resiliency Plan reaches relative to ISPC reaches 73
Table 14. The Risk and hazards: invasive plant spread via streamflow, roads, equipment, hillslope processes, and weather 73
Table 15. The Condensed ac of invasive plants inventoried and treated in 2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 76
Table 16. The Condensed ac of Scotch broom inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 77
Table 17. The Condensed ac of reed canarygrass inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 78
Table 18. The Condensed ac of knotweed inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 79
Table 19. The Condensed ac of herb Robert inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches. 80
Table 20. The Invasive plant survey and treatment timings based on plant life histories, treatment methods, and site and weather conditions 81
LIST OF FIGURES
Figure 1. Exposure of Pleistocene glacial outwash (gravel) and glaciolacustrine (clay) deposits near Elk Creek on the right bank of the main stem channel at RM 19. 6
Figure 2. Ice River Glacier (2010) terminus with exposed pro-glacial and steep lateral moraine sediments (left) and Blue Glacier (2009) lateral moraine and unstable hillside contributing
to rockfall on glacier (right) Courtesy of NPS. 9
Figure 3a. Photo showing relatively stable snag in river where bank erosion recruited a large tree. The snag is deflecting flow away from the bank and helped to slow down bank erosion.
Flow is from right to left (Oct. 1, 2020, at RM 18.8). 14
Figure 3b. Photos illustrating importance of large trees in providing wood to the river that is capable of altering channel hydraulics and slowing bank erosion (top photo of stable 9-ft
diameter Sitka Spruce with 30-ft rootwad in the Queets River, T. Abbe). Bottom photo showing high terrace with young trees (industrial tree farm) where trees reaching the channel are
quickly transported downstream and, consequently, don’t contribute hydraulic roughness to the channel sufficient to slow erosion (Abbe and Brooks 2011). 15
Figure 3d. Example of where recruitment of large trees to the South Fork Hoh and channel response. Slides containing mature trees not only stopped bank erosion but also built a new
floodplain along the toe of the eroded bank. From 1990 to 2006 the river migrated 106 ft into an area of mature timber. Once in the channel the timber halted erosion and by 2013 the
active channel had moved back to the south (Abbe et al. 2016). 17
Figure 3e. Example of where erosion proceeded along the South Fork Hoh due to a lack of large trees. This site is located downstream of the previous figure. Between 1990 and 2006 the
river migrated 153-ft (9.6-ft/yr) into a clear-cut. Bank erosion triggered a landslide that extended 550 ft into the adjacent valley margin. The landslide headscarp retreated at a rate
of 24-ft/yr from 1990 to 2013 (Abbe et al. 2016). 18
Figure 3f. Old growth alluvial valley in Upper Hoh, about 3.6 miles upstream from ONP Hoh Visitor Center. Flow is from right to left. Note extensive cover of old growth across valley
bottom, large number of islands and channels, and patches of old-growth within the active channel migration zone. Small side channels extend through much of valley bottom that aren’t
visible through the forest canopy. Young deciduous forest that occupies most of the valley bottom in areas that were logged only accounts for a relatively small portion of the valley
bottom where there was no logging. The large trees reduce the rates of channel migration and form logjams that provide the foundations of the old-growth patches. Over millennia the
recruitment of large wood allows old-growth to colonize most of the valley bottom. 19
Figure 4. Comparison of Bureau of Reclamation 2004 mapping to updated mapping conditions and revised methodologies. Image at left is 2004 channel location mapping (colored ribbons) and
channel migration zone (black hatched zone). Image at right is updated channel mapping with historic migration zone (red), geomorphic migration zone (dark orange), erosion hazard area
(light orange) and geotechnical setback (yellow), all included in the updated CMZ limit (black boundary). Note that the current channel is outside of the 2004 delineated channel migration
zone. 23
Figure 5. First return 2014 LiDAR DEM of forest vegetation heights at South Fork confluence with main stem Hoh River. Olympic National Park is clearly delineated by the much higher
tree heights. 25
Figure 6. Middle Hoh River riparian forest type cover and height classes. 26
Figure 7. Hoh River riparian forest mosaic of cover types. October 1st, 2020, near RM 20.6 26
Figure 8. Forest typing and channel migration zone (CMZ). 27
Figure 9. Young deciduous (red alder and willow) and older mixed conifer deciduous (red alder and Sitka spruce) forest types. October 1, 2020, near RM 20.6. Note logjam proximity to
larger conifer patch. 29
Figure 10. Mature mixed (Black cottonwood, red alder, Sitka spruce) and conifer forest types (Sitka spruce, Douglas fir). Right bank (flow from right to left), October 1, 2020, near
RM 20.7. 30
Figure 11. Inside CMZ. Riparian Forest Type (Ac) and Height Class (Ft/Ac). 31
Figure 12. Outside CMZ. Riparian Forest Type (Ac) and Height Class (Ft/Ac). 32
Figure 13. Locations of model inflow and outflow locations illustrated over the model domain. 34
Figure 14. Overview of the Hoh River survey area with survey reaches of interest (Piety et al. 2004) and start and end of survey locations. 40
Figure 15. Instantaneous flow (cfs) data at USGS Gage 12041200 – Hoh River at Highway 101 near Forks, WA recorded and reported by USGS. Surveys were conducted between September 28th
and October 1st, 2020. 40
Figure 16. Example of jams delineated in the Morgan’s Crossing Reach (RM 21.9), shown at a 1:600 scale using aerial photography collected in March 2021 (NV5 GeoSpatial 2021). Blue polygons
depict wood in wetted channel (time of photo) and red polygons wood outside wetted channel but within ordinary high water (bankfull) channel. Flow is from right to left. 42
Figure 17. Instantaneous flow (cfs) data at USGS Gage 12041200 – Hoh River at Highway 101 near Forks, WA recorded and reported by USGS. Aerial imagery was collected on March 20th, 2021. 43
Figure 18. Overview of the Middle Hoh River Study Area with the results from the CFS habitat surveys conducted from September 28th to October 1st, 2020, at an average daily flow of 1,614
to 2,652 cfs at the Highway 101 bridge (USGS Gage 12041200). For more detailed mapping see Map 7. Braids and side channels were surveyed as time allowed and were not a full census.
Main stem diversion locations of un-surveyed wetted braids and side channels were identified during surveys and general channel locations were mapped using the aerial imagery in GIS. 46
Figure 19. Overview of the Oxbow Canyon Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the Highway
101 bridge (USGS Gage 12041200). No wetted braid or side-channel habitat was present in this reach. The mapped photo is shown in Figure 20. 47
Figure 20. An example of the bedrock banks present in Oxbow Canyon (location of image shown in Figure 19). The canyon confines the channel and prevents the formation of off-channel
habitat; therefore the reach would be primarily used as a migration corridor for salmon. 48
Figure 21. Overview of the Willoughby Creek Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the
Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as a representative subset of the habitat available as time allowed and were not a full census. Two side
channel networks and two braided areas, shown in gray, were documented as connected to the main channel at the time of the surveys but were not surveyed due to time constraints. Photos
are shown in Figure 22. 50
Figure 22. Reach photos with locations shown in Figure 21. 1) An example of a large stretch of eroding banks with immature industrial forest observed which demonstrates inputs of fine
sediment and clay and small wood into the river. 2) A LWJ formed pool in a side channel with intermittent connectivity. 3) An eroding bank recruiting small Douglas firs into the channel.
4) A location of rip rap where the Upper Hoh Road runs along the right bank and constrains the channel. The riprap provides no complexity, no cover, no hydraulic refugia and no potential
for wood recruitment. 51
Figure 23. Overview of the Morgan’s Crossing Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at
the Highway 101 bridge (USGS Gage 12041200). The Lindner side channel is the dashed line between photo points 4 and 1. The photo was taken in the winter (leaf off) so deciduous trees
stand out as light brown areas compared to green areas of conifer forest. The current valley bottom is dominated by small deciduous trees. Prior to logging of the valley the number
and size of large conifer patches or “islands” would have been much greater. Braids and side channels were surveyed as a representative subset of the habitat available as time allowed
and were not a full census. One side channel complex and three braided areas, shown in gray, were documented as connected to the main channel at the time of the surveys but were not
surveyed due to time constraints. Photos are shown in Figure 24. 54
Figure 24. Reach photos with locations shown in Figure 23. 1) A long pool with ample vegetation cover but lacking large wood jams in lower portion of Lindner Side Channel. 2) The remnants
of a landslide that has begun to be colonized by Red Alder. 3) Another landslide that appears to be actively depositing silt and clay into the channel but is protected by a large wood
jam. 4) A log jam that runs along the bank at the Lindner Side Channel inlet along the main stem (2020). 55
Figure 25. Overview of the Spruce Canyon Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the
Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as time allowed and were not a full census. Two side channel networks, shown in gray, were documented
as connected to the main channel at the time of the surveys but were not surveyed due to time constraints. The mapped photo is shown in Figure 26. 57
Figure 26. Example of a pool in Spruce Canyon with remnant pilings along the left bank. The canyon confines the channel and prevents the formation of off-channel habitat, and therefore
would primarily be used as a migration corridor for salmon. 58
Figure 27. Overview of the Huelsdonk-South Fork Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at
the Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as a representative subset of the habitat available as time allowed and were not a full census. Only
one side channel, shown in gray, was documented as connected to the main channel at the time of the surveys but was not surveyed due to time constraints. Photos are shown in Figure
28. 60
Figure 28. Reach photos with locations shown in Figure 27. 1) An example of a wide braided channel. 2) A side channel with alder banks and frequent channel spanning wood. 3) A side
channel that departs from the main channel with multiple root wads and vegetated banks with slow water refugia. 4) An example of a large log jam creating edge pool habitat in the main
channel. 61
Figure 29. Scotch Broom infestation in the Morgan’s West Reach. 74
Figure 30. Scotch Broom and forest type cover of the Morgan’s West Reach. 75
Figure 31. Summary of Scotch Broom acreage for reaches of the Middle Hoh valley. 77
Figure 32. Summary of Reed Canary grass for reaches of the Middle Hoh Valley. 78
Figure 33. Summary of Knotweed for reaches of the Middle Hoh Valley. 80
Figure 34. Summary of Herb Robert for reaches of the Middle Hoh Valley. 81
Figure 35. Summary of invasive plant treatment timing throughout the year. 82
Figure 36. Illustration of portion of Upper Quinault River Valley (RM 44.3-45.8). The channel traces show the HMZ and active channel migration zone. The area to the south shows a network
of stable side channels flowing through mature forest. Southward migration of the river has eroded important side channel habitat crucial to salmon and the lack of big timber is preventing
the habitat from being reformed. Most recent historical channel is 2002. Since then the river has migrated further to the south. Adapted from QIN (2006). 86
Figure 37. Conceptual geographic framework for restoring large wood cycle, floodplain forests and side channels. The layout shows protective measures (green squares, zone 1) of property
and infrastructure within CMZ, this protection would not be needed in undeveloped areas. The thin blue lines within zone 2-3 represent stable side channels in an area that natural would
be dominated by old-growth (see Figure 3f). The area of more active channel migration (zone 4) the density of ELJs diminishes. Taken from QIN 2006. 87
Figure 38. Implementation of several phases of restoration in the Upper Quinault River. At the site the river had moved several thousand feet to the south, ultimately destroying one
home and threatening the South Shore Road. The black symbols show constructed ELJs constructed between 2013 and 2017. The ELJs allow water flow in-between them but discourage the main
channel from occupying the area (analogous to zone 2-3 in Figure 37). The project has resulted in the main channel moving north while creating new side channel habitat within the treatment
area. The ELJs have also increased the number of new pools with complex cover and created new floodplain for reforestation. 88
Figure 39. Example of large-scale restoration project to restore large wood cycle and side channel habitat in the Cispus River (Lewis County, WA). Photos show before and after. 89
Figure 40. Same reach of the Cispus River, looking downstream on November 24, 2021, after a 100-yr recurrence peak flow. All 21 of the ELJs are intact and undamaged. The project increased
cumulative channel length over 4-fold and increased the number of pools over 10-fold. Photo by Eli Asher, Cowlitz Tribe. 90
Figure 41. Example of engineered logjams constructed in 2020 and 2021 in the Cispus River. The structures were subjected to a 100-yr flood event in November 2021, only weeks after construction.
All 21 of the structures were undamaged and most collected large volumes of wood. Flow is from right to left. Photo 11-24-21 by Eli Asher, Cowlitz Tribe. 91
Figure 3e. Example of where erosion proceeded along the South Fork Hoh due to a lack of large trees. This site is located downstream of the previous figure. Between 1990 and 2006 the
river migrated 153-ft (9.6-ft/yr) into a clear-cut. Bank erosion triggered a landslide that extended 550 ft into the adjacent valley margin. The landslide headscarp retreated at a rate
of 24-ft/yr from 1990 to 2013 (Abbe et al. 2016). 18
Figure 3f. Old growth alluvial valley in Upper Hoh, about 3.6 miles upstream from ONP Hoh Visitor Center. Flow is from right to left. Note extensive cover of old growth across valley
bottom, large number of islands and channels, and patches of old-growth within the active channel migration zone. Small side channels extend through much of valley bottom that aren’t
visible through the forest canopy. Young deciduous forest that occupies most of the valley bottom in areas that were logged only accounts for a relatively small portion of the valley
bottom where there was no logging. The large trees reduce the rates of channel migration and form logjams that provide the foundations of the old-growth patches. Over millennia the
recruitment of large wood allows old-growth to colonize most of the valley bottom. 19
Figure 4. Comparison of Bureau of Reclamation 2004 mapping to updated mapping conditions and revised methodologies. Image at left is 2004 channel location mapping (colored ribbons) and
channel migration zone (black hatched zone). Image at right is updated channel mapping with historic migration zone (red), geomorphic migration zone (dark orange), erosion hazard area
(light orange) and geotechnical setback (yellow), all included in the updated CMZ limit (black boundary). Note that the current channel is outside of the 2004 delineated channel migration
zone. 23
Figure 5. First return 2014 LiDAR DEM of forest vegetation heights at South Fork confluence with main stem Hoh River. Olympic National Park is clearly delineated by the much higher
tree heights. 25
Figure 6. Middle Hoh River riparian forest type cover and height classes. 26
Figure 7. Hoh River riparian forest mosaic of cover types. October 1st, 2020, near RM 20.6 26
Figure 8. Forest typing and channel migration zone (CMZ). 27
Figure 9. Young deciduous (red alder and willow) and older mixed conifer deciduous (red alder and Sitka spruce) forest types. October 1, 2020, near RM 20.6. Note logjam proximity to
larger conifer patch. 29
Figure 10. Mature mixed (Black cottonwood, red alder, Sitka spruce) and conifer forest types (Sitka spruce, Douglas fir). Right bank (flow from right to left), October 1, 2020, near
RM 20.7. 30
Figure 11. Inside CMZ. Riparian Forest Type (Ac) and Height Class (Ft/Ac). 31
Figure 12. Outside CMZ. Riparian Forest Type (Ac) and Height Class (Ft/Ac). 32
Figure 13. Locations of model inflow and outflow locations illustrated over the model domain. 34
Figure 14. Overview of the Hoh River survey area with survey reaches of interest (Piety et al. 2004) and start and end of survey locations. 40
Figure 15. Instantaneous flow (cfs) data at USGS Gage 12041200 – Hoh River at Highway 101 near Forks, WA recorded and reported by USGS. Surveys were conducted between September 28th
and October 1st, 2020. 40
Figure 16. Example of jams delineated in the Morgan’s Crossing Reach (RM 21.9), shown at a 1:600 scale using aerial photography collected in March 2021 (NV5 GeoSpatial 2021). Blue polygons
depict wood in wetted channel (time of photo) and red polygons wood outside wetted channel but within ordinary high water (bankfull) channel. Flow is from right to left. 42
Figure 17. Instantaneous flow (cfs) data at USGS Gage 12041200 – Hoh River at Highway 101 near Forks, WA recorded and reported by USGS. Aerial imagery was collected on March 20th, 2021. 43
Figure 18. Overview of the Middle Hoh River Study Area with the results from the CFS habitat surveys conducted from September 28th to October 1st, 2020, at an average daily flow of 1,614
to 2,652 cfs at the Highway 101 bridge (USGS Gage 12041200). For more detailed mapping see Map 7. Braids and side channels were surveyed as time allowed and were not a full census.
Main stem diversion locations of un-surveyed wetted braids and side channels were identified during surveys and general channel locations were mapped using the aerial imagery in GIS. 46
Figure 19. Overview of the Oxbow Canyon Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the Highway
101 bridge (USGS Gage 12041200). No wetted braid or side-channel habitat was present in this reach. The mapped photo is shown in Figure 20. 47
Figure 20. An example of the bedrock banks present in Oxbow Canyon (location of image shown in Figure 19). The canyon confines the channel and prevents the formation of off-channel
habitat; therefore the reach would be primarily used as a migration corridor for salmon. 48
Figure 21. Overview of the Willoughby Creek Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the
Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as a representative subset of the habitat available as time allowed and were not a full census. Two side
channel networks and two braided areas, shown in gray, were documented as connected to the main channel at the time of the surveys but were not surveyed due to time constraints. Photos
are shown in Figure 22. 50
Figure 22. Reach photos with locations shown in Figure 21. 1) An example of a large stretch of eroding banks with immature industrial forest observed which demonstrates inputs of fine
sediment and clay and small wood into the river. 2) A LWJ formed pool in a side channel with intermittent connectivity. 3) An eroding bank recruiting small Douglas firs into the channel.
4) A location of rip rap where the Upper Hoh Road runs along the right bank and constrains the channel. The riprap provides no complexity, no cover, no hydraulic refugia and no potential
for wood recruitment. 51
Figure 23. Overview of the Morgan’s Crossing Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at
the Highway 101 bridge (USGS Gage 12041200). The Lindner side channel is the dashed line between photo points 4 and 1. The photo was taken in the winter (leaf off) so deciduous trees
stand out as light brown areas compared to green areas of conifer forest. The current valley bottom is dominated by small deciduous trees. Prior to logging of the valley the number
and size of large conifer patches or “islands” would have been much greater. Braids and side channels were surveyed as a representative subset of the habitat available as time allowed
and were not a full census. One side channel complex and three braided areas, shown in gray, were documented as connected to the main channel at the time of the surveys but were not
surveyed due to time constraints. Photos are shown in Figure 24. 54
Figure 24. Reach photos with locations shown in Figure 23. 1) A long pool with ample vegetation cover but lacking large wood jams in lower portion of Lindner Side Channel. 2) The remnants
of a landslide that has begun to be colonized by Red Alder. 3) Another landslide that appears to be actively depositing silt and clay into the channel but is protected by a large wood
jam. 4) A log jam that runs along the bank at the Lindner Side Channel inlet along the main stem (2020). 55
Figure 25. Overview of the Spruce Canyon Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the
Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as time allowed and were not a full census. Two side channel networks, shown in gray, were documented
as connected to the main channel at the time of the surveys but were not surveyed due to time constraints. The mapped photo is shown in Figure 26. 57
Figure 26. Example of a pool in Spruce Canyon with remnant pilings along the left bank. The canyon confines the channel and prevents the formation of off-channel habitat, and therefore
would primarily be used as a migration corridor for salmon. 58
Figure 27. Overview of the Huelsdonk-South Fork Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at
the Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as a representative subset of the habitat available as time allowed and were not a full census. Only
one side channel, shown in gray, was documented as connected to the main channel at the time of the surveys but was not surveyed due to time constraints. Photos are shown in Figure
28. 60
Figure 28. Reach photos with locations shown in Figure 27. 1) An example of a wide braided channel. 2) A side channel with alder banks and frequent channel spanning wood. 3) A side
channel that departs from the main channel with multiple root wads and vegetated banks with slow water refugia. 4) An example of a large log jam creating edge pool habitat in the main
channel. 61
Figure 29. Scotch Broom infestation in the Morgan’s West Reach. 74
Figure 30. Scotch Broom and forest type cover of the Morgan’s West Reach. 75
Figure 31. Summary of Scotch Broom acreage for reaches of the Middle Hoh valley. 77
Figure 32. Summary of Reed Canary grass for reaches of the Middle Hoh Valley. 78
Figure 33. Summary of Knotweed for reaches of the Middle Hoh Valley. 80
Figure 34. Summary of Herb Robert for reaches of the Middle Hoh Valley. 81
Figure 35. Summary of invasive plant treatment timing throughout the year. 82
Figure 36. Illustration of portion of Upper Quinault River Valley (RM 44.3-45.8). The channel traces show the HMZ and active channel migration zone. The area to the south shows a network
of stable side channels flowing through mature forest. Southward migration of the river has eroded important side channel habitat crucial to salmon and the lack of big timber is preventing
the habitat from being reformed. Most recent historical channel is 2002. Since then the river has migrated further to the south. Adapted from QIN (2006). 86
Figure 37. Conceptual geographic framework for restoring large wood cycle, floodplain forests and side channels. The layout shows protective measures (green squares, zone 1) of property
and infrastructure within CMZ, this protection would not be needed in undeveloped areas. The thin blue lines within zone 2-3 represent stable side channels in an area that natural would
be dominated by old-growth (see Figure 3f). The area of more active channel migration (zone 4) the density of ELJs diminishes. Taken from QIN 2006. 87
Figure 38. Implementation of several phases of restoration in the Upper Quinault River. At the site the river had moved several thousand feet to the south, ultimately destroying one
home and threatening the South Shore Road. The black symbols show constructed ELJs constructed between 2013 and 2017. The ELJs allow water flow in-between them but discourage the main
channel from occupying the area (analogous to zone 2-3 in Figure 37). The project has resulted in the main channel moving north while creating new side channel habitat within the treatment
area. The ELJs have also increased the number of new pools with complex cover and created new floodplain for reforestation. 88
Figure 39. Example of large-scale restoration project to restore large wood cycle and side channel habitat in the Cispus River (Lewis County, WA). Photos show before and after. 89
Figure 1. Exposure of Pleistocene glacial outwash (gravel) and glaciolacustrine (clay) deposits near Elk Creek on the right bank of the main stem channel at RM 19. 6
Figure 2. Ice River Glacier (2010) terminus with exposed pro-glacial and steep lateral moraine sediments (left) and Blue Glacier (2009) lateral moraine and unstable hillside contributing
to rockfall on glacier (right) Courtesy of NPS. 9
Figure 3a. Photo showing relatively stable snag in river where bank erosion recruited a large tree. The snag is deflecting flow away from the bank and helped to slow down bank erosion.
Flow is from right to left (Oct. 1, 2020, at RM 18.8). 14
Figure 3b. Photos illustrating importance of large trees in providing wood to the river that is capable of altering channel hydraulics and slowing bank erosion (top photo of stable 9-ft
diameter Sitka Spruce with 30-ft rootwad in the Queets River, T. Abbe). Bottom photo showing high terrace with young trees (industrial tree farm) where trees reaching the channel are
quickly transported downstream and, consequently, don’t contribute hydraulic roughness to the channel sufficient to slow erosion (Abbe and Brooks 2011). 15
Figure 3d. Example of where recruitment of large trees to the South Fork Hoh and channel response. Slides containing mature trees not only stopped bank erosion but also built a new
floodplain along the toe of the eroded bank. From 1990 to 2006 the river migrated 106 ft into an area of mature timber. Once in the channel the timber halted erosion and by 2013 the
active channel had moved back to the south (Abbe et al. 2016). 17
Figure 3e. Example of where erosion proceeded along the South Fork Hoh due to a lack of large trees. This site is located downstream of the previous figure. Between 1990 and 2006 the
river migrated 153-ft (9.6-ft/yr) into a clear-cut. Bank erosion triggered a landslide that extended 550 ft into the adjacent valley margin. The landslide headscarp retreated at a rate
of 24-ft/yr from 1990 to 2013 (Abbe et al. 2016). 18
Figure 3f. Old growth alluvial valley in Upper Hoh, about 3.6 miles upstream from ONP Hoh Visitor Center. Flow is from right to left. Note extensive cover of old growth across valley
bottom, large number of islands and channels, and patches of old-growth within the active channel migration zone. Small side channels extend through much of valley bottom that aren’t
visible through the forest canopy. Young deciduous forest that occupies most of the valley bottom in areas that were logged only accounts for a relatively small portion of the valley
bottom where there was no logging. The large trees reduce the rates of channel migration and form logjams that provide the foundations of the old-growth patches. Over millennia the
recruitment of large wood allows old-growth to colonize most of the valley bottom. 19
Figure 4. Comparison of Bureau of Reclamation 2004 mapping to updated mapping conditions and revised methodologies. Image at left is 2004 channel location mapping (colored ribbons) and
channel migration zone (black hatched zone). Image at right is updated channel mapping with historic migration zone (red), geomorphic migration zone (dark orange), erosion hazard area
(light orange) and geotechnical setback (yellow), all included in the updated CMZ limit (black boundary). Note that the current channel is outside of the 2004 delineated channel migration
zone. 23
Figure 5. First return 2014 LiDAR DEM of forest vegetation heights at South Fork confluence with main stem Hoh River. Olympic National Park is clearly delineated by the much higher
tree heights. 25
Figure 6. Middle Hoh River riparian forest type cover and height classes. 26
Figure 7. Hoh River riparian forest mosaic of cover types. October 1st, 2020, near RM 20.6 26
Figure 8. Forest typing and channel migration zone (CMZ). 27
Figure 9. Young deciduous (red alder and willow) and older mixed conifer deciduous (red alder and Sitka spruce) forest types. October 1, 2020, near RM 20.6. Note logjam proximity to
larger conifer patch. 29
Figure 10. Mature mixed (Black cottonwood, red alder, Sitka spruce) and conifer forest types (Sitka spruce, Douglas fir). Right bank (flow from right to left), October 1, 2020, near
RM 20.7. 30
Figure 11. Inside CMZ. Riparian Forest Type (Ac) and Height Class (Ft/Ac). 31
Figure 12. Outside CMZ. Riparian Forest Type (Ac) and Height Class (Ft/Ac). 32
Figure 13. Locations of model inflow and outflow locations illustrated over the model domain. 34
Figure 14. Overview of the Hoh River survey area with survey reaches of interest (Piety et al. 2004) and start and end of survey locations. 40
Figure 15. Instantaneous flow (cfs) data at USGS Gage 12041200 – Hoh River at Highway 101 near Forks, WA recorded and reported by USGS. Surveys were conducted between September 28th
and October 1st, 2020. 40
Figure 16. Example of jams delineated in the Morgan’s Crossing Reach (RM 21.9), shown at a 1:600 scale using aerial photography collected in March 2021 (NV5 GeoSpatial 2021). Blue polygons
depict wood in wetted channel (time of photo) and red polygons wood outside wetted channel but within ordinary high water (bankfull) channel. Flow is from right to left. 42
Figure 17. Instantaneous flow (cfs) data at USGS Gage 12041200 – Hoh River at Highway 101 near Forks, WA recorded and reported by USGS. Aerial imagery was collected on March 20th, 2021. 43
Figure 18. Overview of the Middle Hoh River Study Area with the results from the CFS habitat surveys conducted from September 28th to October 1st, 2020, at an average daily flow of 1,614
to 2,652 cfs at the Highway 101 bridge (USGS Gage 12041200). For more detailed mapping see Map 7. Braids and side channels were surveyed as time allowed and were not a full census.
Main stem diversion locations of un-surveyed wetted braids and side channels were identified during surveys and general channel locations were mapped using the aerial imagery in GIS. 46
Figure 19. Overview of the Oxbow Canyon Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the Highway
101 bridge (USGS Gage 12041200). No wetted braid or side-channel habitat was present in this reach. The mapped photo is shown in Figure 20. 47
Figure 20. An example of the bedrock banks present in Oxbow Canyon (location of image shown in Figure 19). The canyon confines the channel and prevents the formation of off-channel
habitat; therefore the reach would be primarily used as a migration corridor for salmon. 48
Figure 21. Overview of the Willoughby Creek Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the
Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as a representative subset of the habitat available as time allowed and were not a full census. Two side
channel networks and two braided areas, shown in gray, were documented as connected to the main channel at the time of the surveys but were not surveyed due to time constraints. Photos
are shown in Figure 22. 50
Figure 22. Reach photos with locations shown in Figure 21. 1) An example of a large stretch of eroding banks with immature industrial forest observed which demonstrates inputs of fine
sediment and clay and small wood into the river. 2) A LWJ formed pool in a side channel with intermittent connectivity. 3) An eroding bank recruiting small Douglas firs into the channel.
4) A location of rip rap where the Upper Hoh Road runs along the right bank and constrains the channel. The riprap provides no complexity, no cover, no hydraulic refugia and no potential
for wood recruitment. 51
Figure 23. Overview of the Morgan’s Crossing Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at
the Highway 101 bridge (USGS Gage 12041200). The Lindner side channel is the dashed line between photo points 4 and 1. The photo was taken in the winter (leaf off) so deciduous trees
stand out as light brown areas compared to green areas of conifer forest. The current valley bottom is dominated by small deciduous trees. Prior to logging of the valley the number
and size of large conifer patches or “islands” would have been much greater. Braids and side channels were surveyed as a representative subset of the habitat available as time allowed
and were not a full census. One side channel complex and three braided areas, shown in gray, were documented as connected to the main channel at the time of the surveys but were not
surveyed due to time constraints. Photos are shown in Figure 24. 54
Figure 24. Reach photos with locations shown in Figure 23. 1) A long pool with ample vegetation cover but lacking large wood jams in lower portion of Lindner Side Channel. 2) The remnants
of a landslide that has begun to be colonized by Red Alder. 3) Another landslide that appears to be actively depositing silt and clay into the channel but is protected by a large wood
jam. 4) A log jam that runs along the bank at the Lindner Side Channel inlet along the main stem (2020). 55
Figure 25. Overview of the Spruce Canyon Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the
Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as time allowed and were not a full census. Two side channel networks, shown in gray, were documented
as connected to the main channel at the time of the surveys but were not surveyed due to time constraints. The mapped photo is shown in Figure 26. 57
Figure 26. Example of a pool in Spruce Canyon with remnant pilings along the left bank. The canyon confines the channel and prevents the formation of off-channel habitat, and therefore
would primarily be used as a migration corridor for salmon. 58
Figure 27. Overview of the Huelsdonk-South Fork Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at
the Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as a representative subset of the habitat available as time allowed and were not a full census. Only
one side channel, shown in gray, was documented as connected to the main channel at the time of the surveys but was not surveyed due to time constraints. Photos are shown in Figure
28. 60
Figure 28. Reach photos with locations shown in Figure 27. 1) An example of a wide braided channel. 2) A side channel with alder banks and frequent channel spanning wood. 3) A side
channel that departs from the main channel with multiple root wads and vegetated banks with slow water refugia. 4) An example of a large log jam creating edge pool habitat in the main
channel. 61
Figure 29. Scotch Broom infestation in the Morgan’s West Reach. 74
Figure 30. Scotch Broom and forest type cover of the Morgan’s West Reach. 75
Figure 31. Summary of Scotch Broom acreage for reaches of the Middle Hoh valley. 77
Figure 32. Summary of Reed Canary grass for reaches of the Middle Hoh Valley. 78
Figure 33. Summary of Knotweed for reaches of the Middle Hoh Valley. 80
Figure 34. Summary of Herb Robert for reaches of the Middle Hoh Valley. 81
Figure 35. Summary of invasive plant treatment timing throughout the year. 82
Figure 36. Illustration of portion of Upper Quinault River Valley (RM 44.3-45.8). The channel traces show the HMZ and active channel migration zone. The area to the south shows a network
of stable side channels flowing through mature forest. Southward migration of the river has eroded important side channel habitat crucial to salmon and the lack of big timber is preventing
the habitat from being reformed. Most recent historical channel is 2002. Since then the river has migrated further to the south. Adapted from QIN (2006). 86
Figure 37. Conceptual geographic framework for restoring large wood cycle, floodplain forests and side channels. The layout shows protective measures (green squares, zone 1) of property
and infrastructure within CMZ, this protection would not be needed in undeveloped areas. The thin blue lines within zone 2-3 represent stable side channels in an area that natural would
be dominated by old-growth (see Figure 3f). The area of more active channel migration (zone 4) the density of ELJs diminishes. Taken from QIN 2006. 87
Figure 38. Implementation of several phases of restoration in the Upper Quinault River. At the site the river had moved several thousand feet to the south, ultimately destroying one
home and threatening the South Shore Road. The black symbols show constructed ELJs constructed between 2013 and 2017. The ELJs allow water flow in-between them but discourage the main
channel from occupying the area (analogous to zone 2-3 in Figure 37). The project has resulted in the main channel moving north while creating new side channel habitat within the treatment
area. The ELJs have also increased the number of new pools with complex cover and created new floodplain for reforestation. 88
Figure 39. Example of large-scale restoration project to restore large wood cycle and side channel habitat in the Cispus River (Lewis County, WA). Photos show before and after. 89
Figure 40. Same reach of the Cispus River, looking downstream on November 24, 2021, after a 100-yr recurrence peak flow. All 21 of the ELJs are intact and undamaged. The project increased
cumulative channel length over 4-fold and increased the number of pools over 10-fold. Photo by Eli Asher, Cowlitz Tribe. 90
Figure 41. Example of engineered logjams constructed in 2020 and 2021 in the Cispus River. The structures were subjected to a 100-yr flood event in November 2021, only weeks after construction.
