HomeMy WebLinkAboutNSD_JeffCo_DosewallipsBrinnonAssessment_022125
Tami Pokorny
Natural Resources Coordinator
Jefferson County Public Health Department
615 Sheridan Street
Port Townsend, WA 98368
1900 N. Northlake Way, Suite 211
Seattle, WA 98103
Dosewallips River – Brinnon Reach
Draft Existing Conditions Assessment Memo
February 2025
THIS PAGE INTENTIONALLY LEFT BLANK
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design i
February 2025
TABLE OF CONTENTS
1 Project background ............................................................................................................................................ 1
1.1 Assessment Goals .................................................................................................................................... 2
1.2 Reach History ........................................................................................................................................... 2
1.3 Previous Restoration Efforts .................................................................................................................... 4
2 Current Conditions ............................................................................................................................................. 4
2.1 Flows During 2024 Site Visit ..................................................................................................................... 4
2.2 Reach Geology ......................................................................................................................................... 5
2.3 Channel and Floodplain Morphology ....................................................................................................... 7
2.4 Sediment .................................................................................................................................................. 9
2.5 Channel Migration ................................................................................................................................. 12
2.6 Aquatic Habitat Conditions .................................................................................................................... 15
2.7 Salmonid Use, Life History, and Limiting Factors in the Lower Dosewallips River ................................ 18
2.7.1 Puget Sound Chinook (Oncorhynchus tshawytscha) ................................................................ 19
2.7.2 Hood Canal Summer Chum ESU (Oncorhynchus keta) ............................................................. 19
2.8 Large Wood Distribution and Function .................................................................................................. 20
2.9 Riparian and Wetland Communities ...................................................................................................... 24
2.10 Nearshore and Estuary Habitat .............................................................................................................. 29
3 Hydrology and Hydraulics ................................................................................................................................ 29
3.1 Hydrology ............................................................................................................................................... 29
3.1.1 Tidal Datums ............................................................................................................................. 32
3.1.2 Hydrologic Effects of Climate Change ....................................................................................... 33
3.1.3 Future Tidal Datums .................................................................................................................. 34
3.2 Hydraulics ............................................................................................................................................... 34
3.2.1 Model Development ................................................................................................................. 34
4 Existing Conditions Model Results ................................................................................................................... 38
4.1 Implications for salmonids ..................................................................................................................... 38
4.2 Implications for flood and channel migration risk ................................................................................. 40
4.3 Climate Change ...................................................................................................................................... 43
5 Flood and Channel Migration Risk ................................................................................................................... 45
5.1 Flood Risk Focus Areas ........................................................................................................................... 45
5.2 Projected Changes in Flood Risk with Climate Change .......................................................................... 52
5.2.1 Findings from this Study ........................................................................................................... 52
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design ii
February 2025
5.2.2 Summary of findings in 2023 Jefferson County Sea Level Rise study ....................................... 53
6 Primary Impairments to Fluvial Processes ....................................................................................................... 56
6.1 Disruption of the Floodplain-Large-Wood-Cycle ................................................................................... 56
6.2 Artificial Bank Hardening and Channel Simplification ........................................................................... 57
6.3 Undersized Hydraulic Structures and Lack of Flood Protection ............................................................ 57
7 Proposed Actions to Treat Impaired Processes ............................................................................................... 58
8 References ....................................................................................................................................................... 60
LIST OF TABLES
Table 1. Habitat unit widths, depths, and areas ...................................................................................................... 16
Table 2. Side channel lengths................................................................................................................................... 16
Table 3. Periodicity for key salmonid species within the lower Dosewallips River. ................................................ 19
Table 7. Estimated discharge values ........................................................................................................................ 32
Table 8. Estimated Change and Magnitude of Future Flows ................................................................................... 34
Table 10. Range of Sea-Level Rise Projections for Jefferson County, WA ............................................................... 54
Table 11. Table of impairments and proposed actions ........................................................................................... 58
LIST OF FIGURES
Figure 1. Overview Map of the Dosewallips River Watershed and Project Reach .................................................... 1
Figure 2. Photo of Splash Dam in Early 1900s ............................................................................................................ 2
Figure 3. Historic 1939 aerial photo of the project reach and the surrounding upland forest ................................. 3
Figure 4. Flow conditions for the 2023/2024 water year on the Duckabush River gage USGS 12054000 ................ 5
Figure 5. 1:24k Geologic map of the Brinnon project area ........................................................................................ 6
Figure 6. Deep seated landslides in glacial sediments ............................................................................................... 6
Figure 7. Relative elevation model map .................................................................................................................... 8
Figure 8. Longitudinal profile of the project reach .................................................................................................... 9
Figure 9. Dominant substrate map .......................................................................................................................... 10
Figure 10. Pebble count results relative to preferred chum and chinook spawning ranges ................................... 11
Figure 11. Time series of Dosewallips River delta.................................................................................................... 12
Figure 12. Channel migration history relative to bank modifications and previously constructed ELJs ................. 13
Figure 13. Channel migration zone (CMZ) mapped by the USBOR (2004) .............................................................. 14
Figure 14. Photos of left bank berm feature preventing northward channel migration ......................................... 15
Figure 15. Habitat unit and side channel map ......................................................................................................... 17
Figure 16. Constructed side channels ...................................................................................................................... 18
Figure 17. Large wood observed during the June 2024 field reconnaissance ......................................................... 22
Figure 18. Examples of a key piece (left) and functional wood (right) .................................................................... 23
Figure 19. Examples of an engineered log jam (left) and natural log jam (right) .................................................... 23
Figure 20. Functional ELJ at RM 0.1 ......................................................................................................................... 24
Figure 21. Canopy height and wetland map ............................................................................................................ 26
Figure 22. Constructed channel through native forest; emergent estuary community .......................................... 27
Figure 23. Young alder forest; Alder-Cottonwood community downstream of the US-101 bridge ........................ 27
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design iii
February 2025
Figure 24. Wetlands forming around high flow side channels; narrow band of trees rooted into berm ............... 27
Figure 25. Side channel, pool, and wetland habitat; high gravel bar with wood debris ......................................... 28
Figure 26. Big-leaf maple and red alder forest; and mixed conifer/deciduous forest ............................................ 28
Figure 27. Two seasonal tributaries drain through the southern terrace and flow into the Dosewallips River ..... 28
Figure 28. Key salmonid life history stages and Dosewallips River flows ................................................................ 30
Figure 29. Drainage areas utilized for hydrologic analysis and hydraulic model..................................................... 31
Figure 30. CoSMoS data station availability in the vicinity of the project area. ...................................................... 33
Figure 31. Boundary condition locations within hydraulic model domain. ............................................................. 35
Figure 32. Hydraulic structures included in hydraulic modeling for project reach. ................................................ 36
Figure 33. Existing berm extents based on 2023 LiDAR ........................................................................................... 37
Figure 34. Modeled water depth for existing conditions spawning and juvenile outmigration flow scenarios ..... 38
Figure 35. Modeled flow velocity for spawning and juvenile outmigration flow scenarios .................................... 39
Figure 36. Water depth and velocity distributions during median February exceedance flow .............................. 39
Figure 37. Water depth and velocity distributions during median September exceedance flow ........................... 40
Figure 38. Modeled water depth for existing conditions flood flow scenarios ....................................................... 41
Figure 39. Modeled flow velocity for existing conditions flood flow scenarios ...................................................... 42
Figure 40. Water surface elevations indicating location of backwater influence by MHHW (red line) .................. 42
Figure 41. Water depth and velocity distributions and preferences for key salmonid species .............................. 43
Figure 42. Modeled changes in depth and velocity relative to climate change predictions ................................... 44
Figure 43. Areas of heightened flood risk ................................................................................................................ 45
Figure 44. Berm extent as designed (left) and observed in 2023/2024 (right) ....................................................... 47
Figure 45. Berm cross section at three modeled flows (2-year, 100-year, typical winter flow) ............................. 48
Figure 46. Profile of berm design (WA State Department of Waterways, 1957) .................................................... 48
Figure 47. REM and channel migration map along extent of berm ......................................................................... 49
Figure 48. Modeled water depth along length of berm at 2-year and 100-year flows ........................................... 50
Figure 49. Flooding in Dosewallips State Park Campground ................................................................................... 51
Figure 50. Diagram of the floodplain large-wood cycle (Collins et al., 2012) .......................................................... 53
Figure 51. Still water elevations and total water levels for the 1% event ............................................................... 54
Figure 52. Sea level rise projections (ESA 2023) ...................................................................................................... 55
Figure 53. Erosion effects of sea level rise (Battalio et al., 2016) ............................................................................ 56
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 1
February 2025
1 PROJECT BACKGROUND
Natural Systems Design (NSD) was contracted by Jefferson County to assess the existing conditions in the lower
1.2 miles of the Dosewallips River, where it runs south of the town of Brinnon and into Dabob Bay of the Hood
Canal (Figure 1). The assessment is focused on characterizing the geomorphic, habitat, and hydraulic conditions
in the project reach, along with assessing flood and channel migration risk to the town of Brinnon. Additionally,
the assessment includes insights into how existing conditions are projected to change with climate change. The
habitat assessment is focused on Puget Sound Chinook (Oncorhynchus tshawytscha) and Hood Canal summer-
run chum (Oncorhynchus keta), both listed under the Endangered Species Act as federally threatened species.
This assessment will ultimately be the basis for conceptual designs for aquatic habitat and flood risk
improvements.
Figure 1. Overview Map of the Dosewallips River Watershed and Project Reach
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 2
February 2025
1.1 Assessment Goals
The specific goals for this assessment are as follows:
1. Quantifying existing aquatic and riparian habitat associated with key salmonid species.
2. Assessing channel migration and flooding risk to key infrastructure and property in the town of Brinnon,
US 101, and the Dosewallips State Park Campground.
This assessment is intended to provide reach-scale information on impaired processes and their causal
mechanisms with respect to addressing these two goals. The outcomes of this assessment will inform future
restoration actions. Actions will focus on addressing:
1. Areas of heightened flood and erosion risk, and improving sustainability to existing mitigation structures
2. Increasing the quality and quantity of key aquatic habitats
3. Increasing the structure and function of riparian habitats.
1.2 Reach History
The Dosewallips river valley has a rich history of human occupation and homesteading, with the most
documented accounts during and after the 1900’s. Native Americans inhabited the Dosewallips region for
14,000 years, with the first contact with Europeans occurring in the late 16th century (Bush et al., 2023). In the
early 19th century, Euro-American settlement began and increased after the Donation and Land Claim Act of
1850, and the homestead act of 1862 (Bush et al., 2023). Logging in the area initiated in 1859 which
incorporated railroads near the turn of the 20th century, and later trucks in 1920 (Labbe et al., 2005). Logging
impacts in the floodplain and river were extensive, including the use of a splash dam at the downstream end of
the Rocky Brook reach (~RM 3.5) which was constructed in 1917 and operated for 9-10 years by Sims Logging
Company (Figure 2; Labbe et al., 2005, Baily & Baily, 2011). When water was released from the dam, all wood
was transported from the head of the Dosewallips Canyon to Hood Canal (Labbe et al., 2005) – an erosive event
that was catastrophic for salmon and salmon habitat.
Figure 2. Photo of Splash Dam in Early 1900s
Source: Baily and Baily (2011)
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 3
February 2025
Historic accounts from Vern and Ida Baily, longtime residents of the Dosewallips Valley, indicate that the river
was full of large wood and old growth trees in the early 1900’s, including log jams that were large and stable
enough to cross and fish from (Labbe et al., 2005, p74).
They also stated that some jams appeared to be more than 100 years old. At present, there is little to no wood
in the stream, nor large old growth trees in the valley bottom. Vern and Ida Bailey also recounted an abundance
of king salmon and steelhead, which are rarely seen or caught today:
“I remember when I was eight years old the big king salmon coming up here. There were great
big ones as big as this table. Of course you never see them anymore. My dad was a fisherman
and he would catch them. And he caught lots of steelhead. And now we hardly ever get a
steelhead. I’m sure the steelhead were wild steelhead in those days.” – Ida Bailey (Labbe et al.,
2005, p74).
Historical imagery from the early 1900’s shows that the landscape and channel location have been relatively
stable throughout the last 100 years between RM 0 and RM 1.2 (Figure 3). Forested uplands managed by the
state flank the southern bank and valley, with a mix of pasture, forest, and municipal land to the north in the
town of Brinnon. The project area to the east of Hwy 101 is owned by a mix of the state and private landowners,
with several areas also leased for commercial shellfish production. The stability in planform of the project reach
is owed significantly to human modifications to reduce flood risk and facilitate human settlement (Labbe et al.,
2005). Several left and right bank dikes have been constructed, particularly surrounding and downstream of the
US-101 bridge, where the thalweg has also been dredged repeatedly to maintain the current planform (Labbe et
al., 2005). Part of the bank stabilization throughout the 1900’s was large woody debris removal (Labbe et al.,
2005), the outcomes of which are prevalent in current conditions as of 2024, with limited instream wood
throughout the project reach.