All 21 of the structures were undamaged and most collected large volumes of wood. Flow is from right to left. Photo 11-24-21 by Eli Asher, Cowlitz Tribe. 91
LIST OF MAPS
Map 1 Project Reach
Map 2 Geologic Map
Map 3 Landslide and Slope Stability Map Book
Map 4 Channel Migration Zone Map Book
Map 5 Riparian Forest Map Book
Map 6 Existing Conditions Hydraulic Model Results
Map 7 Aquatic Habitat Map Book
Map 8 Invasive Species Map Book
Map 9 Resiliency Corridor Map Book
LIST OF APPENDICES
Appendix A Aquatic Habitat Field Data
Appendix B Middle Hoh River Action Plan
Appendix C Local Capacity
Appendix D 2021 List of Middle Hoh Steering Committee email list
INTRODUCTION
Problem Statement
River valleys such as the Hoh are the most ecologically rich and complex areas within a watershed. River valleys also are attractive areas for people, offering productive soils for agriculture,
level ground for homes and roads and beautiful vistas. But these attributes come with substantial risks associated with natural river processes: flooding and erosion. Flooding, and
erosion are crucial in forming and sustaining habitat. Building resiliency for human and natural ecological communities helps to ensure they can be sustained over time, something that
has increasing urgency given the impacts of the warming climate. Resiliency for human communities is best achieved by minimizing exposure to risks by focusing development and infrastructure
in areas outside flood and erosion hazard areas. This basic strategy is consistent with providing ecologic communities with the space they need within the productive habitat within
those hazard areas. Where defensive measures to protect infrastructure and development are unavoidable, they should be done to create and sustain the natural habitat of the river.
Ecological resiliency also requires restoring natural conditions impacted by historic development such as the conditions associated old-growth valley forest that once occupied much
of the Middle Hoh River valley. The intent of a resiliency plan is to provide a description of the Middle Hoh and its watershed, layout recommendations for improving resiliency for
the river’s ecosystem and human communities and provide a structure for future communications and decision making. The plan is a living document that is intended to be updated as needed.
In the process of developing this initial edition the project has outreach and engaged a leadership group that have greatly improved communications and transparency in river management.
There are no formal programs for comprehensive management of river valleys. Management typically reflects the diversity of landownership and infrastructure found in a valley; thus, actions
of different entities can frequently conflict with one another. The lack of communication between responsible entities can lead to unintended impacts to ecosystems and human communities.
The Middle Hoh Resiliency Plan is intended to establish a unifying structure to better manage this very important river. In developing the initial edition of the resiliency plan Jefferson
County reached out to a broad community of people with ties to the Middle Hoh valley, forming the Middle Hoh Steering Committee which has been engaged throughout the process. Regular
meetings of the Steering Committee have improved communications regarding resource management and infrastructure with the Hoh Valley. The committee includes representatives from the
Hoh Tribe, Jefferson County, Washington State Departments of Natural Resources (WDNR) and Fish & Wildlife (WDFW), The Nature Conservancy (TNC), Trout Unlimited (TU), the 10,000 Years
Institute (10K), the Coast Salmon Partnership, the Wild Salmon Center, local businesses, river guides, private landowners and federal agencies.
The Hoh Tribe, Chalá·at: People of the Hoh River, lived sustainably within the watershed for millennia, long before Europeans first came to the Olympic Peninsula. On July 1, 1855 the
Hoh Tribe and the United States signed the Quinault Treaty, guaranteeing the Tribe’s Rights to the natural resources of their usual and accustomed lands. Hoh Tribal members continue
to depend economically, culturally, and spiritually upon natural resources. The first European settlers, John and Cornelius Huelsdonk, came to the Hoh valley in 1892. The Hoh River
watershed supports a diverse ecosystem that ranges from alpine glaciers to temperate rainforests and is home to all five species of Pacific salmon (Chinook, Coho, Sockeye, Chum, Pink)
along with Steelhead, Bull Trout and Cutthroat Trout. The native peoples’ deep respect for the river and its natural rhythm of flooding and shifting channel banks was key to sustaining
a balanced human and wildlife ecosystem. Over the last century, the watershed has undergone changes at a scale not seen since the last glaciation nor witnessed by its original inhabitants,
including local and regional flow and temperature changes occurring due to a warming climate, accelerated retreat of major glaciers altering sediment and flow regimes, and extensive
road building and deforestation beginning in the late 1800’s. Human changes in the basin include the loss of primary and
secondary floodplain forest, including most of the largest trees, road building and repairs, installation of long rock revetments and the establishment and spread of numerous invasive
plant species. These actions and activities have altered the natural processes that create and sustain aquatic and riparian habitat, contributing to plunging salmon populations. In
particular the clearing and high-grading of large trees from the riparian corridor and channel banks removed that key structural element responsible for maintaining and sustaining the
rich habitat diversity of the Hoh. With the large trees gone the main stem channel is now free to migrate across the valley bottom unimpeded, frequently recycling floodplains and making
it impossible for a mature riparian corridor to re-establish for generations, if ever. The accelerated bank erosion rates have also led to property loss, elevated flood and erosion
risk and costly, repetitive road repairs and associated invasive plant introductions.
The primary public access in the watershed is the Upper Hoh Road, located parallel to and north of the river. Washouts threaten access for local residents as well as for visitors to
Olympic National Park (ONP); thus, directly impacting the economy of this important local and scenic valley as well as the broader region. Road repairs are expensive and directly destroy
salmon habitat; channel margins are some of the most important zones for juvenile and adult salmon where they expect to find a submerged forest to hide within and feed on a complex
food web. The chronic threat of erosion puts the valley’s homes, roads and other human improvements at risk - and will become more severe in the coming decades. The realization that
the river community is not currently as resilient as it can be prompted Jefferson County to develop a resiliency plan for the Middle Hoh River, extending from the ONP boundary to Oxbow
Canyon (Map 1). This plan will better identify flood and erosion risks to residents, infrastructure and habitat and detail appropriate and necessary measures that can be taken to reduce
these risks, while allowing the Hoh River and its floodplain space to support healthy, self-sustaining salmon and wildlife populations. A multi-disciplinary team of locals, scientists,
conservation groups, county officials and tribal representatives have developed this plan working together with the Hoh Tribe, greater community, state and federal agencies and the
recreational fishing community. The Plan’s goal is to bring forward a collective voice for the river and floodplain with the aim to develop, prioritize and implement actions that are
mutually beneficial to the community and wildlife. This Resiliency Plan will also open additional opportunities for funding as well as reduce regulatory costs and timelines that have
impacted past and current projects.
The Plan establishes an initial template that is intended to be periodically appended as new information is made available. This first edition utilized the best available science at
the time and solicited the input of numerous individuals with intimate knowledge of the Hoh River. The Plan and the leadership team will hopefully be sustained permanently, like commitments
to maintain infrastructure. Most importantly the Plan is intended to bring the community and numerous entities responsible for managing resources in the Middle Hoh River valley together
under one tent.
Plan Assumptions and Context
This Plan assumes the following:
The Middle Hoh Resiliency Plan is “living” document intended to enhance management of the Middle Hoh River valley to the benefit of people and natural resources.
People, governments, non-governmental organizations (NGOs) who live, and work in the Middle Hoh River valley will continue to contribute their time to help ensure a healthy future for
the river and its fish and wildlife. To this end, the Middle Hoh Steering Committee will continue to meet, and the Resiliency Plan will be updated and revised to best manage the river
Infrastructure improvements and repairs by Jefferson County, Olympic National Park and the Federal Highway Administration will be communicated to the Middle Hoh Steering Committee during
initial planning through implementation and be consistent with the Resiliency Plan.
Analyses performed as part of this plan development relied significantly on data five or more years old, such as the 2013/2014 LiDAR topography used for hydraulic modeling and forest
characterization,
Implementation of the accompanying Action Plan will advance the goals of the Resiliency Plan. Each action will only proceed with support of the Hoh Tribe, local residents, the salmon
restoration community and governmental agencies.
Tributaries and fish passage remain important to the successful recovery of salmon stocks in the Hoh River and should be included and considered for measures to improve habitat conditions,
in tandem with efforts in this document focused on the main stem channel and floodplains,
Restoration actions used successfully in similar rivers can be successfully implemented in the Middle Hoh River,
Sitka spruce and other native conifers will maintain their rate of growth on the floodplain,
Control of invasive species in the valley will be continued to ensure the regrowth and resiliency of mature forested floodplains,
Climate change projects are taken from the Intergovernmental Panel on Climate Change (IPCC) Assessment Report (www.ipcc.ch). Regional climate will change over time largely consistent
with University of Washington Climate Impact Group (https://cig.uw.edu) projections,
Rules and regulations governing timber harvest remain in place,
Wildfires will continue to be rare and generally under 1,000-acres (ac),
This planning process began in earnest in early 2020 through monthly in-person meetings which, as a result of the COVID-19 pandemic, were held online beginning in March of that year
and continuing through 2021.
The Resiliency Plan established a structure that has brought together the people who live, work and manage resources within the river valley and provide the following:
an understanding of the processes and conditions that define the river’s ecology and impact the human community,
a source of information and experts (e.g., natural resources, hazards, regulations, infrastructure),
a network for communication regarding management actions,
a safe structure for debate that is respectful of different perspectives and encourages transparency in decision-making that impacts natural resources and people,
works to develop a shared vision for one of the most unique and intact large river ecosystems in the contiguous United States,
provides guidance for protecting and restoring critical salmon habitat as well as supporting the prosperity of the local communities. This plan provides an update to previous work (Piety
er al 2004) on the changes to the geomorphology and hydraulics of this large river and possible human responses, based on the experience of the authors across multiple Washington river
systems, including the Hoh River, over decades.
Plan Goals and Objectives
The primary goal of this plan is to produce a collective voice for the river and floodplain based on a shared scientific foundation. Its objectives are to develop, prioritize and implement
prioritized actions that are mutually beneficial to the community and wildlife as articulated in Appendix B- Middle Hoh River Action Plan. The Action Plan sets out a detailed strategy
for improving community and ecosystem resiliency in the Middle Hoh River
over the next 80+ years (yrs). Due to the size and characteristics of the river itself, the scale and intensity of human alteration, as well as the need to act quickly to restore salmon
habitat and adapt to climate change impacts, the plans necessarily describe an assertive path to restoration. It has been collaboratively developed with the local community, property
owners and agencies responsible for land management and infrastructure.
Plan Overview
This plan is separated into six broad topics covering existing conditions in the project reach, trends and anticipated responses, desired conditions, resiliency opportunities, local
capacity to perform work, and the Action Plan. Each topic is presented as follows: Current Conditions; Anticipated Changes; Desired Conditions; Next Steps to Achieve Desired Conditions;
Capacity to Hire Locally; and Develop Overarching Strategies. Descriptions of existing conditions are provided to establish new “baseline” conditions in the project reach in order to
identify opportunities to improve resiliency. System trends describe how conditions have evolved in more recent history and anticipates future system responses over 80-yrs and accounting
for climate change (IPCC 2021) under a no-action scenario (i.e., no proactive measures to improve system resiliency).
Plan Steering Committee
A project steering committee was convened monthly during the development of this plan to provide a forum for collaboration with the Hoh Tribe, various government agencies, non-government
organizations, local landowners and non-local river users. This diverse group provided information and data resources, coordinated outreach to the community and contributed meaningful
and critical feedback on assessments completed, proposed actions, and interim drafts of this plan. Governments, agencies and organizations represented in the steering committee included:
Hoh Tribe
Jefferson County
10,000 Years Institute
Coast Salmon Partnership
Hoh River Trust
Olympic National Park
Olympic Natural Resources Center
North Pacific Salmon Lead Entity
Pacific Coast Salmon Coalition
The Nature Conservancy
Trout Unlimited
Olympic National Forest
Washington Department of Fish and Wildlife (WDFW)
Washington Department of Natural Resources (DNR)
Wild Salmon Center
Resiliency
In the context of out watersheds, resiliency can be defined as the capacity of an ecosystem or community to accommodate and recover from disturbance and environmental change without
loss of overall function. A flood resilient community is one that can recover from major floods quickly with little damage or harm to the community; achievable by focusing on providing
adequate space for natural flooding processes to occur by limiting exposure of vulnerable assets (homes, roads, utilities, out-buildings, etc.). Likewise, a resilient ecosystem is able
to maintain diverse and productive wildlife habitat despite being subjected to disturbances such as fire, landslides and climate induced increases in air, stream and soil temperature,
and the frequency, duration and magnitude of peak flows, and lower summertime flows.
This Resiliency Plan is intended to provide a scientific foundation and a blueprint for necessary measures and actions that can be made to maximize the resiliency of the community and
ecosystem of the Middle Hoh River valley. When resiliency is lacking, the environment and community suffer significant damage and recover slowly from change, whether coming quickly
like a flood or more gradually like extended drought, invasive species
spread and river aggradation. Greater damage requires more money to fix, oftentimes repeatedly, and can impact vital services. In an era where we are looking ahead to projected increases
in the frequency of disturbances (floods, landslides, etc.) as a result of the current climate crisis, future generations of citizens and wildlife of the Middle Hoh valley will benefit
from a science-based plan for resiliency.
EXISTING CONDITIONS
Landscape Setting
The Hoh River is located on the West Coast of Washington State, originates from glaciers high on Mount Olympus and descends 7,980-feet (ft) over 57-miles (mi) before entering the Pacific
Ocean. The watershed encompasses 298 mi2, the upper 57% of the basin lies within Olympic National Park (ONP) and is considered to be pristine temperate rainforest, home to some of the
largest trees in the United States (McMillan and Starr 2008; NPCLE 2020). Downstream of ONP the Hoh basin is primarily a mix of public open space, state and private forestland, private
residences and local businesses, with forestlands the dominant land cover type (93%) on the valley bottom outside of the active channel. The full-time population is around 20 residents,
with approximately 80,000 vehicles entering ONP via the Upper Hoh Road (FHWA 2021). The watershed receives an average annual precipitation of 155-inches (in) falling as rain and snow
depending on time of year and elevation, with a large precipitation gradient from 240-in above the glaciers near the summit of Mt. Olympus at 7,980-ft, to 93-in at the coast (Lieb and
Perry 2005). The cobble-gravel bed river flows generally east to west gaining flow from several major non-glacial tributaries draining primarily industrial timber lands. River flows
fluctuate throughout the year with the highest monthly averages occurring in November and minimum averages in September, with a sustained freshet typically extending into June (Lieb
and Perry 2005).
The Middle Hoh project reach extends from the ONP boundary to Highway 101, covering 15-river miles (RM) with an average slope of 0.3% (Map 1). The upstream end of the reach begins as
the river transitions from the confined steep upper watershed draining ONP to a broad valley with wide floodplains and lower slope. The valley is approximately 1800-ft wide on average,
with Pleistocene glacial terraces along the margins composed of outwash, drift and glaciolacustrine deposits. The 15-mi project reach is divided into five contiguous distinct geomorphic
reaches, adopted from previous work completed by the Bureau of Reclamation (Piety et. al. 2004).
Geology & Geomorphology
The geologic and glacial history of the Olympic Mountains are primary drivers influencing conditions in the Hoh River basin. The Olympic Mountain Range was and continues to be uplifted
resulting from the Juan de Fuca plate subducting under the North American plate, creating a mélange of oceanic sedimentary rocks accreted onto the North American plate at the convergent
boundary (Pazzaglia et al 2003). These rocks are uplifted, folded, fractured and experience low-grade metamorphism, weakening the rocks and making them highly erodible. These rocks
underlie the watershed, and are exposed primarily at the eastern, highest portion of the watershed in steep mountainous terrain (Map 2). This bedrock is exposed in a few locations within
the study reach on the valley floor, but the predominant geologic units are unconsolidated/poorly-consolidated glacial deposits and post-glacial alluvium/colluvium. Where bedrock is
present, it has significant influence on both lateral channel migration and vertical gradient control. Bedrock controls are present in two locations within the study reach: Spruce Canyon
(RM 26) and Oxbow Canyon (RM 17).
During the Pleistocene from 54,000 – 18,000-yr B.P. alpine glaciers advanced and retreated up and down the Hoh valley at least four times (Thackray 2001). Glacial processes both widened
the valley, scoured the valley, and then deposited a complex set of overlapping glacial sediments within the valley as the glaciers receded. Glacial deposits in the valley include moraines,
tills, alluvial outwash, lacustrine, sub-glacial, ice-marginal, and
ablation drifts. Deposits are present as low to moderately high banks along the margins of the active geomorphic migration zone. Of note is the presence of thick clay deposits in the
channel banks and valley margins in the vicinity of Elk Creek near RM 19 and Owl Creek near RM 27, just upstream of Oxbow and Spruce Canyons, respectively (Figure 1). Oxbow Canyon formed
as the Hoh cut through a mapped terminal moraine, formed at the end of an advancing alpine glacier. Once the glacier retreated a pro-glacial lake formed behind the moraine as meltwater
accumulated, depositing the clay layers observed in the channel banks today. Once the river breached the moraine it carved Oxbow Canyon and drained the lake and were subsequently buried/filled
with coarse outwash deposits. While not formally mapped, we interpret a similar course of events transpired to form Spruce Canyon, leaving behind the abundant and thick clay deposits
found upstream to Canyon Creek.
Figure 1. Exposure of Pleistocene glacial outwash (gravel) and glaciolacustrine (clay) deposits near Elk Creek on the right bank of the main stem channel at RM 19.
Unconsolidated alluvial sediments that range from sand and gravels to floodplain silts comprise more recent Holocene post-glacial deposits. Also included in this geologic group are alluvial
fan deposits, colluvium, and recent mass-wasting deposits. The active geomorphic migration zone is comprised primarily of post-glacial fluvial deposits forming readily erodible low
banks, and is the predominant channel bank material within the study reach.
The Middle Hoh River has a variety of channel morphologies ranging from confined (no floodplain) single thread boulder-bedrock channels (Oxbow and Spruce Canyons) to different unconfined
(floodplain) alluvial channels that include meandering channel reaches, wide braiding (large gravel bars) channel reaches, and multi-channel island dominated reaches. Prior to 1900,
geomorphic processes in the Middle Hoh were closely linked to the extensive floodplain and slope-side forests of mature conifers as they interacted with flow. Natural recruitment into
the channel from bank erosion, windthrow or other means forces the river to move or flow around the tree. If of sufficient size and aided with rootwad and branches, the tree remains
stable in the channel and will alter flow characteristics that initiate geomorphic processes. Flow can be deflected into an adjacent bank initiating erosion, flow can be split forming
an island between channel threads with the tree at the upstream end, flow can be forced downward as the channel impinges on the tree forcing a scour pool to form in the channel bed.
The presence of these trees, particularly those capable of remaining embedded in the channel, were the literal backbone of the system providing the long-term stability that fostered
the development of critical salmon habitat. While the geographic setting differs from the Middle Hoh, the Upper Hoh within ONP provides an excellent example of the importance of large
trees in driving the geomorphic processes, leading to a more stable and resilient system.
Early homesteaders began arriving in the late 1890’s began clearing land for fields and wood for homes and barns. Commercial timber operators began harvesting more large trees from the
riparian forest, and over time high-grading and more clearing continued. The loss of the large trees and widespread clearing ultimately led to a riparian forest largely devoid of the
large trees capable of remaining stable in the channel and responsible for driving the geomorphic processes that maintained the stable channel and supported healthy salmon runs. Once
the channel began to erode channel banks lacking large trees, there was nothing to impede further erosion of the floodplain and the active channel began to widen and freely migrate
across the valley bottom. This resultant highly dynamic system limits the ability for the floodplain forests to recover from erosion, before the channel is back to reclaim the land
and small trees beginning to take hold. The historic air photo record portrays a system gradually unraveling throughout the 20th and now 21st century. The channel through alluvial reaches
of the river has progressively become simplified from an anabranch planform with, multiple channel threads, to an irregular wandering channel type that is less predictable over time
than other forms because of characteristic higher erosion rates and a relatively high potential for avulsions (new channel courses that take completely new path through the floodplain).
The high natural sediment supply to the Middle Hoh further exacerbates channel migration and is expected to increase as the source glaciers recede due to the climate crisis.
Landslides
Landslides are common in the Hoh River watershed. The stability of slopes is controlled by the underlying geology, topography, vegetation, rainfall duration/intensity and natural and
human disturbances. Natural disturbances that alter forest cover can initiate landslides and include fire (historically rare in the Middle Hoh), frequent windstorms that alter forest
cover, and channel migration. Human disturbances such as road, drainage and forest clearing can destabilized slopes. Natural variability across the watershed creates the potential for
three types of mass wasting: rapid shallow landslides, deep-seated rotational landslides and debris flows. Rapid shallow landslides are translational (parallel) slope failures occurring
on a relatively planar surface, typically associated with a zone of weakness in the soil (e.g., limit of root depth) or soil contact with bedrock, and are the most common mass wasting
process in the watershed. These slides are common along ridgelines bordering the Middle Hoh and frequently transform into debris flows once the failed material reaches creek channels,
sometimes continuing for a considerable distance downstream.
Deep seated rotational landslides occur on a curved failure plane within the slope, with the displaced material rotating during failure, leaving a head-scarp at the up-gradient end and
rotated material at the toe. Deep seated slides are typically associated with a deep impermeable layer such as the glacial lacustrine clay deposits common in the Hoh Valley and can
occur slowly or rapidly. Surface exposures of these deep clay layers are commonly visible at the toe of hillslopes along the river such as at RM 26.8 on the left bank, and in many tributaries
to the Hoh. Fine sediment inputs to the Middle Hoh from these exposed clay layers appears limited, however, significant failure could result in catastrophic damage and deliver large
quantities of fine sediment to the river.
The third type are debris flows resembling wet concrete in consistency and can be very damaging where they build up mass and momentum. Debris flows happen when water pore pressure exceeds
the soil strength and gravitational forces trigger a landslide that develops into a debris flow, usually in steep headwater channels or hillslopes. If there is sufficient large wood
in the path of the debris flow, it can slow or halt the flow by limiting the flow from increasing in mass and momentum. Where flows encounter little resistance, they quickly gain
mass and momentum, and scour out channels all the way down to the Hoh valley where they deposit the load of mobilized sediment and wood. Numerous debris flows have occurred in the many
tributaries of the Middle Hoh, commonly resulting in channels scoured of large wood and sediment, including spawning gravels (McMillan and Starr 2008). Runout from debris flows can
travel to the confluence with the main stem channel or floodplain, typically stopping shortly after reaching the valley floor where the gradient diminishes rapidly.
The resiliency plan includes an inventory of landslides that incorporates previous work (Parks 1999, McHenry 2001), as well as identification and inclusion of new landslides using more
recent data. The primary source of previously identified landslides was from Parks (1999), and a hill shade generated from the 2013 LiDAR was the primary data used to identify new,
previously unidentified landslides. Three NSD geologists licensed in the state of Washington reviewed the LiDAR hill shade for surface forms indicating mass wasting had occurred, delineating
the initiation point and runout to the failed material. A total of 202 landslides encompassing 730-ac were delineated by Parks (1999), with an additional 19 landslides mapped by NSD
covering 123-ac (Map 3). Most of the rotational and translational failures occur along the contact between Quaternary alluvial deposits and unconsolidated Pleistocene glacial deposits
at the valley margins. NSD mapped several landslides in the vicinity of Spruce Canyon and upstream to the confluence with the South Fork Hoh River that initiated in fine grained glaciolacustrine
deposits not mapped previously (Map 2). These deposits are prone to deep-seated slope failure (McHenry 2001) and are understood to have been deposited in a proglacial lake impounded
by a terminal moraine that crossed the valley near Spruce Canyon. Meltwater from the receding glaciers would have backed up behind the moraine, depositing the thinly bedded clays and
silts.
In addition to mapping landslides that have occurred over time in the Hoh valley, the results of a predictive model of shallow rapid slope stability are included on the landslide inventory
maps (Map 3) (Montgomery and Dietrich 1994, Shaw and Johnson 1995). The model does not predict the potential for deep-seated landslides or debris flows. The model results from DNR show
classified slope stability based on slope gradient and curvature, landslide inventories, geology, precipitation, etc. into zones of high, medium and low probability for failure. The
models used were SMORPH (Shaw and Johnson 1995) and SHALSTAB (Montgomery and Dietrich 1994, the terrain surface used for the analysis was a 10-meter DEM with an accuracy of +/- 35-ft.
The results show almost all the steep high terrain forming the valley margins are in the medium to high probability zones, as well as the same boundary between Quaternary alluvial deposits
and Pleistocene glacial deposits where NSD mapped several landslides (Map 3).
These results highlight the inherent instability of the watershed, high potential for mass wasting primarily due to geologic and hydrologic forces greatly exacerbated by forest clearing
and road construction. Direct delivery of failed slope sediments to the channel can result in large sediment pulses propagating downstream, or remain chronic high-sediment load contributors
to the channel. Larger failures that could dramatically alter the course of the Hoh are possible, potentially resulting in miles of flooding and erosion up and downstream. Locations
that are known to be susceptible to failure, exhibit indications of impeding failure and/or are located above sensitive areas and populations, should be monitored and appropriate mitigating
measures taken.
Sediment Sources
The sediment supply in the upper watershed of the Hoh River is virtually limitless given the high relief, weak rock and abundant unconsolidated glacial deposits. This sediment is generated
from glacial processes, rainfall, slope failures, road disturbance and hillslope surface erosion. The glaciers descending Mount Olympus are the predominant source of sediment to the
Hoh River as they grind away at the valley floor and margins. As these glaciers recede over the next century, sediment once buttressed by ice against the valley walls will become exposed
on steep unvegetated slopes. These deposits, prone to shallow rapid failure, have the potential to deliver copious amounts of additional sediment to the already sediment-laden pro-glacial
Hoh River. Additionally, rockfall and rockslides deliver sediment to the river and have been increasing in frequency as the
ice buttressing the over-steepened valley walls retreats (Lyons, E. 2003) (Figure 2). All these glacier-derived sediments can be delivered to the river gradually over time or as a sediment
pulse or wedge that will propagate downstream, varying sediment delivery downstream over time. Recent studies have concluded that these processes are occurring at an accelerated rate
due to the changing climate (Riedel et. al. 2015 and East et. al. 2016), increasing the sediment supply downstream, and initiating channel widening and taking on a more braided form
(East et. al/ 2016). Where present, the native forests have played a significant role in limiting wide scale erosion, providing resiliency as sediment flux varies over time. Where absent,
resulting young forests provide little resistance to channel migration as the channel expands to accommodate sediment pulses.
//
Figure 2. Ice River Glacier (2010) terminus with exposed pro-glacial and steep lateral moraine sediments (left) and Blue Glacier (2009) lateral moraine and unstable hillside contributing
to rockfall on glacier (right) Courtesy of NPS.
Landslide susceptibility mapping by the DNR of the contributing basin shows that mass wasting potential in the watershed is high (Map 3). While upper watershed tributaries continuously
transport sediment to the river system, significant volumes of sediment delivered to the channel from mass wasting processes tend to occur episodically and are typically associated
with high intensity precipitation events.
Sediment Budget
Estimates of sediment production and transport were taken from published studies where available for each of the primary sources identified. Nelson (1986) estimated that 630,000 tons
of suspended sediment is transported by the Hoh River (measured at the mouth) annually, and that 60% of that load is from the portion of the watershed draining ONP. Because the Middle
Hoh reach begins at this same location, we have used 60% of the suspended sediment load reported by Nelson (1986) at the mouth, or 378,000-tons/yr, as an appropriate average annual
suspended sediment load entering the project reach. This value, or 810-tons/yr/km2 when scaled
to the drainage area, is in-line with other findings that coastal Pacific Northwest rivers have sediment loads between 100 – 500-tons/yr/km2 and rivers draining the glaciated Mt. Rainier
are on the order of 1,000- tons/yr/km2 (Czuba et. al. 2012). Scaling the suspended sediment load to the downstream end of the project reach at Highway 101 yields a load of 815-tons/yr/km2,
greater than the load per unit area within ONP. The difference in sediment load per unit area is due to a combination of human disturbances over time within the Middle Hoh reach.
Because only the suspended sediment load was measured (Nelson 1986), estimates for bedload are needed to determine a total sediment load (dissolved loads are not accounted for in this
assessment). The fraction of the total sediment load constituting bedload varies as a function of eroding surface characteristics, slope steepness, discharge magnitude and underlying
geology and typically fall within the range of 1 – 15% of the total load, but can be as high as 87% or greater (Babinski 2005). Overall, the percentage of the total load as bedload
increases away from the equator and for terrains sculpted by previous glaciations (Babinski 2005). Due to this uncertainty, estimates of bedload were calculated as a percentage of the
total load between 1 – 30%, or between 381,780 – 491,400-tons/yr at the ONP boundary.
Parks (1999) found that sediment loading within the Middle Hoh from hillslope fine sediment production to be 1,739-tons/yr and 2,147-tons/yr contributing from roads, Logan et. al. (1991)
estimated 94,350-tons/yr of coarse sediment loading from mass wasting and 16,405 tons/yr contributing from tributaries (Czuba et. al. 2012) within the Middle Hoh. To estimate sediment
production within ONP from these same sources we determined the tons/yr/mi2 of sediment produced in the Middle Hoh and scaled to an appropriate value for within ONP due to the difference
in land cover and use. We estimated hillslope fine sediment production within ONP would be 50% that outside of ONP, sediment contribution from roads to be 1% relative to outside ONP,
coarse sediment loading from mass wasting to be 90% and from tributaries to be 75% of that outside ONP per unit area. The resultant estimated sediment production within ONP from hillslope
fine sediment is 1,840-tons/yr, from roads is 45-tons/yr, 179,820-tons/yr from mass wasting and 26,055-tons/yr from tributary inputs.
/
Summing these sources yields 207,762-tons/yr of sediment production within ONP, however this does not account for the contribution of sediment generated from the many glaciers feeding
the Hoh River. Simply subtracting the total sediment production estimated within ONP (207,762-tons/yr) from the total sediment load exiting ONP (381,780 – 491,400-tons/yr depending
on 1 - 3% bedload) yields an estimated sediment production between 174,018 – 283,638-tons/yr from glaciers, accounting for 46 – 58% of the total sediment load entering the reach. At
the downstream end of the project reach the relative contribution of sediment from glaciers diminishes to between 40 – 54%. This approach, equating the sediment load transported by
the Hoh River to a sediment production rate, neglects any change in sediment storage within the project reach (i.e. bank erosion reducing sediment storage, aggradation increasing storage).
Based on these findings glacial processes are the dominant sediment producer within ONP. As the climate warms and the glaciers recede, sediment production should be expected to increase
in the near-term (decades) as sediments previously shielded and buttressed by ice are now exposed to erosive forces. As the glaciers recede and sediment production increases the average
total sediment load will increase in-kind, likely with a larger proportion of the total load as bedload. Once the available sediment has been mobilized and the glacier is reduced to
the extent that it is no longer producing new sediment due to lack the mass, sediment stored in the system will begin to evacuate, and will continue to do so
ultimately leading to widespread incision as the system adjusts. Further study into the timing of transitions from excess available sediment to lack thereof would further assist planning
efforts.
Channel Migration Zone
This chapter summarizes the methods and analyses utilized to develop a Channel Migration Zone (CMZ, Map 4) map for the Middle Hoh River. Over the last century the watershed and river
valley have been extensively manipulated to meet human needs and expectations of the population, war time needs (large spruce) and commercial industry. Local, and more expansive changes
have affected flow, channel banks, floodplains, vegetation, sediment, air and soil temperatures, precipitation patterns and land use. New landscape conditions, in turn, affect the physical
processes that create and sustain aquatic and riparian habitat essential for healthy salmon populations. New conditions also place human communities at risk with regards to flooding,
channel incision and bank erosion. This assessment is intended to provide the best available science to better define fluvial-related risks and opportunities to residents, infrastructure
and habitat.
Methods
The Washington State Department of Ecology Publication (#03-06-027) “A Framework for Delineating Channel Migration Zones” was the general guidance used to develop a new CMZ. The erosion
and avulsion hazards analysis “takes into account trends in channel movement, context of disturbance history and changes in boundary conditions, as well as topography, bank erodibility,
hydrology, sediment supply and woody debris loading” (Rapp et al, 2003).
The purpose of mapping the CMZ for the Middle Hoh was necessitated by both riverine hazard planning and ecosystem recovery planning. The intent of this study is to provide a technical
background document and set of maps to help guide decision makers in adopting a CMZ and developing comprehensive flood hazard and ecological planning. To the extent possible, CMZ mapping
for this analysis builds upon existing knowledge and information. This included previous mapping efforts and studies. In particular, the Geomorphic Assessment of Hoh River in Washington
State ((Piety et al 2004), LiDAR topography, geologic mapping, other geomorphic studies, air photo analysis, interviews, anecdotal accounts, and reconnaissance-level field investigation
provided the foundation of information for this analysis.
Definitions
The definitions used in this analysis build on the definitions in regulatory and guidance documents but have been modified to fit the conditions and processes specific to the Middle
Hoh River. The definitions used are:
Historic Migration Zone (HMZ): The HMZ is a composite of the historic locations of the river as determined from historic information and interpretation, over some length of historic
record.
Active Geomorphic Migration Zone (GMZ): The GMZ is “the geographic area where a stream or river has been and will be susceptible to channel erosion and/or channel occupation” (Rapp et
al, 2003); the GMZ considers a time period greater than the historic photo record.
Erosion Hazard Area (EHA): The EHA in this analysis refers to the hazards resulting from lateral migration of a channel through bank erosion that occurs from channel expansion, channel
meandering, channel course changes, or channel bank and fluvially related slope failures. The EHA is a predicted horizontal channel migration potential area. Lateral erosion is not
necessarily limited to the floodplain or areas inundated during the 100-yr flood event.
Avulsion Hazard Area (AHZ): The AHZ describes an area susceptible to multiple alluvial hazards associated with rapid channel course changes or temporary channelization of flow that in
addition to having flooding hazards, instantaneously becomes an erosion hazard.
Geotechnical Hazard Area (GHA): The GHA describes an area of potential slope instability driven by channel migration processes. For this study the fluvial-related geotechnical hazards
are shown as an overlay on the maps.
Alluvial Fan Hazard Area (AFH): The AFH describes an area where alluvial fan processes have occurred based on landform interpretation. Alluvial fan hazards include flooding, scour, erosion,
avulsion, deposition, and debris impacts. For this study, AFHs are shown as an overlay on the maps.
Channel Migration Zone (CMZ): The cumulative area of the HMZ, GMZ, EHA, AHA, and GHA is the CMZ. The State of Washington, in [WAC 173-26-221(2)(c)(iv)(3)(b)] describes this concept further
as: “The dynamic physical processes of rivers, including the movement of water, sediment and wood, cause the river channel in some areas to move laterally, or "migrate," over time.
This is a natural process in response to gravity and topography and allows the river to release energy and distribute its sediment load.”
A thorough review of previous data and literature was conducted as part of the development of this plan. The intent was to build extensively from the existing information. The Geomorphic
Assessment of Hoh River in Washington State (Piety et al 2004) was identified as a key document. From this review, data gaps and updates were identified, and additional analyses were
conducted to supplement and update the existing information. A synthesis of the key findings from this study is provided below.
The Bureau of Reclamation study was a detailed geomorphic analysis of the Middle Hoh River between Oxbow Canyon (RM 17) to Mount Tom Creek (RM 40). The purpose of the analysis was to
improve the understanding of channel processes and human impacts on the channel conditions. The analysis included channel migration mapping, hydrologic analysis, woody debris conditions,
sediment estimates, and management considerations. Key findings were:
The frequency and magnitude of floods has been increasing over the past 50-yrs
Channel migration and erosion rates increased with increased flood magnitude
Channel planform changes, particularly from avulsion, caused the highest localized erosion rates in the photo record (1939 – 2002)
Human impacts, particularly from loss of mature riparian woody vegetation and stable instream logjams, increased erosion rates in the study area
Sediment supply and transport appeared balanced based on geomorphic indicators.
Field and Desktop Observations
Channel migration is lateral movement of the channel and was observed to occur in two primary ways on the Middle Hoh River: 1) progressive lateral erosion of channel banks and valley
margins, and 2) relict channel or topographic capture and channel realignment (avulsions). Dynamic rates of channel migration from lateral erosion were observed in field and desktop
analysis. Evidence of regular avulsions, or rapid changes in channel location typically to a shorter flow path, was also apparent in both desktop and field observations. Mapping and
interpreting lateral erosion and avulsion potential included analysis of different physical conditions that tend to drive these processes. Primary drivers and conditions observed influencing
channel migration evaluated were:
valley geologic composition
valley vegetation
sediment deposition trends
channel forms.
Observed Valley Geologic Composition
Field observations of geologic composition revealed that post-glacial alluvial and glacial geology (low- moderate bank) groups are the predominant bank composition within the geomorphic
migration zone. A recent maximum annual lateral erosion rate of 75-ft/yr was measured in post-glacial alluvial group sediments (low bank, pasture) at one location in the reach. Anecdotally,
maximum annual erosion rates are reported to be double or triple that rate. The glacial group sediments were observed to have more variable erosion rates depending upon the geologic
composition which ranged from lacustrine silts/clays to coarse outwash. The lacustrine silts appear to have lower erosion rates than the outwash but had higher potential for slope instability
than the outwash deposits. The outwash deposits likely have erosion rates similar to the erosion rates of the post-glacial alluvial group deposits. The bedrock was metamorphosed marine
silt/sandstone and was relatively erosion resistant over the planning-horizon of this study. For the mapping analysis, we considered three geologic groups for erosion potential mapping:
Post-glacial (Holocene) Alluvial Group and Glacial) Group (Low-Moderate Bank Height)
Glacial Group (High Bank)
Bedrock Group
While short term (< 5-yrs) erosion rates can be close to 100-ft/yr, longer term rates are influenced by the duration of erosion in any one place. Given rapid avulsion and lateral migration
processes result in relatively short durations of channel occupation in a single place, we estimated lower long-term erosion potential rates.
Observed Valley Vegetation Groups
Mature Conifer Dominant Stands
Immature/Pioneering Trees and Shrubs
Herbaceous
Immature/pioneering trees and shrubs were observed to be the dominant riparian condition within the geomorphic migration zone. Where large, mature trees were observed in riparian areas,
erosion rates appeared lower, concurring with other research (Abbe et al. 2003, Abbe and Brooks 2011). Mature forest’s deeper, denser root mass creates increased soil cohesion and roughness
which reduce erodibility. We observed that where large trees were recruited to the channel, they were relatively stable and influenced channel migration processes (Figure 3a, 3b). In
contrast, immature/pioneering trees and shrub zones are recruited regularly to the channel and were observed to provide little erosion resistance and had shallow rooting that was often
undercut by lateral erosion (Figure 3b). Immature/pioneering trees and shrubs recruited to the channel appear to have transported downstream and did not directly influence channel migration
processes. Herbaceous vegetation offered the least erosion resistance and bank cohesion due to limited root depths. Erosion rates are expected to be higher in the herbaceous and immature/pioneering
trees and shrubs riparian areas (Figure 3c, Abbe and Brooks 2011). Abbe et al. 2016 presented two examples illustrating the influence of tree size on channel migration on the South
Fork Hoh, one showing how the recruitment of large trees retarded erosion and initiated new floodplain development (Figure 3d) and one showing that the lack of large trees failed to
limit erosion (Figure 3e).
Figure 3a. Photo showing relatively stable snag in river where bank erosion recruited a large tree. The snag is deflecting flow away from the bank and helped to slow down bank erosion.
Flow is from right to left (Oct. 1, 2020, at RM 18.8).
/
Figure 3b. Photos illustrating importance of large trees in providing wood to the river that is capable of altering channel hydraulics and slowing bank erosion (top photo of stable 9-ft
diameter Sitka Spruce with 30-ft rootwad in the Queets River, T. Abbe). Bottom photo showing high terrace with young trees (industrial tree farm) where trees reaching the channel are
quickly transported downstream and, consequently, don’t contribute hydraulic roughness to the channel sufficient to slow erosion (Abbe and Brooks 2011).
/
Figure 3c. Comparison of erosion rates in the Queets and Hoh Rivers as a function of tree size. Riparian areas with trees greater than 21-in (0.53m) erode at less than half the rate
of areas with trees less than 21-in (Abbe et al. 2003, Abbe and Brooks 2011).
/
Figure 3d. Example of where recruitment of large trees to the South Fork Hoh and channel response. Slides containing mature trees not only stopped bank erosion but also built a new
floodplain along the toe of the eroded bank. From 1990 to 2006 the river migrated 106 ft into an area of mature timber. Once in the channel the timber halted erosion and by 2013 the
active channel had moved back to the south (Abbe et al. 2016).
/
Figure 3e. Example of where erosion proceeded along the South Fork Hoh due to a lack of large trees. This site is located downstream of the previous figure. Between 1990 and 2006 the
river migrated 153-ft (9.6-ft/yr) into a clear-cut. Bank erosion triggered a landslide that extended 550 ft into the adjacent valley margin. The landslide headscarp retreated at a rate
of 24-ft/yr from 1990 to 2013 (Abbe et al. 2016).
/
Figure 3f. Old growth alluvial valley in Upper Hoh, about 3.6 miles upstream from ONP Hoh Visitor Center. Flow is from right to left. Note extensive cover of old growth across valley
bottom, large number of islands and channels, and patches of old-growth within the active channel migration zone. Small side channels extend through much of valley bottom that aren’t
visible through the forest canopy. Young deciduous forest that occupies most of the valley bottom in areas that were logged only accounts for a relatively small portion of the valley
bottom where there was no logging. The large trees reduce the rates of channel migration and form logjams that provide the foundations of the old-growth patches. Over millennia the
recruitment of large wood allows old-growth to colonize most of the valley bottom.
Hazards and Risks
Hazards: To understand the potential impacts related to fluvial process hazards in the project reach, it is necessary to define hazard and risk. Hazard is defined as the source of danger.
For this analysis, the channel migration associated hazards are:
Erosion and scour hazards
Flooding, debris impacts, and deposition hazards
Landslide/slope instability hazards
Risk: Risk is defined as the product of probability of a hazard and the consequences it will have. For example, even if the probability of erosion reaching a home is low but the consequences
are high (losing the house), the risk is high. Risk is typically applied to human development, but it can also be applied to ecological systems. For example, in an old-growth valley
there may be high probability of bank erosion, but the consequences are actually beneficial to habitat because of large wood recruitment. If the old-growth timber was removed from the
same valley the probability of erosion would increase dramatically and the consequences would have significant impact to habitat by creating wider more dynamic channels with large gravel
bars subject to invasive plant colonization. Thus removing big trees from a valley bottom poses very high risks to aquatic and riparian habitat. Building infrastructure such as roads
outside a channel migration zone lowers the probability of erosion to zero, demonstrating why long-term planning should strive to get roads out of CMZs. The probability of flood inundation
is dependent on river flow, topography and channel conditions. Flooding risk can thus be lowered by either moving out of flood prone areas or raising a structure. While flooding poses
risks to development it is a beneficial process to ecological communities. A chart illustrating different levels of relative risk based on probability of occurrence and resulting consequences
is presented below.
/
Mapping for this analysis presents areas in which erosion, scour, debris impacts, and scour related to fluvial process may occur within a 100-yr planning horizon, incorporating changes
to flow and sediment regimes resulting from predicted climate change (IPCC 2021). No distinction of consequences from hazards is provided in this analysis, therefore relative risk levels
have not been assigned. It is our opinion that specific risk assessments, targeted either geographically or toward a particular concern such as boater safety, would be best informed
by a more detailed analysis.
Mapping Results
The mapping assessment provides the following hazard information and is provided in Map 4:
Historic Migration Zone (HMZ) Mapping: The HMZ mapping in this analysis shows channel locations occurring between 1939 and 2018. The HMZ should not be confused with a channel migration
zone (CMZ). The HMZ is simply where channels have been during historic times whereas the CMZ delineates where channels may move in the future. Since HMZ is the area occupied by a river
in recent times it is always low-lying areas of the valley – a major attribute distinguishing it from a CMZ which can include high ground that can be eroded by the river, a common occurrence
in the Middle Hoh. As described earlier, areas with large trees erode at less than half the
rate as areas with small trees (<21 inches in diameter), thus timber harvest can accelerate channel migration, even into high terraces and valley margins. Current forest practices prohibit
timber harvest in CMZs. Mapping of channel locations conducted by the Bureau of Reclamation (2004) consisted of the visible wetted channel area. The 2017-18 channel mapping consisted
of the visible “high flow” channel (wetted channel plus unvegetated bar areas) and low flow wetted channel. GPS mapping of the October 2020 channel thalweg was also included. Not all
channel locations during this historic air photo interval were captured by the air photo record, so the true HMZ may extend beyond the mapped HMZ area. The HMZ is an area with erosion
and avulsion hazards that has a high probability of occurrence.
Active Geomorphic Migration Zone (GMZ) Mapping: The GMZ mapping was based on LiDAR interpretation, REM surface interpretation and FEMA flood mapping. The GMZ extends beyond the HMZ and
includes an area where channel activity has recently occurred and at certain flows or with changes in channel bed elevation or wood loading, may be reactivated and experience fluvial
processes. The GMZ is an area with erosion and avulsion hazards that has a moderate to high probability of occurrence.
100-Yr Erosion Hazard Area (EHA) Mapping: The EHA refers to an area in which erosion hazards may be realized within the 100-yr planning horizon. The consequences of erosion hazards being
realized within the designated EHA have not been determined for this analysis, and therefore risk has not been evaluated. Structures and infrastructure impacted by erosion hazards may
be well above river water surface elevations or a long distance from today’s current channel location. Included in the erosion hazard areas are landslides that are triggered by lateral
channel migration and expand up-gradient from the channel margin (see Geotechnical Hazard Area). Professional opinion of erosion rates was applied. This opinion considered bank height,
geologic composition (younger alluvium, older alluvium, glacial diamicts, or bedrock), vegetation type and age. Geology type was inferred by mapped geology, LiDAR, and field observations.
The most erodible banks are low height, younger alluvium, with grass vegetation. Short-term erosion rates in this bank type may be on the order of 100-ft/yr, perhaps higher. Erosion
rates over a 100-yr time frame must also consider the duration of time the river channel is located at a bank. In most cases on the Hoh, channel migration processes limit the duration
of time a channel is actively eroding a bank to a few years before the channel migrates to a new location. The frequency that a channel returns to a bank over the assessment timeframe
must also be considered. For this analysis, the 100-yr erosion rate for low banks with younger alluvium and immature or non-native vegetation was a minimum of 500-ft/yr. Bedrock banks
were assumed to have a 100-yr erosion rate of 0-ft/yr. The EHA is an area with erosion and avulsion hazards that has a lower probability of occurrence.
Avulsion Hazard Area (AHZ) Mapping: The AHZ describes an area within the HMZ and GMZ that is susceptible to rapid channel course changes or temporary channelization of flow that would
become an erosion hazard. While erosion hazards are often realized incrementally and avoidance or retreat is an option to manage risk, avulsion hazards may be realized abruptly and
potentially catastrophically, thereby potentially increasing consequences and decreasing viability of avoidance or retreat management strategies. Avulsions are common on the Hoh River,
thus exacerbating risk. Once a relict channel is reclaimed in an avulsion, rapid lateral erosion and channel widening of the new channel results. Over the course of the 100-yr planning
horizon of this delineation, the GMZ can be considered synonymous with the AHZ, as any of the relic channels defining the GMZ could pose an avulsion risk over time as the channel migrates.
Therefore, the AHZ is not specifically identified in the delineation.
Geotechnical Hazard Setback Area (GHA) Mapping: The GHA mapped area has potential slope instability driven by channel migration processes. For this study the fluvial-related geotechnical
hazards are considered if the 100-yr EHA contacts steep slopes. The GHA is shown as an overlay on the maps. Site specific geotechnical conditions should be considered for risk analyses
within the GHA. The GHA is an area with erosion and landslide hazards
that has a low to high probability of occurrence based on the location of the channel relative to the mapped GHA.
Alluvial Fan Hazard Areas (AFH) Mapping: The AFH defines an area where a sharp decrease in channel slope causes a wedge of sediment to deposit as stream energy is rapidly lost, creating
a characteristic fan shape in plan view. Alluvial fans are common at the margins of valley bottoms and are highly dynamic environments prone to rapid channel avulsions and aggradation.
The AFH has the potential to be an erosion hazard, aggradation hazard, flooding hazard and/or debris impact hazard or any combination thereof and may change over time as the fan evolves.
The AFH is an area with erosion, aggradation, flooding, and debris impact hazards that has a high probability of occurrence based on the high energy of the environment.
Channel Migration Zone (CMZ) Mapping: The cumulative area and hazards associated with the HMZ, GMZ, EHA and GHA areas define the extent of the CMZ.
Summary and Conclusion
The Middle Hoh is a dynamic landscape that has undergone historic disturbances that have altered and limited natural processes, and faces a future of additional changes that will further
alter natural processes. Historic clearing of the riparian forests altered the rate of channel migration in the reach, making the river more dynamic and less predictable. Future climate
changes will alter the flow regime, sediment supply and transport capacity of the river, with the potential to fundamentally change the character of the river. Given the uncertainty
in what changes will occur and when they will be realized we recommend planning conservatively, as the past is becoming less and less a predictor of what to is to come. An example
of this is show in Figure 4, where the river had migrated beyond the 2004 CMZ delineation in 2006, continuing to erode through 2018 expanding 100-ft past the boundary. As the river
continues to adjust to future changes in sediment supply, land use, vegetation cover, and the timing and magnitude of flooding, we can expect the river to occupy and flood areas that
have been safe since early homesteading. Planning ahead and adaptative management strategies are key to mitigating risks resulting from future channel migration and increased peak flows.
Figure 4. Comparison of Bureau of Reclamation 2004 mapping to updated mapping conditions and revised methodologies. Image at left is 2004 channel location mapping (colored ribbons) and
channel migration zone (black hatched zone). Image at right is updated channel mapping with historic migration zone (red), geomorphic migration zone (dark orange), erosion hazard area
(light orange) and geotechnical setback (yellow), all included in the updated CMZ limit (black boundary). Note that the current channel is outside of the 2004 delineated channel migration
zone.
Limitations
Limitations of this analysis include:
Mapping scale – temporal and spatial: The mapping efforts for this analysis were done on a reach scale. Site-specific investigation may be necessary. Site-specific investigations and
validation of mapping was conducted only sparingly because of a limitation of time and resources.
Temporal changes, disturbances, or recovering conditions, such as riparian development or wood jams, are variables that will impact channel response. Predicting the local response and
impacts over the time scale of the study is not considered feasible because of the unpredictable nature, time scale, and extent of change. For example, predicting the location of the
channel, size and duration of a wood jam, and riparian conditions at a site even in 5-yrs is felt to have a low degree of assurance. Analysis looking at a smaller timeframe context
and a more site-specific or sub-reach perspective should include such conditions and potential channel response.
New information and technology – new interpretations: Improvements in technology, new information and geologic interpretations may allow for a higher confidence in mapping or reinterpretation
of mapped hazards. Maps should be updated as new information and technology becomes available.
Hazards that exist but were not mapped for this assessment: The hazards mapped for this analysis were limited to lateral erosion and avulsion potential on the Hoh River. Other natural
hazards in the valley exist, including but not limited to flooding, landslides, tectonic deformation.
Riparian and Floodplain Native Forests
The Hoh River valley floodplains, terraces and adjacent hillslopes within Olympic National Park are the home of the extraordinary Olympic rain forest, including riparian forest four
to seven hundred years old (Fonda 1974; Kramer et al., 2020). The Middle Hoh River floodplain is predominantly second and third growth forest and recently established red alder. Remnant
patches of mature forest hundreds of years old comprise just 1,271-ac, approximately 24% of its former extent. The Olympic rain forest provides few limits to forest growth and old growth
trees reach diameters of 6 – 10-ft and up to 260 – 300-fts in height (Fonda 1974; Kramer et al., 2020). The structural size and complexity of the primordial Hoh River riparian forest
provides the river with in-channel large wood that the river hydraulically sorts into stable wood jams forming the structural skeleton of river channel splits, mid-channel islands and
patchwork floodplains (Montgomery and Abbe 2006).
The numerous gravel bars present in the Middle Hoh River floodplain are nursery sites for establishment of herbaceous and woody plants. Typically, young riparian forest consists of early
successional willows (Salix L.), red alder (Alnus rubra), and black cottonwood (Populus balsamifera ssp. trichocarpa). As the young pioneer forest matures (25 – 50-yrs), incipient soil
development allows conifers to also colonize the floodplain–primarily Sitka spruce (Picea sitchensis)–with minor elements of western hemlock (Tsuga heterophylla), Douglas fir (Pseudotsuga
menziesii) and western red cedar (Thuja plicata). Later colonizers include deciduous trees–vine maple (Acer circinatum) and big leaf maple (Acer macrophyllum). As the riparian forest
matures over 50 – 100-yrs, Sitka spruce and black cottonwood reach key member size diameters of 3-ft or more. Circa 100-yrs, and older, the riparian forest becomes the source pool for
recruitment of large trees to the river channel in a phenomenon called the floodplain large wood cycle (Collins et al. 2012). The floodplain large wood cycle is the natural process
that generates Hoh River channel complexity wherever mature conifer floodplain forests predominate; mid-channel islands, pools, salmon aquatic habitat and overall valley complexity
and habitat diversity.
An essential foundation of Hoh River’s diverse geomorphology and aquatic and riparian habitats is the preponderance of mature conifer forests, the occasional exceptionally large trees
and the ecological processes that generate these trees. The interaction of abundant precipitation, temperate climate, rapid tree growth, and geology produce the wide range of habitat
types that favor Hoh River salmon and steelhead productivity, and thriving populations of bear, cougar, osprey, eagle, and a myriad of other creatures.
The goal of the Middle Hoh River riparian forest study was to characterize the current riparian forest composition, structure and trends to inform development of a forest conservation
and restoration and conservation strategy. Key findings include: (1) the Hoh River valley riparian forest was largely cut during the 20th century, (2) inside and outside the CMZ only
24% of the area is in mature conifer deciduous forest >125-ft tall, (3) young deciduous forest < 75-ft tall dominates 41% of the study area, and (4) riparian key member sized trees
(≥5-ft diameter), available for channel recruitment, are significantly reduced in number compared to historic or reference conditions.
Methods & Study Area
Forest Mapping & Characterization
The Middle Hoh River study area includes river floodplains and terraces along the riparian corridor from Olympic National Park to Highway 101. Riparian forest mapping was conducted in
a GIS (ArcMap) environment using 2015 NAIP imagery and a 2014 first return LiDAR digital surface model (DSM). The first return LiDAR DEM provides a mapped measurement of vegetation
height (Figure 5). The NAIP color imagery allows mapping of coarse scale monotypic forest stands. Forests were typed in three cover classes and four height classes (Figures 6, 7 & 8).
Cover classes included: (1) > 75% cover deciduous trees (willow, red alder, black cottonwood, big leaf
maple, and vine maple; (2) > 75% cover conifers (Sitka spruce, Douglas fir, western hemlock and western red cedar); and cleared land (pasture, development); and (3) mixed conifer deciduous
(<75% deciduous & <75% conifers). Five height classes were measured with the LiDAR DEM: <5-ft, <25-ft, 25-75-ft, 75-125-ft, and >125-ft.
/
Figure 5. First return 2014 LiDAR DEM of forest vegetation heights at South Fork confluence with main stem Hoh River. Olympic National Park is clearly delineated by the much higher
tree heights.
/
Figure 6. Middle Hoh River riparian forest type cover and height classes.
/
Figure 7. Hoh River riparian forest mosaic of cover types. October 1st, 2020, near RM 20.6
/
Figure 8. Forest typing and channel migration zone (CMZ).
Results & Discussion
The Middle Hoh River riparian forest was mapped within and outside the CMZ to capture channel disturbance dynamics within the CMZ as compared with older terrace and adjacent hillslope
surface processes (Map 5, Figure 6). The Hoh River CMZ riparian forest, as mapped, is a mosaic of patches of young pioneer red alder flats to mature mixed conifer deciduous forest stands
(Figures 9, 10 & 11; Table 1). Channel migration, through erosion and sedimentation, generates new surfaces that are colonized by willows, red alder and black cottonwood. As the forest
matures, within 25-yrs, conifer colonization begins in earnest and these trees will grow to maturity over hundreds of years, unless harvested or succumbing to a natural disturbance
such as erosion by the main stem river, as yet unexperienced climate extremes, major wildfire or wind events, or competition by invasive species.
Deciduous forest (red alder, willow, black cottonwood) makes up 41% of the CMZ cover types, all of which are <125-ft in height, descriptive of young to mature red alder floodplain forests
less than 80-yrs (Table 1). Mixed conifer deciduous type comprises 32% of the total CMZ forest cover (Table 1). Together the deciduous and mixed cover types total 73% of the entire
CMZ forest cover, while coniferous cover type comprises 19% of the total CMZ forest (Table 1). The height structure of the riparian forest generally reflects age, with shorter trees
being younger and taller trees older. The >125-ft coniferous and mixed classes comprise 24% of the total forest cover, while <75-ft classes comprise 44%. This height distribution is
reflective of a younger CMZ red alder dominated forest with patches of mature conifer and mixed conifer deciduous stands dispersed throughout. Within the CMZ past timber harvest has
led to diminished floodplain roughness and to an accelerated channel disturbance regime, the primary control of existing riparian forest composition and structure. The loss of large
key member size trees, and stable wood jams that generate mid-channel islands and stable floodplain surfaces,
results in a loss of refugia sites for mature floodplain forests to develop (Montgomery and Abbe 2006). Acknowledging the key role of mature riparian forests in providing in-channel
aquatic habitat, Washington Forest Practice regulations limit the number of harvestable trees within the Riparian Management Zone (WAC 222-30-021).
Wind in the coastal Pacific Northwest also plays a role in structuring mature forest stands outside the CMZ where forest cover is dominated by Mixed (44%) and Coniferous (33%) cover
types (Figure 12; Table 2). Deciduous forest makes up only 16% of the total non-CMZ cover as compared to CMZ’s 41%. Forest stands >125-ft comprise 25% of the total forest cover indicating
forest stands reaching a century in age. Extensive deforestation is the primary driver of forest type and structure both within and outside the CMZ. However, changes in forest practices
allow for forests to regrow in the CMZ where channel migration rates are conducive to long-term growth. Unfortunately, a threshold appears to have been reached where floodplain stability
continues to decline and regrowth of large trees is likely limited to sites with large stable log jams accumulated over years with rare “key pieces” of in channel wood where erosion
has locally claimed exceptional trees, and also in areas protected by human interventions such as the placement of rip rap. Outside the CMZ clearing, timber operations and development
continue to drive forest type and structure. The implication of these findings for the Hoh River resiliency strategy is that the existing mature conifer and deciduous riparian forests
need protection, and new floodplain refugia sites need to be strategically re-created in a manner that protects them from main stem channel erosion.
Given climate change projections for the next century the Olympic peninsula forest will likely have an increase in fire frequency and experience more frequently periods of heat stress
(Halofsky et al. 2011). In the Olympic peninsula river valley forests, a changing climate over the next century is not likely to change the species composition, however, individual
trees more adapted to the changing climate will be at a selective advantage as compared to those that are not as resilient. In the face of these increased temperature projections foresters
throughout North America have begun to develop new seed transfer guidelines for forestry replanting operations (Kilkenny et. al. 2013). Seed transfer guidelines are developed from genecological
garden plot studies throughout the region located at various latitudinal and longitudinal positions allowing assessment of which tree ecotypes are most favorable across a species range
and projected climatic conditions. The new seed transfer zones will form the basis of assisted migration efforts to meet the challenges of our changing climate. Indeed, this ongoing
research will result in new sets of seed transfer zones for the Olympic peninsula that will be important in selecting climate adapted tree ecotypes for reforestation operations. Although
the coastal Olympic peninsula riparian forests are more resilient than the upland forests they will be impacted as the overall climate continues to warm. The results of forestry genecological
studies will be forthcoming over the next decade informing forest restoration efforts in the Hoh River valley.
/
Figure 9. Young deciduous (red alder and willow) and older mixed conifer deciduous (red alder and Sitka spruce) forest types. October 1, 2020, near RM 20.6. Note logjam proximity to
larger conifer patch.
/
Figure 10. Mature mixed (Black cottonwood, red alder, Sitka spruce) and conifer forest types (Sitka spruce, Douglas fir). Right bank (flow from right to left), October 1, 2020, near
RM 20.7.
/
Figure 11. Inside CMZ. Riparian Forest Type (Ac) and Height Class (Ft/Ac).
Table 1. Inside the CMZ. Forest type and height class.
HEIGHT (FT)
CLEARED (AC)
CONIFEROUS (AC)
DECIDUOUS (AC)
MIXED (AC)
GRAND TOTAL (AC)
<25
165
16
76
257
25-75
116
476
142
734
75-125
28
362
334
725
>125
272
260
532
Grand Total
165
432
914
737
2248
/
Figure 12. Outside CMZ. Riparian Forest Type (Ac) and Height Class (Ft/Ac).
Table 2. Outside the CMZ. Forest classes and types.
HEIGHT (FT)
CLEARED (AC)
CONIFEROUS (AC)
DECIDUOUS (AC)
MIXED (AC)
GRAND TOTAL (AC)
<25
222
175
67
44
507
25-75
162
287
311
760
75-125
277
91
632
1000
>125
376
24
339
739
Grand Total
222
990
468
1325
3006
Hydrology & Hydraulics
A hydraulic model of the Middle Hoh River was developed using Hydronia’s RiverFlow-2D Plus GPU and Aquaveo SMS v13.0 computer software. RiverFlow-2D is a two-dimensional finite volume
computer model that provides depth averaged hydraulic parameters at centroids within a triangular mesh model domain. The model domain encompasses 15-RM, with the upper boundaries on
the Hoh River and the South Fork Hoh River upstream of the National Park boundary, and the lower boundary just downstream of the Highway 101 bridge.
Hydrology and Boundary Conditions
To develop estimated inflows for the Middle Hoh River model, NSD performed a hydrologic analysis of the region. This analysis focused on two USGS gages: USGS gage #12041200, Hoh River
at US Highway 101 near Forks, WA (henceforth referred to as the Highway 101 gage) and USGS gage #12041000, Hoh River near Forks, WA (henceforth referred to as the upstream gage). The
downstream gage’s period of record is 1960-2021, but the upstream gage was retired in 1964, though it has data all the way back to 1921. Since the model domain is so large, both gages
were analyzed to help determine model hydrology. NSD performed a flow frequency analysis on both gages using the Log-Pearson schedule 3b methodology (USGS 1981), the results of which
are shown in Table 3.
Table 3. Estimated peak flows at each gage in the model domain.
USGS GAGE
DRAINAGE AREA (SQ MI)
1-YR FLOOD (CFS)
2-YR FLOOD (CFS)
10-YR FLOOD (CFS)
100-YR FLOOD (CFS)
# 12041200 (Highway 101 gage)
253
12,260
32,700
52,300
73,600
# 12041000 (Upstream gage)
208
8,190
19,000
30,800
46,000
The model was run for a series of three representative flow scenarios – the 1-yr, 10-yr, and 100-yr floods – to evaluate hydraulic parameters at the project site. There were 10 inflow
locations to this model – their locations are illustrated in Figure 13. The magnitude of all modeled flows at each of 10 inflow locations is shown in Table 4.
The main stem inflow was determined using the peak flow analysis of the upstream gage, scaled by drainage area to the main stem inflow location using the weighted USGS scaling method
presented in Mastin et al 2016. Discharge at the remaining inflow locations was determined using Streamstats where available, and when Streamstats was not available the discharge was
calculated using the USGS regional regression equations. Because only the 9 largest tributaries were explicitly modeled, the flow contributed by several smaller tributaries as well
as seepage and hillslope runoff within the model domain needed to be accounted for. This additional flow to the model discharge is needed to approximate the 10-yr and 100-yr peak flows
estimated from the downstream gage. This was achieved by adding a small amount of additional discharge to each inflow location proportional to its drainage area. For the 1-yr flood,
the inflows were reduced in proportion to their drainage area to achieve the same goal. For each inflow location, the model inflow which has been adjusted to account for hillslope
runoff is shown Figure 13, Table 4.