Figure 3. Historic 1939 aerial photo of the project reach and the surrounding upland forest
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 4
February 2025
1.3 Previous Restoration Efforts
From the early-2000’s through 2022, the Wild Fish Conservancy and Washington State Parks partnered to
implement habitat restoration and park facility updates at the Dosewallips State Park downstream of the
Highway 101 bridge, an effort called the Lower Dosewallips Floodplain and Estuary Restoration project. The
overall goal of the project was to increase floodplain and estuarine ecological processes in the lower Dosewallips
River and Estuary through dike, revetment, and levee removal, the construction of ELJs and distributary
channels, invasive plant removal and riparian plantings, and re-location of park facilities out of the lower
floodplain. By the spring of 2022 the removal of 1000 ft of revetment along the main river shoreline, and the
removal 2820 linear feet of asphalt roads in the floodplain was completed, including 500 feet of road that was
buried 2 feet underneath the existing surface and was likely abandoned by parks after a flood buried it in the
1960's or 1970's (Figure 12). The project also created over 3500 feet of new distributary channel and
constructed 3 engineered log jams at distributary junctions with the main river, to work with the 5 ELJs that
were constructed during a previous phase. Invasive plant species (Himalayan blackberry) was removed across an
area of 2 acres, and riparian plantings occurred throughout the entire area disturbed by construction.
2 CURRENT CONDITIONS
To assess current conditions in the project reach, NSD used a combination of newly acquired lidar, GIS, hydraulic
modeling, field reconnaissance, and previous design records of bank and instream modifications. Lidar was
collected by NV5G (October 31st, 2023) and was used to develop the hydraulic model and additional geospatial
products, such as a relative elevation map. Hydraulic modeling was completed at multiple flood and fish life
history flows, across both an existing conditions scenario and with projected mid-century hydrologic and sea
level conditions (For more details see Sections 3 and 4).
During the field reconnaissance we collected information on 1) channel and floodplain morphology 2)
streambed sediment, 3) riparian habitat, 4) large wood quantification and function, and 5) instream aquatic
habitat features. Along with field assessments of current conditions, the team also assessed reach geology,
channel migration history, and reviewed documentation on nearshore/estuarine environments. An additional
focus of the reach assessment was observing critical hydraulic structures within the project reach, including
culverts, bridge crossings, and a berm on the northern bank upstream of US-101. We discuss the findings of each
of these observations below.
2.1 Flows During 2024 Site Visit
The flows leading up to the 2024 site visit were reflective of the tail end of winter rainfall events and the onset
of spring snowmelt on the eastern Olympic Peninsula. As discussed in more detail in section 3.1 below, the gage
on the Dosewallips River (USGS 12053000, drainage area 93.5 square miles) has been inactive since 1951,
however the Duckabush gage in the basin immediately to the south (USGS 1205400, drainage area 66.5 square
miles) can be used to approximate current flow conditions. Figure 4 below shows the 2023/2024 water year on
the Duckabush River, with the dates of lidar acquisition and the 2024 site visits highlighted. The 2023 lidar
dataset was acquired between flashy fall rain events, and the 2024 site visit occurred at the onset of spring
snowmelt, at a slightly higher flow than when the lidar was flown.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 5
February 2025
Figure 4. Flow conditions for the 2023/2024 water year on the Duckabush River gage USGS 12054000
Lidar acquisition and 2024 field visit dates are highlighted.
2.2 Reach Geology
The Dosewallips watershed lies on the eastern slope of the Olympic Mountains which are composed of uplifted
and folded oceanic crust formed by subduction of the Juan de Fuca plate under the North American plate
offshore of the Pacific coast. The upper portion of the watershed is composed of slightly metamorphosed and
highly erodible marine sedimentary lithologies formed during uplift processes. Because of their highly erodible
nature, these uplifted marine sedimentary materials provide the Dosewallips River with a high, fine-grained
sediment load (Labbe et al., 2005). Much of this material is transported as suspended load, with the majority of
substrate within the Dosewallips active channel characterized as a gravel/cobble.
The middle and lower portions of the watershed are composed of less erodible Crescent formation basalts
(Figure 5) which are exposed at the uppermost extent of the project reach where the river exits a bedrock
constriction. The valley bottom below the constriction at RM 1.2 is filled with Holocene alluvium (river-derived
sediment) that has been deposited since the retreat of the Cordilleran ice sheet over the last ~10,000 years.
Hillslopes to the south in the Dosewallips state park are overlain with glacial outwash and till that have been
compacted by ice, along with isolated glacial lacustrine deposits. These glacial sediments are unconsolidated and
highly erodible, which makes them susceptible to mass-wasting processes such as landslides. There are two
deep-seated landslide deposits within the project reach which are located northwest of the Dosewallips State
Park Campground, between RM 0.7 and RM 1 (Figure 6).
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 6
February 2025
Figure 5. 1:24k Geologic map of the Brinnon project area
Figure 6. Deep seated landslides in glacial sediments
Source: WDNR Landslides Database (accessed January 2025)
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 7
February 2025
2.3 Channel and Floodplain Morphology
The project reach has four distinct morphological zones (MZ) (Figure 7). These include, from upstream to
downstream: 1) a simplified pool-riffle reach between RM 0.9-1.2 below the bedrock constriction, 2) a pool-riffle
reach with a large right-bank side channel complex between RM 0.6 and RM 0.9, 3) an armored, constricted
section under the US-101 bridge crossing, between RM 0.3 and RM 0.6, and 4) a highly modified and tidally
influenced zone with one main channel and several right and left bank distributaries downstream of RM 0.3.
Generally, the mainstem channel has a pool-riffle morphology throughout, with limited influence of instream
wood. The tidal influence was apparent starting at/downstream of the US-101 bridge, mainly in the left bank
distributary channels. The right bank side channel complex within the state park west of the bridge (MZ2) was
only contained ponded waters during the site visit, with evidence of overbank flow during flood events. This
indicates that these side channels are only active during higher flow events. Besides this area, the floodplain was
generally disconnected from frequent flood flows due to a lack large log jams that can work to connect side
channel and floodplain areas, and intentional bank modifications throughout the 1900’s (Labbe et al., 2005) to
keep the channel within the existing planform.
The side channel complex throughout MZ2 has an ephemeral morphology, with frequent migration of the
mainstem active channel through this area and back towards the current position along the northern bank (see
section 2.5 below). The frequent migration through this area has led to erosion of developing floodplain
surfaces, limiting the growth of mature forest and reproduction of large wood to stabilize banks and provide
cover for salmonids. The high channel migration rate in this area is in a positive feedback cycle, fed by a
combination of a lack of mature trees to stabilize banks and sediment aggradation (see section 2.4 below). The
slope of the channel profile throughout the 1.2 mile long reach is relatively stable at 0.4% (Figure 8) indicating
that the relatively higher sediment aggradation is most likely attributed to a combination of local hillslope
sources, such as the right bank landslides immediately adjacent (Figure 6), and backwater from the US-101
bridge constriction, as found by a larger regional assessment by Aspect Consulting (2009). The reach just
upstream of the bedrock constriction at the top of the project reach, referred to as Powerlines in other reports
(i.e. NSD 2021) has high aggradation and channel migration rates as well, owing similarly to backwater behind
the natural bedrock constriction. Significant sediment deposition in the Powerlines reach may also be
contributing to the lower aggradation and channel migration rates seen in MZ1, which his comparatively much
more stable than MZ2, and has limited artificial bank armoring.
Downstream of MZ2, MZ3 and MZ4 are also comparatively much more stable, owing mainly to bank
modifications. A long history of bank hardening, channel dredging, and large wood removal throughout the
1900’s kept the channel straight downstream of RM 0.6 (Labbe et al., 2005), including dikes downstream of the
US-101 bridge, eliminating connectivity of distributary channels. Currently, bank modifications include the left-
bank armoring/berm between RM 0.4 and RM 0.7, which prevents channel migration towards the town of
Brinnon, and previously placed engineered log jams on both banks downstream of the US-101 bridge, which
encourage flow into distributary channels. The armoring of the left bank intentionally removes hydrologic
connectivity with the northern floodplain, and the distributary channels downstream of the bridge are
contributing to 1) floodplain connectivity and additional estuary habitat for rearing summer chum, and 2) a
stable morphology, where flow and shear stress is distributed across multiple stable channels as opposed to a
single mainstem.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 8
February 2025
Figure 7. Relative elevation model map
Elevations are relative to the low flow water surface. MZ= Morphology Zone
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 9
February 2025
Figure 8. Longitudinal profile of the project reach
2.4 Sediment
We classified the dominant and subdominant grain size per habitat unit throughout the project reach and
collected two pebble counts (Wolman 1954) to characterize patterns in sediment deposition and transport
(Figure 9, Figure 10). Generally, the substrate retained in this reach is influenced by channel morphology, with
coarser sediment in riffle areas and finer sediment in pools as expected (Montgomery and Buffington, 1997). In
MZ1 and MZ2 (Figure 7), the pool riffle sections downstream of the bedrock constriction, the substrate has a
pattern alternating between coarse gravel and coarse cobble, aligned with pools and riffles respectively.
Downstream of RM 0.6, where bank modifications become most prevalent and the river channel is constricted
to flow under the US 101 bridge, the grain size distribution in the channel coarsens, and is large-cobble-
dominated throughout most of the mapped habitat units. Pockets of sand and medium gravel were apparent in
deeper pools, which are forced by both wood and left-bank armoring features. There is also evidence of bed
armoring when comparing surface and subsurface grain size distributions, such as at PC1, where subsurface
sediment has a much finer distribution than the surface at the same location (Figure 10).
Sediment inputs into the project reach come from erodible marine sedimentary bedrock upstream in the
watershed, along with glacial outwash, till, and lacustrine deposits. Importantly, glacial sediments on hillslopes
are susceptible to deep-seated landslides (Figure 6), which have the potential to input a significant amount of
sediment in distinct events.
Pebble counts on representative gravel bars (Figure 9) showed grain size distributions within that preferred by
Chinook salmon, and are slightly coarser than the sizes preferred by chum (Figure 10). The subsurface grain size
distribution indicates that smaller grain sizes preferred by chum (versus Chinook) is entering the reach, however
it is not retained due to the reach hydraulic conditions.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 10
February 2025
Figure 9. Dominant substrate map
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 11
February 2025
Figure 10. Pebble count results relative to preferred chum and chinook spawning ranges
See pebble count locations in Figure 9.
The high sediment load entering the project reach is readily transported upstream of RM 0.9 and downstream of
RM 0.5, as indicated by the stable morphology (described in further detail in section 2.5 below) and coarse,
armored bed in these locations. The area between RM 0.5 and 0.9, in contrast, shows evidence of frequent
channel migration and sediment deposition, in both historic imagery and in the current morphology of several
dry overbank/side channels. This indicates that the lower 1.2 miles are primarily a “transport” reach where the
channel has the capacity to transport the sediment load input, with one dynamic segment of sediment
deposition and channel response. Section 2.3 above discusses how local sediment sources and the impact of
backwatering behind the US-101 bridge during peak events (Aspect Consulting, 2009) is contributing to higher
aggradation rates between RM 0.5 and RM 0.9.
The downstream extent of the delta at the outlet of the Dosewallips River has remained generally stable since
1939. The lack of progradation is likely attributed to the steep, glacially carved bathymetry of Hood Canal, which
drops to -122 meters (400 feet) in depth less than a half mile from shore (NOAA 1969), rather than a lack of
sediment transport through the project reach. High transport loads at the delta are indicated by the frequently
meandering planform downstream of RM 0.1, outside of where bank modifications were prevalent throughout
the 1900’s (Labbe et al., 2005; Figure 11).
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 12
February 2025
Figure 11. Time series of Dosewallips River delta
2.5 Channel Migration
To assess channel migration history, we traced the wetted channel area in eight historical images between 1939
and 2023 (Figure 12). We found that the channel is generally stable upstream of RM 0.9 and downstream of RM
0.5, with a very dynamic sub-reach in between. The causal mechanisms of channel migration in this area are a
combination of 1) high sediment input into the reach (Labbe et al., 2005) 2) a decrease in channel gradient
(Aspect Consulting, 2009), 3) backwater at the US-101 bridge crossing during high flows (Aspect Consulting,
2009), and 4) limited large wood both in the stream and on the floodplain to stabilize banks (more detail in
section 3.7 below). The floodplain between RM 0.5 and RM 0.9 was dry during the site visit, with early
succession woody vegetation colonization (i.e. willow and red Alder). This type of floodplain surface is
susceptible to bank erosion, with limited root cohesion and hydraulic roughness. It is expected that the channel
will likely continue to migrate frequently within this zone without the influence of stabilizing features such as
large wood. Channel migration was limited near areas of constructed bank revetment.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 13
February 2025
Figure 12. Channel migration history relative to bank modifications and previously constructed ELJs
Photos taken June 20th, 2024
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 14
February 2025
Downstream of RM 0.5 the channel has migrated very little over the last 85 years (Figure 12). This is due to a
combination of bank modifications upstream and downstream of the US-101 bridge (Labbe et al., 2005), and the
tidal influence of the estuary further downstream. A channel migration zone study completed by the US Bureau
of reclamation in 2004 (USBOR 2004) discusses how channel migration in this area is significantly impacted by
these features. They found that the entire alluvial valley north of the mainstem channel and Dosewallips State
Park Campground is within a defined avulsion hazard area, however this area is disconnected from channel
migration, with a high dependency on existing bank revetment features (Disconnected Migration Area (DMA)
area in Figure 13). The bank armoring on the southern bank along Dosewallips State Park was eroding during
the 2004 assessment (USBOR 2004), and few exposures were observed during the June 2024 site visit (Figure
14). The degradation of these revetment features upstream of the park indicates that the features are providing
less resistance to erosion than originally intended. The left-bank erosion revetment berm feature was generally
intact, with significant vegetation establishment on top of the feature (Figure 14). This berm feature is discussed
in greater detail in section 5.1. USBOR (2004) found that channel migration rates were low upstream of RM 0.9,
even though there are no revetment features, which is corroborated by this analysis.