The models were run in a steady-state (constant discharge) and have only one outflow boundary located just downstream of the Highway 101 bridge and the downstream gage. The outflow boundary
condition was set to a constant water surface elevation (WSE) corresponding to each flow event. These WSEs were based on analysis of the rating curve of the Highway 101 gage. Downstream
WSE was set to 175.03, 185.53, and 189.81 for the 1-yr flood, 10-yr flood, and 100-yr flood respectively.
/
Figure 13. Locations of model inflow and outflow locations illustrated over the model domain.
Table 4. Modeled Discharge Values at Inflow Locations
INFLOW LOCATION
DRAINAGE AREA (SQ MI)
1-YR
10-YR
100-YR
Hoh River
124.9
7,700
27,200
35,900
SF Hoh River
53.8
4,080
13,900
17,700
Winfield Creek
11.6
60
2,100
3,950
Clear Creek
6.9
20
480
880
Maple Creek
9.5
100
1,480
2,620
Owl Creek
2.4
200
2,200
3,760
Hell Roaring Creek
5.5
30
1,020
1,920
Alder Creek
8.3
90
1,670
3,050
Tower Creek
4.5
-
1,140
1,920
Spruce Creek
4.5
-
1,140
1,920
Total
12,280
52,330
73,620
Highway 101 Gage Estimate
12,260
52,300
73,600
Mesh Development and Roughness Categories
The topographic data was developed by combining information from LiDAR data collected in 2012 and 2013, as well as bathymetry data that was provided by Wild Fish Conservancy from a 2013
survey. The model mesh was created with fine mesh spacing in the main channel and in floodplain channels, with expanded mesh spacing in less topographically complex areas further from
the stream.
Hydraulic resistance is characterized in the model by polygons representing differing surface roughness types such as main stem channel, forest, or bare earth. The full list of roughness
categories and their associated Manning’s n values is included in the calibration section in Table 5. Due to the large domain of this model, roughness categories were assigned in as
automated a fashion as possible. First, to attain a base of vegetation roughness, NSD subtracted the bare earth LiDAR from the highest hit LiDAR to attain a 15-ft x 15-ft raster of
vegetation height split into five categories: <1’, 1-4’, 4’-6’, 6’-12’, and >12’. To map roads onto this vegetation base, the lines in the DNR roads shapefile were given a 10’ width
and overlaid. The locations of the roads were spot-checked against aerial imagery to confirm their accuracy; when discrepancies were identified, the road locations were edited to reflect
aerial imagery. The low flow channel was then delineated using a wetted perimeter polygon derived from the combined surface, and on top of that the tributaries, side channels, and logjam
roughness areas were manually delineated using 2013 aerial imagery.
Modeled Infrastructure
NSD identified five road crossings that are directly engaged by the Middle Hoh River or by the modeled tributaries: 3 near the Clear Creek confluence, one upstream of the Clear Creek
confluence, and one on Spruce Creek before the confluence with the main stem. At these locations, NSD modeled the road crossings using the Riverflow2D culvert tool, which accounts for
culvert shape, size, material, slope, and inlet configuration using the FHWA procedure (Norman et al. 1985). Details on the three culverts near Clear Creek were provided by the 10,000
Years Institute from field observation; details on the two remaining culverts were estimated based on the surrounding LiDAR data.
Calibration
The 2D model was calibrated by comparing the modeled 100-yr inundation area to the digitized FEMA base flood map, effective June 7, 2019, and adjusting the model to bring it as close
to the FEMA floodplain as possible. The main tools available to calibrate the 2D model were adjusting the hydraulic roughness parameters using the Manning roughness coefficient (n-value)
for various land cover types. Initial values were set for the roughness coefficients based on recommended values from previous studies and professional judgement. After the initial
run, results were compared with the FEMA base flood and minor adjustments were made to optimize calibration. Final calibrated roughness values are shown in Table 5.
Map 6 includes a series of maps comparing the calibrated model to the FEMA floodplain. The areas in which the two do not agree can be explained by differences in tributary modeling –
for instance at small tributaries where the model does not include a specific inflow, the FEMA floodplain shows more inundation than the model, and conversely where tributary inflows
were inflated to account for surrounding drainage area, the FEMA floodplain shows less inundation than the model. The focus in calibration was on flooding related to main stem flow.
Table 5. Calibrated Manning's n roughness values for each roughness category.
CATEGORY
MANNING'S N VALUE
Veg <1'
0.04
Vegetation 1'-4'
0.07
Vegetation 4'-6'
0.08
Vegetation 6'-12'
0.08
Vegetation >12'
0.1
Road
0.016
Building
0.5
Active Channel
0.035
Tributary
0.045
Revetment
0.1
Results
Maps of hydraulic model results for depth, velocity, and shear stress during the 1-yr, the 10-yr, and the 100-yr floods are included in Map 6. The depth and velocity results were spatially
separated based on location in the floodplain vs. the active channel and then summarized for each of five reaches within the model domain. The resulting depth and velocity averages
are shown in Table 6 and 7, respectively.
Table 6. Average depth by reach for the modeled peak floods.
AREA
FLOOD EVENT
MEAN DEPTH (FT)
HUELSDONK-SOUTH FORK REACH
SPRUCE CANYON REACH
MORGAN’S CROSSING REACH
WILLOUGHBY CREEK REACH
OXBOW CANYON REACH
Channel
1-yr
4.4
7.3
5.1
4.6
8.5
10-yr
6.9
15.8
8.7
8.9
19.8
100-yr
7.9
19.0
9.9
11.5
24.2
Floodplain
1-yr
2.6
3.9
3.1
3.1
5.0
10-yr
3.7
6.6
3.6
5.2
10.0
100-yr
4.3
8.2
4.3
7.6
12.2
Table 7. Average velocity by reach for the modeled peak floods.
AREA
FLOOD EVENT
MEAN VELOCITY (FPS)
HUELSDONK-SOUTH FORK REACH
SPRUCE CANYON REACH
MORGAN’S CROSSING REACH
WILLOUGHBY CREEK REACH
OXBOW CANYON REACH
Channel
1-yr
5.1
8.0
5.6
4.8
7.4
10-yr
6.9
12.3
7.7
5.9
12.4
100-yr
7.4
13.2
8.3
5.9
13.8
Floodplain
1-yr
2.9
4.0
3.2
3.2
3.8
10-yr
3.6
3.7
2.8
2.3
4.7
100-yr
3.9
3.7
3.1
2.3
4.9
The average depth in the main stem ranges from 4.4-8.5-ft in the 1-yr event, 6.9 - 19.8-ft in the 10-yr event, and 7.9 - 24.2-ft in the 100-yr event. The range of average depth at high
flows is large due to the geomorphic differences between the reaches: the main stem is tightly confined in Oxbow Canyon reach and part of Spruce Canyon reach, resulting in high depths
compared to the other reaches like Huelsdonk-South Fork reach, where flow is spread out across the floodplain
The average velocity in the main stem ranges from 4.8 - 8.0-ft per second (fps) in the 1-yr event, 5.9 - 12.4-fps during the 10-yr event, and 5.9 - 13.8-fps in the 100-yr event. Floodplain
velocities remain in the range of 2.3 - 4.9-fps regardless of flood event recurrence interval. As with depth, the highest main stem velocities are in the confined reaches while the
lower velocities are associated with reaches where flow can overtop the banks and engage the floodplain.
Oxbow Canyon Reach
Flow conditions in the Oxbow Canyon Reach are the deepest on average for the entire 15-mi project reach due to near complete confinement of flow in the channel up to the 100-yr flood
(Table 6). Velocities rival those found in Spruce Canyon with the greatest average velocities in Spruce Canyon during the 1-yr flood, transitioning to Oxbow Canyon as flows approach
and exceed the 10-yr flood. A small floodplain is present on the right bank downstream of RM 17, and backwater eddy at RM 16 on the left bank that provide high flow refugia within the
canyon during floods exceeding the 5 – 10-yr flood (Map 6).
Willoughby Creek Reach
Upstream of the Oxbow Canyon Reach the floodplain abruptly widens and flows begin to spread out as the channel traverses the Willoughby Creek Reach. At the downstream end of the reach,
as the channel enters Oxbow Canyon, a backwater forms that rapidly floods the adjacent floodplain and diminishes instream velocities. The influence of this backwater propagates upstream
with increasing flow, extends over a mile upstream during the 100-yr flood, and influences instream and off-channel flow velocities and depths (Map 6). The effect of this backwater
also explains the lack of an increase in the average channel velocity from the 10-yr to the 100-yr flood (Table 6).
During the 1-yr flood side channels are activated in the Elk Creek Floodplain area as well as 2.1-mi long side channel following the valley margin on the left bank floodplain from Peterson’s
Bottom to Elk Creek. Velocities in the main stem channel are greatest where the channel is a single thread and diminish rapidly as the channel bifurcates in more and more channels (Map
6).
Morgan’s Crossing Reach
The Morgan’s Crossing Reach extends from Peterson Bottom to the downstream end of Spruce Canyon and is naturally unconfined with a broad floodplain through the entire reach. Downstream
of Rock Creek the channel bifurcates into multiple threads during the 1-yr flood and floodplain side channels are activated in the Lindner and Clear Creek floodplains. A backwater extends
approximately 2000-ft up an abandoned main stem channel at the Rock Creek confluence, upstream of which flow is confined to the main stem channel to the upstream end of the reach.
As flows increase to the 10-yr flood, water exits the channel banks and there is broad inundation of floodplains throughout the reach (Map 6). Between RM 21 – 22.6 overbank flooding
crosses the Upper Hoh Road near Lindner Creek Lane during the 10-yr flood and continues downstream north of the road following a historic side channel down to Lindner Creek, crossing
the road back to the channel next to the Peak 6 Adventure Store. When flow approaches the 100-yr flood the length of the Upper Hoh Road inundated increases and extends up to the Morgan’s
Crossing floodplain below the Tower Creek confluence.
Spruce Canyon Reach
The shortest reach in the Middle Hoh is the Spruce Canyon Reach, starting below the Owl Creek confluence at Spruce Island, ending at the downstream end of Spruce Canyon. Flows in the
Spruce Canyon are similar in nature to that of Oxbow Canyon, the biggest difference being flows are in general deeper in Oxbow Canyon. The channel is confined to the bedrock canyon
throughout the reach, minus the split around Spruce Island at the upstream end of the reach. Overall, the Spruce Canyon Reach has the highest average velocity in the Middle Hoh for
more frequent high flows less that the 10-yr flood. Near the 10-yr flood overbank flow enters an off-channel habitat pond (Dismal Pond, WDFW) constructed from a gravel pit on the right
bank floodplain at the upstream end of the reach, exiting the other side and continuing downstream on the floodplain to the Spruce Canyon inlet (Map 6).
Huelsdonk-South Fork Reach
At the upstream end of the Middle Hoh is the Huelsdonk-South Fork Reach, extending from Spruce Island to ONP. Average channel flow depths in this reach are the lowest of all reaches
and are the shallowest or close to on the floodplains (Table 6). Velocities are consistently second lowest for all flows, behind the Willoughby Creek Reach (dampened velocity due to
backwater above Oxbow Canyon).
The main stem channel flows unconfined through the reach, meandering across the valley bottom at low flows. As flows increase historic main stem flow paths begin to activate forming
large islands that remain dry during the 1-yr flood, short of a few side channels traversing the islands (Map 6). With increasing flow, the islands and adjacent floodplains become submerged
but do not flood the Upper Hoh Road. The channel has however expanded into private property in the Brandeberry and Fletcher Ranch areas, eroding land and damaging or engulfing structures
in the process. The main stem channel through this reach has been actively avulsing and migrating in recent history, frequently adjusting the nature of hydraulics in the reach. The
topographic surface available for this modeling does not represent the current condition, however predicted flow characteristics in the channel and floodplain are anticipated to be
similar. The location of the main stem channel will dictate where high velocity areas are located, and recent avulsions and bank erosion will expand the area inundated and increase
flow depths.
Main Stem Aquatic Habitat
The goal of this habitat assessment is to quantify and qualify the current habitat conditions of the main stem Hoh river, including braid, side, and main channel habitats and to describe
their use by salmon populations in order to evaluate the current habitat available and to identify impairments and restoration opportunities.
The Hoh River hosts five salmon species, including populations of spring and fall Chinook Oncorhynchus tshawytscha, fall coho O. kisutch, and winter steelhead O. mykiss (McHenry et al.
1996). The Hoh River salmon stocks are not listed under the Endangered Species Act (ESA) and are considered to be among the last healthy wild populations in the contiguous United States,
however recent declines in adult returns have been documented (Huntington et al. 2004; Hoh Tribe 2016; Cram et al. 2018). The runs of interest differ in their timing and use of riverine
habitats, but Chinook, coho, and steelhead all rely on the main stem Hoh for both rearing and spawning (McHenry 2001; McMillan and Starr 2008). Salmon spawning occurs in the main stem
throughout the year, but the timing depends on the species. Spring Chinook generally spawn in August and September, fall Chinook and coho spawn from October through December, and winter
steelhead spawn from December through July (McMillan and Starr 2008). These species also utilize the main stem habitat differently as juveniles. Most fall Chinook and many spring Chinook
out-migrate at age-0, while coho, steelhead, spring Chinook may overwinter and remain in the river for a full year. Coho and steelhead are also documented to use more lateral main stem
(braids and side channels) and tributary habitats during rearing than Chinook, which
primarily rely on main channels of the main stem (McMillan and Starr 2008; Quinn 2018). However, all species are documented to use braid and side channels for spawning.
Adequate freshwater migration, spawning, incubation, and rearing habitat is important for all these species (Quinn 2018). All three species require suitable sized gravel and cobble for
spawning and rearing with low levels of fine sediment. Winter rainfalls and the resulting flooding can exacerbate erosion and landslides, scour and damage redds, and increase infiltration
of fine sediment reducing oxygen levels available to embryos, which in turn decreases the chance of survival to emergence (Quinn 2018). Furthermore, overwinter rearing habitat is particularly
important for juvenile coho, steelhead, and to an extent spring Chinook due to their extended stay in freshwater compared to fall Chinook. Lack of adequate summer flows, water quality
and temperatures, channel complexity, slow water habitat, and off-channel and floodplain connectivity can contribute to low salmon rearing and spawning success (Ames et al. 2000; Hoh
Tribe 2016).
While the status of salmon stocks in the Hoh River are generally considered to be healthy, declines over the past two decades have prompted cause for evaluation (Smith 2003; Hoh Tribe
2016). The Middle Hoh has been impacted by the loss of riparian old growth forest and reduction of large wood inputs, erosion, landslides and increased fine sediment pulses, and decreased
water quality (Parks 1999; Hoh Tribe 2016). Additionally dramatic channel migration has occurred within the study area in recent years, with movements of over 60-ft a year (Chadd 1997;
Piety 2004; Hoh Tribe 2016) and local landowners and the county are concerned about the loss of property and road infrastructure, as well as the impacts on salmon and instream habitat
quality. In the below summary, we characterize the summer low flow (i.e., average daily flow of 1,614 to 2,652 cfs at the Highway 101 bridge (USGS Gage 12041200) main stem aquatic habitat
conditions as surveyed from September 28th to October 1st, 2020, from RM 16 to 31, including channel conditions, habitat quantities, and bank conditions.
Methods
Low flow habitat surveys on the Hoh River main stem were completed by Cramer Fish Sciences (CFS) from upstream of the Highway 101 Bridge (RM 15-16) to the National Park Boundary (RM
30-31) from September 28th to October 1st (Figure 14). Average daily flows during the surveys ranged from 1,614 to 2,652 cfs at the Highway 101 bridge (USGS Gage 12041200; Figure 15).
All main channel watercourses were surveyed; and braid and side channels were surveyed as time allowed to achieve a representative subsample of the habitat quality and quantity (Leopold
and Wolman 1957; Peterson and Reid 1984). Wetted side channels and braids that were unable to be surveyed because of access or time constraints were captured at their diversion point
from the main channel with photos and GPS coordinates.
/
Figure 14. Overview of the Hoh River survey area with survey reaches of interest (Piety et al. 2004) and start and end of survey locations.
/
Figure 15. Instantaneous flow (cfs) data at USGS Gage 12041200 – Hoh River at Highway 101 near Forks, WA recorded and reported by USGS. Surveys were conducted between September 28th
and October 1st, 2020.
Main stem surveys were completed moving downstream by boat. Reaches were determined by geomorphic characteristics (Piety et al. 2004). Channel type was recorded as main, braid, or side
channel; main channels of the main stem were identified as channels with the most flow; braids were identified as one or more anastomosing channels connected to the main channel, and
side channels identified as channels within the main
stem floodplain separated by islands with stable woody vegetation (Leopold and Wolman 1957; Peterson and Reid 1984). Habitat units were identified as pools (non-turbulent), riffles (fast-turbulent),
or glides (fast non-turbulent) (Bisson et al. 1982; Beechie et al. 2005; CHaMP 2016). Pool type (e.g., plunge, scour, dam) and pool-forming feature were recorded for pool units (Bisson
et al. 1982). Pool frequency was calculated as the pools per mile of channel length and pool spacing (channel widths per pool) was calculated using the Montgomery et al. (1995) formula
as the channel length per wetted channel width (Wc) divided by the number of pools. Pool frequency targets were identified using NMFS (1996) (Table 8). Note that while the number of
pools per mile goes down in larger channels, when normalized by channel width, pool frequency (#/Wc) increases as channel size increases. Lengths and wetted widths were recorded in
meters using a laser rangefinder. We were unable to measure depths given the survey scope and funding, as measuring depth by boat requires a sonar or depth rod, which is difficult and
time consuming in a large swift water river. All metric units were converted to English units in processing. The GPS coordinates of the top and the bottom of each dominant unit were
recorded, GPS units were also used to record tracks of the channels surveyed. Recorded data outputs are available in Appendix A.
Table 8. Pool frequency standards for functioning rivers developed by NMFS (1996).
CHANNEL WIDTH (FT)
# POOLS/MILE
# pools/Wc
5
184
0.17
10
96
0.18
15
70
0.20
20
56
0.21
25
47
0.22
50
26
0.25
75
23
0.33
100
18
0.34
For each riverbank of each unit, the percent of length occupied by each edge type at the wetted edge was estimated. Edge types were recorded as bank edge (natural or hydro-modified)
or bar edge (Hayman et al. 1996). Hydro-modified banks were identified as banks where modifications were visually observed, such as levees or riprap, however the quality was not recorded.
If banks were modified but the modification could not be identified, they were categorized as “Hydro-modified unknown.” Bar edges were assumed to be naturally formed (Hayman et al.
1996). The width of slow water was recorded if present for each edge type. Slow water area was defined as present when a boundary between the edge type and mid-channel was visible as
a current shear line; slow water edge width was assumed to be a minimum of 1-m wide for all banks.
Large wood jams (LWJs) were mapped using aerial photography collected in March 2021 (NV5 GeoSpatial 2021; Figure 16). The average daily flow during collection was 3,366 cfs at the Highway
101 bridge (USGS Gage 12041200; Figure 17). We digitized all jams visible within the bankfull channel, including wood visible in the water, on gravel bars, and on vegetated islands.
All jams composed of three or more wood pieces 7.6 m in length by 0.3 m in diameter were manually delineated using QGIS at a zoom level of 1:600 or higher, and the full perimeter of
the jam was recorded (Leif et al. 2004). The location of the jam in the main stem (main channel and braids, side channel, or islands), the channel location (bar, bank, or island), and
whether the jam was wet or dry were recorded for each jam. Jam locations were also recorded during field surveys, however, considerable channel movement led to changes in jam positions
and areas, therefore all reported jam metrics are from the aerial imagery analysis.
/
Figure 16. Example of jams delineated in the Morgan’s Crossing Reach (RM 21.9), shown at a 1:600 scale using aerial photography collected in March 2021 (NV5 GeoSpatial 2021). Blue polygons
depict wood in wetted channel (time of photo) and red polygons wood outside wetted channel but within ordinary high water (bankfull) channel. Flow is from right to left.
/
Figure 17. Instantaneous flow (cfs) data at USGS Gage 12041200 – Hoh River at Highway 101 near Forks, WA recorded and reported by USGS. Aerial imagery was collected on March 20th, 2021.
Results
A total of 28.8 linear miles of wetted main stem habitat were surveyed in the Middle Hoh River Study Area from RM 16, at the Oxbow Campground to RM 31 at the National Park boundary (Appendix
A, Map 7). This included 15.8 miles of main channel, 8.0 miles of braided channel, and 4.9 miles of side-channel habitat. Wetted braids and side channels were present in all reaches
except in the Oxbow Canyon Reach (Figure 18, Map 7). Additional braid and side-channel habitats wetted at the time of the survey were not surveyed because of time constraints, however
representative samples of these habitats were surveyed to qualify the habitat present. Braid and side-channel habitat was the most abundant in the Huelsdonk-South Fork Reach, followed
by the Morgan’s Crossing Reach (Map 7).
Overall, glides were the dominant habitat unit with 48% of the surveyed wetted channel area observed in glide units followed by riffles and rapids which made up 31% of the surveyed wetted
channel area, and pools and backwater units, which accounted for the remaining 21% of wetted channel area (Figure 18, Map 7). It should be noted, that because we were unable to measure
depths and calculate residual depths, pools were identified by the presence of channel (e.g.., confluences, channel bends, bedrock canyons) and large wood features or changes in surface
flow and hydrology (i.e., the presence of slow water), and therefore pools may have been classified as glides and underestimated if no visual clues were present. However, slow water
edge area was also documented, as slow water edge habitats are more utilized by juvenile salmon in large rivers than main channels (Beechie et al. 2005). Riffles and rapids were the
most frequently observed units, with eight units per mile, followed by pools and backwater units (six units per mile), and glides (five units per mile). Of the reaches
surveyed, the Huelsdonk-South Fork Reach had the highest total number of pools, total area in pools, and the highest pool frequency (pools per mile), but also had the most channel length
surveyed (Appendix A). The Spruce Canyon Reach had the fewest pools and lowest pool frequency (pools per mile).
Trench, scour, and backwater pool units were observed (Figure 18). Trench pools were formed by bedrock and canyon features and were primarily observed in the Oxbow Canyon Reach (Figure
19). Large wood frequency was not significantly correlated to pool frequency across the reaches but there was a slight trend indicating that there was a positive relationship between
LWJ frequency and pool frequency. This is expected because LW can drive pool formation, however in the main channel, meanders were the primary pool forming features, followed by bedrock
features. We observed that many jams provided edge cover in summer low flow conditions but were not large enough to drive pool formation in the main channel.
The majority of bank habitat in the Hoh main stem was made up of bar edges (55% of edge length) and natural bank edges (42% of edge length) (Appendix A, Table 4). However, hydro-modifications
including rip rap, roads, and other bank armoring and residential bank modifications were documented in all reaches except the Oxbow Canyon (Figure 18). Rip rap and bank armoring were
most prevalent in the Huelsdonk-South Fork Reach, with 1.3 miles of hydro-modified banks present over ten locations, followed by the Willoughby Creek Reach with 0.8 miles of hydro-modified
banks present over five locations (Appendix A, Table 4). Slow water edge areas associated with natural bank edges are important rearing habitats for juvenile salmon, especially age-0
and age-1 steelhead during winter flows (Beamer et al. 2005; Beechie et al. 2005). Bar edges are also used by rearing salmon, but densities are lower in these habitats due to their
slightly higher velocities and tendency to have less cover than natural bank edges (Beechie et al 2005). Rip rap and diked banks have been documented to have reduced slow water area
and lower Chinook and steelhead densities (Schmetterling et al. 2001), however a study in the Hoh River observed that these habitats can support comparable densities to LWJ, if they
create cover and complex habitat (Peters et al. 2012). Studies have suggested that rearing salmon density is correlated to the area and velocity of slow water present (Beechie et al.
2005; Peters et al. 2012), we observed bar edges and natural bank edges to have the greatest associated average slow water edge width (4.1-ft) and total area, followed by hydro-modified
bank edges (3.7-ft).
Wood was mapped using aerial imagery collected six months after the low flow habitat surveys, during which three flow events over 20,000 cfs occurred leading to multiple changes in the
channel form. Wood values are reported as they were observed in the March 2021 aerial imagery, however, the field data recorded in the low flow 2020 habitat surveys suggest trends in
wood distribution were consistent between the surveys. The majority (87%) of the 2,573,637-ft2 of mapped LWJ area was located in the main channel and braids, followed by side channels
and forested islands. Of the total LWJs mapped in the main channel, 54% were dry on bars, 28% were wetted and on bars, 11% were wetted and bank attached, and 7% were dry and bank attached
(Appendix A). In side channels and on islands, 31% of the jams were dry on bars, 25% were wet and bank attached, 28% were wet and bar attached, 18% were dry on islands, and 8% were
dry on bars. While the majority of the LWJs mapped were dry, LWJs with some wetted area accounted for most of the mapped area and the mean wetted jam size was approximately twice as
large as the dry (Appendix A, Table 6). The Huelsdonk-South Fork Reach contained the most LWJs and total overall, wetted, and dry LWJ area, but the Morgan’s Crossing Reach had the highest
LWJ frequency (LWJ/mi) (Appendix A, Table 7). The mean and median LWJ sizes were above the full study area value (5,066-ft2 and 2,813-ft2, respectively) in the Huelsdonk-South Fork,
Spruce Canyon, and Morgan’s Crossing reaches (Appendix A, Table 7). Jam sizes were smaller than the study area average in the Willoughby Creek and Oxbow Canyon reaches, likely due to
the fact that majority of large wood transport from the ONP would be racked in the upstream reaches (Welber et al. 2013; Kramer and Wohl 2017) and the riparian area within the study
area is dominated by deciduous and second-growth conifer forests. While wood was not the primary driver of pool formation in the main channel and wood sizes have decreased in the last
century as a result of logging, the large wood present drives the generation of slow water edge habitat and eddies and
provides cover for rearing salmon in the main channel (Peters et al. 2012). Additionally, wood cover is generally thought to be essential for predator avoidance for rearing salmon, but
it has been observed to be less so in turbid rivers where visibility is already limited (Peters et al. 2012). The LWJ size was larger relative to the channel wetted width in side channels,
which drove the formation of the majority of mapped pools, this indicates that more complex rearing habitat is available in side channels for steelhead, coho, and spring Chinook (Sedell
1982).
The Huelsdonk-South Fork Reach, Morgan’s Crossing Reach, and Willoughby Creek Reach contained the most braid and side-channel habitat, though the confinement of the Spruce Canyon Reach
and Oxbow Canyon Reach was driven by natural canyon features and not infrastructure encroachment. Hydro-modified banks were infrequent throughout the study area as a whole. However,
the Upper Hoh Road flanks the right bank of the main stem and is directly adjacent to the river in the Huelsdonk-South Fork Reach, Morgan’s Crossing Reach, and Willoughby Creek Reach.
Rocky bank armoring and rip rap was documented along the road without vegetative cover during the summer low flow surveys and presents an opportunity for enhancement and roughening
to improve the habitat. Additionally, the main stem is eroding multiple residential properties in the Huelsdonk-South Fork Reach.
Substrate was visually characterized throughout the study area and was largely gravel and cobble dominated, with numerous high quality spawning habitats in riffles and pool tails. However,
substrate was not characterized at the unit level and this effort did not include mapping spawning habitat or redds. Future efforts aimed at incorporating this information would be
informative for identifying adult salmon use throughout the reaches (Montgomery et al. 1999). Several landslides and eroding banks were observed throughout the reaches, which could
lead to fine sediment pulses during heavy rainfall events (Hoh Tribe 2016). However, fines were not visible as the dominant sediment in any of the reaches at the time of survey during
low flow conditions.
The reaches of the Middle Hoh Study Area generally represent high quality salmon habitat, with abundant braid and side-channel habitat, a large number of pools and slow water glides
and edges, and frequent large wood (17.6 jams/mile). Pool frequencies were below reported targets for both main channels (over 100-ft in wetted width), for which the National Marine
Fisheries Service (NMFS) considers 18 pools per mile to be adequate, and braids and side channels (wetted width of 25 – 50-ft), for which NMFS has a target of 47 pools per mile (NMFS
1996; Table 8), however, there was a large amount of slow water edge habitat observed, both with wood cover and without, and a large number of slow water glide habitats that would provide
rearing and holding habitat for juvenile and adult migrating salmon, respectively. While the Hoh River has a large amount of wood, the vast majority is mobile and has limited function
in forming stable logjams that persist over time. Actions to emulate the large trees that once were common in the valley are needed to increase the number of stable logjams. This is
necessary not only to sustain pool frequency over time, but more importantly to sustain side channels and allow mature conifer patches to develop that can sustain the large wood cycle
(Abbe and Montgomery 1996; Collins et al. 2012; Abbe et al. 2016).
/
Figure 18. Overview of the Middle Hoh River Study Area with the results from the CFS habitat surveys conducted from September 28th to October 1st, 2020, at an average daily flow of 1,614
to 2,652 cfs at the Highway 101 bridge (USGS Gage 12041200). For more detailed mapping see Map 7. Braids and side channels were surveyed as time allowed and were not a full census.
Main stem diversion locations of un-surveyed wetted braids and side channels were identified during surveys and general channel locations were mapped using the aerial imagery in GIS.
Oxbow Canyon Reach
The Oxbow Canyon Reach is the downstream extent of the study area and was surveyed from the Oxbow Campground near RM 16 to downstream of RM 17.5 at the confluence of Winfield Creek and
the main stem Hoh River (Figure 19). The reach is heavily confined by a natural canyon that prevents lateral channel migration. No braided channels were wetted at the time of survey
and no side channels were present in this reach (Appendix A). Glides were the dominant habitat unit in this reach, and accounted for 49% of the wetted channel area, followed by pools,
which accounted for 28% of the wetted channel area (Figure 19). One rapid and four riffles were also observed, accounting for the remaining 23% of surveyed wetted channel area. Natural
bank edges made up 69% of the edge length in this reach, while bar edges made up the remaining 31%. No hydro-modified bank edges were present in this reach. The average slow water edge
width for both natural bank edges and bar edges was 4.1-ft (Figure 19).
The Oxbow Canyon Reach had the lowest main channel pool spacing (8.5 channel widths per pool) and highest main channel pool frequency (4.2 pools per mile) of the reaches surveyed, with
eight pools observed in the 1.9 miles of channel length (Appendix A, Table 5). There were seven trench pools and one scour pool documented in this reach, all of which were formed by
bedrock features (Figure 20).
Wood was infrequent in this reach, with only two wetted jams and no dry jams mapped (Appendix A, Table 6). Large wood jams were relatively small in this reach, with an average jam size
of 2,270-ft2 (Appendix A, Table 7). However, given the confinement and bedrock present in this reach, opportunities for natural wood racking and accumulation are minimal.
Given the confined nature of this reach, the presence of bedrock and boulders, and the lack of off channel habitats, this reach likely serves as a migratory corridor rather than rearing
or spawning habitat for juvenile and adult Chinook, steelhead, and coho (Figure 19). The reach contains numerous deep pools and boulders which provide year-round holding habitat for
the upstream migration of adults and out-migration of juveniles. The lack of side channels in this reach suggests that this reach would not be used heavily for rearing for juvenile
spring
steelhead, coho, or Chinook. Additional large wood placement would not benefit this reach as pool formation was driven entirely by bedrock.
/
Figure 19. Overview of the Oxbow Canyon Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the Highway
101 bridge (USGS Gage 12041200). No wetted braid or side-channel habitat was present in this reach. The mapped photo is shown in Figure 20.
/
Figure 20. An example of the bedrock banks present in Oxbow Canyon (location of image shown in Figure 19). The canyon confines the channel and prevents the formation of off-channel
habitat; therefore the reach would be primarily used as a migration corridor for salmon.
Willoughby Creek Reach
The Willoughby Creek Reach extends from RM 17.5 at the Winfield Creek confluence with the main stem up to upstream of Willoughby Creek near RM 20.5 (Figure 18; Figure 21). The Upper
Hoh Road flanks the right bank of the main channel at the upper extent of the reach. The Willoughby Creek Reach is a relatively unconfined sinuous gravel-cobble reach. Near the middle
of the reach, the main channel demonstrates substantial migration, and has shifted over the last 30-yrs away from a large meander bend on the left bank to a more linear channel form.
The lower extent of the reach is more naturally confined and exhibits less migration.
There were 3.3 miles of main channels, 1.9 miles of braided channel, and 0.5 miles of side-channel habitat surveyed in this reach. Two additional braided networks and two additional
side-channel networks were observed as connected to the main channel the time of survey but were not surveyed due to time constraints (Figure 21). Glides were the dominant main channel
unit in in this reach and accounted for 58% of the wetted channel area (Appendix A). Riffles and rapids made up 32% and 1% of the wetted main channel area, respectively. Pools were
infrequent in the main channel and only accounted for 9% of the total wetted main channel area.
Pools were more frequent in braid and side-channel habitats and made up 25% and 40% of wetted channel area, respectively (Figure 21). Glides made up 47% and 31% of braid and side-channel
wetted habitat area, respectively, and riffles made up the remaining 28% of channel area for both braids and side channels. The additional braided channels present were observed to
be glide and riffle dominated. Given the ephemeral nature of braids in alluvial channels, these habitats would be impacted by winter floods.
Main stem edges were primarily composed of bar edges, which made up 63% of the surveyed edge length. Bank edges made up the other 37% of main channel length, of which 18% were hydro-modified
(Appendix A). The Willoughby Creek Reach had the second highest total length of hydro-modified bank edges after the Huelsdonk-South Fork Reach of the reaches surveyed (Figure 18). Two
areas on the river right bank were noted to have substantial erosion and trees actively sloughing into the channel (Figure 22). Most of the slow water edge area observed was associated
with bar edges, however, slow water edge width was the greatest in natural bank edges, with an average slow water edge width of 4.1-ft, followed by bar edges, with an average edge width
of 3.6-ft (Appendix A). Hydro-modified bank edges had an average slow water edge area of 3.3-ft (Figure 22).
Total percent pool area and pool frequency were low and pool spacing was high in this reach compared to other reaches, with only 13% pool area, 1.5 pools per mile, and 21.8 channel widths
per pool (Appendix A, Table 5). Twenty-eight pools were documented in the 5.7 miles of channel length surveyed in this reach. Large wood jams and pieces were the primary pool forming
feature, followed by channel features, including meanders and confluences (Figure 21).
The Willoughby Creek Reach contained the highest LW frequency for wet jams and second highest overall. However, the median jam size was smaller than other reaches at 2,685-ft2, and the
reach had the third lowest total jam area (Appendix A, Table 7). The majority of jams were located in the main channel on gravel bars (97 jams comprising 377,247-ft2), while only 16
bank-attached jams were mapped totaling 47,564-ft2 of area (Appendix A, Table 6). Jams in this reach were smaller than the upstream Huelsdonk-South Fork and Morgan’s Crossing reaches,
which suggests the larger old growth conifers entering the system from Spruce Flats/Island and ONP are being deposited in the upstream reaches (Welber et al. 2013; Kramer and Wohl 2017)
and the wood inputs to the Willoughby Reach are primarily from the deciduous and second-growth conifer trees located in the Middle Hoh Study Area riparian.
Pool frequency and total pool area was low compared to other reaches and to NMFS targets (NMFS 1996; Table 8). However, slow water estimates that include glide habitat and slow water
edge areas suggest that there is abundant habitat in this reach for juvenile rearing and adult holding for coho, steelhead, and spring and fall Chinook (Appendix A). Additionally, there
were numerous off channel habitats for rearing coho, steelhead, and spring Chinook and spawning coho, steelhead, and Chinook in this reach.
The jam size in this reach is smaller than the study area average, which is likely a result of and a facilitator to the high degree of channel migration during winter and spring floods.
Chinook, steelhead, and coho redds may be at risk of scour during these large channel migration events and the impact of scour on egg-to-fry survival should be evaluated in future studies
(Roni et al. 2016). Fall Chinook, coho, and steelhead, would be particularly impacted by potential scour events as fry would not yet have emerged by spring floods (Montgomery et al.
1999). This reach contained abundant gravel throughout, which suggests suitable spawning conditions for adult Chinook, steelhead, and to a lesser extent coho in the main channel if
redds are buried below the scour line (Montgomery et al. 1999). Additionally, side channels and the tributary confluences were also noted to contain gravel, which would provide spawning
habitat more protected from scour events.
The abundance of wood in this reach, the presence of numerous off channel habitats and large confluence side channels and tributaries within the main stem floodplain (e.g., the Elk Creek
Confluence Side Channel), and the large area of slow water edge and main channel habitats suggests that this reach would provide abundant habitat for juvenile steelhead and coho overwinter
rearing, and juvenile spring and fall Chinook rearing and out-migration (Figure 21; Beechie et al. 2005). Wood in side channels was frequently observed to be channel spanning and pool
forming, creating slow water habitat area and providing cover to rearing salmon (Figure 22). Additionally, it is presumed that the terrace-tributary Elk Creek Confluence Side Channel
provides relatively stable habitat (less impacted by channel migration) that would provide high quality rearing and spawning habitat in this reach (Sedell et al. 1982).
Restoration efforts in this reach should target the 0.8 miles of road that runs along the right bank of the main stem (Figure 22), which would benefit from enhancement actions such as
raising the road prism and protecting the bank toe with large wood structures to create a more natural hillslope form and increase bank attached wood jam area to create slow water main
channel habitat and provide shading and predator protection. Western Federal Lands Highway Division, in partnership with Jefferson County, completed construction of a log-dolo revetment
at the upstream end of the reach in 2021 to replace the riprap along the road. Additional actions aimed at protecting riparian forests with the long-term goal of establishing forests
similar to those in the ONP and planting meander bends with conifers to increase roughness would also benefit this reach.
/
Figure 21. Overview of the Willoughby Creek Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the
Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as a representative subset of the habitat available as time allowed and were not a full census. Two side
channel networks and two braided areas, shown in gray, were documented as connected to the main channel at the time of the surveys but were not surveyed due to time constraints. Photos
are shown in Figure 22.
/
Figure 22. Reach photos with locations shown in Figure 21. 1) An example of a large stretch of eroding banks with immature industrial forest observed which demonstrates inputs of fine
sediment and clay and small wood into the river. 2) A LWJ formed pool in a side channel with intermittent connectivity. 3) An eroding bank recruiting small Douglas firs into the channel.
4) A location of rip rap where the Upper Hoh Road runs along the right bank and constrains the channel. The riprap provides no complexity, no cover, no hydraulic refugia and no potential
for wood recruitment.
Morgan’s Crossing Reach
The Morgan’s Crossing Reach extends from RM 20.5 upstream of the Willoughby Creek confluence with the main stem to RM 25.5 (Figure 18). The Upper Hoh Road runs adjacent to this reach
and confines the floodplain on the right bank. The road runs directly adjacent to the main channel right bank at the downstream and upstream end of this reach and flanks the right bank
of a side channel in the middle of the reach. A substantial amount of channel migration has occurred in the middle of this reach, with the channel straightening and moving away from
the road on the right bank. Most of the sinuosity in this reach occurs in the middle extent, with minimal braid and side-channel habitat or main channel meandering occurring in the
upstream and downstream ends of the reach. The Morgan’s Crossing Reach was the longest reach surveyed and represents important spawning and rearing habitat for Chinook and coho and
steelhead, respectively in both the main channel and off-channel and tributary habitats.
A total of 7.7 miles of wetted main stem channel length was mapped in the reach, including 4.8 miles of main channel, 1.8 miles of braided channel, and 1.1 miles of side-channel length
(Appendix A). The majority of the Lindner side channel complex was wetted during the low flow surveys but was not surveyed due to time constraints and difficulty accessing deep non-wadeable
or boatable channels, however a representative sub-section was surveyed (Figure 23). Three additional braid complexes were not surveyed due to time constraints, these habitats appeared
consistent with the braid complexes that were surveyed. Glides accounted for approximately 60% of the total wetted channel area, followed by riffles which made up 31%, pools which made
up 9%, and backwater units that made up 1% (Appendix A). This reach had the lowest main channel percent pool area, and second lowest total pool channel area.
Pools accounted for more wetted channel area in braids and side channels than in the main channel in this reach. In braids, glides accounted for 46% of wetted channel area, riffles accounted
for 28% of wetted channel area, and pools accounted for 26% of wetted channel area. In side channels, pools accounted for 60% of the wetted channel area, followed by glides which made
up 22% of wetted channel area, and riffles which made up 18% of wetted channel area (Figure 23).
Approximately 60% of the total edge length was bar edges, 38% was natural bank edges, and 2% of banks were hydro-modified (Figure 23). Bank hydro-modifications, including rip rap and
residential development, were only documented in main channels in this reach, however, a road also runs along the Lindner side channel and confines lateral movement on the right bank
(Figure 23; Map 7). The average slow water edge width 4.1-ft for bar edges, 3.9-ft for hydro-modified bank edges, and 3.6-ft for natural bank edges. Slow water edge habitat associated
with main channel bare bar edges and bar edges with LW racking provides an additional 100,000-ft2 of slow water area in addition to the pool area, which would be important low velocity
habitat for rearing steelhead, Chinook, and coho (Beechie et al. 2005).
The Morgan’s Crossing Reach had the second highest total main stem (main channels, braids, and side channels combined) pool area of the reaches surveyed, after the Huelsdonk-South Fork
Reach (Appendix A, Table 5). While the percent pool area and pool frequency (pools per mile) in the main channel were the lowest of the reaches surveyed, in braids and side channels,
percent pool area and pool frequency were amongst the highest of the reaches (Figure 18). Main channel pools were primarily formed by channel features, while large wood primarily drove
pool formation in braids (Figure 23). Side-channel pools were formed by both large wood and channel features (Figure 24).
The Morgan’s Crossing Reach had the highest overall (wet and dry) LWJ frequency and second highest total LWJ area (Appendix A, Table 7). Main channel and side channel LWJs were abundant
in this reach (Appendix A, Table 6). There were 120 jams located on gravel bars, comprising 634,488-ft2 of total jam area, and 31 jams located attached to banks, comprising 184,083-ft2
of jam area (Appendix A, Table 6). This reach had the second largest mean jam size of the reaches surveyed (Appendix A, Table 7).
Total pool area was low in the main channel of Morgan’s Crossing Reach but there was abundant slow water habitat in side channels for juvenile rearing and for coho, steelhead, and spring
and fall Chinook spawning. The Upper Hoh Road runs directly adjacent to this reach in multiple locations which presents an opportunity to enhance bank habitat. Additionally, three locations
along the right bank were noted to have severe erosion and landslide risk. These locations likely provide pulses of fine sediment and clay after storm, but further evaluation is needed
to understand their impact on egg-to-fry and rearing salmon survival (Figure 24). Bank enhancement actions such as raising the road prism, planting, and placing large wood benefit salmon,
by roughening banks and providing cover, and help to protect infrastructure. Floodplain plantings of conifers in the larger alder groves on older bars and meander bends could speed
up riparian succession and provide bank stabilization. This reach would also benefit from continued floodplain forest protection to encourage the establishment of mature
conifer forests, similar to those present in the ONP, to provide long-term large wood inputs to create large functional jams and stabilize the channel.
The Morgan’s Crossing Reach contains the Lindner Side Channel, which is a large and important side channel complex for juvenile salmon. In the subset of the Lindner Side Channel surveyed,
we observed abundant pool-riffle habitat, which provides high quality spawning substrate and slow-water refuges for adult and juvenile salmon, respectively. One large jam was present
in the surveyed extent of the Lindner Side Channel, which would provide cover for rearing Chinook, steelhead, and coho. Additionally, vegetated natural bank edges were prevalent in
this reach, which also provide slow water habitat with cover. There was little evidence of recent channel migration in the surveyed extent of Lindner Side Channel, suggesting that the
side channel is relatively resistant to recent winter flooding and scour events. Wood frequency (jams per mile) was lower in the Lindner Side Channel than many of the other side channels
surveyed, which may limit cover for coho and steelhead (Figure 23; Figure 24). Cover in the Lindner Side Channel would be especially important since the side channel appeared less turbid
the main channel during the low flow surveys, meaning juveniles would be more at risk to predators with the absence of other vegetative or wood cover.
/
Figure 23. Overview of the Morgan’s Crossing Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at
the Highway 101 bridge (USGS Gage 12041200). The Lindner side channel is the dashed line between photo points 4 and 1. The photo was taken in the winter (leaf off) so deciduous trees
stand out as light brown areas compared to green areas of conifer forest. The current valley bottom is dominated by small deciduous trees. Prior to logging of the valley the number
and size of large conifer patches or “islands” would have been much greater. Braids and side channels were surveyed as a representative subset of the habitat available as time allowed
and were not a full census. One side channel complex and three braided areas, shown in gray, were documented as connected to the main channel at the time of the surveys but were not
surveyed due to time constraints. Photos are shown in Figure 24.
/
Figure 24. Reach photos with locations shown in Figure 23. 1) A long pool with ample vegetation cover but lacking large wood jams in lower portion of Lindner Side Channel. 2) The remnants
of a landslide that has begun to be colonized by Red Alder. 3) Another landslide that appears to be actively depositing silt and clay into the channel but is protected by a large wood
jam. 4) A log jam that runs along the bank at the Lindner Side Channel inlet along the main stem (2020).
Spruce Canyon Reach
The Spruce Canyon Reach runs from approximately RM 25.5 to RM 27 and is the shortest reach in the Middle Hoh Study Area (Figure 18; Map 7). The left bank is confined by the valley wall,
and the right bank is confined by intermittent bedrock and boulders. This reach experiences little channel migration, and the channel is largely confined to the main channel with little
braid or side-channel habitat present, apart from one side channel in the upper extent.
We surveyed 1.7 miles of main channel and 0.1 miles of braided channel length in the Spruce Canyon Reach (Appendix A). One additional small side-channel complex was wetted at the time
of survey but was not surveyed due to time constraints (Figure 25). Pools accounted for approximately 33% of the surveyed wetted channel area, followed by glides, riffles, and rapids,
which accounted for 29%, 20% and 17% of the surveyed wetted channel area, respectively (Figure 18; Figure 26). A long section of rapids ran upstream and into the canyon, and one rapid
spanning 472-ft and made up all of the surveyed braided channel. Natural bank edges accounted for 78% of the total edge length surveyed in this reach and bar edges made up 20% (Figure
26). One section of rip
rap was present on river right that made up the other 2% of the edge length. Natural bank edges had an average slow water edge width of 5.4-ft, followed by bar edges which had an average
slow water width of 5.2-ft (Figure 25). The rip rap bank section had a slow water edge width of 3.3-ft. Slow water edge habitat was minimal in this reach for riffle and rapid units
and these features would likely be avoided by rearing coho and steelhead.
Total LWJ area and frequency were low in this reach, with only 15 jams mapped, however, the mean and median wetted jam size in this reach was the highest of the reaches surveyed (Appendix
A, Table 7). This was caused by the presence of six large bank-attached wetted LWJs at upstream end of the reach, upstream of the canyon. Minimal gravel bar habitat was present downstream
in the canyon which prevented natural LWJ accumulation (Appendix A, Table 6).
Four pools were observed in the Spruce Canyon Reach, including three pools formed by channel features and one formed by bedrock (Figure 25 Figure 26). The main channel of the Spruce
Canyon Reach had the highest percent pool area, though pool frequency and pool spacing were lower and higher than other reaches, respectively (Figure 18). This reach is largely confined
by natural features and presents little opportunity for restoration. Given the presence of bedrock throughout the reach, there is not much of an opportunity for wood recruitment, however,
management actions to protect the floodplain forest in the upper extent of the reach, upstream of the canyon to establish large conifers would increase long-term large wood inputs.
Spruce Canyon is largely confined, lacks off-channel habitat and is dominated by bedrock and boulders, indicating it likely serves as a migratory corridor rather than rearing or spawning
habitat for juvenile and adult salmon (Figure 26). This reach contains a long pool that runs through most of the canyon which would provide year-round holding habitat for the upstream
migration of adults and out-migration of juveniles.
/
Figure 25. Overview of the Spruce Canyon Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at the
Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as time allowed and were not a full census. Two side channel networks, shown in gray, were documented
as connected to the main channel at the time of the surveys but were not surveyed due to time constraints. The mapped photo is shown in Figure 26.
/
Figure 26. Example of a pool in Spruce Canyon with remnant pilings along the left bank. The canyon confines the channel and prevents the formation of off-channel habitat, and therefore
would primarily be used as a migration corridor for salmon.
Huelsdonk-South Fork Reach
The Huelsdonk-South Fork Reach extends from RM 27 to RM 30.5 below the confluence of the South Fork Hoh River and National Park Boundary (Figure 18; Map 7). The Huelsdonk-South Fork
Reach is a meandering and unconfined reach with abundant braid and side-channel habitat. The Upper Hoh Road runs along the main channel Hoh in two places in this reach on right bank.
Multiple residential properties exist on the left bank and various armoring techniques were present that attempt to confine lateral movement and reduce erosion. The reach has experienced
substantial lateral channel movement in the recent years and the main stem migration zone has widened considerably (Figure 28). Two side channels were observed, one of which was surveyed
that appeared to recently be the path of the main channel, demonstrated by the presence of large open gravel bars and dispersed LWJs throughout. Two of the side channels, including
the Lewis Homestead Side Channel Complex appeared to be more stable with more confined bankfull and wetted channels.
We surveyed 4.1 miles of main channel, 4.2 miles of braided channel, and 3.4 miles of side channels in the Huelsdonk-South Fork Reach (Appendix A). All wetted braided channels connected
to the main channel at the time of survey were surveyed, however, one large side-channel complex was wetted at the time of survey but was not surveyed due to time constraints (Figure
27). In main channels of the main stem, glides were the predominant habitat type and accounted for 43% of the total wetted habitat area, followed by pools and riffles which accounted
for 30% and 27%, respectively (Figure 18). In braided channels, riffles accounted for 46% of the wetted channel area, followed by glides, pools, and backwaters, which made up 27%, 26%,
and 1% of the wetted habitat area, respectively. In side channels, pools accounted for 37% of the total wetted channel area, followed by riffles and glides, which accounted for the
other 32% and 31% of the wetted channel area, respectively.
Edge habitat in the Huelsdonk-South Fork Reach was primarily made up of natural banks and bar edges, however, the Huelsdonk-South Fork Reach had the most hydro-modified edge length of
the reaches surveyed (Appendix A). Right bank modifications included rip rap along the Upper Hoh Road, and left bank modifications included bank armoring to protect residential property
and an active erosion zone along the Huelsdonk Property. Hydro-modified banks mostly occurred along the main channel of the main stem. The average slow water edge width associated with
bar edges was 4.3-ft. Natural bank edges and hydro-modified bank edges had average slow water edge widths of 4.1 and 3.9-ft, respectively.
The Huelsdonk-South Fork Reach had the most total pool area (325,878-ft2) of the reaches surveyed (Appendix A, Table 5). The total percent main channel pool area in this reach was the
highest of the reaches surveyed, however, main channel pool frequency was low with only 3.2 pools per mile (Figure 18; Table 8). Pools were abundant in braids and side channels; 44
and 40 pools were documented in braided channels and side channels, respectively. Channel features, including meanders and confluences, were main pool forming features in main channel
pools, but large wood jams and pieces drove pool formation in braids and side channels (Figure 28).
There were 223 LWJs mapped in the Huelsdonk-South Fork Reach, accounting for 1,170,159-ft2 of jam area, which was the largest total area of the mapped reaches (Appendix A, Table 7).
The reach also contained the largest amount side channel and island total LWJ area (Appendix A, Table 6). The Huelsdonk-South Fork Reach contained the largest jam mapped in the study
area; however, the median jam size was smaller than the Spruce Canyon and Morgan’s Crossing reaches (Appendix A, Table 7). The large total number and area of jams in this reach is likely
due to the proximity to the ONP, which provides a source of large old growth conifers to the river, as well as the abundant gravel bar and roughened bank habitat present to rack large
wood accumulations (Figure 28).
While pool frequency was low in this reach, pool area was high, and the abundance of braid and side-channel habitat provides abundant off-channel slow water habitat for juvenile salmon
rearing and refuge. Cobbles and gravels were the dominant substrate throughout the reach, which suggests adequate spawning habitat for coho, Chinook, and steelhead. The Huelsdonk-South
Fork Reach contains the most private property in the channel migration zone of the reaches surveyed and had multiple zones of erosion that are concerning both for their impacts on private
property and as sources of sediment and infrastructure debris into the river which could lead to fine sediment pulses and scour which could damage Chinook, coho, and steelhead redds
downstream, additionally fine sediment and clay can have negative impacts on rearing salmon. The large wood present in this reach does not appear to deter channel movement, which also
scours and damage redds (Montgomery et al. 1999). The main stem is also very wide across most of this reach and a large amount of shallow glide and riffle habitat without cover was
present (Figure 28). Even though slow water edge habitat widths were wider than other reaches for natural bank and bar edges, these large shallow expanses represent a potential risk
to juvenile steelhead and coho rearing by increasing risk to predation. They also represent large expanses of the reach that the lack cool deep slow water area that is preferred by
steelhead, in particular (Beechie et al. 2005).
Side channels in this reach contained a high pool frequency (37%) and would be the preferred rearing habitat of coho, steelhead, and spring Chinook. Cover, from large wood and overhanging
vegetation was observed throughout the side channels that would create complex habitat for feeding and predator evasion (Figure 28).
Due to the more moderated flows in side channels, they have more stable substrate. They also tend to have a higher frequency of pools and habitat diversity. For these reasons side channels
can provide preferred spawning in large river valleys. Since they are also associated with floodplain wetlands and slower velocities during peak flows, side channels are crucial rearing
habitat. Logjams can be the dominate mechanism in forming and sustaining side channels (Abbe and Montgomery 1996, 2003), and the presence of large trees in riparian/floodplain forests
is critical in the formation of logjams (Abbe and Montgomery 1996, Collins et al.
2012). These linkages demonstrate that critical salmon habitat (side channels) in large alluvial rivers is closely linked to riparian forests and wood.
Restoration actions, such as raising the road prism, planting, and placing large wood, aimed at enhancing the 0.4 miles of the Upper Hoh Road, which runs adjacent to the main stem of
this reach in two locations, would provide cover and create slow water habitat for salmon as well as protect infrastructure. Additionally, floodplain fencing and other bank stabilization
methods, including planting, could serve to prevent further erosion in the short-term. Numerous alder forested islands and meander bends were present in this reach, which present an
opportunity for conifer planting to speed up riparian succession and establish stable riparian forests. Lastly, this reach appears to have the most floodplain potential and relic side
channels could be identified for reoccupation to reduce pressure on priority banks, however, further assessment would be needed to assess the feasibility of this restoration action.
/
Figure 27. Overview of the Huelsdonk-South Fork Reach with results from CFS habitat surveys from September 28th to October 1st, 2020, at an average daily flow of 1,614 to 2,652 cfs at
the Highway 101 bridge (USGS Gage 12041200). Braids and side channels were surveyed as a representative subset of the habitat available as time allowed and were not a full census. Only
one side channel, shown in gray, was documented as connected to the main channel at the time of the surveys but was not surveyed due to time constraints. Photos are shown in Figure
28.
/
Figure 28. Reach photos with locations shown in Figure 27. 1) An example of a wide braided channel. 2) A side channel with alder banks and frequent channel spanning wood. 3) A side
channel that departs from the main channel with multiple root wads and vegetated banks with slow water refugia. 4) An example of a large log jam creating edge pool habitat in the main
channel.
Future Fish Habitat Survey Needs
That habitat surveys found abundant braid and side channel and main channel slow water habitat throughout the study area with adequate gravel, which suggests there is ample habitat for
juvenile Chinook, coho, and steelhead rearing as well as adult spawning. Logging efforts over the past decades have led to a decrease in old growth forest and thus the loss of long-term
stable forested islands and a more mobile channel bed (personal communication, Tami Pokorny), however, further studies are needed to fully understand the impacts on salmon reproduction
and survival. This survey aimed at evaluating the low flow habitat as of October 2020, with regards to meeting salmon needs, but further surveys aimed at quantifying historical and
present fish abundance and identifying limiting factors would be beneficial to further understand the changing conditions in the Hoh River. Additionally, evaluating egg-to-fry survival
and smolt trapping efforts would be useful to understand if overwinter survival of eggs or juvenile salmon is limiting success in the river (Roni et al. 2016).
Anticipated Trends
Aquatic habitat in the Middle Hoh River will change and redistribute within the reach as the channel and riparian corridor respond to changes in the flow and sediment regimes. As a result
of the changing climate the channel is
anticipated to have a higher and increasing channel migration and avulsion rates due to changes in hydrology. As these processes become more dominant it will lead to a wider channel,
as mature forests are lacking along the channel banks that could slow erosion. This wider channel, coupled with higher sediment loads associated with receding glaciers, will increase
the potential for channel aggradation and a more braided planform. The historic loss of large trees from the Middle Hoh valley has had the most significant impacts:
increased rates of channel migration,
reduced the probability of mature forests to develop due to frequency reoccupation by river,
increased the unvegetated width of the river, increasing solar radiation and decreasing flow depths
reduced the number and lengths of stable side channels
reduced complex edge habitat
reduced stable logjams and thus number pools and forested islands (and secondary channels)
increased the valley’s exposure to invasive plants (large gravel bars and seed source from nearby timber harvest units)
The cumulative effect of these changes will be a more simplified channel, with fewer and more shallow pools and unstable off-channel habitat as the channel freely migrates and avulses
through the floodplain. The transition of the riparian forests to younger and more deciduous dominant will reduce local LWD inputs as the few remaining small patches of mature timber
are eroded. Fewer large trees in the channel will further reduce the number of forced pools with cover, smaller and more few stable logjams in the channel and an overall more simplified
channel lacking hydraulic complexity.
All these changes will negatively impact habitat conditions in the Middle Hoh for salmon at the most critical times in their entire life history: rearing and spawning. Destabilized floodplains
frequently eroding cannot provide the off channel refugia of more mature floodplains with established forests and side channel networks connected to hyporheic flow, all needed for rearing
juveniles. Diminished instream complexity reduces hydraulic complexity and slow water refugia in the channel and the loss of large trees reduces instream cover, creating an unhospitable
main stem channel for rearing juveniles. Fewer and more shallow pools will further reduce juvenile refugia as well as holding spots for migrating adults. Spawning adults will also find
it harder to locate preferred spawning areas as sediment will tend to segregate by size less as hydraulic complexity diminishes and riffles are buried in sediment. Cumulatively these
impacts to aquatic habitat will most impact rearing juveniles as off-channel and instream habitats become increasingly degraded.
It is imperative to restore the large wood cycle in the Middle Hoh Valley to restore and sustain critical salmon spawning and rearing habitat, particularly found in floodplain side channels.
Transportation
Roads have been built throughout much of the Middle Hoh River valley. The primary paved transportation corridor, the Upper Hoh Road, provides access to Olympic National Park from State
Route 101 for hundreds of thousands of visitors each year along with the businesses that serve the tourists. A combination of gravel and paved roads also supports logging and mining
operations (primarily for road construction), farms, recreation and residential access. The development of this road network is the result of cumulative additions over decades that
largely did not consider the many geologic hazards present nor the dynamic natural processes associated with flooding and erosion. Maintaining vehicle access to residents, businesses,
ONP and to the river is crucial to the local community. The current alignment of several road segments within the CMZ have chronic maintenance
issues and impair natural processes and habitat formation and creation, limiting the resiliency of the transportation network that is relied upon while degrading habitat conditions.
To improve transportation and ecological resiliency roads planning should be a high priority. The Upper Hoh Road should be relocated outside the Resiliency Corridor where possible.
When it can’t be and is threatened by erosion, protection measures should create complex large-scale roughness IELJs) along the bank along with a narrow buffer that can be reforested.
Unimproved roads providing river and property access within the Resiliency Corridor can be maintained but will be at risk of washouts. Where new river access is needed it should be
planned to minimize risk by building in more stable areas less likely to flood or erode.
The purpose of this assessment is to inventory road segments vulnerable to erosion and flooding and to identify opportunities for redesign or realignment that would improve resiliency
consistent with the local community needs and desires. Road segments (DNR 2013) were identified within the resiliency corridor, delineated CMZ and the FEMA 100-yr floodplain in GIS
and are summarized in Table 9.
Table 9. Identified road segments within the Middle Hoh CMZ, Resiliency Corridor and FEMA 100-yr floodplain.
ROAD SEGMENTS
LENGTH IN CMZ (MILES)
LENGTH IN RESTORATION OCRRIDOR (MILES)
LENGTH IN FEMA SFHA (MI)
Unnamed
7.80
3.54
2.19
Upper Hoh Road and spurs
8.58
5.62
2.46
Old Milwaukee Rd
3.53
2.19
1.95
FR-H-1088 and spurs
1.33
0.04
-
FR-H-1009 and spurs
1.91
0.65
0.51
FR-H-1011 and spur
0.57
-
-
Lewis Ponds and spur
0.58
0.14
0.09
FR-H-1062 and spurs
0.11
-
-
FR-H-3700
0.06
0.00
-
FR-H-1060
0.04
-
-
FR-H-3900
0.02
-
-
Road Before Allen's Cutoff
0.21
0.21
-
Pit Road
0.01
-
-
Hoh Oxbow Boat Launch
0.02
0.02
0.02
Minnie Peterson Campground
0.14
0.14
0.14
Grand Total
24.92
12.56
7.36
These summary tables indicate there are a number of road segments that are currently at risk from flooding and/or erosion and that limit natural processes on the Middle Hoh. Strategies
for addressing these risks and their impact on the environment should consider the location of the segment relative to the resiliency corridor; segments within the resiliency corridor
should be prioritized for relocation or revisioning as they pose the greatest risk to both the human and ecologic communities. Segments within the CMZ and/or FEMA 100-yr floodplain
that are outside of the resiliency corridor could consider other alternatives to limit risk to the road, however planned relocation to a safer position on the landscape provides the
greatest resiliency as local protection typically requires maintenance and can fail.
Opportunities for improving resiliency within the Middle Hoh through road redesign and relocation are limited as much of the valley bottom is within the CMZ and/or FEMA 100-yr floodplain
and the surrounding hillsides are naturally unstable (Map 3). As such, many options need to be considered and assessed in more detail to identify preferred alignments or, in the long-term,
road alternatives. High traffic volumes and high demand for campsites are additional sources of stress and impacts to the area landowners and properties. During meetings with the Middle
Hoh steering committee and local landowners several ideas have been brought forward that illustrate the complexities and “out-of-the-box” thinking needed to consider current and future
risks, local and visitor access and safety, all while restoring, preserving and protecting ecologic function. Some examples include:
Move visitor access to the Middle Hoh to the Hoh Mainline on the south side of the valley, crossing the river at Spruce Canyon where the CMZ narrows with a new bridge. Once over the
new bridge on the north side of the river the route climbs out of the valley bottom to mid-valley slope elevations, heading east until gradually descending back to the valley floor
and joining the Upper Hoh Road east of Canyon Creek. This alternative would improve existing roads where possible to create the new alignment, allow decommissioning segments of road
that are within the CMZ and/or FEMA 100-yr floodplain that have been bypassed by the new road, and provide an alternative route in and out of the valley for emergency purposes. The
required bridging of the Hoh for this option would detract from ecologic resiliency as the span would need to be within the resiliency corridor, there would be impacts to upland ecosystems
establishing and improving the new alignment, and the cost would be very high.
Relocate those segments of the Upper Hoh Road that are within the CMZ and/or FEMA 100-yr floodplain away from the river, out of the valley bottom where necessary. New road alignments
traverse terrain with a low probability for slope failure if they ascend up the valley hillsides, subsequently descending to rejoin the Upper Hoh Road once the relocated segment has
been bypassed.
Establish alternative transportation methods for visitors to access the Middle Hoh and ONP. Creation of a new parking area near Highway 101 and creation of a shuttle bus or 4WD service
taking passengers on the Upper Hoh Road would reduce traffic and improve safety. Building recreational trails along the river corridor and/or through the hillsides leading up to ONP,
using a similar parking area near Highway 101 would serve a similar purpose. A bike lane could be added to the Upper Hoh Road providing yet another alternative method to access the
Middle Hoh and ONP. As new technologies come online in the coming decades, additional alternatives for moving visitors up the Middle Hoh valley will be realized that should be considered
as well. Any transportation method that does not rely on contact with or disturbance of the ground surface would provide the least impact and provide the greatest resiliency, if not
cost prohibitive. Low-altitude aircraft requiring minimal infrastructure of take-off and landing and tunnels are examples of technologies that are currently not developed sufficient
to be cost effective but may be in the future and should remain under consideration as implementation timelines allow.
A more complete evaluation and assessment of alternatives to improve transportation resiliency is needed to identify a preferred solution. Alternatives need to incorporate the needs
and desires of the local community, account for potential hazards associated with the alternative while improving ecologic and community resiliency in the Middle Hoh.