Figure 13. Channel migration zone (CMZ) mapped by the USBOR (2004)
Modified from US BOR (2004). “High” and “Moderate” indicate risk of channel migration. DMA = Disconnected Migration Area.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 15
February 2025
Figure 14. Photos of left bank berm feature preventing northward channel migration
The southwestern bank between RM 0.5 and 0.9 also shows evidence of deep-seated landslides, as mapped in
the WDNR landslides database (Figure 12). It is likely that these landslides were partially induced by toe erosion
by the channel into the hillslopes of glacial sediment, as occurs throughout many river valleys in western
Washington (e.g. Booth et al., 2017). While there is limited infrastructure directly adjacent to these southern
hillslopes, a future landslide in this area would deposit a significant amount of sediment in a discrete pulse to
the channel, which may lead to subsequent rapid channel changes including channel migration and avulsions.
The coincidence with the landslide locations and low-lying avulsion pathways on the north side of the active
channel highlights this area as one of particular concern as sediment deposited from a landslide could
encourage channel migration and a potential avulsion towards the north (away from the landslide), putting
pressure on the left-bank berm.
ELJs constructed between the early 2000’s and 2022 (see section 1.3 above) were placed on the left and right
banks of the mainstem Dosewallips, downstream of the US-101 bridge (Figure 12). This ELJ implementation was
completed in tandem with rock bank revetment/dike removal to reconnect the mainstem to existing distributary
channels on the left bank and newly excavated side channels on the right bank. Discussed in further detail in
sections 2.6 and 2.8 below, these features are adding bank stability to the mainstem while also connecting the
floodplain and estuary by encouraging flow into secondary channels. There has not been any significant channel
changes associated with the ELJ placement, rock bank revetment/dike removal, and side channel excavation,
with the channel remaining in a similar location since implementation. The lack of channel change associated
with the ELJ construction is inherent to where and how the structures were placed – with most being tied into
the banks, and as such not significantly obstructing mainstem flow.
2.6 Aquatic Habitat Conditions
The project reach was highlighted by Labbe et al., (2005) to have relatively higher habitat quality for Chinook
and Hood Canal summer chum than upstream reaches of the Dosewallips River, despite being highly modified to
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 16
February 2025
reduce flood hazard and facilitate human settlement. The reach is lower gradient, with relatively more seasonal
floodplain engagement and LWD on the floodplain than upstream reaches. Additional key aquatic habitat
features identified by Labbe et al., (2005) include right bank side channels within Dosewallips State Park west of
Hwy 101, which showed signs of seasonal connectivity in the early 2000’s, along with pools and isolated large
wood jams upstream of the US-101 bridge. Additional features include the Walcott Slough network in the
northern end of Brinnon which is tidally influenced but disconnected from the river, and State Park Creek, which
is a seasonal channel that is important for Chum spawning (Figure 15). Although these habitat features exist,
there is room for improving the functionality of habitat features within the project reach, specifically engaging
the right bank floodplain side channels within MZ2 for a longer percentage of time during the year.
To update the findings by Labbe et al., (2005) with current conditions, we mapped habitat units and side
channels throughout the project reach at a coarse scale (Figure 15). In the field, we collected the length,
dominant/subdominant substrate size, and unit type. We then used the lidar and hydraulic modeling tools to
determine geometric attributes like wetted width and depth (Table 1). Side channels were mapped using a
combination of field observations, the relative elevation model map (Figure 7), and hydraulic modeling results
(discussed in section 4).
The distribution of different habitat unit types aligns with the pool-riffle morphology observed at a broader
scale, with a mix of pools, riffles, and glides (Table 1). We mapped 24 pools, 22 of which were wood-forced, and
two were related to bedrock and bank modifications. Some of the wood-forced pools were directly linked to
engineered log jams from the WFC Lower Dosewallips Floodplain and Estuary Restoration project (see section
1.3). During the June 2024 site visit, juvenile salmonids were prevalent in isolated pools in the right bank side
channel complex between RM 0.5 and 0.9.
Table 1. Habitat unit widths, depths, and areas
HABITAT UNIT TYPE POOL RIFFLE GLIDE
Percent of reach area 17% 45% 39%
Mean low flow wetted width (ft)1 - 87 98
Mean low flow depth (ft)2 4.2 2.9 3.0
Mean pool area (sqft) 3,850 - -
1: Mean wetted width is based on flow during lidar flight (2023).
2: Mean low flow depth is based on modeled low flow depth.
We mapped a total of 2.7 miles of side channels with variable degrees of hydraulic and hyporheic flow
connectivity to the mainstem Dosewallips River (Table 2). The largest proportion of side channels are tidal
distributary channels that are connected and prevalent east of the US-101 bridge crossing. Of the remaining
length of mapped channels, 0.5 miles were either directly connected to the mainstem or were wetted with a
hyporheic connection during the June 2024 site visit and are located in MZ2. Seasonal and flood overflow
channels were prevalent as well, primarily in the area of high channel migration rates between RM 0.5 and RM
0.9.
Table 2. Side channel lengths
SIDE CHANNEL TYPE LENGTH (FT) LENGTH (MILES)
Connected to mainstem 1,530 0.3
Hyporheic connection 1,170 0.2
Tidal/distributary channel 7,410 1.4
Seasonal overflow channel 1,090 0.2
Flood overflow channel 2,850 0.5
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 17
February 2025
Connectivity of tidal distributary channels is a recent habitat feature as of the early 2000’s, as the area
downstream of the bridge was highly modified and straightened throughout the 1900’s (Labbe et al., 2005).
These side channels were reconnected through the Lower Dosewallips Floodplain and Estuary Restoration
project (see section 1.3) and were observed to be sustaining flows while showing some bank and channel
deformity and head-cut erosion during the 2024 site visit indicating that they are continuing to evolve following
implementation of the restoration actions.
Figure 15. Habitat unit and side channel map
Imagery source: 2023 NAIP
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 18
February 2025
Figure 16. Constructed side channels
As part of the Lower Dosewallips Floodplain and Estuary Restoration project.
Along with mainstem and side channel habitat, several small tributaries enter the Dosewallips River floodplain in
the project reach (Figure 15). A small tributary at RM 0.95 drains from the southern hillslope, across the right
bank terrace, forming a small alluvial fan at the toe of the hillslope. During the field reconnaissance there were
several channels through this fan area with exposed gravel, but no surface water. Upstream from this tributary
at RM 1.0 a large tributary channel drains from the southern hillslope. This channel has incised 8 feet down
through the terrace to form an open channel connection with the main Dosewallips River channel. This channel
was also dry during the field reconnaissance and appears to flow powerfully with seasonal variation and storm
events. A third tributary identified by Labbe et al. (2005) flows along the floodplain to the southwest of the
reach, crossing beneath US-101 near the main entrance of the State Park Campground (referred to as State Park
Creek).
2.7 Salmonid Use, Life History, and Limiting Factors in the Lower
Dosewallips River
The focal species in the Dosewallips River, fall Chinook, Hood Canal summer chum, spawn in the fall and spend
hours to months rearing in freshwater before out-migrating to estuarine and nearshore habitat. Both fall
Chinook and Hood Canal summer chum are listed as federally threatened species under the Endangered Species
Act. The life stages of these species most affected by the quality, quantity, and diversity of aquatic habitats are
spawning, incubation, emergence, and fry rearing. The following sections focus on these three species, but the
Dosewallips River also supports Pink Salmon (Oncorhynchus gorbuscha), and steelhead and rainbow trout
(Oncorhynchus mykiss).
Critical life stages for chum and Chinook in the Dosewallips River have been identified as spawning, incubation
and adult holding (Shared Strategy 2007). Incubation and rearing success are driven by large wood and
moderate peak flows, absence of excessive fines in spawning gravels, moderate or low levels of scour, and
access to off-channel and slow-water floodplain habitats (Brewer et al. 2005). Loss of riparian forest has also
been noted for the Dosewallips watershed. The floodplain in the lower reaches of the river has been converted
to agriculture, forestry, urban commercial and rural residential uses (Correa 2003). These changes in land use
have reduced the side channel and floodplain wetland habitat in the project reach and reduced the recruitment
of large wood that provides channel stability, sorts and retains spawning gravels, creates pools, and forces side
channel creation.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 19
February 2025
Table 3 presents the periodicity of habitat uses for summer chum, and Chinook in the Dosewallips River.
Additional details on life history and periodicity of these key species are listed in sections 2.7.1, and 2.72 below,
along with limiting factors to survival.
Table 3. Periodicity for key salmonid species within the lower Dosewallips River.
2.7.1 Puget Sound Chinook (Oncorhynchus tshawytscha)
Chinook, as a larger species, require larger substrate for spawning and deep holding pools with cover for adult
migration. Egg incubation is affected by bed scour during high flows and excessive fine sediment deposition
which can smother redds. Juvenile Chinook remain in the river for approximately 4 months (Table 3) and depend
on low velocity habitat and cover for rearing. The majority of juvenile Chinook in the Dosewallips and other
coastal rivers exhibit the ocean-type life history and out-migrate after about 4 months of rearing in freshwater,
completing the rest of their growth in estuarine or nearshore habitat. A small proportion of Chinook maintain a
stream-type life history, spending up to a year in fresh water before outmigration. The proportion of stream-
type Chinook in the Dosewallips is unknown but expected to be small. Low velocity habitat includes river
margins, alcoves, back watered areas, and side channels, where small fish can escape the force of the current,
have sufficient hiding cover for protection from predators, and be able to rest and feed.
The factors most limiting Chinook production in the Dosewallips River are estuarine degradation, habitat
complexity and channel conditions, high water scour and fine sediment and floodplain disconnection. These
mostly occur in the lower reaches of the river and the estuary habitat since the upper watershed is less
developed (Shared Strategy 2007).
2.7.2 Hood Canal Summer Chum ESU (Oncorhynchus keta)
Chum salmon have similar requirements for deep pools with cover for adult holding, but they use slightly
smaller gravels for spawning than Chinook (Figure 10). While most Chinook spawn in mainstem river channels,
chum salmon are more likely to spawn in lower velocity areas with smaller substrate, which may include
mainstem habitat or smaller creeks and side channels. Incubation for chum is also limited by bed scour from
high flows, perhaps to a greater extent than Chinook based on the chum behavior of mass spawning and redd
superimposition. Spawning success for chum is linked to suitable spawning gravel, adequate stream flows and
water temperatures, along with habitat quality in the form of large wood for cover and holding pools for
returning adults to rest. Excessive fine sediment is also a concern for chum eggs in terms of suffocation. Since
chum fry out-migrate upon emergence (Table 3), their dependance on adequate riverine rearing habitat is less
than that of Chinook, but they still need safe pathways to out-migrate through the mainstem channel and have a
higher need for sufficient estuarine and nearshore habitat. Chum spawning in the Dosewallips is limited to the
lower 4.3 miles of river, with the greatest concentrations below RM 2.5 (Brewer et al. 2005), which includes the
project reach.
Chinook Salmon Life History
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Spawning
Emergence
Rearing
Outmigration
Summer Chum Salmon Life History
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Spawning
Emergence
Rearing
Outmigration
Coho Salmon Life History
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 20
February 2025
Several factors have been identified in the Dosewallips River as affecting summer chum populations.
Unfavorable stream flows in 1975 and 1976 caused a crash in chum populations across Washington State, but
Hood Canal populations remained low while other populations recovered. High levels of fishing in the 1980s
had further impacts on the population and coincided with a shift in ocean conditions in 1986 to patterns less
favorable to chum (Brewer et al, 2005). In the Dosewallips River, much of the lower few miles of the river has
been simplified, with the construction of dikes, placement of riprap, splash damming and the removal of wood.
The surrounding floodplain in the area has also been converted to pastureland and residential development
(Brewer et al. 2005). Channelization and diking are also noted as a problem for spawning and incubation.
Logging of forests, specifically old growth areas, has resulted in loss of recruitment sources for large wood into
the river, with most trees in the floodplain below RM 4.5, including in the project reach, being 12 inches or less
in diameter (Brewer et al. 2005). Loss of side channel habitat and channel instability have also been noted as
limiting factors for salmon habitat. Estuarine habitat degradation was also noted as a leading limiting factor for
juvenile Hood Canal summer chum rearing.
2.8 Large Wood Distribution and Function
Large stable wood provides critical functions for sustaining river systems. For example, it partitions shear stress
(i.e., stream energy) across the channel bed and banks, and thus lowers the available energy for bank erosion,
channel migration, and channel incision (Manga & Kirchner, 2000; May & Gresswell, 2003). Instream wood also
increases channel complexity by forming pools and promoting multi-thread (anabranching) channel patterns,
which increases the diversity of high-quality habitat for aquatic species.
During the 2024 field survey, NSD counted, measured (through visual estimation), and geo-located functional
wood pieces, natural log jams, engineered log jams, and key pieces within the project reach (Figure 17). The
total number of key pieces for each data point were also identified and used to compare to reference
conditions, and for comparison NSD also evaluated the condition of the existing installed ELJs as part of the
Dosewallips State Park restoration project. The following definitions were used to evaluate the large wood
distributions and guide the data collection:
Key pieces: Pieces of wood estimated to be stable during flood events (Figure 18). Key pieces serve as the
forming members of log jams. A general rule for key piece sizing is the length equal to ½ bankfull width and the
diameter at breast height (DBH) equal to ½ bankfull depth (Abbe & Montgomery, 2003). We identified all pieces
greater than 30” DBH to be a key piece within the Dosewallips River.