The current national trend of National Parks moving toward reservation and timed entry systems for day-use may change transportation needs in the future, potentially changing the number
of visitors to the area and where they look to recreate. Future anticipated changes in precipitation patterns need to be included in planning as they will alter the frequency and locations
of hazards posing risk to any alternative route under consideration. These include more extensive flooding and accelerated erosion adjacent to the river and more frequent slope failures
on steep valley margins. Site specific investigations can resolve these concerns once a preferred alternative has been identified.
The Hoh River Bridge on Highway 101 defining the downstream boundary of the Middle Hoh project reach was constructed in 1931 and is a vital link in the transportation network that serves
the Olympic Peninsula. The near 500-ft long span is only 20-ft wide, making for dangerous conditions as this is a major transportation route with numerous logging trucks, motorhomes
and trailers crossing every day. A replacement for this span should be included in any planning to improve resiliency for the community as it is a vital and vulnerable piece of infrastructure.
TRENDS & ANTICIPATED CHANGES
While the Middle Hoh River has experienced significant encroachment and manipulation, it remains relatively wild and undeveloped. The Middle Hoh also has the benefit of the Upper Hoh
River being in the largely undisturbed and protected ONP. These factors have given the Middle Hoh River resiliency from the other disturbances that have impacted habitat conditions,
primarily logging of the floodplain forests. Even with the loss of the old-growth forest within most of the Middle Hoh, large wood persists in the channel largely due to inputs from
the ONP. Projecting future conditions based on the best available science and observed indicators suggests changes to the Middle Hoh are coming and should be planned for accordingly.
Climate Change
Scientific studies in the Pacific Northwest region have concluded that the frequency and magnitude of peak flows will increase over the next 100-yrs while summer/fall low flows will
diminish (Hamlet et al. 2013; Mauger et al. 2016; Warner, Mass, and Salathé 2015; Hamlet and Lettenmaier 2007; Elsner et al. 2010, Isaak et al. 2017). Stream flow data for the Hoh River
from the USGS gage at the Highway 101 corroborate this finding of increased peak flows and diminished low flows (Hoh Tribe 2016). Impacts from the warming climate on flooding, vegetation,
soils and the length and severity of droughts (impacting low flows) are projected are occurring and are projected to worsen. Numerous studies for western Washington have indicated increases
in peak flows as early as the 2020s (inclusive of 2010-2039; Elsner et al. 2010, Mantua, Tohver and Hamlet 2010). Increases in peak flows continued for model extending to the 2040s
(inclusive of 2030-2059; Hamlet et al. 2013), 2050s (inclusive of 2040-2069; Mauger et al. 2016), and beyond. The warming climate effects are therefore relevant to the consideration
of all geomorphic and hydraulic processes related to flow duration, frequency, and magnitude, as well as forest composition, regeneration, rate of growth and risk from wildfire. Two
“heat dome” events linked to a slowing jet stream caused record heat in western Washington including the coast, as recorded at the Quillayute Airport (106(F) and Sol Duc River (108(F)
in June 2021, that led to anecdotal reports of widespread die off of south-facing conifer limb ends and terminal bud die back on young Douglas fir.
For the Hoh River watershed, extreme variability in rainfall intensity, duration, and seasonality can be expected. High rainfall in the watershed currently occurs primarily during the
fall and winter when Pacific cyclones cause prolonged, orographically enhanced precipitation. These storms can last for several days and are often linked to unseasonal warmth and flooding
in the Pacific Northwest. Colder temperatures at the higher elevations can lock some of this precipitation if it falls as snow, or if the snowpack has capacity to store rainfall. If
storms bring rapid rises in freezing level from warm air, or rain falls in areas with less snowpack storage, the amount of runoff the Hoh River receives can be dramatically higher,
compounding flooding and magnifying channel responses.
The impact of the climate crisis on river dynamics in the project area were analyzed by projecting peak flow estimates into the future based on the work done by the University of Washington
Climate Impacts Group project (Tohver et al. 2014) and are included in Table 10. Projections for increased peak flow magnitudes with the warming climate are driven by projections for
more intense precipitation, particularly in the fall and winter months. Higher precipitation rates will increase the magnitude of annual peak and average winter flows and contribute
to increases in mass-wasting and sediment delivery to the river network, both upstream and within
the project area. These increases in flow will expand flooding extent in the Middle Hoh and the increased sediment loads may aggrade the channel bed, further exacerbating flooding as
the channel bed and water levels rise. The increase in annual peak flow magnitude will also contribute to increases in channel migration rates and erosion as more flow equates to more
stream energy and capacity to do work. The result of this increased erosion will be to further limit the availability of off-channel habitats, as they will be re-worked by the river
more frequently, impacts to infrastructure will be more frequent and of greater magnitude, and private property loss should be expected to continue without mitigating measures.
Table 10. The magnitude of future peak flows in the Hoh River for 2070-2099, projected as result of the climate crisis under IPCC A1B scenario (broadly representing “business as usual”
through 2050, IPCC 2000).
RECURRENCE INTERVAL
PRESENT DISCHARGE ESTIMATE
(CFS)
PERCENT INCREASE DUE TO CLIMATE CHANGE
FUTURE (2070-2099) DISCHARGE ESTIMATE
(CFS)
1-yr
12,280
14 – 34%
13,975 – 16,427
10-yr
52,300
14 – 34%
58,574 – 68,850
100-yr
73,600
14 – 34%
82,140 – 96,551
The warming climate will also result in lower base flows in the late summer and fall. This will diminish the wetted channel area and depths that will decrease available habitat. Warmer
air temperatures will also result in warmer water in the river, further stressing fish (Mantua et al., 2010; Isaak et al. 2017, Winkowski 2020) and making riparian shade and in-stream
cover even more important for salmon. While these changes were not quantitatively evaluated by the scope of this study, they further underscore the importance of increasing the ecological
resiliency of the project area. This could be accomplished by increasing stream shade, the quantity of off-channel habitats and deep pools, as well as increased surface water/ground
water interactions through increased floodplain connectivity and added roughness, so that the predicted increases in stream and air temperature can be partially offset.
Sediment Sources
Sediment transport changes and trends can manifest in changes to the channel and associated geomorphic processes, and can be divided into two general categories:
Supply Limited (leading to incision and straighter channel with lower erosion rates)
Transport Limited (leading to deposition and braided channel/wander channel forms with higher erosion rates)
Channel forms observed in the air photo record suggest that past century sediment transport conditions and trends are neither predominantly supply nor transport limited. The 2004 BOR
hypothesizes that sediment transport conditions were generally balanced. Field conditions identified that while the sediment supply is robust, the Hoh River has the ability to transport
this supply over time. Field observations indicated that some localized areas are experiencing net deposition (transport limited) conditions, whereas observations at other areas indicated
that the river can be locally supply limited. Areas upstream of confined reaches, such as above Spruce and Oxbow Canyons show evidence of sediment aggradation leading to a more dynamic
channel, whereas other areas, like below Spruce Canyon the channel is stable or degrading, due in part to confinement from the Upper Hoh Road. We expect the general trend to be the
expanding of and new aggregational areas, as new sediment is liberated due to glacial recession and transported downstream.
The long-term trend since deglaciation (end of the Pleistocene) has been one of net supply limited conditions which has driven incision and resulted in stranded alluvial terraces. Vertical
changes in the channel can influence
channel migration. Primary drivers in vertical changes in the channel are sediment supply, roughness, and hydrology. Within the study reach, the net, long-term Holocene trend was incision
and evidence of the downward vertical movement of the channel is supported by the presence of numerous terraces within the valley. The incision is controlled in part by the bedrock-lined
canyons. Over time, the channel slowly incises into the bedrock grade controls and the channel lowers. Periods of aggradation (deposition) have likely occurred within the overall long-term
incision trend. During these periods, channel migration areas would be expected to widen. Based on the stream gradient and active erosion observed within the bedrock- controlled areas,
the long-term incision trend is likely continuing. However, this trend is very slow and channel migration processes over the 100-yr planning horizon will be more influenced by periods
of aggradation, channel widening and instability.
Rapid sediment delivery to the channel can occur from large storms, landslides, and glacial retreat associated outwash events. These short-lived, events delivering sediment to the channel
often take the form of a pulse of sediment that may take years propagate downstream. Once in the fluvial system, the sediment is temporarily stored in the channel and floodplain and
periodically remobilized through scour and lateral erosion processes, creating temporal increases in sediment load that can further supplement sediment pulses as they migrate episodically
downstream. Conversely, longer term stabilization of some sediment supply is possible with vegetation establishment. This is true within the channel migration zone as well as in the
sediments exposed as glaciers retreat. The variability of event driven processes and the uncertainty of future hydrologic conditions makes predicting sediment conditions problematic;
however the trends point to a shift to increased sediment loads as the climate warms, driven by more mass wasting and glacial retreat exposing and releasing more available sediment.
The fate of this sediment, where it will accumulate in the system at a given time and the magnitude, will be a function of a number of variables that are dynamic in nature. Intermediate
and longer-term planning (5 – 25-yrs) should consider a wide range of variability in sediment loading and transport over time. The overarching trend in the short to intermediate-term
is for higher sediment loads that could elicit a channel response in the Middle Hoh. Indicators that suggest sediment storage is growing include an increase in the rate of avulsions,
channel migration, channel braiding and an increase in flood elevations and extent.
All aspects of geomorphic (landslides, channel avulsions, channel migration) and hydrologic response to the changing climate will provide an advantage to non-native invasive plants,
which thrive on disturbed and nutrient-poor soils. (Naiman et al 2010, Carter et al 2019). Wider flood inundation carries invasive propagules farther out in the river floodplain and
braided channel complexes (Martinez et al. 2016). In these habitats knotweed (Polygonum spp.) and reed canarygrass (Phalarus arundinacea) thrive, altering structure and flow, litter
fall, and insect prey production, and arresting growth of trees for large wood and shade. Where these disturbances erode riverbanks and damage roads, Scotch broom seeds introduced during
reconstruction of river adjacent roads and restoration projects spread to adjacent gravel bars, where the leguminousnitrogen-fixing species thrives with deep roots reaching hyporheic
flow (Carter et al. 2019). Landslides initiating in infested roads and harvest units transport propagules of Scotch broom and other non-natives (e.g. foxglove (Digitalus purpurea),
tansy ragwort (Jacobaea vulgaris), European blackberry (Rubus bifrons, R. lacinatus) downstream channels to the main stem river. Reed canarygrass seeds introduced in hay and straw used
for erosion control are transported down ditches into streams, side-channels, and the river. Where invasive species establish and expand, a continuing feedback loop replacing native
species reduces shade, humidity, large wood, leaf litter, insect prey, and other services provided by mature resilient native floodplain and riparian forests.
Forests
Restoration of native forests, particularly the old-growth patch mosaic that once characterized the Middle Hoh River valley is essential to sustaining salmon habitat and building climate
resiliency. Forest restoration within the river valley requires restoring the large wood cycle that depended on the big trees the valley was famous for. Those big trees formed stable
logjams that formed “islands” within and along the margins of the channel
migration zone on which trees could grow old and sustain the supply of big wood (Figures 3b, 3d). Since big trees are rare, it will take hundreds of years to restore the river’s ecosystem
without intervention to emulate the function of the big trees. Restoration needs to include construction of stable logjams to restore the forest patch mosaic and side channel network
of the river. This has been successfully done in the Upper Quinault River (FIGURE), Lower Elwha River, Dungeness River, South and North Forks of the Nooksack Rivers, and other rivers.
Restoration actions on these rivers have not only restored channel anabranching and riparian forests, but have helped protect local communities and infrastructure. It must be accompanied
by native conifer planting and invasive species control. All these actions are needed to restore a resilient river corridor that will sustain keystone species such as salmon as well
as protecting the local community.
Climate change is impacting forests; increasing temperatures, reducing snowpack and changing west side Olympic Peninsula precipitation patterns resulting in more winter flooding, and
lower summer stream flows (Halofsky et al. 2011). More extreme temperature regimes and low snowpack have resulted in severe drought years such as in 2015, when the Paradise Fire,
a rare extensive coastal temperate rain forest fire, occurred in the Upper Queets River basin burning more than four square miles of rainforest (2,796 ac) (Current Fire Status - Olympic
National Park (U.S. National Park Service) (nps.gov)). And as this document was under preparation the H-1500 Maple Creek fire burned 70-ac in August 2021. The potential for wildfires
within the temperate rain forests of the Olympic Peninsula are projected to increase (Perry et al. 2015; Halofsky et al. 2011). Forested slopes at middle elevations are particularly
vulnerable to fires initiated by lightning during dry conditions and have the potential to spread rapidly upslope. Extensive burns reduce slope roughness and increase surface water
runoff that, in turn, may impact water quality and geomorphic processes in the floodplain below. Fortunately, where they exist mature riparian forests are naturally resilient to adjacent
forest fires due to the locally higher levels of humidity. Mature forested floodplain side channel networks are essential to provide cooler thermal refugia for salmon as the climate
warms. In addition to climate increased fire frequency, channel migration rates are projected to increase that threaten to erode and carry invasive plant propagules into the riparian
forests.
Removal of much of the Middle Hoh River riparian forest in the 20th century has reduced the area of mature floodplain forest and associated side channel network–the area of mature side
channel aquatic habitat. Mature riparian forests are the source of large conifer riparian trees–keystones of mature riparian and aquatic ecosystems. Logging of the mature riparian forest
resulted in a dramatic acceleration of channel migration rates and replacement by red alder dominate riparian forests with few conifers to replace the large trees that were, prior to
logging, common along the banks of the Middle Hoh. The loss of large keystone coniferous trees has further resulted in loss of large wood jams, in-channel habitat, and forested islands,
all contributing to the decline in salmon populations. More recent conversion to Scotch broom in reaches downstream of bank revetments and landslides from invaded timberlands has reduced
early successional species necessary to provide organic matter and soil nutrients for mature riparian forest development (10KYI, Carter et al., Grove et al.).
Wood jam islands are stable refugia for establishment of riparian conifers. Alder-dominated floodplains without wood jam hard points, together with climate driven increases in channel
migration, portrays a future floodplain forest continuously dominated by red alder or Scotch broom. Without large wood jams very few floodplain conifers will become established as the
channel migrates freely across the valley bottom. Climate increased channel migration will in turn require larger trees to create stable wood jams. In effect the current and future
river and riparian forest ecosystem has shifted from an historic state of mid-channel mature forested islands; anastomosing primary and secondary channel network and patchwork of mature
floodplain forests; to a simplified more braided floodplain state dominated by young red alder and characterized by a lack of stable large wood. This scenario will result in a future
Middle Hoh River–likely on the order of hundreds of years to
recover–in which large wood generated patchwork channel and floodplain forest processes are lost resulting in continuous degradation and loss of both in-channel salmon habitat and riparian
forest fish and wildlife habitat.
Invasive Species Trends
Invasive non-native plants are a threat to forest succession, aquatic habitat formation and climate resiliency on the Hoh River. After two decades of action to prevent and control knotweeds
(Fallopia spp.), Scotch broom (Cytisus scoparius), reed canarygrass (Phalarus arundinacea), herb Robert (Geranium Robertianum), and Canada thistle (Circium arvense), significant seedbanks
and new introductions by a variety of activities in the watershed require river-wide persistent and continuous prevention and control activities.
The climatic, hillslope, and riverine disturbance regimes that impact the watershed’s natural and built infrastructure create conditions ideal for the invasion of non-native invasive
plants, which out-compete native plants – especially the early successional species – willow (Salix L.) and red alder (Alnus rubra). These invasions delay and impede floodplain forest
development, degrade forest function and services, and require significant, persistent, and continued investment and implementation of active control practices.
Persistent work has been conducted over the past two decades to prevent the spread of non-native invasive Eurasian plants and habitats can be resilient to these impacts where protected
and restored. Successive projects have been implemented beginning in 2003, resulting in benefits to the local ecosystem through increased protection from competition for space, nutrients,
and water; and a river corridor that continues to provide foundational ecosystem services of native plant communities.
Invasive plants move between multiple ownerships with differing capacity, interest, legal authority or responsibility for control in each watershed on the coast. Layered on that ownership,
rivers and roads connect between watersheds, and invasive species move down these pathways. Each entity conducts some weed control, but all lack sufficient resources, interest, or
strategies to address species moving across their ownership boundaries via wind, water, construction, or traffic. When small Scotch broom, knotweed, reed canarygrass, everlasting peavine
(Lathyrus latifolia), and others are eliminated from roadsides, source populations are prevented from traveling down ditches and through culverts to streams, where water transports
each seed to bare gravel and banks which are ideal environments for invasion.
Invasive species degrade all habitats and restoration strategies. A common assumption is dense forest shade will control invasive plants. Our experience shows that none are ever fully
eliminated by shade, and a single plant that contributes its load of seed to wind, water, tires or hooves will start new populations that must be located and prevented from seed production
once again.
“Lag time” as described by Crooks et al., is the process of introduction and explosion of invasive species into ecosystems, and those where environmental conditions are deteriorating
or disrupted are expected to experience an increased rate of invasion over time. Decades may pass before there exist enough propagules in a watershed such as the Hoh to cause an explosion.
Knotweed was stable in planted locations in the upper Middle Hoh until a channel avulsion related to a high flow event in or around 1999 and after 20-yrs of treatment, continues to
be found along the entire corridor in single stems and clumps. European blackberry has been present since homesteading in the early 1900s, and seeds are spread by frugivorous birds,
generalists such as American Robin (Turdus migratorius) and European starling (Sturnus vulgaris). The observed ‘lag time’ since introduction for European blackberry, knotweeds, and
Canada thistle in the Hoh watershed is approximately a century post-homesteading, and for Scotch broom, herb Robert and reed canarygrass, two decades since introduction through road
and river revetment construction, mowing and hay introductions.
Invasive Species Introductions and Persistence in the Hoh Watershed
Funding for invasive species other than knotweed did not become available until 2011, after multiple introductions, especially during years of Upper Hoh Road bank protection and repair.
With multiple species in multiple sites spreading for a decade unimpeded, googles of propagules (seeds and fragments of root, rhizome, or stem) capable of producing a new plant) ‘banked’
in the substrates of terraces, floodplains, river bars and banks, and along roads in the middle Hoh watershed. These propagule banks are continual sources of all species, and are easily
moved in river migration, scour, deposition, or by equipment.
Programmatic invasive plant prevention and control in the watershed
10KYI has worked for two decades to reduce the spread of non-native invasive Eurasian plants so that habitats can be resilient to these impacts where protected and restored. Work conducted
since 2003 has resulted in benefits to the local ecosystem through increased protection from competition for space, nutrients, and water; and has resulted in a river corridor that continues
to provide the ecosystem services of native plant communities.
Invasive plants move between multiple ownerships with differing capacity, interest, legal authority or responsibility for control in each watershed on the coast. Layered on that ownership,
rivers and roads connect between watersheds, and invasive species move down these pathways. Each entity conducts some weed control, but all lack sufficient resources, interest, or
strategies to address species moving across their ownership boundaries via wind, water, construction, or traffic. When small Scotch broom, knotweed, reed canarygrass, everlasting peavine
(Lathyrus latifolia), spotted jewelweed (Impatiens capensis) and others are eliminated from roadsides, source populations are prevented from traveling down ditches and through culverts
to streams, where water transports each seed to bare gravel and banks that offer ideal environments for invasion.
Invasive species degrade all habitats and restoration strategies. A common assumption is dense forest shade will control invasive plants. 10KYI’s experience demonstrates that none
are ever fully eliminated by shade, and that shade is temporary on the dynamic Hoh River. A single plant that contributes its load of seed to wind, water, tires or hooves will start
new populations that must be located and prevented from seed production once again.
Table 11. Objectives and strategies necessary to achieve the resiliency goal of restoring native vegetation.
/
Table 12. The Invasive plant species propagation, seed number, seed viability, and allelopathy (with sources)
INVASIVE PLANT SPECIES
PROPAGATES VIA
ANNUAL NUMBER OF SEEDS
PER PLANT
SEED VIABILITY
ALLEOPATHIC
Scotch broom
Seeds
Average: 9,650 seeds1
Maximum: 18,000 seeds2
Average: 5 – 20-yrs1
Maximum: 60-80-yrs2,1
Yes9
Knotweed
Seeds*
Rhizomes
Nodes
Average: 50,000 seeds per stem8
Maximum: 150,000 seeds per stem8
Average: 1 – 2-yrs8
Maximum: 15-yrs7
Yes10
Reed canarygrass
Seeds
Rhizomes
Nodes
Average: Varies between populations14
Maximum: 600 seeds3
Average: 1 – 6-yrs14
Maximum: 30-yrs14
Suspected11; more research is needed
Herb Robert
Seeds
Average: 500 - 1,000 seeds4
Maximum: 1,550 seeds16
Average: 3 – 5-yrs15
Maximum: 6-yrs16
Yes4
Canada thistle
Seeds
Rhizomes
Average: 1,000 - 1,500 seeds5,6
Maximum 5,300 seeds6
Average: 3-5-yrs17
Maximum: 20-yrs5
Suspected12; more research is needed
*Bohemian knotweed13
Invasive Plant Control in the MHRP by River Reaches
The 10KYI Hoh River Invasives Species Prevention & Control Program (ISPCP) divides the river into floodplain complexes and reaches associated with tributaries. Over the past two decades
of invasives control on the Hoh River, data has been collected based on reaches and floodplains named for tributaries or other features as follows:
Table 13. Middle Hoh River Resiliency Plan reaches relative to ISPC reaches
MIDDLE HOH RIVER RESILIENCY PLAN REACHES
HOH RIVER ISPC PROJECT REACHES
Oxbow Canyon
Oxbow Canyon, Hell Roaring Creek
Willoughby Creek
Alder Creek, Winfield Creek, Elk Creek and Elk Creek Floodplain, Schmidt Bar, Peterson’s Floodplain, Lindner Bar (mid to lower)
Morgan’s Crossing
Upper Lindner Bar, High Bluffs, Rock Creek, Clear Creek, Tower Creek
Spruce Canyon
Canyon, Maple Creek, Pole Creek, Dismal Creek
Huelsdonk - South Fork
Owl Creek, Spruce Creek, Spruce Flats, Canyon Creek, Fletcher Island, Lewis Channel, Fletcher Ranch, Brandeberry Lots, Richmond Island, Upper Brandeberry Island, Crippen Homestead
Invasive Plants in the Hoh Watershed – Risks and Hazards by Site and Activity:
Table 14. The Risk and hazards: invasive plant spread via streamflow, roads, equipment, hillslope processes, and weather
/
*Level of risk and hazard: High (H), Moderate (M), Low (L), Not applicable (N/A)
**Propagules: Rhizome (), Fragment-Stem (F/S), Fragment-Root (F/R), Seed (S)
Invasive Plants in the Hoh Watershed – Species, Sources, Impacts
Species in the watershed
Knotweeds, Scotch broom, reed canarygrass, herb Robert, Canada thistle, tansy ragwort (Jacobaea vulgaris), non-native blackberry Himalayan or Armenian (Rubus bifrons) and Evergreen or
cut-leaf (Rubus lacinatus), and other invasive plants and noxious weeds are widely distributed at different densities throughout the Hoh watershed.
Points and polygons collected during field surveys document plant species, phenology, site type, treatment and more, collated into maps and databases since 2003. To support resiliency
planning and restoration project
implementation in the Middle Hoh, layers of invasive species that impact growth of native forest have been overlain onto the MHRP Riparian and Floodplain Forest maps (Map 5, Map 8),
with examples from Morgan’s Crossing reach provided in Figure 30 and in true color aerial imagery (Figure 29):
/
Figure 29. Scotch Broom infestation in the Morgan’s West Reach.
/
Figure 30. Scotch Broom and forest type cover of the Morgan’s West Reach.
Sources
Knotweed, Canada thistle and the blackberry species have been present since homesteading in the early 1900’s. Knotweed was originally planted at homesteads as an ornamental and to provide
forage for honeybees late into the fall. Canada thistle and tansy ragwort were most likely accidental introductions in livestock feed in earlier landings of Europeans to the coast (NRCS).
The European blackberry species – Himalayan (Rubus bifrons), and Evergreen (R. lacinatus) were planted for their large and prolific berries. Some reports suggest that reed canarygrass
was introduced at homesteads to provide shade-tolerant and fast-growing forage for livestock among the huge stumps from cleared old-growth forest (NRCS). Reed canarygrass was observed
in the river corridor in 2009 (Map 8, Figure 32) and is believed to have been more recently introduced in hay or straw used for livestock or roadside erosion control, or field and roadside
mowing equipment used in other watersheds and then in Hoh valley hayfields. Herb Robert was first observed in the mid-90’s in Olympic National Park at the
Snider Creek equipment yard – likely unintentionally brought from the north and east Peninsula where it has been established. Until treatment is moved upriver, this species will continue
to move into MHRP reaches.
Recent introductions in road and bank protection work in the middle and upper river are everlasting peavine (Lathyrus latifolius), spotted jewelweed (Impatiens capenisis), wild chervil
(Anthriscus sylvestris), and Queen Anne’s lace (Dacus carota). Each of these species are quickly caught and treated, but boots, vehicles, mowing and construction equipment continuously
introduce propagules that must be controlled before seed production.
Table 15. The Condensed ac of invasive plants inventoried and treated in 2020 within the Middle Hoh Resiliency Plan (MHRP) reaches.
MHRP REACHES
KNOTWEED
SCOTCH BROOM
REED CANARYGRASS
HERB ROBERT
CANADA THISTLE
Oxbow Canyon
0.0004 ac
0.0004 ac
0.00 ac
0.00 ac
0.00 ac
Willoughby Creek
0.01 ac
8.62 ac
0.11 ac
0.007 ac
0.09 ac
Morgan’s
0.004 ac
4.19 ac
1.54 ac
0.90 ac
0.17 ac
Spruce Canyon
0.00 ac
0.00 ac
0.0002 ac
0.02 ac
0.01 ac
Huelsdonk/South Fork Hoh
0.01 ac
2.28 ac
3.83 ac
1.35 ac
0.22 ac
Upper Hoh Road
0.00 ac
0.01 ac
0.02 ac
0.06 ac
0.00 ac
Total
0.03 ac
15.1 ac
5.51 ac
2.33 ac
0.49 ac
Impacts by these species to the development and resiliency of native forest communities in the Hoh watershed
Invasive plants impede and interfere with the germination, growth, and health of native plant species. Through a variety of adaptations, each of the listed non-native Eurasian species
establish and spread in the varying degrees of moisture, flooding, and low nutrient disturbance zones in gravel bars. Each species form monocultures that arrest the growth of native
species, and reduce the space, soil nutrients, water availability, and light, replacing the functions and services provided by native plants including food, structure, and shade.
Four species are particularly impactful in the Hoh watershed’s managed forests and river floodplains to a range of native plants, fish, and wildlife:
Scotch broom (Cytisus scoparius) – Arrests the growth of early successional riparian plants which set the ecological stage for conifer forests. Outcompetes and replaces native shrubs
used by pollinators, doesn’t feed native insects, reducing forage for frugivorous and insectivorous small mammals and birds. Alters soil chemistry with nitrogen fixation and additional
requirement for phosphorus. Toxic to grazers – elk, deer, livestock. Reduction in organic contribution for soil development (loss of leaf litter from red alder). Flammable volatile
oils are reported to cause allelopathy in soils.
A SB monoculture grows to 10-ft tall, producing 10,000 seeds per plant per year, which may remain viable for 80-yrs. SB is shown to arrests native forest succession through a variety
of influences, replacing early successional native willow, alder, and in time, the conifer forests providing forage, shade, nutrients, and other services including large woody debris,
bank stability and sediment filtration to keep streams and rivers and forests naturally healthy (Grove et al. 2017 and 2017b, Slesak et al. 2016). SB alters the composition of several
important soil nutrients and disrupts the mycorrhizal fungi critical to healthy forest growth (Grove). It is mildly toxic to herbivores, which avoid it. It is also more flammable than
native shrubs and trees. Scotch broom survives both floods and drought, influencing sediment transport and storage, and morphology of bars and terraces. Scotch broom possesses several
traits that make it tolerant and avoidant of drought conditions: high root length density, low leaf area to root mass ratio, low specific leaf area, photosynthetic stems, and a drought-
deciduous phenology (Carter et.al. 2019, Bannister 1986; Bossard and Rejmanek 1994; Matı´as et al. 2012; Boldrin et al. 2017). The bushy branches and long roots anchor the plants in
mobile river bars, reaching hyporheic below the bar surface, while collecting fine sediment in high water events.
A report of the Hoh Tribe paddling canoes across the river to the Huelsdonk homestead on the north side in the 1940’s to put out a fire in Scotch broom is an indication that species
was present, but it was not widely observed in the river itself until after large scale road revetment work began in 1996 through 2007 on the Upper Hoh Road, and 1999 and ongoing on
Oil City Road, and 2004 on SR 101. All these construction sites introduced Scotch broom to the river, which started to establish by 2007, and continues to produce googols of seed, despite
continuous control effort.
Table 16. The Condensed ac of Scotch broom inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches.
MHRP REACHES
SCOTCH BROOM
2017
2018
2019
2020
Oxbow Canyon
NS
0.006 ac
NS
0.0004 ac
Willoughby Creek
4.24 ac
14.85 ac
3.91 ac
8.62 ac
Morgan’s
1.20 ac
4.99 ac
2.69 ac
4.19 ac
Spruce Canyon
NS
NS
0.00 ac
0.00 ac
Huelsdonk/South Fork Hoh
0.0006 ac
1.12 ac
2.82 ac
2.28 ac
Upper Hoh Road
0.16 ac
0.04 ac
0.02 ac
0.01 ac
Total
5.61 ac
21.00 ac
9.45 ac
15.10 ac
NS = Reach not surveyed
/
Figure 31. Summary of Scotch Broom acreage for reaches of the Middle Hoh valley.
Reed canarygrass (Phalarus arundinacea) Impedes access to off-channel habitats and fills floodplain side-channels, wetlands and channel margins. Slows and warms water, reducing dissolved
oxygen and cold water refugia. Reduces production of native insect and amphibian prey species utilized by juvenile salmon (Weilhofer et al., WDFW).
Table 17. The Condensed ac of reed canarygrass inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches.
MHRP REACHES
REED CANARYGRASS
2017
2018
2019
2020
Oxbow Canyon
NS
0.002 ac
NS
0.00 ac
Willoughby Creek
0.18 ac
0.19 ac
0.03 ac
0.11 ac
Morgan’s
1.76 ac
0.002 ac
1.57 ac
1.54 ac
Spruce Canyon
NS
NS
0.00001 ac
0.0002 ac
Huelsdonk/South Fork Hoh
0.0001 ac
0.0008 ac
0.01 ac
3.83 ac
Upper Hoh Road
0.04 ac
0.002 ac
0.008 ac
0.02 ac
Total
1.98 ac
0.20 ac
1.62 ac
5.51 ac
NS = Reach not surveyed
/
Figure 32. Summary of Reed Canary grass for reaches of the Middle Hoh Valley.
Knotweeds (Polygonum spp.) Limits growth of early and later successional native riparian plant species through alteration of soil nutrients, dense shade during the growing season (Urgenson,
2009, Claeson et al. 2013)), and loss of woody structures to slow winter flow and protect banks from erosion. Causes loss of litterfall benefitting macroinvertebrates, and thus stream
biota, and loss of forage for grazing mammals. However, European honeybees and native bald-faced hornets are heavy users of late fall blossoms.
Knotweed rhizomes are documented to extend 5 meters into substrates and 10 meters laterally from each clump or stem. A 2cm fragment of rhizome can produce a new plant. Each node on
the cane (stem) will root and sprout in contact with water or damp soil. Rhizomes 4 meters long and 1 meter wide have been observed on the Hoh River, exposed by river scour. Treating
small leaves growing from these huge rhizomes does not translocate enough herbicide to effectively control them. Plants may emerge as late as October from deeply buried fragments, which
is too late to treat effectively in wet conditions and higher water. Thus, nearly two decades after treatment began, knotweed persists in the watershed at very low levels – 2020 surveys
found scattered plants and clumps totaling 0.12-ac in area over 3400-ac of floodplain, river bars and banks searched between RM 30 and RM 0.5. Three sites where knotweed had not been
found in 2 to 7-yrs are Brandeberry Island, Elk Creek side-channel, and Clear Creek side-channel, and all have knotweed again in 2021 (Map 8).
Table 18. The Condensed ac of knotweed inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches.