Natural Log jams: Wood accumulations that showed evidence of remaining stable under high flow conditions,
such as the presence of multiple piece wood accumulations in the river, sediment accumulation behind or
vegetation establishment within or around the jam (Figure 19).
Engineered Log jams: Wood accumulations that have been engineered to provide in-channel habitat and side
channel enhancement have been constructed within the project reach as part of previous restoration efforts
associated with the Lower Dosewallips Floodplain and Estuary Restoration project (Figure 19). The total number
of natural key piece analogs was determined for each engineered log jam type for use in reference condition
comparisons.
Other Functional wood: Wood pieces that exert an influence on bed topography and sediment distributions but
may not necessarily be stable during high flows (i.e. not a key piece, Figure 18). These wood pieces are
important habitat features (e.g., providing cover) despite not always exerting a strong geomorphic influence on
channel processes (<30” dbh).
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 21
February 2025
Table 4. Count of functional instream or floodplain wood
STANDALONE KEY PIECES NATURAL LOG JAMS ENGINEERED LOG JAMS OTHER FUNCTIONAL
WOOD
12 24 8 19
Table 5. Number of key pieces in study reach, compared to reference conditions (Fox and Bolton, 2007)
N KEY PIECES STANDALONE KEY
PIECES
KEY PIECES IN
NATURAL JAMS
KEY PIECES IN ELJS TOTAL
Number of Key Pieces 12 31 19 62
Reference number of
Key Pieces
>77 >77 >77 >77 total
Reference key pieces were calculated using the 75th percentile for reference reaches in Western Washington (Fox and Bolton, 2007)
Table 6. Primary functions of instream and floodplain wood
FUNCTION1 STANDALONE KEY
PIECES
NATURAL JAMS ELJS OTHER FUNCTIONAL
WOOD
Pool Formation2 83% 67% 50% 11%
Bar/Island Formation 83% 25% 25% 26%
Grade Control 25% 8% 0% 21%
Substrate Sorting 8% 67% 50% 32%
1: Percentages reflect functions that were identified as primary or secondary for each observation of LWD.
2: includes pools/depressions that were not wetted/engaged with flow during the June 2024 site visit.
NSD made 61 observations of functional instream and floodplain wood, including 12 standalone key pieces, 24
natural log jams, 8 engineered log jams, and 19 other accumulations of functional wood (Table 4, Figure 17). The
total amount of key pieces within the active channel and floodplain, combining all classifications, was 62 pieces
which is on the same order of magnitude as expected in reference condition channels in western Washington
(Table 5, Fox and Bolton, 2007) however less than the reference amount of >77 pieces. The majority of key
pieces were included in natural or engineered log jams (Table 5).
Table 6 above highlights the wood functions that were observed to be the primary or secondary function of each
LWD observation. The most common primary or secondary function was pool formation, followed by bar
formation and substrate sorting, with the majority of LWD on the banks of the active channel. Grade control was
observed with a few pieces of mostly buried wood on the mainstem and within side channels. While pools were
formed at the majority of functional wood, many of these pools were dry depressions on the right bank
floodplain, not providing functional habitat for aquatic species during the field visit (e.g. Figure 18).
All of the ELJs installed as a part of the Dosewallips Floodplain and Estuary Restoration Project were constructed
at the inlets of side channels downstream of US-101. The structures are currently functioning to stabilize the
heads of islands between the mainstem and secondary channels, as well as encourage flow into both natural
and excavated channels where they are engaged with the active channel. Hydraulic conditions where ELJs are
engaged with flow also create scour pools, encourage substrate sorting, and encourage island/bar development
in their lee. Many of the ELJs, however, were disengaged from active flow, mainly serving to stabilize the heads
of right bank islands between the mainstem and excavated side channels.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 22
February 2025
Figure 17. Large wood observed during the June 2024 field reconnaissance
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 23
February 2025
Figure 18. Examples of a key piece (left) and functional wood (right)
All ELJs were intact during the June 2024 survey, however the most downstream structure at RM 0.1 was not
positioned in the middle of the channel as depicted in the original design sheets. The structure was highly
functional, creating a deep pool, and development of a vegetated island, and split flow onto the right bank
floodplain (Figure 20). This was the most functional ELJ observed, highlighting the efficacy of wood that is highly
engaged with active flow. Tieing this observation back to the impairments of the project reach, while there are
amounts of large wood close to reference conditions within the floodplain of the active channel (Table 5), the
majority of pieces are either tied into ELJs that are not engaged with flow, or distributed throughout the right
bank floodplain where they are also not engaged. Additional functional large wood that is directly obstructing
flow of the mainstem channel would have a greater impact on engaging the right bank floodplain upstream of
Dosewallips State Park, and distributing the shear stress of flows across multiple channels.
Figure 19. Examples of an engineered log jam (left) and natural log jam (right)
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 24
February 2025
Figure 20. Functional ELJ at RM 0.1
2.9 Riparian and Wetland Communities
NSD used tree height data from Lidar, aerial imagery, and a field reconnaissance to characterize the riparian
composition and structure within the project reach. This analysis concluded that the adjacent floodplain riparian
communities within the project reach are characterized by mature mixed coniferous and deciduous forested
communities typical of wet forests within the Pacific Northwest. However, their current composition and
condition has been impacted by past and current land clearing. The North Pacific Lowland Riparian Forest and
Woodland Group typical of the lower Dosewallips River floodplain, is found on low-elevation, alluvial floodplains
that are confined by valleys and inlets throughout the coastal regions of the Pacific Northwest (NatureServe
2015). Although geographically widespread, over half of this system is estimated to have been lost and the
remaining majority degraded as a result of land uses within the riparian zone as well as surrounding uplands
(e.g. logging, dams, road construction, agriculture) (Rocchio and Crawford 2015).
Riverine flooding and subsequent successional processes following flood disturbance are major factors creating
the species composition and successional diversity of riparian systems. Major broadleaf dominant species that
would be expected within the Dosewallips River riparian areas are bigleaf maple (Acer macrophyllum), red alder
(Alnus rubra), and black cottonwood (Populus balsamifera). These early successional species would give way to
conifers such as grand fir (Abies grandis), Sitka spruce (Picea sitchensis), western red cedar (Thuja plicata), and
Douglas fir (Pseudotsuga menziesii) with succession in the absence of major disturbance (Rocchio and Crawford
2015).
Throughout the project reach the clearing for timber harvest, roads, home construction, Highway 101, and the
Dosewallips River State Park has altered much of the riparian community within the floodplain. Impacts
associated with past clearing and channel migration can be seen in the canopy height model in Figure 21.
Riparian communities located directly adjacent to the riverbanks and on the lower floodplain are dominated by
western red cedar, red alder, big-leaf maple, and black cottonwood in the overstory, with vine maple (Acer
circinatum), salmonberry (Rubus spectabilis), and sword fern (Polystichum munitum) in the understory. The
overstory in these communities are primarily dominated by deciduous tree species, with multiple age class of
coniferous species growing within the understory. This is an indicator that the riparian forest in these locations is
recovering from past human and natural disturbances, with young conifer starting to establish.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 25
February 2025
Low-lying gravel bars show signs of frequent flooding (i.e. wood debris and scour patterns) and support pioneer
species such as sapling red alder, and the invasive species Scotch broom (Cytisus scoparius), and Himalayan
blackberry (Rubus discolor). The mosaic of overflow channels on river right upstream of the Highway 101 bridge,
supports a mix of flood-tolerant species such as willow (Salix sp.), sedge (Carex sp.), and the invasive reed
canarygrass (Phalaris arundinacea). Wetland and waters occupy these low overflow channels, however the
higher terraces do not display any wetland characteristics although some of the areas are labeled as wetlands in
the National Wetlands and Jefferson County inventories (USFWS 2020, Jefferson County 2017). These higher
floodplains dominated by red alder, western red cedar, and big-leaf maple tend to be wet forests rather than
jurisdictional wetlands (Figure 24, Figure 25, Figure 26). Two seasonal tributary channels flow through the high
floodplain terrace to the south of the river upstream of the State Park (Figure 15). Both channels showed signs
of seasonal flow with areas of fan-like deposition and incision through the terrace materials. These channels do
not provide fish habitat but do provide inputs of flow and gravels seasonally (Figure 27).
Below the Highway 101 bridge the riparian community continues to be affected by past land clearing associated
with the State Park, revegetation efforts, and a change to estuarine conditions. The river left floodplain is a
mosaic of higher wet forest dominated by red alder, black cottonwood, big-leaf maple, western red cedar, and
dense Himalayan blackberry, and lower wetland forest dominated by red alder, salmonberry, and sedge
communities. Previous restoration efforts associated with the Lower Dosewallips Floodplain and Estuary
Restoration project have initiated riparian plantings with the relocation of Park facilities and construction of
distributary channels. Many of the willow stake plantings have been either lost to bank erosion or
browse/beaver activity. A few planted sapling red alder and Sitka spruce have survived and are now providing
shade to the stream channels (Figure 23).
The lower Dosewallips River estuary is characterized by the main channel, multiple distributary channels, and
vegetation and wetland communities adapted to salinity and periodic inundation from tidal exchange.
Vegetation includes common saltmarsh species of Puget Sound including seaside arrowgrass (Triglochin
maritima), plantain (Plantago lanceolata), Pacific silverweed (Potentilla anserina), redtop (Agrostis gigantea),
and saltgrass (Distichlis spicata) (Figure 22).
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 26
February 2025
Figure 21. Canopy height and wetland map
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 27
February 2025
Figure 22. Constructed channel through native forest; emergent estuary community
Figure 23. Young alder forest; Alder-Cottonwood community downstream of the US-101 bridge
Figure 24. Wetlands forming around high flow side channels; narrow band of trees rooted into berm
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 28
February 2025
Figure 25. Side channel, pool, and wetland habitat; high gravel bar with wood debris
Figure 26. Big-leaf maple and red alder forest; and mixed conifer/deciduous forest
Figure 27. Two seasonal tributaries drain through the southern terrace and flow into the Dosewallips River
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 29
February 2025
2.10 Nearshore and Estuary Habitat
The Dosewallips estuary and the rest of Jefferson County experiences mixed-semidiurnal tides, with two daily
high and low tides of differing elevations (ESA 2023). Spring-neap tidal cycles are also prevalent, occurring
approximately twice per month with the lunar cycle. Sea level at the Port Townsend gage since 1972 has an
upward trend in elevation, indicating a 0.59’ rise over 100 years, which is expected to accelerate with climate
change (ESA 2023). Still water elevations during storm events are generally 1’ higher in Dabob Bay and
surrounding the Dosewallips estuary than elsewhere in Jefferson County, with elevations up to 12.7’ NAVD88 in
Dabob Bay, compared to 11.6’ NAVD88 in Port Townsend (FEMA 2019).
Sandy and rocky beaches throughout Jefferson County, including the Dosewallips estuary, create important
nearshore and intertidal habitat for salmonids including Chum, Chinook, and Coho, along with steelhead and
cutthroat trout (ESA 2023). Additional organisms that depend on estuary and nearshore habitat include forage
fish, shellfish, shore and sea birds, and marine mammals for feeding, breeding and migration, which are
expected to be impacted by retreating coastlines projected with climate change (Krueger et al. 2011; Miller et al.
2013; Smith and Liedtke 2022). The influence of tidal channels occurs downstream of the US-101 bridge on the
mainstem Dosewallips (see Figure 34 in section 4.1), with two inlet channels crossing the highway at the north
end of town near the Community Center (Walcott Slough, Figure 15). Additional assessments of nearshore
habitats were not a part of this project scope.
3 HYDROLOGY AND HYDRAULICS
3.1 Hydrology
The lower 1.2 miles of the Dosewallips runs through private, county, and state park managed land, from a
bedrock constriction at RM 1.3, south of the town of Brinnon, and into a tidal delta downstream of US 101. The
Dosewallips River flows eastward from the central Olympic Mountains and has a watershed area of 116 square
miles (mi2) and a relief of 7,770 feet. The watershed receives an average annual precipitation of 77.6 inches
which falls as both rain and snow during the fall and winter months (October-February). The annual hydrology at
the outlet of the Dosewallips River is governed by this mixed rain-snow basin hydrology, with flashy flooding
events throughout the winter and a spring freshet from high elevation snow melt (Figure 28). Most of the basin
is forested, with 63% of the area covered by tree canopy (USGS, 2019).
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 30
February 2025
Figure 28. Key salmonid life history stages and Dosewallips River flows
To model relevant hydraulic conditions in the project reach, we developed discharge estimates for five flow
scenarios: two that are relevant to salmonid life history stages (Figure 28), and three peak flow scenarios
including the 2-year, 10-year, and 100-year floods (Table 7). The hydraulic model used for this analysis is a
subsection of a larger model which extends from RM 6.7, at the USFS bridge, to the river outlet at Hood Canal.
The model includes three inflow points: an upstream inflow located at RM 6.7, an inflow for Rocky Brook Creek
near RM 3.7, and an intermediate inflow point applied at a right bank tributary near RM 2.8. All inflows are
applied upstream of the project reach.
Flow estimates were made based on the historical discharge record available from the now inactive USGS gage
that was located on the Dosewallips River just upstream of the project site (USGS gage #12053000). Accurately
representing the magnitude of peak flows within the project reach is challenging because the Dosewallips River
gage operated for a relatively short period, from 1930-1951, with no other available gage in the basin. The active
USGS gage on the Duckabush River (USGS gage #120454000), which has been operating from 1938 to present,
was considered as an alternate source of peak flow data, as the Duckabush River watershed is located
immediately to the south of the Dosewallips River watershed and has similar characteristics for slope, relief, and
precipitation. A Bulletin 17B peak flow analysis (USGS 1982) of both the Dosewallips gage and the Duckabush
gage data was performed and scaled to the project area by drainage area as described in Mastin et al., (2016).
However, since the drainage area ratio of the Dosewallips project area to the Duckabush River gage area is
greater than the recommended limit of 1.5, the flow estimates from the shorter-duration Dosewallips River gage
were chosen for use in hydraulic model. A flow duration analysis was computed for the Dosewallips gage to
estimate low flows relevant to salmonid life stages. The median September exceedance flow was selected to
represent typical flows for Hood Canal Summer Chum spawning, and the median daily February exceedance
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 31
February 2025
flow was selected to represent typical flows for Hood Canal Summer Chum outmigration. Inflows for Rocky
Brook were based on USGS regression estimates obtained from the USGS StreamStats (Mastin et al., 2016).
The drainage basins for the project reach were delineated using the USGS Streamstats tool (USGS, 2019) and are
displayed in Figure 29. The total drainage area for the Dosewallips basin at Hood Canal is approximately 116
square miles. The drainage area attributed to the upstream Dosewallips River inflow is approximately 101
square miles while the drainage area for Rocky Brook is approximately 9 square miles. Scaled flows for two
nearby, unnamed tributaries, located on the right and left banks of the Dosewallips River just downstream of
Rocky Brook, were added to the Rocky Brook inflow due to their relatively low magnitude and proximity to
Rocky Brook. The total drainage area included at the Rocky Brook inflow location is approximately 10.6 square
miles. The intermediate inflow applied at RM 2.8 accounts for flow that accrues within the model domain
downstream of Rocky Brook Creek, with a drainage area of 4.4 square miles.
Figure 29. Drainage areas utilized for hydrologic analysis and hydraulic model.
Smaller tributaries were not explicitly included as separate hydrologic inputs in the hydrologic analysis and
hydraulic model but were considered in the upper Dosewallips River and RM 2.8 intermediate inflows, due to
their relatively low flow contributions at the scale of the hydraulic model.
Discharge values used for current conditions hydraulic modeling and associated analyses are shown in Table 7.
The 2-, 10- and 100-year recurrence peak flows are also referred to as the 2-, 10- and 100-year floods. We also
calculated two flows that are critical to the timing of Hood Canal summer chum adult spawning (median
September flow) and juvenile outmigration (median February).
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 32
February 2025
Table 7. Estimated discharge values
RECURRENCE INTERVAL DOSEWALLIPS
RIVER DISCHARGE
ESTIMATE
(CFS)
ROCKY BROOK
DISCHARGE
ESTIMATE
(CFS)
RM 2.8
INTERMEDIATE
DISCHARGE
ESTIMATE (CFS)
HCSC Spawning (median Sept
exceedance)
168 18 7
HCSC Outmigration (median Feb
exceedance)
327 34 14
2-year (50% probability in any
given year)
4,689 412 192
10-year (10% probability in any
given year)
8,776 783 352
100-year (1% probability in any
given year)
14,469 1,279 589
HCSC = Hood Canal Summer Chum
3.1.1 Tidal Datums
The downstream boundary for the hydraulic model extends into Hood Canal, into the nearshore zone. The
hydraulic model utilizes a constant stage hydrograph downstream boundary condition based on the tidal
datums at Hood Canal. Mean higher high water (MHHW), defined as the average level of the highest tide for
each day over a period of 19-years, was chosen as the tidal condition for all modeled flows. MHHW was
computed using outputs from the Coastal Storm Modeling System (CoSMoS) over the period from 1996 to 2015
(Grossman, 2023). Available CoSMoS data station locations are shown in Figure 30. Data station 259 (located at -
122.786, 47.691) was chosen to compute MHHW for the project site. MHHW for this station location was
computed as 9.15 feet NAVD88. This value was applied as the downstream stage elevation for all modeled
flows.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 33
February 2025
Figure 30. CoSMoS data station availability in the vicinity of the project area.
Full hydraulic model domain and Brinnon project reach endpoints shown for reference.
3.1.2 Hydrologic Effects of Climate Change
Once peak flows were determined for the study reach, it was necessary to estimate the impact that climate
change would have on these flows to understand and model future flow events. The Columbia Basin Climate
Change Scenarios (CBCCS) project summarizes climate change projections for many watersheds, the closest to
the Dosewallips being the Skokomish River (Hamlet, 2010). For each basin, the CBCCS projects the 20-year, 50-
year, and 100-year recurrence interval floods into the future to estimate their magnitude in the years 2070-2099
based on one of two climate change scenarios. For this project the A1B climate change scenario was used, which
is the higher of the two scenarios and the closest to current climate change projections. In addition, monthly
mean model predictions for September and February were extracted from the model and compared to historic
conditions to predict change during the Hood Canal summer chum spawning and outmigration periods.
Data were downloaded from the CBCCS Project website (Hamlet, 2010). These materials were produced by the
Climate Impacts Group at the University of Washington in collaboration with the WA State Department of
Ecology, Bonneville Power Administration, Northwest Power and Conservation Council, Oregon Water
Resources Department, and the B.C. Ministry of the Environment. The percent change in the Skokomish River
floods from 2020 to 2070-2099 as estimated by the CBCCS and then applied as a multiplier to the Dosewallips
floods to estimate climate change flows (Table 8). An important note is that the lowest flood flow for which the
CBCCS makes estimates is the 20-year flood; therefore, the multiplier applied to the Dosewallips 2-year and 10-
year floods is the CBCCS-estimated 20-year flood percent increase.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 34
February 2025
Table 8. Estimated Change and Magnitude of Future Flows
FLOW PROJECTED
CHANGE (%)
FUTURE (2070-2099)
DOSEWALLIPS RIVER
DISCHARGE ESTIMATE
(CFS)
FUTURE (2070-2099)
ROCKY BROOK DISCHARGE
ESTIMATE (CFS)
FUTURE (2070-2099) RM
2.8 INTERMEDIATE
DISCHARGE ESTIMATE
(CFS)
HCSC Spawning -30 118 12 5
HCSC
Outmigration 10 360 38 15
2-year 18 5,533 487 226
10-year 18 10,355 924 416
100-year 23 17,797 1,573 724
HCSC = Hood Canal Summer Chum; Projected Climate Change Impacts for 2070-2099
3.1.3 Future Tidal Datums
To assess future conditions, relative sea level rise (RSLR) was estimated for Hood Canal at the mouth of the
Dosewallips River using Miller er al., (2018). Sea level rise for the 1% exceedance RCP 8.5 emissions scenario was
estimated as 1.3 feet for 2050. This value was applied to the MHHW elevation, resulting in an estimated future
MHHW elevation of 10.45 ft NAVD88. This future conditions tidal elevation was coupled with the estimated
future flows shown in Table 8.
3.2 Hydraulics
3.2.1 Model Development
A two-dimensional (2D) hydraulic model of the Dosewallips River was developed using the U.S. Army Corps of
Engineers (USACE) Hydrologic Engineering Center modeling platform, River Analysis System (HEC-RAS), version
6.4.1 (USACE, 2023). The model was developed to support characterization of existing reach conditions and
future design phases and flood risk, with model outputs including flow depth, velocity, and shear stress. The
hydraulic model used for this analysis is a subsection of a larger model which extends from RM 6.7, at the USFS
bridge, to the river mouth at Hood Canal. The full hydraulic model includes three inflow locations, at RM 6.7 of
the Dosewallips River, at Rocky Brook Creek, and near RM 2.8 of the Dosewallips River. The downstream
boundary condition is located within the zone of tidal influence at Hood Canal. The hydraulic model domain and
boundary condition locations are shown in Figure 31. The underlying terrain for the hydraulic model was
developed using 2023 bathymetric bare earth LiDAR collected and processed by NV5 (NV5, 2024).
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 35
February 2025
Figure 31. Boundary condition locations within hydraulic model domain.
Hydraulic Structures
Several hydraulic structures are included in the full hydraulic model domain. Outside of the project reach limits,
the existing Dosewallips Road bridge over Rocky Brook is included in the hydraulic model as a 2D-connection in
the model mesh. Bridge geometry data (high and low chord elevations) were estimated based on 2023 LiDAR
data and the WDFW Fish Passage Inventory (Site ID 420135). A small opening was graded into the terrain to
allow flow through a fill prism over the historical Dosewallips River flow path near RM 5.2.
Figure 32 shows hydraulic structures modeled within the Brinnon project reach. Six crossings with US-101 are
included as 2D-connection structures in the hydraulic model mesh. Culvert and bridge geometry data were
estimated based on crossing information included in the WDFW Fish Passage Inventory, field measurements,
and field observations. Several additional crossings exist within the estuary on private property and were not
explicitly included in the hydraulic model due to lack of available data. Mesh cells were modified where data was
available, or modeled to be intentionally large to avoid seeing these unknown crossing locations as hydraulic
barriers.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 36
February 2025
Figure 32. Hydraulic structures included in hydraulic modeling for project reach.
The hydraulic model mesh was created with finer mesh spacing in main channels and presumed flow paths and
coarser mesh spacing in less topographically complex areas. The existing berm (Crossing ID 7 on Figure 32) on
the left bank of the Dosewallips River upstream of the US-101 bridge was represented with a breakline with a
mesh spacing of 5 feet. The existing vegetated berm is made up of rock/rip rap and is not a formally certified
levee. The existing berm, as represented in the 2023 LiDAR terrain, is shown in Figure 33. Photos of the berm are
included in Figure 14 in section 2.5 above.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 37
February 2025
Figure 33. Existing berm extents based on 2023 LiDAR
Hydraulic resistance is characterized in the model by polygons representing differing surface roughness types,
with Manning’s n roughness coefficients assigned using Chow (1959) and engineering judgment. Roughness
polygons for this model were delineated using 2023 LiDAR data and 2023 NAIP imagery. The Manning’s n values
for each land cover category are shown in Table 9.
Table 9. Manning’s n Roughness Values
CATEGORY MANNING'S N
Forested floodplain 0.08
Grass floodplain 0.03
Grazed floodplain 0.03
Compacted surface 0.025
Impervious surface 0.015
Dosewallips River Main Channel 0.04
Rocky Brook Main Channel 0.045
Gravel bar 0.045
Forested bar 0.08
Existing wood 0.15
Buildings 0.99
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 38
February 2025
The hydraulic model was run in a quasi-steady state for the 2-year, 10-year, and 100-year peak flows and for the
Hood Canal summer chum spawning (median September exceedance) and HCSC outmigration (median February
exceedance) flows.
4 EXISTING CONDITIONS MODEL RESULTS
Here we present flow depth and velocity output for four modeled flow scenarios and highlight implications for
fish habitat and flood risk. Water depths and velocities during September (spawning life history stage) and
February (outmigration life history stage), indicate that the channel is confined to a single thread throughout the
entire project reach, with the exception of the distributary channels downstream of the US-101 bridge. Right
bank floodplain features become more engaged at the 2-year flow, particularly in the area of high channel
migration rates between RM 0.5 and RM 0.9. The 100-year flow shows inundation across the lowest areas of the
alluvial valley, with significant impoundment on the upstream side of US-101, and an overflow channel spilling
water eastward towards community infrastructure at RM 1.
4.1 Implications for salmonids
With regards to aquatic habitat, the confined, single thread nature of the channel at lower flows highlights
limited off-channel connectivity in the project reach, which, combined with limited instream wood to provide
cover, makes the habitat for juvenile salmonid rearing very limited (Figure 34; Figure 35). This is important for
critical life history stages for Puget Sound Chinook (see section 2.7), while Hood Canal Summer Chum exit the
channel immediately after emergence.
Figure 34. Modeled water depth for existing conditions spawning and juvenile outmigration flow scenarios
Gray areas indicate extent of MHHW (9.13ft)
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 39
February 2025
Figure 35. Modeled flow velocity for spawning and juvenile outmigration flow scenarios
Gray areas indicate extent of MHHW (9.13ft)
Figure 36 below shows a comparison of the distribution of depths and velocities in the project reach during
February compared to reported preferences for juvenile Chinook rearing (Beecher et al., 2022). While it is
important to consider how depth, velocity, and sediment overlap, a comparison of depth and velocity in
isolation with habitat preferences indicates that while there are sufficient areas with water depth that are
suitable for juvenile rearing, flow velocities are higher than suitable. This is due to a combination of limited
connection to floodplain side channels during outmigration flows, along with a lack of perennially engaged
stable large wood.
Figure 36. Water depth and velocity distributions during median February exceedance flow
Rearing preferences for Chinook overlain
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 40
February 2025
Figure 37 below shows depth and velocity distributions during September, in comparison to preferred depth and
velocity conditions for Chinook and Chum spawning. In contrast to juvenile habitat availability, there is some
available depth, velocity, and substrate conditions that are suitable for adult Chinook and Chum. This is due to
pool-riffle morphology forced by the relatively low gradient in the reach. This pool-riffle morphology produces
deeper pools suitable for velocity refuge along with substrate sorting suitable for spawning (e.g. pebble counts
in Figure 10).