MHRP REACHES
KNOTWEED
2017
2018
2019
2020
Oxbow Canyon
NS
0.00002 ac
NS
0.0004 ac
Willoughby Creek
0.0006 ac
0.0001 ac
0.006 ac
0.01 ac
Morgan’s
0.009 ac
0.02 ac
0.07 ac
0.004 ac
Spruce Canyon
NS
NS
0.00 ac
0.00 ac
Huelsdonk/South Fork Hoh
0.0005 ac
0.00 ac
0.01 ac
0.01 ac
Upper Hoh Road
0.000006 ac
0.0001 ac
0.0006 ac
0.00 ac
Total
0.01 ac
0.02 ac
0.09 ac
0.03 ac
NS = Reach not surveyed
/
Figure 33. Summary of Knotweed for reaches of the Middle Hoh Valley.
Herb Robert Weakly perennial – largely annual. Limits growth of native forest understory plants through strong allelopathy (Mardani et.al. 2016), reducing browse for herbivores and
nectar for native pollinators. Research is needed to evaluate effect on native woody species. Sticky seeds remain dormant for 5 yrs (Jones and Reichard 2009). Seeds are transported
by water, animals, and boots. Seedlings germinate as mature plants are removed. The species begins flowering in February, forms seeds by April, and continues growth and seed production
through November.
Table 19. The Condensed ac of herb Robert inventoried and treated between 2017-2020 within the Middle Hoh Resiliency Plan (MHRP) reaches.
MHRP REACHES
HERB ROBERT
2017
2018
2019
2020
Oxbow Canyon
NS
0.00 ac
NS
0.00 ac
Willoughby Creek
0.23 ac
0.92 ac
0.0003 ac
0.007 ac
Morgan’s
0.76 ac
0.86 ac
4.44 ac
0.90 ac
Spruce Canyon
NS
NS
0.01 ac
0.02 ac
Huelsdonk/South Fork Hoh
0.24 ac
0.46 ac
2.45 ac
1.35 ac
Upper Hoh Road
0.17 ac
0.02 ac
0.75 ac
0.06 ac
Total
1.40 ac
2.27 ac
7.65 ac
2.33 ac
NS = Reach not surveyed
/
Figure 34. Summary of Herb Robert for reaches of the Middle Hoh Valley.
These non-native species each tolerate a range of conditions in shade or sun, moist or dry; growing at varying rates and to different heights depending on site characteristics. Similarly,
in different site conditions, each can reach maturity at different times during each year, necessitating repeat surveys and treatments each year, depending on river flow, season, seed
bank, and weather
Table 20. The Invasive plant survey and treatment timings based on plant life histories, treatment methods, and site and weather conditions
INVASIVE PLANT SPECIES
SITE TYPE
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
DEC
Scotch broom
Upland
X
X
X
X
X
X
X
X
X
X
X
X
Riparian
X
X
X
X
X
X
X
X
Knotweed
Upland
S
S
S
X
X
Riparian
S
S
S
X
X
Reed canarygrass
Upland
S
S
S
S
X
X
X
X
X
X
S
S
Riparian
S
S
X
X
X
X
X
X
Herb Robert
Upland
X
X
X
X
X
X
X
X
X
X
X
X
Riparian
X
X
X
X
X
X
X
X
Canada thistle
Upland
X
X
X
X
X
X
Riparian
X
X
X
X
X
X
S = Plant survey only
X = Plant survey and treatment
/
Figure 35. Summary of invasive plant treatment timing throughout the year.
Desired Future Conditions
A key element to the development of this resiliency plan is to identify desired future conditions, to provide a target through which we can develop a series of steps to achieve. Incorporating
our understanding of the reach with projected and anticipated changes provides an opportunity to improve resiliency in the Middle Hoh. Of particular interest are how restoration of
riverine processes, improvements to roads or changes in landuse and public access could improve resiliency for the community and environment. Short-, intermediate- and long-term desired
future conditions statements for each of these categories are provided that include overarching goals and specific objectives to meet these conditions.
Resiliency Corridor
The core of our approach to the development of resiliency planning for the reaches of the Middle Hoh is the delineation of a resiliency corridor, which estimates the minimum spatial
footprint required to sustain habitat-forming processes along the river corridor. The processes that form and sustain aquatic habitat in a western Washington river system – lateral
channel migration, channel avulsion (i.e., rapid channel relocation), flow splitting, overbank inundation – are dynamic and require both sufficient space and a healthy riparian forest
to function (Collins et al., 2003; 2012).
NSD utilized an examination of the geomorphic, hydraulic, and aquatic habitat parameters to define the Resiliency Corridor for the study reach. The resiliency corridor was defined as
the entire valley bottom throughout the project reach based on following:
The 100-yr floodplain which includes areas frequently subject to flood inundation,
The channel migration zone,
Floodplain topography and geomorphology, including areas with clear evidence of historical alluvial channels (GMZ),
The meander bend amplitude and frequency.
The boundary of the resiliency corridor should be viewed as a planning tool to guide decision making for actions occurring within the corridor. Actions taken within the corridor should
consider their impact on natural process functions contributing to available high-quality habitat. Areas within the corridor that are undeveloped should remain so, be evaluated for
invasive plant species presence and need for restoration. Where natural processes are actively creating and supporting aquatic habitat and/or climax riparian communities exist the area
should be targeted for preservation and invasive species monitoring. Where development exists within the corridor there is an opportunity to improve resiliency for the Hoh River, and
if possible, restoration should be the preferred alternative. If restoration is not achievable, alternatives that minimize impacts and/or mitigate for impairments to natural processes
should be pursued to the extent possible.
Acquisition and land conservation provide for the greatest opportunity to develop long-term resiliency, that is naturally sustained and provides for people and many other dependent species.
The Middle Hoh is unique in that there are currently 194-ac of land in conservation and 374-ac of federal or state ownership within the resiliency corridor that would inhibit future
development. This represents 74% of the corridor, a wonderful foundation to continue to build upon with future acquisitions and conservation easements. When private property comes up
for sale in these areas it should be purchased. Acquisition will eliminate future flood damages and enable habitat restoration. Landowners that are receptive can consider conservation
easements as a means of preserving natural areas while retaining ownership. Funding programs also exist to help homeowners relocate outside flood prone areas. This will also reduce
flood damage liabilities to individuals and the public and consistent with current guidance by the Association of State Floodplain Managers (ASFPM) and the Federal Emergency Management
Agency (FEMA).
Long-Term Desired Conditions (include overarching goals and specific objectives)
Looking beyond 20-yrs the goal is to have completed all of items in the action plan, and focus the effort on continued passive restoration, monitoring and maintaining community engagement
to ensure this vision for the Middle Hoh can persist well into the future. The river will appear different that it does today, with more channel threads and instream islands resulting
in narrower and deeper channels on average. Most of the energy of the river is partitioned interacting with stable instream wood, instead of being released eroding channel banks, resulting
in diminished channel migration rates that have allowed the floodplain riparian forest to mature. It will take more than a century for these forests to produce trees of sufficient size
to form stable logjams in the river once recruited, however conifer release and planting are accelerating natural succession of the riparian forest. The ELJs constructed in the river
and floodplains mimic the role of these large trees once played in the channel, allowing the system to recover and begin producing new trees to replace the ELJs once they deteriorate.
The expected lifespan of the ELJs suggests that additional phases of ELJs construction will be needed to ensure the system remains stable long enough for the riparian forests to mature
sufficiently to become self-maintaining.
The findings of the Upper Hoh Road Plan successfully identified an alternative route for the road that meets the needs of the community, is outside of the resiliency corridor, CMZ, and
is on low-landslide risk terrain (or the risk has been otherwise mitigated). This new route maintains access to the river, providing landowner and recreational access, while problematic
chronic washout and side-prone sections of the road have been decommissioned and removed. Traffic heading to and from ONP on the new route has diminished overall traffic close to the
river and homes, with individuals looking to access the river locally are directed on spur roads to established sites and local businesses.
Recreation opportunities in the Middle Hoh River have been expanded to include a trail network connecting Highway 101 and the Hoh Visitor Center, new interpretive trails along the river
educating the public about the
ecosystem, bike lanes along the new Upper Hoh Road to reduce traffic and noise and improved stabilized boat launch sites that persist following high water. Working with entities that
manage, own property or enter into conservation easements for ecosystem preservation to consider maintaining existing public access to the river so individuals can return to their favorite
places.
Establish, sustain and protect the Middle Hoh River Resiliency Corridor and all elements of the Action Plan. Restore the large wood cycle to the valley and its old-growth forests. Ensure
residents and visitors have safe access and their exposure to flood and erosion damages are minimal. Restoration of native salmon populations and native plant species. Continue updates
to resiliency planning and community communications.
Intermediate-Term Desired Conditions
For the purposes of this discussion, intermediate-term refers to the next 5 - 20-yrs and is a time when active restoration efforts are underway and wrapping up, having set the stage
for a self-sustaining ecosystem where passive actions can be used to accommodate unanticipated shifts due to climate change. Active restoration actions include continued focus on restoration
of off-channel rearing habitat and partitioning flow to reduce stream energy. Restoration projects have been implemented at the sites ranked as having the highest priority for restoration,
and progress is continuing down the priority list toward the bottom. Monitoring is showing more juveniles out-migrating to the sea, indicating restoration actions are having a measured
effect that will only continue to improve. Returning adults encounter more deep pools as they migrate upstream from instream ELJs forcing scour, creating resting spots as they move
upstream. All these changes are making it easier for the salmon that use the Middle Hoh to adapt to other changes brought about by the changing climate. Known patches of invasive species
have been eradicated and continued monitoring has limited spreading within the Middle Hoh.
Work is beginning after years of planning on a new road that will reduce the impact on the environment and is safer for all users and the local community. Most traffic has been routed
further from local homes and the river, while maintaining access. Refinements to the alignments continue as new information is gathered and/or conditions change. The chronic road washout
locations on the Upper Hoh Road have been stabilized and continue to provide protection from erosion as the new alignment is constructed.
A series of trails connecting Highway 101 to the Hoh Visitor Center are under construction providing alternative access to ONP, or an alternative to visiting ONP during high-demand times.
Additional interpretive trails along the river are being built that highlight the restoration work underway and the vision for the future.
Short-Term Desired Conditions
For the purposes of this discussion, short-term refers to the next 5-yrs. Actions that can be achieved in the short-term include those such as riparian conifer inter-planting, invasive
species monitoring and treatment, thinning, stabilize known chronic road washout locations, opportunistic acquisition of property within the resiliency corridor, maintaining the leadership
group that will work to realize the goals of this plan. Floodplain protection from channel migration, establishing islands in the channel that are protected from erosion.
Over the next 5-yrs we envision the establishment of a Middle Hoh working group consisting of a range of stakeholders who work to move the plan forward. A riparian forest management
plan for the Middle Hoh was developed that has guided the initiation of silviculture actions aimed at acceleration the conversion to a mature forest, including conifer interplanting,
thinning, understory management, invasive species monitoring and control, as well as formation of volunteer groups. Instream restoration projects have been implemented where high-quality
floodplain habitat was at most risk of loss through erosion, utilizing ELJs strategically placed in the channel, banks and floodplain to mimic the role large wood once played in the
Middle Hoh. These ELJs are used
to limit avulsion potential through side and relic channels, reduce channel migration rates through young floodplain forests and split the main stem channel, forming instream islands
that can persist over time. As these projects begin to be implemented new islands are forming in the river and vegetation is starting to colonize. Conifers are emerging from the dense
alder stands where thinning and planting was completed, and understory invasive species have been eliminated and/or are receiving continual prevention and control efforts.
In response to a broader national strategy to reduce the resource burden on the National Park Service, a reservation system was established for ONP visitors to the Hoh Rain Forest. This
change in policy has increased demand for recreation and engagement with nature in the region, and decreased traffic along the Upper Hoh Road as a result. While overall traffic is down,
more recreational users are looking to the Middle Hoh
Chronic washout locations along the Upper Hoh Road have been stabilized using wood-based structures, and a plan has been developed to consider realignment where the road is within the
Resiliency Corridor and/or CMZ. Improvements to the stability of existing boat launches have been improved or have been moved to limit susceptibility to erosion during high flows. Boat
launches that have been lost to erosion over time are under evaluation to determine need and appropriate locations.
Resiliency Plan actions will likely take several decades even if they are aggressively funded. The actions will likely be needed into perpetuity. The greater the geographic extent and
rate that actions are implemented, the faster the goals of the Resiliency Plan will be achieved. The magnitude of actions needed would eventually dimension. But if the Actions are not
implemented, the severity of impacts will continue to grow and become more expensive to treat. Summary of Resiliency Plan actions, all of which can be implemented concurrently, are
listed below (presented earlier in document):
Conifer floodplain planting within the Resiliency Corridor
Invasive plant control throughout the valley
Construction of stable engineered logjams (ELJs) within and along the margins of active channel migration
Relocation of Upper Hoh Road outside the CMZ
Construction of large roughness elements (ELJs) and riparian buffers along road where it can’t be moved
Acquisition or conservation easements of land within the Resiliency Corridor.
Relocate residents at high risk of erosion and flooding to safe areas outside the Resiliency Corridor
Focus new development in upland areas outside the Resiliency Corridor
A foundational framework of the Resiliency Plan is taken from the successful work that has been done on the Upper Quinault River (QIN 2006). Like the Hoh (Figure 3f), the Upper Quinault
Valley historically had extensive old-growth forest areas with networks of stable side channels (Figure 36). Historic clearing led to more active channel migration which has eroded
these areas and prevent similar habitats from re-forming (QIN 2006). This consists of linking stable roughness elements (ELJs) to restoring forests and side channels. The basic concept
is illustrated in Figure 37. The geographic framework is to protect margins of resiliency corridor so that side channels can form and be sustained. To do this the density of ELJs increases
toward the corridor margins with lower density further inside the corridor. The Upper Quinault Restoration program led by the Quinault Indian Nation has been implemented in phases almost
every year since 2008. At the PA5 site (Figure 38) several phases of ELJ construction have been successful in shifting the mainstem river away from development and are restoring side
channels and floodplain forest. ELJs built in the Cispus River in 2021 were subjected to a 100-yr recurrence flood weeks after construction and they all survived and achieved goals
to create restore anabranching and inundate side channels in existing floodplain forest (Figures 39, 40 and 41).
/
Figure 36. Illustration of portion of Upper Quinault River Valley (RM 44.3-45.8). The channel traces show the HMZ and active channel migration zone. The area to the south shows a network
of stable side channels flowing through mature forest. Southward migration of the river has eroded important side channel habitat crucial to salmon and the lack of big timber is preventing
the habitat from being reformed. Most recent historical channel is 2002. Since then the river has migrated further to the south. Adapted from QIN (2006).
/
Figure 37. Conceptual geographic framework for restoring large wood cycle, floodplain forests and side channels. The layout shows protective measures (green squares, zone 1) of property
and infrastructure within CMZ, this protection would not be needed in undeveloped areas. The thin blue lines within zone 2-3 represent stable side channels in an area that natural would
be dominated by old-growth (see Figure 3f). The area of more active channel migration (zone 4) the density of ELJs diminishes. Taken from QIN 2006.
/
Figure 38. Implementation of several phases of restoration in the Upper Quinault River. At the site the river had moved several thousand feet to the south, ultimately destroying one
home and threatening the South Shore Road. The black symbols show constructed ELJs constructed between 2013 and 2017. The ELJs allow water flow in-between them but discourage the main
channel from occupying the area (analogous to zone 2-3 in Figure 37). The project has resulted in the main channel moving north while creating new side channel habitat within the treatment
area. The ELJs have also increased the number of new pools with complex cover and created new floodplain for reforestation.
/
/
Figure 39. Example of large-scale restoration project to restore large wood cycle and side channel habitat in the Cispus River (Lewis County, WA). Photos show before and after.
/
Figure 40. Same reach of the Cispus River, looking downstream on November 24, 2021, after a 100-yr recurrence peak flow. All 24 of the ELJs are intact and undamaged. The project increased
cumulative channel length over 4-fold and increased the number of pools over 10-fold. Photo by Eli Asher, Cowlitz Tribe.
/
Figure 41. Example of engineered logjams constructed in 2020 and 2021 in the Cispus River. The structures were subjected to a 100-yr flood event in November 2021, only weeks after construction.
All 21 of the structures were undamaged and most collected large volumes of wood. Flow is from right to left. Photo 11-24-21 by Eli Asher, Cowlitz Tribe.
LOCAL CAPACITY TOSUPPLY RESTORATION NEEDS
Introduction
The goal of this chapter is to provide a summary of the local capacity currently available to support restoration needs in the region and to give insight to the feasibility of reaching
the objectives in the Middle Hoh Resiliency Plan (MHRP). This chapter also serves to provide an understanding of where there may be gaps in local capacity impeding restoration progress.
Identifying gaps in capacity during Phase 1 of this project will inform local stakeholders, practitioners, decision makers, and funders where investment is likely needed to help build
local capacity to fulfill restoration objectives of the MHRP.
The information provided in this chapter is a snapshot of the current status of local capacity, and current strengths and limitations in capacity are anticipated to change with time.
Looking ahead, it is important to note that relatively recent increases in restoration funding to the region are providing ample opportunities build restoration capacity (e.g., restoration-related
education programs, jobs, businesses, etc.).
For the purposes of this chapter, we are discussing the local capacity to complete Ecological Restoration of natural processes in the river and floodplain environments. According to
the Society for Ecological Restoration, an international non-profit organization that advances the science, practice and policy of ecological restoration to sustain biodiversity, improve
resilience in a changing climate, and re-establish an ecologically healthy relationship between nature and culture, Ecological Restoration is defined as the process of assisting the
recovery of an ecosystem that has been degraded, damaged, or destroyed (www.ser-rrc.org).
By definition ecological restoration encompasses a very wide range of projects. The practice of ecological restoration requires a high degree of ecological knowledge that can be drawn
from practitioner experience, Traditional Ecological Knowledge (TEK), Local Ecological Knowledge (LEK), and scientific discovery. Practitioner knowledge is derived from experience in
repairing ecosystems, and from information from a spectrum of disciplines (e.g., restoration ecology, agronomy and seed production, forestry, horticulture, botany, wildlife science,
zoology, hydrology, soil science, engineering, landscape design, conservation biology, natural resource management, etc.). Additionally, LEK and TEK experts, who are typically members
of a local community, can provide extensive and detailed information about sites and ecosystems drawn from their long-term relationships and connections to these sites. When integrated
into restoration projects, these multiple forms of knowledge provide opportunities to improve restoration outcomes for ecological, social, and cultural benefits. (Gann, et al, 2019).
In the following chapter sections, the term restoration is used when referring to ecological restoration as defined above.
In the context of this chapter, we are referring to restoration projects that aim to restore river and floodplain environments, and this scope includes a wide diversity of project types
and capacity needs. Restoration project types considered in this discussion include riparian and floodplain restoration as well as instream restoration.
Generally speaking, floodplain restoration improves floodplain structure, function, connection, and natural processes. Floodplain restoration actions include on the ground projects as
well as efforts to promote land conservation (e.g., conservation easements, land transactions, etc.). Ground-based projects may strive to reconnect side channels and off channel habitats
via culvert upgrade or removal, channel excavation, and engineered log jams. Ground-based projects can also include more passive actions that promote healthy forest plant communities
through planting native plants, managing invasive plant species, and forest thinning. Additionally, floodplain restoration projects could include efforts to address land development
and/or land use in the floodplain. For example, roads, levees, and water diversion infrastructure can have significant negative impacts to natural floodplain processes. Floodplain restoration
projects could include actions such as removing or upgrading levees; removing or upgrading roads; removing or upgrading water diversion structures, culverts, etc.
Instream restoration actions include projects that promote natural morphological processes that in turn support salmon habitat forming processes. Instream restoration projects often
overlap with floodplain restoration, as the stream channel and floodplain are a single, usually complex, system. Instream restoration actions could include more passive work, using
hand tools, or more aggressive actions using large machinery, or even helicopters. Examples of passive instream restoration projects include: installation of PALs (Post Assisted Log
structures), BDAs (Beaver Dam Analogs), direct felling of trees into the stream, LTPBR (Low Tech Process Based Restoration) applications, etc. Larger scale instream restoration could
include the use of excavation, fill, and the import of native materials lost due to degradation, such as large wood and log jams. Instream restoration projects also include projects
aimed to reconnect fragmented habitats caused by anthropogenic migration barriers to fish. For the most part, addressing these migration barriers entails an upgrade to road infrastructure
where a road crosses a fish-bearing stream. These ‘fish passage projects’, when implemented, usually involve the replacement of the existing infrastructure causing the fish barrier
(e.g., culvert conveying streamflow under
road) with an appropriate structure (e.g., countersunk culvert, bottomless arch, or bridge) that allows unimpeded fish passage and natural stream processes, such as sediment and wood
transport.
All of these restoration projects require more than the physical actions to implement the plan and/or design. The capacity needed to complete restoration projects also require cultural
resources review, permitting work, engineering, project management, etc. This chapter and the MHRP acknowledges the essential role of cultural resource protection and the needed capacity
to ensure cultural resource compliance.
Recent Restoration – a Snapshot of Restoration Types and Funding Over Past Five Years.
The timeline to develop and implement these various kinds of restoration projects vary widely. The lifespan of restoration projects can be relatively long and complex, relying on multiple
disciplines and expertise, but they can also be relatively short and straight forward, applying only one or two disciplines to meet individual project goals and objectives. In researching
for this chapter to gain an understanding of the current local capacity to supply restoration needs, we queried five years of completed and currently active restoration projects in
the region. Project data was collected for all riparian, floodplain, and instream restoration projects in the local Watershed Resource Inventory Area (WRIA) 20 and the neighboring WRIAs
19 and 21. See Attachment C for the project list, project type, and associated costs. All of these projects were funded through the WA Salmon Recovery Funding Board, WA Coast Restoration
and Resiliency Initiative Program, Family Forest Fish Passage Program, or the Brian Abbott Fish Barrier Removal Board. All of these restoration funding programs are managed by the Washington
Recreation and Conservation Office (RCO), and RCO staff supplied the data in early 2021.
Understandably, this list of restoration projects and associated costs do not include all of the funding and capacity applied to complete each project. Capacity considered, but not quantified
here, includes: funding, staff time, and capacity from local sources (e.g., Counties, Cities, Conservation Districts, etc.), federal sources (e.g., EPA, NRCS, Title II, etc.), tribal
sources (e.g., BIA, NWIFC, etc.), foundations, and private donors. This inventory of recent restoration projects does not include protection and acquisition projects (e.g., WA RCO ALEA
projects, Hoh River Trust, The Nature Conservancy, etc.), however these types of land acquisition and protection projects should still be considered a critical tool in the restoration
toolbox.
Elements of Restoration Projects and Associated Capacity Needs
To understand the capacity needed to carry a restoration project from concept to implementation, we must first understand the steps or phases a typical restoration project goes through
to be successful. Gaps in capacity at any point in project phases can lead to project delays, funding shortfalls, incomplete projects, or poorly implemented projects. Therefore, it
is critical to assure the needed capacity is available to fit the project’s needs early in the project development process.
In general, restoration projects go through the following phases:
Project Concept – project need and/or idea is identified. Conceptual project ideas can originate from previously completed assessments, studies, strategic plans, etc. Project ideas can
also come from local landowners, stakeholders, scientists, user groups, restoration practitioners, agency staff, or just about anyone with knowledge about a particular issue or location/site.
Project Development – Once a conceptual project has been identified and a project sponsor or project steward decides to pursue the project, the project must be further developed before
it is ready to pursue funding.
Project development is a general term, and it can include more elements and processes than can be discussed in detail here. In general, project development usually entails the following:
Identify project goals and objectives; Develop project scope; Develop project timeline; Develop project budget; Continue outreach and collaboration with local stakeholders, agency staff,
and tribe staff; and finally confirm project feasibility based on regulatory compliance, landowner support, and local stakeholder support.
Project Funding – After project has been developed, is deemed feasible, and stakeholders have expressed support, it is time for the project sponsor and project partners to apply for
project funding via restoration-focused funding sources (e.g., grant programs, foundations, donors, etc.). At this phase, the project sponsor and partners identify the most likely funding
source(s) where the proposed project will compete well (against other restoration project funding proposals), and the project team drafts and submits grant proposal(s) to secure project
funding. This phase often entails securing funding from multiple funding sources, as many restoration project funding programs require matching funds. Grant writing skills, persistence,
and creativity may all be required to be successful in this phase of the project.
Project Planning and/or Design – Now the project is funded, and this is when the project elements identified in previous phases come to life. At this point, the project scope, timeline,
and budget are put into action, stakeholder outreach continues, and the project begins reaching some of its objectives. Multi-phase restoration projects may complete a planning and/or
design phase as a standalone Phase 1 project. These Phase 1 planning projects usually require additional funding for future phase(s) to complete the project (e.g., Phase 2 implementation).
Project Implementation – This is often the most expensive, but also the most exciting phase of the project. This is where restoration actions take place on the landscape. The implementation
phase can entail manual labor, hand tools, large machinery, and even helicopters. Implementation can occur relatively quickly and take just a matter of days, but on the other end of
the spectrum, implementation can take months or even years and multiple implementation phases to complete.
Project Monitoring – Ideally, all restoration projects will have some level of pre-project monitoring and post-project monitoring to track project effectiveness. Unfortunately, many
restoration funding programs do not allow for funding this monitoring phase, and the project team may need to find funding elsewhere or collaborate with other entities that may be able
to contribute monitoring efforts (e.g., agencies, tribes, landowners, etc.).
In summary, carrying a project successfully through all of these phases in a timely and consistent manner is no easy task and requires collaboration and support from local stakeholders.
Leveraging the skills, knowledge, expertise, and funding from an array of sources is often the key to success. Please see Attachment A for a list of restoration-related capacities often
required to complete a project.
Survey Results from Restoration Practitioners
In December 2020 MHRP project partner, Trout Unlimited, solicited local restoration practitioners, agency staff, tribe staff, and stakeholders to complete a short survey. The intent
of this survey was to engage those experienced in local restoration efforts to gain insights regarding restoration history, current restoration capacity status, and likely restoration
capacity needs in the future.
Below are the four questions posed in the survey, and respondents had approximately six weeks to respond. Respondents had the option to complete the survey online, or they could submit
survey responses directly via emailed word document. We received eight responses to the survey, and a summary of the responses is displayed in bullets below each question (in no particular
order):
What are current strengths in local capacity to implement restoration projects?
Landowner knowledge of watershed, history, historic condition and change over time
Existing strong knowledge base (e.g., agencies, tribes, NGOs, landowners, etc.)
Experienced organizations and resource managers have been and continue to complete restoration projects
Multiple funding programs currently exist (WA Restoration & Resiliency Initiative, Salmon Recovery Funding Board, Brian Abbott Fish Barrier Removal Board, etc.)
Heavy equipment expertise and availability
Wood locally available, especially racked on bars, but also standing in forests
The participation and contribution to restoration, monitoring, and science from University of WA Olympic Natural Resource Center (UW ONRC)
Good coordination and strong restoration project partnerships currently exist
Existing project sponsors; local knowledge and relationships; project prioritization; functioning Lead Entity Group incorporating technical and community members
Where are gaps or challenges in local capacity to implement restoration projects?
Lack of public outreach and landowner relationships
Data gaps. Need to prioritize restoration actions on good data, not just available data. Need data for non-salmon species (to inform restoration actions).
Limited staff or staff time available to steward additional projects through the pipeline
Grant writing (to secure project funding)
Lack of local certified engineers
Over-engineering and cost for some types of projects
Funding, especially for higher cost projects (e.g., road related issues, reach scale projects, etc.)
Materials such as large wood with root wads or locally sourced native plants/seeds can be challenging
Many projects are large in scale, equipment-intensive, and impactful (not necessarily positive)
Post-project lessons learned not shared regularly
Change occurs regularly – and project funding can be too late to implement as designed or for the function intended.
Where do opportunities exist to increase capacity to implement restoration projects?
Continue/increase project partnerships (e.g., agencies, tribes, NGOS, landowners, stakeholders, etc.)
Increase project coordination at the ground level
Conduct focused assessments to close data gaps, promote proactive restoration project development, and more shovel ready projects prepared for funding opportunities
Planning projects, like this MHRP Phase 1 project, and organized workgroups where many shareholders can get together to focus in on specific areas and share knowledge increases capacity
Increase landowner outreach to educate them about potential opportunities, like funding and support, to restore their property
Leveraging current capacity through strategic collaboration; developing and/or advancing strategic restoration action plans; more shovel ready projects; streamlined permitting; untapped
funding sources
Work with UW ONRC, Peninsula College, Tribes, Counties, and NGOs to fund internships, fellowships, class projects, etc.
Increase engagement and data sharing with UW ONRC T3 project and DNR’s Olympic Experimental State Forest program
More studies, data collection, and monitoring needed to better understand priority restoration actions
What strategy is needed to pursue these opportunities and increase restoration capacity?
Need dedicated staff capacity for landowner outreach. Applicable outreach strategies currently exist (Coast Salmon Partnership Outreach Strategy), however more capacity needs to be focused
to increase outreach efforts and meet outreach objectives.
Strategic coordination for long term data collection and management (e.g., tribes, agencies, local stakeholders, etc.)
Endowment fellowships and technical positions focused on the Western Olympic Peninsula
Education pathways; training & education incentives; awareness about restoration employment and/or funding opportunity(s)
Stable funding to support a resilient restoration economy; local support; engaging volunteers
Conservation easements for protected areas (that allow natural ecological processes)
Increased monitoring to track trends and change over time
Move infrastructure (e.g., roads, buildings, etc.) out of floodplains, where possible, or build the riverbanks back (buffers) to maintain natural riparian function
Collate data sets into a single searchable location to inform future work
Build advocacy for the watershed that recognizes all the contributions made and interests involved
Coordinate a strategic outreach effort to educate communities about the less obvious benefits of restoration, such as positive impacts to ecosystem services, fisheries, tourism, water
quality, summer stream flows/water quantity, etc.
Inventory - Local Capacity to Supply Restoration Needs
Please see Attachment A for a list of restoration-related capacity needs and Attachment B for an inventory of available restoration-related businesses in the region. Attachment B is
meant to be a living document, and we anticipate it will be revised and become more conclusive over time. In compiling this inventory of local business, we focused on the communities
within WRIAs 20 and 21, and we consulted the following resources: Forks Chamber of Commerce; WA Secretary of State; Google search engine; Personal communication with local stakeholders
and practitioners.
Things to Consider
Difficult to Quantify All Capacity
There are many capacity needs not discussed in detail in this chapter or associated attachments. Nearly all restoration projects require administrative resources such as office space
and equipment, internet connectivity, travel/per diem expenses, fuel, meeting venues, meeting facilitation, etc. Restoration efforts can also call for acquisition-related professionals
for land appraisals, legal support, land conservation designations (e.g., conservation easements), environmental service technicians (for site assessments and testing), water quality
monitoring, etc.
Additionally, some projects require materials that are unique, specialized, and/or not easy to procure. For example, small- and large-scale projects often require a source for native
plants and/or seeds appropriate for the local ecosystem being restored. Another example would be Large wood and engineered log jam projects, that aim to replicate the function of large
fallen wood in the channel and floodplain. These large wood restoration projects depend on a supply of mature trees (often with root wads attached), and the acquisition of timber rights
to conduct specialty harvest of mature second growth forests may be necessary to complete these kinds of projects. In summary, it is important to consider the vast array of funding,
infrastructure, materials, administration, etc., beyond project-specific requirements when estimating the capacity needed to fulfill restoration projects’ needs.
Climate Change
The ecology of the Olympic Peninsula’s West End is a unique and relatively remote, but it is also highly vulnerable to the changing climate. Recent forest fires in a second growth and
old growth forests of the Queets
and Quinault Watersheds surprised many. Heat domes and altered precipitation patterns are stressing trees with implications for forest extent and productivity, river hydrology, and
habitat quantity and quality for fish. During June 2021, temperatures reached 98 degrees at the ONP Visitor Center and a record breaking 109.9 degrees at the nearby Quillayute Airport
weather station. Along the Hoh Main Line and Maple Creek Road, the new top growth of many Douglas fir trees were killed during this heat wave. Currently, along many south-facing road
corridors, western red cedars, maples and other forest vegetation appeared heavily singed or desiccated. These heat impacts were noticeable on July 24th2021, the date of the Middle
Hoh River resiliency plan launch event at the Fletcher Ranch.