Figure 37. Water depth and velocity distributions during median September exceedance flow
Spawning preferences for key salmonid species (HCSC and Chinook) overlain
4.2 Implications for flood and channel migration risk
With regards to flooding, the 2-year and 100-year depth and velocity results highlight significant inundation on
the southern side of the mainstem at both infrequent (100-year) and relatively common (2-year) flows (Figure
38; Figure 39). The majority of infrastructure in the town of Brinnon is located to the north of the mainstem,
which is not significantly impacted by flooding at the 2-year flow indicating the effects of the berm construction ,
but has widespread inundation at the 100-year flow. The berm upstream of the US-101 bridge appears to
prevent flooding to the north of it during the 2-year event but has limited function during the 100-year flow.
Discussed in further detail in section 5 below, this is consistent with the assumed design intent of the berm,
which was to provide erosion control rather than flood prevention. The lack of floodplain connectivity to the
north at the 2-year flood is also an indicator of channel incision upstream of the US-101 bridge.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 41
February 2025
Figure 38. Modeled water depth for existing conditions flood flow scenarios
Gray areas indicate extent of MHHW (9.13ft)
An important observation in the 100-year model output is impoundment on the western side of US-101. Figure
40 below highlights the upstream extent of tides during MHHW, indicating that backwater behind the road
prism at the US-101 bridge is not likely influenced by tides in the hydraulic model and thus is coming from river
flow. This impoundment, along with that observed at US-101 crossings to the north and south, emphasizes the
importance of adequate width of hydraulic structures designed to convey water beneath the highway,
particularly with a projected increase in peak flows associated with climate change models (USACE, 2023). This
topic is discussed in further detail in section 5 below.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 42
February 2025
Figure 39. Modeled flow velocity for existing conditions flood flow scenarios
Gray areas indicate extent of MHHW (9.13ft)
Figure 40. Water surface elevations indicating location of backwater influence by MHHW (red line)
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 43
February 2025
4.3 Climate Change
Historic flow records and climate change projections indicate a trajectory of both sea level rise and increased
peak flows in the Olympic Peninsula region (Hamlet, 2010; USACE, 2023). Along with higher intensity flows
produced by rainfall, there is also a projection of lower flows in the summer, due to a combination of glacial
recession and more water falling as rain than as snow in winter months, reducing the overall snowpack. These
observations are captured in the mid-century scenarios of the hydraulic model (Figure 42), and are significant for
both fish and community infrastructure. Lower flows in the summer will have the highest impact on key
salmonid species, with reduced areas of the depths and velocities preferred by returning adult fish to spawn
(Figure 41).
Figure 41. Water depth and velocity distributions and preferences for key salmonid species
For both existing conditions and Mid-Century. Spawning preferences for key salmonid species overlain (HCSC and Chinook).
Higher peak flows will have a greater impact on the community and Park infrastructure, with an inherent
increase in flood depths and extents with higher magnitude inputs of water from both the watershed and sea
level rise. The most significant changes in flooding are in flood depth, which has the greatest magnitude in areas
influenced by hydraulic structures (i.e. culverts and bridges) that are undersized for projected flood flows. In
particular, there is greater impoundment of flood and tidal waters near key hydraulic structures that cross US-
101, such as the area near the Community Center (Figure 42). Along with projected changes in flood risk,
channel migration risk will also likely increase with climate change due a combination of changes in flow and
associated sediment supply in areas that are already susceptible to channel migration (upstream of the US-101
Dosewallips River bridge crossing). Further discussion of these impacts are included in section 5.2 below.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 44
February 2025
Figure 42. Modeled changes in depth and velocity relative to climate change predictions
During the 2-year and 100-year floods; Gray areas indicate extent of MHHW (10.43ft)
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 45
February 2025
5 FLOOD AND CHANNEL MIGRATION RISK
Along with assessing the hydrogeomorphic conditions of the project reach with regards to habitat suitability for
threatened salmonid species, we also examined flood and erosion risk. This assessment used a combination of
topography, hydraulic model outputs, historical imagery, along with previously mapped FEMA flood hazard
zones. Five areas of heightened flood risk were highlighted through this process (Figure 43). These areas include:
1. An overflow channel that flows east through town, north of the fire station
2. Impoundment of water behind US-101 on the north end of town, surrounding the Community Center
3. The confined reach just upstream of the US-101 bridge on the mainstem Dosewallips (RM 0.4-RM 0.7)
4. Impoundment of water behind the US-101 bridge within the Dosewallips Campground.
5. Mass wasting hazard area to the southwest of the main channel.
Below we walk through the modeled impacts and hydraulic context for each of these areas, and we follow up
with an assessment regarding how flood risk is projected to change by mid-century.
5.1 Flood Risk Focus Areas
Figure 43. Areas of heightened flood risk
1: Overflow channel north of fire station, 2: low area near community center, 3: berm area (RM 0.4-0.7), 4: low area between State
Park Campground and US-101, 5: hillslopes susceptible to landslides on southwestern bank.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 46
February 2025
Area 1 - Overflow Channel North of Fire Station
A low-lying overflow channel is present north of the fire station, which connects to the mainstem Dosewallips at
RM 1.0 and flows eastward, connecting to the tidal outlet under US-101 at the Community Center. This channel
is dry at typical flows (e.g. median September exceedance and median February exceedance), and at the 2-year
flood. The lack of flows reaching this left bank floodplain area at a 2-year flow also indicates that the channel
itself is slightly incised within this upper part of the project reach (Figure 38).
At the 100-year event, however, the channel is connected to the mainstem, with 136 cfs of water diverted
towards community infrastructure, with flow velocities up to 3ft/second. This water is impounded where it
crosses Schoolhouse Road as there is no hydraulic structure (i.e. culvert or bridge) to convey water beyond it.
The hydraulic model shows flows overtopping Schoolhouse Road during this 100-year flood event, and then
impounding downslope behind US-101 as well, interacting with tidally influenced backwater and box culverts at
the US-101 highway prism.
While the channel poses a flood risk during infrequent, high-intensity flood events, the risk of the mainstem
channel avulsing into it are low. Arguments for relatively low avulsion risk include 1) the channel inlet is 5-7 feet
above the low flow water surface elevation (Figure 7), 2) the channel is on the inside of a bend, where water
depths and velocities are lowest during peak events, and 3) the floodplain between the mainstem and the
overflow channel inlet is colonized by a relatively mature mixed forest, with trees up to 150’ in height (Figure
21). However, these conditions could change as the channel receives more frequent flow with increasing flood
frequency and magnitude with climate change.
Area 2 - Low Area Near the Community Center
The northern extent of the alluvial valley that confines the town Brinnon has two tidal inlets that are
hydraulically connected via culverts under US-101. This area is also at the downstream end of the overflow
channel described above and adjacent to the Community Center. A combination of water entering the inlets
from both tides and this overflow channel causes significant inundation near the Community Center during high
tide and flooding events, with flood depths up to 4-6 feet in depth outside of the main tidal channel (Figure 43).
The two crossings at US-101 in this area are box culverts that have flows of ~240 cfs (northern crossing) and
~210 cfs (southern crossing), which is more than triple the amount of water that is diverted the mainstem via
the overflow channel north of the fire station (~140 cfs). This indicates that a large proportion of the water in
this area is sourced from both local runoff and tidal backwater from Dabob Bay. It is likely that water backing up
behind the constriction from the US-101 bridge over the Dosewallips River (Area 3) flows northward to this
crossing as well, as indicated by flow paths in Figure 35. The depth of flooding is projected to be higher with
climate change, with this area being the zone with greatest change (Figure 42).
Area 3 - Left Bank Berm Upstream of US-101
An area of particular focus in this assessment is the constructed berm on the northern bank upstream of the US-
101 bridge crossing of the Dosewallips River. The Brinnon community and Jefferson County requested an
evaluation of the efficacy of the berm feature to protect against flooding and channel migration. Through
assessment of the 1957 design drawing (Washington State Department of Waterways, 1957) and channel
migration history (Figure 12), we determined that the intention of the berm was to provide erosion protection
for community infrastructure to the north, rather than prevent flooding during significant flood events (i.e. 100-
year flood). This assumption is supported by the non-continuous extent of the design, which ends 280 ft
upstream of the bridge crossing, at a natural low-lying area that routes flood water around the berm and
towards Easy Street and the highway prism (Figure 44). A more robust flood protection structure would have
been more continuous across this low feature upstream.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 47
February 2025
Figure 44. Berm extent as designed (left) and observed in 2023/2024 (right)
When comparing the berm with the original design plans (Washington State Department of Waterways, 1957)
the height, extent, and condition of the berm are consistent with the original plans, and are succeeding at
preventing northward channel migration along the existing length of the feature. The original design shows that
the berm is approximately 540 ft long, and 7 ft high above the low-flow water surface (Figure 45, Figure 46).
Field observations combined with newly collected lidar (NV5G 2024) indicate 640 ft more feet of berm extend
upstream to the west, which is also functioning to prevent channel migration towards the town (Figure 44). The
berm is constructed of rip-rap with diameter between 3-5’ (Figure 46), and is becoming vegetated on the top.
We were unable to assess the toe at the time of survey, but cross sections of the topobathymetric lidar dataset
(NV5G 2023) indicate that the shape of the berm is generally consistent with the original design plans (Figure 45,
Figure 46).
There is a low-lying meander scar that pre-dates available historic imagery upstream of the western end of the
berm, which is unprotected by existing armoring and poses a risk of channel migration/avulsion towards the
town (Figure 47). Avulsion into this feature would significantly alter current risk conditions, with a higher chance
of both flooding and erosion towards existing infrastructure, private property, and homes. It is important to
note that there is historic evidence of high migration rates throughout the floodplain on the opposite bank of
the berm, due to a combination of sediment aggradation and a lack of stable large wood (see sections 2.3-2.5).
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 48
February 2025
Figure 45. Berm cross section at three modeled flows (2-year, 100-year, typical winter flow)
Cross section location indicated in Figure 44
Figure 46. Profile of berm design (WA State Department of Waterways, 1957)
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 49
February 2025
Figure 47. REM and channel migration map along extent of berm
Blue circle indicates low-lying avulsion pathway
At the 2-year flood interval the berm feature is effective at preventing inundation behind it within the footprint
of the constructed length (Figure 48), with the exception of the downstream-most 25%, where water pools on
the northern side. The freeboard between the water surface elevations at the 2-year flood and the top of the
berm is approximately 1.5-2.5’ (Figure 45). At the 100-year flood, the berm provides minimal flood protection,
with water both overtopping the feature and flowing behind it from both upstream around the berm and
downstream via backwater behind the US-101 bridge crossing. When comparing the modeled discharge of
16,337 cfs entering the project reach with the amount flowing out through the mainstem bridge crossing, we
find that 13,150 cfs (80)% of water flows through the US-101 bridge, with 2200 cfs (14)% flowing collectively
through the two crossings immediately to the north, and 490 cfs (3)% flowing through the crossing to the south
(see next section). The remaining 3% of inflow flows out of the crossings near the Community Center. This
partitioning indicates that 17% of flow during the 100 year event is flowing to the north of the berm, highlighting
the low-efficacy of it to provide flood protection for the town.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 50
February 2025
Figure 48. Modeled water depth along length of berm at 2-year and 100-year flows
A previous study (Aspect Consulting, 2009) found that flooding and channel migration in this area is in part due
to backwater and sediment aggradation caused by a constriction at the US-101 bridge crossing. The backwater
observed here is a common theme along the length of US-101 where it crosses the alluvial valley of the
Dosewallips River (as discussed in sections above and below), highlighting a primary causal mechanism of flood
and erosion risk to the town of Brinnon.
An additional factor to consider with regards to the condition of the berm is the level of woody vegetation
establishment on top of it. While the mature trees provide habitat benefits, root expansion within the riprap
that the berm is constructed of may weaken the designed integrity. Vegetation maintenance to reduce berm
deterioration due to tree roots is recommended to ensure that the berm is functioning as intended into the
future.
Area 4 - Flooding in Dosewallips State Park Campground
The south-eastern end of the Dosewallips State Park Campground is located within a low spot that is confined to
the northwest by alluvial deposits and to the southeast by US-101. The combination of low topography and lack
of flow capacity within the US-101 highway prism show the potential for significant ponding of water in this
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 51
February 2025
area, during both the 2-year and 100-year flood (Figure 49). A box culvert is present at the very southern end of
this area, which is documented to cross State Park Creek in the WDFW Fish Passage Database, which enters the
alluvial valley within the state park campground (Figure 49). The ponding in this area indicates that the box
culvert, which is modeled to have ~490 cfs flowing through it during the 100 year flood, is insufficient to drain
the water that enters the area. Water comes from a combination of State Park Creek, and backwater behind the
US-101 bridge constriction spilling to the south. The depth of flooding is projected to increase with climate
change (Figure 42).
Channel migration risk into the state park campground is also prevalent, with the park located within the
downstream translation path of channel meanders between RM 0.6 and RM 0.9. Previously installed rip rap and
bank revetments in this area are degraded (Aspect et al., 2009; Figure 49), and likely have less capacity to
prevent bank migration towards the campground into the future.
Figure 49. Flooding in Dosewallips State Park Campground
Area 5 - Landslide Susceptibility on Southwestern Bank
The Washington state department of natural resources has two mapped, deep seated landslides on the
hillslopes to the south west of the active channel in Dosewallips State Park (Figure 6). The morphology of these
features is similar to that observed in recent landslides that that are known to be caused by cut bank hillslope
toe erosion, such as the Oso landslide in 2014 (LaHusen et al., 2016). The proximity of historic channel planforms
to these hillslopes indicates a high likelihood of future channel migration towards this area, posing a potential
risk of future landslide activity. A landslide on this hillslope would add a large pulse of sediment to the river,
which might cause both immediate change in channel planform, moving it more northward towards Brinnon and
increasing the risk of channel avulsion as the channel reaches a new sediment equilibrium. Landslides also have
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 52
February 2025
a negative impact on fish habitat, both disrupting navigation of returning spawners and suffocating redds with a
large blanket of fine grained sediments (e.g. NSD 2024).
5.2 Projected Changes in Flood Risk with Climate Change
5.2.1 Findings from this Study
As discussed in section 4.3 above, both peak river flows and high tide water levels within Hood Canal are
projected to increase by mid-century. This poses an inherent increase in flood risk, with a greater supply of
water to inundate low-lying, flood prone areas within the Dosewallips estuary and the town of Brinnon. A major
finding within most focus areas in section 5.1 above is the impoundment of water behind road crossings,
particularly at US-101 conveyance structures within the backwater influence of high tides. These impounded
areas are projected to have the greatest magnitude of change in flooding by mid-century, with up to 2.5 ft
increase in water depth near the community center. Water flooding this area is coming from a combination of
tidal levels in Hood Canal and water flooding from the mainstem Dosewallips River, particularly just upstream of
the mainstem bridge on US-101 around the berm.
Along with projected changes in flood risk, channel migration risk will also likely increase with climate change
due to a combination of changes in flow and associated sediment supply. These increases exacerbate existing
channel migration processes in areas that are already susceptible to channel migration. The projected increase
in peak flows will increase water depths and flow velocities and therefore shear stress on the bed and channel
banks, increasing the erosion risk. Additionally, a high sediment supply is a large driver of channel change, as
deposition and aggradation can cause the channel to change course. Sediment supply is likely to increase in the
project reach due to a combination of glacial recession and erosion from increased peak flows higher in the
drainage basin. These two drivers of change are likely to compound with existing conditions that are susceptible
to bank erosion. In particular, the confined, single threaded morphology and lack of stable wood engaged with
the active channel upstream of the US 101 bridge have led to a condition where the channel frequently changes
course, preventing the capacity for mature forest to establish on the floodplain (i.e. inner cycle in Figure 50).
This creates a positive feedback cycle where erosion and channel migration is likely to continue.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 53
February 2025
Figure 50. Diagram of the floodplain large-wood cycle (Collins et al., 2012)
In forested river valleys of the North Pacific coastal ecoregion.
5.2.2 Summary of findings in 2023 Jefferson County Sea Level Rise study
A comprehensive study of the impacts of sea level rise on the coastlines in Jefferson County was completed by
ESA in 2023 (ESA 2023). In this work, ESA focused on a combination of flood and erosion risk along the entire
coastline of the county, including the Dosewallips estuary and the town of Brinnon. The study applied FEMA
base flood elevations (FEMA et al., 2019; total water levels from the 1% exceedance flood) along coastlines to
map areas exposed to flood and erosion risk given existing sea level conditions. The team further delineated
how flood and erosion risks will change with sea level rise projected for different time horizons (e.g. mid-
century, late century) following Miller et al., (2018) (Table 10). They found that low-lying areas connected to
freshwater tributaries were at the highest risk of flood inundation, with areas defined by bluff coastline
morphology having a higher risk of erosion and bluff retreat than flooding.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 54
February 2025
Table 10. Range of Sea-Level Rise Projections for Jefferson County, WA
ANTICIPATED TIMELINE LIKELIHOOD (% CHANCE
EXCEEDANCE)
SEA-LEVEL RISE (FT)A,B SEA-LEVEL RISE (FT)
SELECTED FOR STUDY
2023 NA 0 0
2040 1% 0.8 - 1.1 1
2060 1% 1.7 - 2.1 2
2100 1% 4.6 - 5.2 5
Table has data drawn from Miller et al (2018), and is modified from Table 3 in ESA (2023).
A: The range of sea-level rise projections above are all for the RCP 8.5 (high) emissions and summarize the range of projections for
each stretch of shoreline throughout Jefferson County.
B: The sea-level rise projections account for vertical land movement.
The still water elevations and total water levels in Dabob Bay and at the outlet of the Dosewallips River for a
100-year storm event are higher than those elsewhere in Jefferson County (Figure 51, FEMA 2019), with still
water elevations at the Dosewallips Estuary ranging between 12-12.5’ NAVD88, and total water levels as high as
13-16’ NAVD88 in Dabob Bay. The extent of inundation by total water level at Brinnon for the 100-year event,
which is equivalent to still water and wave height combined, is shown in dark green in Figure 52 below. The
orange and red colors highlight the flood levels with an added 1’, 2’, and 5’ sea level rise, representing 2040,
2060, and 2100 projections respectively. The figure visually highlights that the majority of existing buildings and
the main transportation route in and out of town (US-101) are likely to be impacted by sea level rise over the
next century.
Figure 51. Still water elevations and total water levels for the 1% event
Source: modified from Figures 3 and 4 in ESA (2023)
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 55
February 2025
Along with delineating areas of flood and erosion exposure, ESA completed a vulnerability assessment, which
combines exposure to flooding with sensitivity of existing infrastructure to that exposure, and existing adaptive
capacity to react to a given flooding event. They found that the key features of risk in the town of Brinnon
include the Fire Station, Elementary School, US-101, and the homes and septic systems within the existing and
projected extents of sea water inundation. While the School and Fire Station are outside of projected end of
century sea level inundation, they are near the maximum extent of flooding (Figure 52), which with incorporated
uncertainty, highlights these areas as exposed to future sea level conditions. US-101 will likely be overtopped
with projected sea level rise by 2100, but was not categorized as particularly vulnerable due to adaptive capacity
(ESA 2023). Buildings and septic systems were highlighted as having the greatest vulnerability to sea level
changes, with septic being the main wastewater infrastructure for private parties within the town, and many
buildings and septic systems within the extent of projected seawater inundation (Figure 52).
Figure 52. Sea level rise projections (ESA 2023)
Green indicates areas within the current FEMA Base Flood Elevation (BFE), which is between 13-16’
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 56
February 2025
Along with a greater extent of flood inundation with sea level rise, there is also an expected erosional impact
with higher water levels, leading to a retreating coastline (ESA 2023; Figure 53). Battalio et al., (2016) found that
as sea level elevations increase, so do the elevations where waves dissipate their power. This change in
elevation can lead to inland shore migration that is one to two orders of magnitude greater than the elevational
change in sea level (e.g. 1’ of sea level rise = 10-100’ of lateral coastline retreat). This erosional process is an
important consideration with projected changes in sea level. For the Brinnon project area, some degree of this
erosional impact is expected in all areas, corresponding with the likelihood of different amounts of sea level rise.
Figure 53. Erosion effects of sea level rise (Battalio et al., 2016)
Source: Figure 7 in ESA (2023)
6 PRIMARY IMPAIRMENTS TO FLUVIAL PROCESSES
Impairments to fluvial processes within the project reach influence both aquatic habitat for key salmonid species
and flood and erosion risk to community infrastructure. The over-arching causal mechanisms of these
impairments are:
1. The lack of large wood and disruption of the floodplain large wood cycle (Collins et al., 2012)
2. The long history of bank hardening to keep the channel in a single thread planform (Labbe et al., 2005)
3. Undersized hydraulic structures (i.e. culverts and bridges) that create impoundments during peak
flooding events.
Additionally, the low-lying elevation of Brinnon and the Dosewallips State Park within the estuary has inherent
exposure to impacts from sea level rise, as discussed in detail by ESA (2023) (summarized in section 5.2.2
above). Here we outline the outcomes of each of the three primary impairments to fluvial processes and suggest
applicable restoration actions (Table 11).
6.1 Disruption of the Floodplain-Large-Wood-Cycle
Reference conditions of unmodified rivers in western Washington indicate that large, old-growth trees both on
the floodplain and within the stream play a large role in channel morphology, by promoting a stable,
multithread planform with abundant cover and complexity ideal for salmonid life-history stages (Collins et al.,
2012; Fox and Bolton 2007; Abbe and Montgomery 1996, 2003). Abundant large wood and mature floodplain
forests promote a cycle of channel stability, through a combination of distributing flow and therefore shear
stress across multiple channels, and bank stability through root cohesion and velocity reduction from instream
LWD (Figure 50, e.g. O’Connor et al., 2003; Abbe and Montgomery 2003). The combination of velocity reduction
and flow complexity also leads to sediment aggradation and storage, which further reduces the degree of
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 57
February 2025
incision often seen in modified rivers, further promoting connectivity to floodplain side channels, and promoting
water storage and resiliency to summer low flows.
Removal of large wood in this project reach, through both logging on the floodplain, splash damming upstream
in the Rocky Brook reach in the late 1800’s, and direct removal of log jams, have disrupted the floodplain large
wood cycle (Figure 50). As such, the channel above the US 101 bridge is now characterized by high channel
migration rates, and bank hardening and modifications to prevent erosion towards community infrastructure.
The resulting impacts as discussed in this existing conditions assessment include: 1) high channel migration
rates, 2) lack of floodplain connectivity, 3) channel incision and confinement, 4) low complexity and cover for
aquatic habitat, and 5) erosion of developing floodplain forest, preventing the establishment and growth of
mature trees.
Appropriate actions to mitigate the disruption of the floodplain large wood cycle and the resulting impairments
to channel migration and aquatic habitats include installation of engineered log jams and riparian floodplain
silvicultural treatments. ELJ designs would be targeted towards both stabilizing banks, promoting the
establishment of forested islands, diverting flow into floodplain side channels, and creating habitat diversity for
Puget Sound Chinook and Hood Canal Summer Chum. Silvicultural treatments will include the removal of
invasive species and planting of conifers to accelerate the establishment of mature mixed coniferous forest.
6.2 Artificial Bank Hardening and Channel Simplification
As discussed in section 6.1 above, erosive and flood-prone conditions have led to bank armoring to protect the
Brinnon community, the Dosewallips State Park, and transportation routes. While these modifications have
been often successful at protecting localized bank erosion, they have also straightened and simplified the
mainstem Dosewallips River, disconnecting the river from previously engaged floodplain side channels and tidal
distributary channels, particularly on the north side of the active channel. The impacts on rearing and spawning
salmonids are paramount with these changes, limiting the available habitat for both adult and juvenile fish
through both the disconnection to rearing channels, and armored and coarse substrate conditions from
increased shear stress on the channel bed. Removal of floodplain connectivity also has negative implications for
erosion and flooding, with only one channel to convey peak flows as opposed to a network of channels, leading
to higher erosion risk and less predictable floodwater conveyance.
In the long-term, the effects associated with the confined, single thread channel adds to many of the negative
impacts influenced by a lack of large wood described above, including 1) a lack of floodplain connectivity, 2)
channel incision and confinement, and 3) low complexity and cover for aquatic habitat. While mitigated in the
short term through engineered bank armoring features (e.g. the left bank berm upstream of the highway),
channel migration and floodplain erosion would be heightened as well without continual maintenance of these
features (BOR 2005). Actions to restore floodplain connection while also maintaining bank stability near critical
infrastructure include: improving the habitat value of bank revetment features by adding wood complexity, and
connecting the floodplain and distributary channels within Dosewallips State Park by diverting flow with ELJs.
This work would expand the restoration of the tidal distributary network that was undertaken with the Lower
Dosewallips Floodplain and Estuary Restoration project (see section 1.3), via side channel excavation and ELJ
installation downstream of the US-101 crossing.
6.3 Undersized Hydraulic Structures and Lack of Flood Protection
The areas that are most susceptible to flooding, as shown through hydraulic modeling, are along the western
embankment of US-101 near box culverts and bridges designed to convey water beneath the highway (Figure
32; section 5.1). With projected increases in both the magnitude and frequency of peak flows with climate
change, these crossings will need to convey more water during storm events. Comparisons between existing
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 58
February 2025
conditions and future mid-century hydraulic modeling results (Figure 42) indicate that the box culverts across
Walcott Slough near the Community Center are projected to have the greatest increase in flood depth.
Increased backwater behind the US-101 bridge across the mainstem Dosewallips River and the two bridges to
the north is also expected. The lack of capacity of these structures, and the lack of flood protection provided by
the berm upstream of the US-101 bridge also influences the flood flows in the direction to the Community
Center, compounding the impact to that area. Suggested treatments include increasing the width and therefore
capacity for these structures to convey water beyond the highway.
7 PROPOSED ACTIONS TO TREAT IMPAIRED PROCESSES
This assessment is the first step in identifying opportunities within the framework of landowner and stakeholder
input to address the geomorphic, hydraulic, and habitat impairments observed within the Dosewallips River
Brinnon project reach. These impairments are typical of Pacific Northwest rivers and floodplains affected by
historical logging, channel manipulation, and consequent disconnection of rivers and their floodplains. The
impairment of fundamental river migration and flood processes has reduced the quantity and quality of habitats
that are essential to supporting native salmonids and has increased the flood and erosion risk to the local
Brinnon community, the Dosewallips State Park, and transportation infrastructure.