The Middle Hoh River is famous for its rainfall, mushrooms and mosses, but increasingly the area experiences long periods of drought, with low summer river flows and higher water temperatures,
as well as higher flood flows and frequencies in winter. The glaciers that feed the Hoh River are receding rapidly and the risk of wildfire, once almost unthinkable, is becoming a serious
concern. Projects that anticipate and respond to new ecological hazards associated with erratic and extreme weather events linked to climate change, and local expertise to implement
them, are needed to meet the challenges of an increasingly uncertain future. Local capacity to respond to these needs would ideally be paired with economic activities designed to sequester
carbon and participate in larger efforts to rapidly reduce the earth’s temperature while carbon emission reductions continue to accelerate. As a primary example, the grassroots MEER:ReflEction
(meerreflection.com) Framework developed by Dr. Ye Tao at Harvard Rowland Institute offers a scalable approach to reducing the global temperature to help slow the loss of ice, and associated
methane emissions (a greenhouse gas twenty times more powerful than carbon dioxide), in a manner that may also produce benefits for industrial agriculture. Ultimately, the success of
restoration activities on the West End will depend upon the world community’s embrace of actions to restore the planet’s climate systems.
Increased Funding
As previously mentioned, the Washington Coast has experienced a relatively recent increase in restoration funding from multiple agencies, grant programs, and/or foundations. As a result,
our capacity to fill jobs locally is in transition, and currently, much more is being planned, prioritized, and implemented than ever before. This is largely due to the WA Coast Restoration
and Resiliency Initiative grant program, and there are also substantial federal funding increases on the horizon. This increase in funding is already providing opportunity to grow local
restoration capacity, restoration-related jobs and businesses, and increase the pace and scale of restoration projects. These opportunities are anticipated to grow as available funding
continues to increase.
Discussion
For the first time in the Middle Hoh River, local stakeholders, restoration experts, agency staff, local and regional jurisdictions, and vested interests have come together at monthly
intervals for nearly two years to collaborate and develop a plan to promote a vision for the long term resiliency of the Middle Hoh River corridor. Without a doubt, the process of developing
the draft plan has been a challenging task with many questions yet to be answered. Regardless of these challenges, much progress has been made in identifying a path forward to restore
and/or protect the Middle Hoh in the face of constant change. Through many meetings, site visits, data sharing, personal communications, etc., we now have a draft plan and strategic
actions identified to realize a productive and sustainable Middle Hoh River for the ecosystem and communities that call this special place home.
Reaching the objectives laid out in the recommended actions in MHRP will provide multiple benefits. For example, restoration actions are intended to address threats to property within
the corridor while also promoting improved habitat for salmon, steelhead, and the surrounding ecosystem. Additionally, a healthier and resilient Middle Hoh ecosystem will allow local
communities and wildlife a greater ability to adapt to climate change impacts, such as increased peak flows and mainstem Hoh River migration and avulsion events. It is
important to reiterate that the restoration actions recommended are intended to promote natural ecosystem processes. Aligning with and promoting these natural processes is critical for
creating and maintaining fish and wildlife habitat and allowing the natural ecosystem the space and time needed to adapt to disturbance as well as climate change. In other words, if
we understand how to restore and/or take care of the Middle Hoh ecosystem, the Middle Hoh ecosystem will be able to take care of itself in the long run. Looking ahead, a resilient Middle
Hoh will result in rebounding salmon and steelhead stocks and a healthy, mature riparian forest ecosystem. This uplift has the ability to contribute significantly to the fast-growing
tourism and recreation economy the region is experiencing.
According to a 2015 report by the University of Oregon: Assessing, Planning, and Monitoring to Increase Local Economic Opportunities from Restoration: “Increased interest in restoration
on both public and private lands has led to opportunities for advancing a more robust restoration economy in some rural communities. However, achieving local economic benefits for businesses
and workers may require deliberate strategies that are carefully matched to local strengths and limitations.” To identify these strengths and limitations the report recommends conducting
a restoration industry assessment, similar to the research completed to inform this chapter. The report explains: “A restoration industry assessment collects information about the state
of the restoration industry in a particular locale. It can include information about the types and amount of work undertaken in an area, the businesses contracted to perform that work,
and characteristics of the workforce implementing the work. This information can act as a foundation for developing and implementing strategies that links forest and watershed restoration
work with the people and businesses skilled and equipped to undertake this work.”
Once a restoration industry assessment has been conducted, the report recommends the development of a restoration jobs action plan as part of the strategy to link restoration projects
to employers and businesses. The report notes: “The action plan is likely to involve a number of strategies involving different groups and organizations. For example, the action plan
might include a federal lands strategy, a nonprofit contracting strategy, and a contractor capacity strategy. Some strategies may be implemented quickly, whereas others might take months
or even years. Implementation is most likely to succeed when it becomes a regular part of the work plan of local collaborative groups, nonprofits, and land managers.”
As a result of these recommendations, and the increased understanding as to the local restoration capacity strengths and limitations, we are recommending the formation of a Restoration
Economy Workgroup as a subcommittee of the Middle Hoh Resiliency Leadership Team. This subcommittee workgroup would require participation from multiple individuals and groups already
active in developing the MHRP, but most importantly, it would need leadership from the local businesses and employers to drive the economic development strategies to foster long-term
economic uplift through a new and essential sector. Over time, this subcommittee could coordinate with practitioners working in other major projects to foster a West End O.P. Ecologic
Restoration Program (or Sector) with representation on the West End Business and Professional Association.
The local business and expertise currently available to conduct restoration actions are certainly a strength already being applied to work in the Middle Hoh and the West End. There are
many restoration groups already implementing successful restoration projects, and these projects are leveraging the skills of local contractors, machine operators, forestry experts,
etc. That said, there does seem to be limitations of locally available businesses needed to develop, design, and implement restoration projects. Where these restoration capacity limitations
exist, local project sponsors generally have to contract businesses located outside of the region to accomplish their restoration projects’ objectives. An example of this can be seen
in the limitations of local certified engineers to design and oversee implementation and construction of restoration projects. A likely reason for these capacity limitations is often
due to the relatively small and inconsistent restoration-related work available in the region. In other words, there simply isn’t enough ecosystem restoration work available for an
engineering firm to set up shop locally and keep full time staff employed via WRIA 20 restoration work alone.
In researching this chapter, several interviewees noted that currently much of the resources/services needed to complete a restoration project are imported into WRIA 20, and therefore
much of the economics are exported. To change this trend and promote local restoration economic growth would require an increase in restoration-related work to the point that it could
justify and sustain new local businesses (while also supporting existing businesses). In short, the demand and funding for restoration jobs and expertise needs to exist locally at a
level to justify the supply of services (e.g., certified engineers).
A respected coastal economist provided feedback in the interview process, and they noted that currently, restoration work itself is likely having minimal impact on the local economy
as the work is temporary and labor is often imported from out of the region. They also recommended focusing on the indirect impacts of restoration work as having the greatest benefits
to the local economy – restoring salmon habitat can positively impact the local economy through increased recreational opportunities, commercial opportunities, as well as cultural benefits.
These elements also attract tourism seeking those attributes to the region, which can bring additional dollars to the economy.
Another potential restoration capacity limitation appears to be clear, locally available career path opportunities for those interested in education, training, experience, and employment
in the field of ecosystem restoration. The region is rich in history of commercial forestry and commercial fisheries, however these industries have had to adapt to changing markets
and limited availability of natural resources (e.g., depressed fish stocks, loss of mature forest stands). The businesses, organizations, families, and individuals that historically
did this work are very skilled and knowledgeable of working on the landscape, and many of these skills could be applied to restoration work with available incentives and/or opportunities.
Working with these industries in a collaborative manner to reach ecosystem restoration goals is an incredible opportunity worth pursuing. For example, several recent salmon and steelhead
habitat restoration projects have contracted construction contractors that were formerly timber harvest contractors. These contractors have intimate knowledge of the area and the experience
to move about and work on the landscape. Awareness and consistency of available work/contract opportunities, as well as awareness of training, certification, and career path opportunities,
appears to be a limiting factor for further developing these relationships between skilled local businesses and residents and the field of ecosystem restoration.
Available funding for restoration projects is another factor limiting the local capacity to supply restoration needs. As previously noted, the relatively small and inconsistent nature
of restoration work in the region is not likely having a significant impact on the local economy. Increased and consistent funding aimed at restoring the region’s ecosystem function
would likely have positive influence on several other restoration capacity limiting factors. For example, consistent restoration funding would increase the rate and scale of restoration
projects implemented in the area, and therefore increase the justification for restoration-related businesses to operate and be housed locally. The current Infrastructure Investment
and Jobs Act currently (as of August 2021) moving through the House and Senate includes $1 billion to correct culvert fish barriers and $172 million for the Pacific Coastal Salmon Recovery
Fund. If this bill is passed, this would be a noticeable increase in restoration funding to the region and provide increased opportunity to grow local restoration capacity in WRIA 20.
Many restoration projects currently being implemented in the region are labor intensive. Related to the need for an increase in consistent funding is the need for more skilled labor
trained and capable to do the work needed to reach restoration objectives. Local nonprofit organization, 10,000 Years Institute, is currently working to develop and fund a Coastal Conservation
Corps. The goal of this program is to create stable employment provided by a continuously funded program to conduct place-based, persistent, year-round work via trained and skilled
resource workers. The program would aim to support resilient natural resources industries and ecosystems, promoting stable employment and communities. Types of work the Coastal Conservation
Corps aims to engage in include: invasive species detection, monitoring, and management; forest stand treatments
and biochar pyrolysis; habitat restoration, such as revegetation, wood placement, fish barrier correction, etc.; recreation via trail and facilities maintenance, signage, etc. This is
an example of how consistent restoration-related funding would increase the number of restoration-related jobs in the area while also increasing the range of positive benefits from
restoration actions.
Recommendations to Increase Local Restoration Capacity
Assess the desire and feasibility to increase restoration-related jobs and business (capacity) in the region in collaboration with local stakeholders, restoration practitioners, and
the local business community.
Develop a Restoration Economy Workgroup as a subcommittee of the Middle Hoh Resiliency Leadership Team. If developed, the Restoration Economy Workgroup could work towards a program and
funding of a comprehensive West End O.P. Restoration Program. This is a similar idea to the Coastal Conservation Corps proposal that 10,000 Years Institute is working towards. The West
End O.P. Restoration Program could work with the local education programs to develop clear career pathways and incentives for those interested in the field of restoration. The program
could also work with local groups to promote economic growth and stability (e.g., West End Business and Professional Association), and this would act to attract new restoration-related
businesses and services to the local communities.
Develop outreach, training, incentives, and opportunities for local contractors not yet experienced in restoration actions that take place within stream channels and floodplains (e.g.,
installation of engineered log jams).
Develop strategy(s) to take advantage of increasing restoration project funding opportunities. This strategy(s) would promote a proactive approach and be prepared with high priority,
locally vetted, shovel ready projects before funding opportunities arrive (e.g., West O.P. Barrier Prioritization Project and Wild Salmon Center, Coast Salmon Partnership, and Trout
Unlimited’s Cold Water Connection Campaign).
Increase the capacity and ability to incorporate climate change resiliency into restoration project design and implementation. Increased connection between climate change trends and
restoration objectives will promote more climate resilient restoration outcomes. Projects that anticipate and respond to new ecological hazards associated with erratic and extreme weather
events linked to climate change, and local expertise to implement them, are needed to meet the challenges of an increasingly uncertain future.
PHASE II APPROACH
To realize the potential of this Resiliency Plan, the formation of a permanent group of individuals (Middle Hoh Resiliency Initiative) is needed to ensure all groups working in the community
are working together in a coordinated effort toward the common vision(s) articulated in the desired conditions. This group should include at a minimum several local landowners, the
Hoh Tribe, Jefferson Co., local non-profits and conservation organizations, recreational users and agency representatives to further ensure any potential actions are vetted and considered
openly and news can be communicated. The group will be needed to initiate and develop the additional plans needed to realize the desired future conditions, as well as coordinate to
help locate funding sources for individual projects.
While the Action Plan (Appendix B) identifies specific opportunities and prioritizes their implementation, there is additional, more detailed planning and study needed to move specific
elements of the resiliency plan forward in a coordinated and informed manner. Completion of these studies and plans will provide a framework for moving forward larger components of
the Action Plan that require additional information and/or coordination. This additional work described as part of Phase II should proceed significant implementation of the Action Plan.
Each of the 5 project reaches of the Middle Hoh should have a more detailed Action Plan developed that could provide a more detailed and site-specific plan for restoring the specific
reach of the river. Reach specific Action Plans would be collaboratively developed with the Hoh Tribe, local landowners, and resource agencies to find a shared vision for each reach
that addresses the needs for salmon and the surrounding community. Each plan would prioritize work areas, or phases, for sequenced implementation in a matter that optimized resources
and local constraints.
The Lower Hoh River would benefit from having a similar Resiliency and Action Plan to better prepare the community and river corridor ecosystem from the current and worsening climate
crisis. A Resiliency Plan for the Lower Hoh River would be led by the Hoh Tribe, rely and build upon the findings of this Plan, and include similar inventories to characterize existing
conditions and evaluation of specific trends and associated impacts. Formation of a steering committee including a similar spectrum of representatives as this Middle Hoh Resiliency
Plan will greatly benefit development and ultimately adoption of the Plan. The protection of existing salmon habitat, restoration of degraded habitat, and actions aimed to improve ecologic
resiliency from impending changes to the river corridor should be a primary theme throughout the document. Improving and/or eliminating interactions with the river corridor and Highway
101 to the south and Oil City Road to the north should be a major component of the Lower Hoh River Resiliency Plan. Because the lower river is within tidal influence additional considerations
for predicted sea level rise should be incorporated into the evaluation of trends and anticipated changes. Predicted increases in flood magnitude and frequency, sediment supply and
rising sea levels will result in increasing flood and erosion risk to the community as well as alter floodplain and instream salmon habitat. Development of a Lower Hoh Action Plan would
outline specific actions to meet the goals of the Resiliency Plan, and be developed collaboratively with all plan stakeholders.
The anticipated changes to the sediment and flow regime of the Hoh River resulting from climate change is not well constrained and should be a primary consideration for planning purposes.
One of the more pressing studies needed is an in-depth evaluation of the anticipated flow and sediment regime changes in the Middle Hoh related to climate change. Warming temperatures
coupled with changing weather patterns will alter the rate of surficial processes and sediment production, both of which are anticipated to increase the sediment load entering the Middle
Hoh. Flood flows are predicted to increase in magnitude and frequency, and summer base-flows are anticipated to diminish. A better understanding of how these changes will manifest on
the landscape over time would greatly benefit the work completed and inform future planning efforts.
A comprehensive survey of side channels within the Middle Hoh would further inform and refine prioritization of floodplain habitat protection and enhancement within the reach. A complete
inventory of the side channels present and their habitat and hydrologic character, fish usage and riparian conditions would facilitate a more informed decision about the properties
to target for preservation and where and what restoration actions are needed. Having functional off-channel habitat is critical to maintain anadromous salmon populations in the Middle
Hoh and needs to be a critical component of the overall restoration strategy.
The state of the riparian forest corridor along the Middle Hoh largely dictates the health of the fish using the river. While the Hoh is known internationally for hosting ancient old-growth
forests, these are largely confined within the boundaries of ONP. Along the Middle Hoh, most of the riparian corridor has been logged at some time in the past and is in some state of
forest succession. A more detailed and complete Riparian Forest Plan could more deeply integrate silviculture actions with invasive species management and restoration actions into a
cohesive strategy that can be implemented throughout the reach.
A comprehensive plan for the Upper Hoh Road is needed to layout a shared vision for what the road should be, that best serves the people that use it and the surrounding environment.
The plan should explore opportunities for relocation, abandonment, improvements or some combination(s) thereof to the road that could improve the user experience and reduce environmental
impacts and risks. Identifying an improved alternative route out of
the northern side of the valley should be included in the plan for the Upper Hoh Road to improve community resiliency. Hoh Tribe, Jefferson County, ONP representatives and the local
community are critical stakeholders in developing any future plan(s) for improvements or relocation of the Upper Hoh Road.
Recreation Plan to explore opportunities for larger recreation features (trails), ensure visitor use does not detract from the environment (garbage?), and the number, location and types
of day-use areas along the river (including boat launches). Bring together local community, guides, day-users in community, NPS, county and landowners to explore alternatives that can
improve the experience for everyone while maintaining the wildness of the river (limited paved parking areas).
While outside the geographic scope of this plan, it would be advantageous for there to be an agreement with ONP and the Hoh Tribe to allow restoration actions within the park, and to
establish emergency road repair protocols to minimize the need for mitigation.
REFERENCES
Abbe, T. and D.R. Montgomery. 1996. Large woody debris jams, channel hydraulics, and habitat formation in large rivers. Regulated Rivers: Research and Management, 12, 201-221.
Abbe, T.B. and D.R. Montgomery, 2003, Patterns and Processes of Wood Accumulation in the Queets River Basin, Washington, Geomorphology 51:81-107.
Abbe, T., J. Bountry, L. Piety, G. Ward, M. McBride, and P. Kennard. 2003. Forest Influence on Floodplain Development and Channel Migration Zones. Geological Society of America Abstracts.
Abbe, T. and A. Brooks. 2011. Geomorphic, Engineering and ecological considerations when using wood in river restoration. In Simon, A., Bennett, S., and Castro (Eds.), Stream Restoration
in Dynamic Fluvial Systems: Scientific Approaches, Analyses, and Tools. Geophysical Monograph Series 194. AGU. pp.419-451.
Abbe, T., B. Belby, M. Fox and D. Shields. 2016. Geomorphology and Hydrology Considerations. Chapter 4, National Large Wood Manual: Assessment, Planning, Design, and Maintenance of
Large Wood in Fluvial Ecosystems: Restoring Process, Function, and Structure. pp.234-326. Bureau of Reclamation and U.S. Army Corps of Engineers. 2016. National Large Wood Manual:
Assessment, Planning, Design and Maintenance of Large Wood in Fluvial Ecosystems: Restoring Process, Function, and Structure. 628 pages + Appendix.
Babinski, Z. 2005 The relationship between suspended and bed load transport in river channels. Sediment Budgets 1. (Proceedings of symposium S1 held during the Seventh IAHS Scientific
Assembly at Foz do Iguacu, Brazil, April 2005) IAHS Publ. 291.
Beechie, T. J., M. Liermann, E. M. Beamer, and R. Henderson. 2005. A classification of habitat types in a large river and their use by juvenile salmon. Transactions of the American Fisheries
Society 134: 717–729.
Beechie, T. J., and T. H. Sibley. 1997. Relationships between channel characteristics, woody debris, and fish habitat northwestern Washington streams. Transactions of the American Fisheries
Society 126:217-229.
Beechie, T. J., O. Stefankiv, B. Timpane-Padgham, J. Hall, G. R. Pess, M. Rowse, M. Liermann, K. Fresh, and M. Ford. 2017. Monitoring salmon habitat status and trends in Puget Sound.
NOAA Technical Memorandum NMFS-NWFSC-137. https://doi.org/10.7289/V5/TM-NWFSC-137.
Bisson, P. A., J. L. Nielsen, R. A. Palmason, and L. E. Grove. 1982. A system of naming habitat types in small streams, with examples of habitat utilization by salmon during low stream
flow. Pages 62-73 in N. B. Armantrout, editor. Acquisition and salmon. Environmental Biology of Fishes 8:203-216.
CHaMP (Columbia Habitat Monitoring Program). 2016. Scientific protocol for salmon habitat surveys within the Columbia Habitat Monitoring Program. Prepared by CHaMP for the Bonneville
Power Administration.
Carter, David R., Robert A. Slesak. Timothy B. Harrington. David H. Peter. Anthony W. D’Amato. 2019. Scotch broom (Cytisus scoparius) modifies microenvironment to promote nonnative plant
communities. Biol Invasions (2019) 21:1055–1073, https://doi.org/10.1007/s10530-018-1885-y.
Collins, B.D., D.R. Montgomery, K.L. Fetherston and T. Abbe. 2012. The Floodplain large-wood cycle hypothesis: a mechanism for the physical and biotic structuring of temperate forested
alluvial valleys in the North Pacific Coastal Ecoregion. Geomorphology 139/140, 460-470.
Cram, J., N. Kendall, A. Marshall, T. Buehrens, T. Seamons, B. Leland, K. Ryding, and E. Neatherlin. 2018. Steelhead at risk report: Assessment of Washington’s steelhead populations.
DNR (Washington State Department of Natural Resources). 2013. Geospatial coverage of DNR managed roads. WA State Department of Natural Resources. https://fortress.wa.gov/dnr/adminsa/gisdata/datadownl
oad/state_roads.zip
Fausch, K. D., and T. G. Northcote. 1992. Large woody debris and salmon habitat in a small coastal British Columbia stream. Canadian Journal of Fisheries and Aquatic Sciences 49(4):682-693.
Forks WA, Chamber of Commerce, https://forkswa.com/listing/forks-chamber-of-commerce/
Frissell, C. A., S. B. Adams, and N. H. Hitt. 2000. Identifying priority areas for salmon conservation in Puget Sound basin. Flathead Lake Biological Station Open File Report, The University
of Montana, Polson, MT. 155 p.
Gann GD, McDonald T, Walder B, Aronson J, Nelson CR, Jonson J, Hallett JG, Eisenberg C, Guariguata MR, Liu J, Hua F, Echeverria C, Gonzalez E, Shaw N, Decleer K, Dixon KW. 2019. International
principles and standards for the practice of ecological restoration. Second edition. Restoration Ecology S1-S46
Grove, S., K.A. Haubensak, and I.M. Parker. 2017b. Do impacts of an invasive nitrogen-fixing shrub on Douglas-fir and its ectomycorrhizal mutualism changeover time following invasion?
Journal of Ecology, 105 (6). DOI: https://doi.org/10.1111/1365-2745.12764
Halofsky, J.E., D.L. Peterson, K. O’Halloran, and C. Hawkins Hoffman. 2011. Adapting to climate change at Olympic National Forest and Olympic National Park. USDA Forest Service General
Technical Report PNW-GTR-844. Pacific Northwest Research Station, Portland.
Hoh Tribe. 2016. State of Our Watersheds Report. Hoh River Basin. Available at: https://geo.nwifc.org/SOW/SOW2016_Report/Hoh.pdf
Huntington, C., W. Nehlsen, and J. Bowers. 1994. Healthy stocks of anadromous salmon in the Pacific Northwest and California. Oregon Trout, Portland, Oregon.
Jones, Chad C. and Sarah Reichard. 2009. Current and potential distributions of three non-native invasive plants in the contiguous USA. Natural Areas Journal 29:332–343. https://bioone.org/journals/N
atural-Areas-Journal
Isaak, D., S. Wenger, E. Peterson, J. Ver Hoef, D. Nagel, C. Luce, S. Hostetler, J. Dunham, B. Roper, S. Wollrab, G. Chandler, D. Horan, S. Parkes-Payne. 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. https://doi.org/10.1002/2017WR020969
IPCC. 2000. Special Report on emissions Scenarios. A Special Report of Working Group III of the Intergovernmental Panel on Climate Change, Nakićenović and R. Swart (Editors). Cam-
bridge University Press, Cambridge, United Kingdom and New York City, New York, 599 pp.
IPCC. 2021. Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte,
V., P. Zhai, A. Pirani, S. L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M. I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T. K. Maycock, T. Waterfield,
O. Yelekçi, R. Yu and B. Zhou (eds.)]. Cambridge University Press. In Press.
Kilkenny F, Clair BS, and Horning M. 2013. Climate change and the future of seed zones. USDA Forest Service Proceedings, RMRS. 2013.
Kramer, N., and E. Wohl. 2017. Rules of the road: A qualitative and quantitative synthesis of large wood transport through drainage networks. Geomorphology. 279:74-97.
Lavergne, S. and J. Molofsky. 2004. Reed canary grass (Phalaris arundinacea) as a biological model in the study of plant invasions. Critical Reviews in Plant Science, 23 (5), 415-429.
DOI: https://doi.org/10.1080/07352680490505934
Leopold, L. B. and M. G. Wolman. 1957. River channel patterns: braided, meandering and straight. U.S. Geological Survey Professional Papers 262B:39-85, Washington, D.C.
Lieb, A., T. Perry. 2005. Seasonal Variability for Select Streams within WRIA 20, Olympic Peninsula, Washington. Bureau of Reclamation Technical Services Center.
Lyons, E.W. 2003. Mass wasting in the Upper Hoh watershed, Olympic National Park, Washington: Interpreted from 1939 and 2001 Aerial Photography. U.S. Bureau of Reclamation.
Mardani, Hossein, Elena Kazantseva, Vladimir Onipchenko, Yoshiharu Fujii. 2016. Evaluation of allelopathic activity of 178 Caucasian plant species. International Journal of Basic and
Applied Sciences, 5 (1) (2016) 75-81. doi: 10.14419/ijbas.v5i1.5631Website:www.sciencepubco.com/index.php/IJBASMartinez, A. E. and P.F. McDowell. 2016. Invasive reed canarygrass (Phalaris
arundinacea) and native vegetation channel roughness. Invasive Plant Science and Management, 2016 (9), 12–21. DOI: https://doi.org/10.1614/IPSM-D-15-00046.1
McHenry, M.L. 2001. Fisheries habitat module. Middle Hoh River Watershed Analysis, Washington State Department of Natural Resources, Forks, WA.
McHenry, M.L., J. Lichatowich, and R. Hagaman. 1996. Status of Pacific Salmon and their habitats on the Olympic Peninsula watersheds. Washington Department of Ecology, Olympia.
McMillan, J. R., and J. C. Starr. 2008. Identification and prioritization of salmon tributaries for conservation in the Hoh River Basin, Washington State. Wild Salmon Center, Portland,
Oregon.
MEER:Reflection, https://www.meerreflection.com/
Montgomery, D.R. and T.B. Abbe. 2006. Influence of logjam-formed hard points on the formation of valley-bottom landforms in an old-growth forest valley, Queets River, Washington, USA.
Quaternary Research 65: 147-155.
Montgomery, D. R., E. M. Beamer, G. R. Pess, T. P. Quinn. 1999. Channel type and salmon spawning distribution and abundance. Canadian Journal of Fisheries and Aquatic Sciences 56:377-387.
Montgomery, D. R., J. M. Buffington, R. D. Smith, K. M. Schmidt, and G. Pess. 1995. Pool spacing in forest channels. Water Resources Research 31(4): 1097-1105.
Moseley C, Ellison A, Davis EJ. 2015. Quick Guide: Assessing, Planning, and Monitoring to Increase Local Economic Opportunities from Restoration. University of Oregon, Ecosystem Workforce
Program.
Naiman, Robert J., J. Scott Bechtold, Timothy J. Beechie, Joshua J. Latterell, and Robert Van Pelt. (2010). A process-based view of floodplain forest patterns in coastal river valleys
of the pacific northwest. Ecosystems (2010) 13: 1–31 DOI: 10.1007/s10021-009-9298-5.
National Parks Conservation Association. 20XX. Park on the Edge – Funding Shortfalls at Olympic National Park. https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwi3g4-nhaDzAhWM
Hc0KHYnFDkAQFnoECBYQAQ&url=https%3A%2F%2Fnpca.s3.amazonaws.com%2Fdocuments%2F3329%2Fd4ef7f57-8b93-4385-9af2-4d342509762b.pdf%3F1449154692&usg=AOvVaw2b6qfFvwQHcPtvJqwwX
Nelson, L. 1986. Fluvial Sediment in the Hoh River. In Reconnaissance of the Water Resources of the Hoh Indian Reservation and the Hoh River Basin, Washington. USGS Water-Resources Investigations
Report 85-4018.
North Pacific Coast Lead Entity. 2020. North Pacific Coast (WRIA 20) salmon restoration strategy.
NV5 GeoSpatial. 2021. Hoh River, Washington orthophotography, technical data report. Technical Report to Jefferson County, Port Townsend, Washington.
Parks, D. 1999. Mass wasting module, in Hoh River Watershed Analysis, Middle Hoh and Rainforest Watershed Analysis Units. Level 2. Washington Department of Natural Resources. Appendix
A.
Pazzagli, F., G. Thackray, M. Brandon, K. Wegmann, J. Gosse, E. McDonald, A. Garcia, D. Prothero. 2003. Tectonic geomorphology and the record of Quaternary plate boundary deformation
in the Olympic Mountains. Geological Society of America Field Guide 4. https://cpb-us-w2.wpmucdn.com/campuspress.yale.edu/dist/7/538/files/2016/01/Pazzaglia2003_Tectonic-geomorphology-and-the-record
-of-Quaternary-plate-boundary-deformation-in-the-Olympic-Mountains-1fr2k22.pdf
Peters, et al. 2012. Fish abundance, habitat, and habitat use at two stabilized banks in the Hoh River, Washington. Report to the State of Washington Department of Transportation, Lacey,
Washington.
Peterson, N. P., and L. M. Reid. 1984. Coho riverine habitat: Its evolution, distribution, and use. Pages 215-225 in Proceedings of the Olympic Wild Fish Conference, Fisheries Technology
Program, Peninsula College, Port Angeles, Washington.
Perry, L.G., L.V. Reynolds, T.J. Beechie, M.J. Collins and P.B. Shafroth. 2015. Incorporating climate change projections into riparian restoration planning and design. Ecohydrology.
Piety, L. A., J. A. Bountry, T. J. Randle, and E. W. Lyon. 2004. Summary report for geomorphic assessment of Hoh River in Washington State. River Miles 17 to 40 between Oxbow Canyon
and Mount Tom Creek. Report to the U.S. Department of the Interior Bureau of Reclamation.
Quinault Indian Nation (QIN). 2006. Salmon Habitat Restoration Plan, Upper Quinault River. 440 p.
Quinn, T. P. 2018. The behavior and ecology of Pacific salmon and trout. University of Washington Press, Seattle.
Slesak, R.A., Harrington, T.B., and D’Amato, A.W. (2016). Invasive scotch broom alters soil chemical properties in Douglas-fir forests of the Pacific Northwest, USA. Plant and Soil,
398(1). DOI: 10.1007/s11104-015-2662-7
Roni, P., C. Johnson, T. DeBoer, and G. Pess. 2016. Interannual variability in the effects of physical habitat and parentage on Chinook salmon egg-to-fry survival. Canadian Journal of
Fisheries and Aquatic Sciences 73:1-13
Schmetterling, D.A., C.G. Clancy, and T.M. Brandt. 2001. Effects of riprap bank reinforcement on stream salmon in the western United States. Fisheries 26(7):6-13.
Sedell, J. R., P. A. Bisson, J. A. June, and R. W. Speaker. 1982. Ecology and habitat requirements of fish populations in South Fork Hoh River, Olympic National Park. Ecological Research
in National Parks of the Pacific Northwest 35-42.
Smith, C. J. 2003. Salmon habitat limiting factors in Washington State. Washington State Conservation Commission, Olympia, Washington.
Society for Ecological Restoration, www.ser-rrc.org
Thackray, G. 2001. Extensive Early and Middle Wisconsin Glaciation on the Western Olympic Peninsula, Washington, and the Variability of Pacific Moisture Delivery to the Northwestern
United States. Quaternary Research 55(3): 257-270
Urgenson, L.S., S.H Reichard, and C.B. Halpern. (2009). Community and ecosystem consequences of giant knotweed (Polygonum sachalinense) invasion into riparian forests of western Washington,
USA. Biological Conservation 142, 1536-1541. DOI: 10.1016/j.biocon.2009.02.023.
Weilhofer, C. L., D. Williams, I. Nguyen, K. Jakstis., and C. Fischer. 2016. The effects of reed canary grass (Phalaris arundinacea L.) on wetland habitat and arthropod community composition
in an urban freshwater wetland. Wetlands Ecol Manage. DOI 10.1007/s11273-016-9507-x. Web. Accessed January 22, 2019.
Western Federal Lands. Upper Hoh River Road Project Information. Unpublished poster accesses 2021. https://highways.dot.gov/sites/fhwa.dot.gov/files/Board1.pdf
WDFW (Washington Department of Fish and Wildlife) and NWIFC (Northwest Indian Fisheries Commission). 2014. Statewide Washington Integrated Fish Distribution (SWIFD) Schema and data dictionary.
Pages 5 in a. t. N. I. F. C. Washington Department of Fish and Wildlife, editor. Olympia, Washington.
Washington Recreation and Conservation Office, https://rco.wa.gov/
Washington Secretary of State, https://www.sos.wa.gov/corps/
Washington Small Business Administration, https://www.sba.gov
Welber, M., W. Bertoldi, and M. Tubino. 2013. Wood dispersal in braided streams: Results from physical modeling. Water Resources Research 49:7388-7400.
Winkowski, J. 2020. The Olympic Peninsula Thermalscape Model: A Spatial Stream Network Model for August Stream Temperature on the Outer Olympic Peninsula, WA (WRIA 20 and 21). Washington
Department of Fish and Wildlife Coast Ecology and Life Cycle Monitoring Unit.