NSD considered restoration actions that will 1) directly address the causal mechanisms of impairment, 2)
improve the quality and quantity of habitats essential for Hood Canal summer chum and Puget Sound Chinook in
the lower Dosewallips River and 3) reduce the flood and erosion risk the community and infrastructure. Table 11
provides an overview of the proposed restoration actions, their link to the causal mechanisms of impairment,
and the expected results of each action. Future work will apply these proposed actions to create conceptual
restoration plans to allow for continued community and stakeholder input.
Table 11. Table of impairments and proposed actions
CAUSAL
MECHANISM RESULTING IMPAIRMENTS PROPOSED ACTIONS EXPECTED RESULTS
Disruption of
Floodplain-
Large-Wood
Cycle
1. Reduced channel stability
and increased channel
migration rates
2. Increased channel incision
and reduced floodplain
connectivity
3. Reduced aquatic
complexity and cover (i.e.
pools and large wood)
4. Increased flow velocities
with corresponding
coarsening of bed
material and reduction of
suitable juvenile rearing
1. Conifer underplanting and
invasive plant removal
2. Construct Engineered Log
Jams
3. Excavation of side channels
1. Promote conifer
succession and restore
the long-term large wood
cycle.
2. Increase channel stability
in near-term with ELJs
and long-term with
conifer succession
3. Encourage split flows into
side and distributary
channels
4. Increase flow partitioning
to reduce channel
velocities.
5. Increase in flood water
capacity.
6. Creation of new juvenile
salmonid rearing habitat
Bank Hardening
and Channel
Simplification
1. Channel confinement
2. Reduced floodplain
connectivity
1. Conifer underplanting and
invasive plant removal
2. Construct Engineered Log
Jams
1. Stabilize banks, near term
with large wood, and
long term with growth of
planted conifers.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 59
February 2025
CAUSAL
MECHANISM RESULTING IMPAIRMENTS PROPOSED ACTIONS EXPECTED RESULTS
3. Reduced recruitment of
large wood and
associated reduction in
aquatic habitat
complexity
4. Increased flow velocities
with corresponding
coarsening of bed
material and reduction of
suitable juvenile rearing
5. Flood and erosion risk
without continued
maintenance of bank
armoring features
3. Removal of relict bank
structures.
4. Addition of large wood to
existing bank structures.
2. Aggrade sediment in
intentional areas
3. Add habitat complexity
features to existing bank
armoring
Undersized
hydraulic
structures and
lack of flood
protection
1. Impoundments to west of
US-101 during flood
events.
2. Increased risk to Brinnon
community, State Park,
and US-101 with climate
change.
1. Implement a Comprehensive
Flood Planning Process
1. Evaluate measures to
increase flood
conveyance and protect
critical community
infrastructure during
future flood events.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 60
February 2025
8 REFERENCES
Abbe, T.B. 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.
Ames, J., G. Graves, and C. Weller. 2000. Summer chum salmon conservation initiative. Report to the
Washington Department of Fish and Wildlife and Point-No-Point Treaty Tribes.
Aspect Consulting 2009. River and Stream Impairment Analysis; WRIA 16 and 14b, Skokomish-Dosewallips
Planning Area. Prepared for WRIA 16 Planning Unit, June 30, 2009. Project Number 080261-001.
Battalio, R.T., P. D. Bromirski, D. R. Cayan, L. A. White, 2016. Relating Future Coastal Conditions to Existing FEMA
Flood Hazard Maps: Technical Methods Manual. Prepared for California Department of Water Resources
and California Ocean Science Trust, Prepared by Environmental Science Associates (ESA), pp. 114.
Beecher, H., Caldwell, B., Kohr, J., Granger, R., Gates, K., Covert., J. 2022. Instream Flow Study Guidelines:
Technical and Habitat Suitability Issues Including Fish Preference Curves. Washington Department of Fish
and Wildlife and Washington State Department of Ecology #04-11-007.
Booth, A. M., LaHusen, S.R., Duvall, A.R., Montgomery, D.R. 2017. Holocene history of deep-seated landsliding in
the North Fork Stillaguamish River valley from surface roughness analysis, radiocarbon dating, and
numerical landscape evolution modeling. Journal of Geophysical Research: Earth Surface. v122(2) p456-
472.
Brewer, S., J. Watson, D. Christensen, R. Brocksmith. 2005. Hood Canal and Eastern Strait of Juan De Fuca
Summer Chum Salmon Recovery Plan. Hood Canal Coordinating Council, 339 p.
Bush, K.R. and Dubois, E.S. 2023. Cultural Resource Overview: Dosewallips River Aquatic Habitat Restoration
Project, Jefferson County, Washington. Prepared by Equinox Consulting on behalf of Natural Systems
Design and Jefferson County on August 10th, 2023.
Correa, G. 2003. Salmon and Steelhead Habitat limiting Factors. Water Resource Area 16 Dosewallips-
Skokomish Basin. Washington State Conservation Commission. 267 pp.
Environmental Science Associates (ESA) 2023. Jefferson County Sea Level Rise Study. Prepared for Jefferson
County Department of Community Development, June 2023. Funded by WA DOE 2021-2023 Shoreline
Master Program Competitive Grant (No. SEASMPC-2123-JCDCD-00011).
Federal Emergency Management Agency (FEMA). 2019. Flood Insurance Study (FIS) for Jefferson County,
Washington and Incorporated Areas. June, 2019.
Grossman, E.E., Tehranirad, B., Stevens, A.W., VanArendonk, N.R., Crosby, S., and Nederhoff, K., 2023, Salish Sea
Hydrodynamic Model: U.S. Geological Survey data release, https://doi.org/10.5066/P946SC3L.
Hamlet, A.F., P. Carrasco, J. Deems, M.M. Elsner, T. Kamstra, C. Lee, S-Y Lee, G. Mauger, E. P. Salathe, I. Tohver,
L. Whitely Binder, 2010, Final Project Report for the Columbia Basin Climate Change Scenarios Project,
http://warm.atmos.washington.edu/2860/report/.
Jefferson County 2017. Jefferson County WA Wetlands. Published June 22, 2017. URL: https://gisdata-
jeffcowa.opendata.arcgis.com/datasets/JeffCoWA::wetlands-1/about
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 61
February 2025
Kondolf, G.M. and Wolman, G.M., 1993. The Sizes of Salmonid Spawning Gravels. Water Resources Research.
v29(7): p2275-2285.
Krueger, K.L., K.B. Pierce, T. Quinn, and D.E. Penttila. 2011. Anticipated Effects of Sea Level Rise in Puget Sound
on Two Beach-Spawning Fishes. Puget Sound Shorelines and the Impacts of Armoring—Proceedings of a
State of the Science Workshop.
https://wdfw.wa.gov/sites/default/files/publications/01210/wdfw01210.pdf
Labbe, T., Grotefendt, R., Carter-Mortimer, A., Jones, J.L. 2005. Dosewallips River Habitat Assessment: Coupling
High-Resolution Remote Sensing and Ground Surveys to Prioritize Aquatic Conservation, Olympic
Mountains, Washington State. Prepared for USDI- Bureau of Indian Affairs by Port Gamble S’Kallam Tribe
Natural Resources Department. December 6th, 2023.
LaHusen, S.R., Duvall, A.R., Booth, A.M., Montgomery, D.R. 2016. Surface roughness dating of lon-runout
landslides near Oso, Washington (USA), reveals persistent postglacial hillslope instability. Geology. v44(2),
p111-114, doi:10.1130/g37267.1
Manga, M., & Kirchner, J. W. 2000 Stress partitioning in streams by large woody debris. Water Resources
Research, 36(8), 2373–2379.
Mastin, M.C., Konrad, C.P., Veilleux, A.G., and Tecca, A.E., 2016, Magnitude, frequency, and trends of floods at
gaged and ungaged sites in Washington, based on data through water year 2014 (ver 1.2, November 2017):
U.S. Geological Survey Scientific Investigations Report 2016–5118, 70 p.,
http://dx.doi.org/10.3133/sir20165118.
May, C. L., & Gresswell, R. E. 2003 Processes and rates of sediment and wood accumulation in headwater
streams of the Oregon Coast Range, USA. Earth Surface Processes and Landforms, 28(4), 409–424.
Miller, I.M., C. Shishido, L. Antrim, and C.E. Bowlby, 2013. Climate Change and the Olympic Coast National
Marine Sanctuary: Interpreting Potential Futures. Marine Sanctuaries Conservation Series ONMS-13-01.
U.S. Department of Commerce, National Oceanic and Atmospheric Administration, Office of National
Marine Sanctuaries, Silver Spring, MD. 238 pp.
Miller, I.M., Morgan, H., Mauger, G., Newton, T., Weldon, R., Schmidt, D., Welch, M., Grossman, E.E., 2018.
Projected Sea Level Rise for Washington State - A 2018 Assessment. A collaboration of Washington Sea
Grant, University of Washington Climate Impacts Group, University of Oregon, University of Washington,
and US Geological Survey. Prepared for the Washington Coastal Resilience Project.
Montgomery, D.R., Buffington, J.M. 1997. Channel-reach morphology in mountain drainage basins. GSA Bulletin.
v109(5), p 596-611.
NatureServe. 2015. International Ecological Classification Standard: Terrestrial Ecological Classifications.
NatureServe Central Databases. Arlington, VA. U.S.A. Data current as of 26 June 2015.
National Oceanic and Atmospheric Administration (NOAA) 1969. Bathymetry Report H09036: Hood Point to
Quatsap Point, Hood Canal, Washington. Point Soundings, acquired between February 25, 1969 and April
15th, 1969.
NSD 2021. Dosewallips River Resiliency Plan: Powerlines/Lazy C Reach. Prepared for Jefferson County
Department of Public Health by Natural Systems Design. June, 2021.
NSD 2024. Lower Quinault River Landslide Assessment. Prepared for Quinault Indian Nation by natural systems
design. November 2024.
JEFFERSON COUNTY BRINNON REACH EXISTING CONDITIONS ASSESSMENT
Natural Systems Design 62
February 2025
NV5, 2024. Dosewallips, Washington 2023 Topobathymetric Lidar Technical Data Report. January, 26, 2024.
O’Connor, J.E., Jones, M.A., Haluska, T.L. 2003. Flood plain and channel dynamics of the Quinault and Queets
Rivers, Washington, USA. Geomorphology. v51(1-3); p31-59.
Quinn, T. P. 2018. The behavior and ecology of Pacific salmon and trout. University of Washington Press, Seattle.
Rocchio, F.J. and R. Crawford. 2015. Ecological Systems of Washington State. A Guide to Classification.
Washington Natural Heritage Program, Department of Natural Resources, Olympia, WA.
Shared Strategy. 2007. Puget Sound Salmon Recovery Plan. Seattle WA. 550 pages.
Simenstad, C. A. 2000. Estuarine landscape impacts on Hood Canal and Strait of Juan de Fuca summer chum
salmon and recommended actions. Appendix Report 3.5, pp. A3.111-A3.132 in J. Ames, G. Graves, and C.
Weller (eds.), Summer Chum Salmon Conservation Initiative: An Implementation Plan to Recover Summer
Chum in the Hood Canal and Strait of Juan de Fuca Region. Wash. Dept. Fish. Wildl.., and Point-No-Point
Treaty Tribes, Olympia, WA.
Smith, C.D., and T.L. Liedtke, 2022. Potential effects of sea level rise on nearshore habitat availability for surf
smelt (Hypomesus pretiosus) and eelgrass (Zostera marina), Puget Sound, Washington: U.S. Geological
Survey Open-File Report 2022–1054, 17 p., https://doi.org/10.3133/ofr20221054.
USACE (2023). HEC-RAS, River Analysis System. Version 6.4.1. U.S. Army Corps of Engineers, Hydrologic
Engineering Center. Davis CA, Hydrology Engineering Center.
U.S. Bureau of Reclamation (US BOR) 2004. Channel Migration Zone Study for the Duckabush, Dosewallips, Big
Quilcene, and Little Quilcene Rivers, Jefferson County, Washington. Technical Service Center Flood
Hydrology Group D-8530, author Jeanne E. Klawon. September 2004.
U. S. Fish and Wildlife Service (USFWS) 2020. National Wetlands Inventory website. U.S. Department of the
Interior, Fish and Wildlife Service, Washington, D.C.
U.S. Geological Survey (USGS) 2019, The StreamStats program, online at https://streamstats.usgs.gov/ss/,
accessed April 2024.
U.S. Geological Survey (USGS), 1982. Guidelines for Determining Flood Flow Frequency. Bulletin #17B of the
Hydrology Subcommittee, U.S. Department of the Interior Geologic Survey, March 1982, Reston, VA.
Washington Department of Fish and Wildlife and Point No Point Treaty (WDFW and PNPTT). 2003. Summer
Chum Salmon Conservation Initiative-An Implementation Plan to Recover Summer Chum in the Hood Canal
and Strait of Juan de Fuca Region. Supplemental Report No. 4. Report on Summer Chum Salmon Stock
Assessment and Management Activities for 2001 and 2002. Washington Department of Fish and Wildlife.
Olympia, WA. 219 p.
Washington State Department of Waterways, 1957. Primary State Highway SR 100 Dosewallips River Bridge.
Jefferson County. Approved July 2nd, 1957.
Wolman, M.G. 1954. A method of sampling coarse river-bed material. Transactions American Geophysical
Union. Volume 35. Number 6. Pp. 951-956.