From FY07-09 Solicitation - Objective 3: Restore hydraulic connectivity to the greatest extent possible in priority geographic areas by 2012.

Klickitat River (RM 18 to 32) Floodplain Conservation and Restoration (Haul Road) Project   (2009-2013)

Background:  The project addresses limiting features (channel confinement) identified for the Klickitat River between river miles 18.3 and 32.2 by the Klickitat Subbasin Plan and Klickitat Lead Entity Salmon Recovery Strategy.  This portion of the river has the greatest habitat complexity of any reach in the lower Klickitat River and provides critical spawning, migration and rearing habitat for winter and summer steelhead (ESA-“Threatened”), Chinook salmon (spring and fall runs), and coho salmon.  This reach provides a high proportion of the basinwide spawning habitat for all three species, accounting for on average 18% (7-34%), 31% (10-58%), and 38% (5-37%) of the annually observed basinwide spawning for steelhead, fall Chinook, and coho, respectively (2002-2008).  Riparian and floodplain conditions have been degraded by a combination of channel encroachment and floodplain isolation by road fill as well as 1996 flood deposits.  The absence of other floodplain development coupled with somewhat less-confined valley conditions affords this reach greater resiliency than downstream reaches.    The project is occurring in two stages: acquisition (Phase 1 funding) and restoration (all subsequent phases of funding).  Columbia Land Trust (CLT) is the lead for acquisition and also sponsors the SRFB grant for the initial phase of restoration.  KWEP is the technical lead for design and construction oversight of restoration actions as well as assisting planning activities.

Project Goal:  Restore connectivity of riverine, floodplain, and hillslope processes to the Klickitat River between river miles 19.0 and 31.6.

 Phase 1

• CLT acquires approximately 480 acres of floodplain, riparian and associated upland as well as the road itself
• 570 lineal-feet of cross-valley embankment removed from Dead Canyon Creek
• Railroad trestle removed and treated wood hauled off-site
• Pullback of 2,800 lineal-feet of embankment along south approach to Dead Canyon
• Funded by:  Salmon Recovery Funding Board # 04-1715 and Bonneville Power Administration Project # 1997-056-00

Figure 1.  Dead Canyon Ck pre- (left) and post-removal of railroad trestle and associated embankment (right).

Phase 2

• Approximately 6,700 l.f. of embankment graded to enhance riverine and floodplain function:

•  ~1780’ of floodplain channel constructed
•  Construction of 11 woody debris jams
• Restore deformability of channel margins to permit lateral channel migration and serve as longterm LWD source
• Restore hillslope interaction

• Removed asphalt from 4.5 miles of floodplain road
• Funded by: Salmon Recovery Funding Board Project # 05-1594 and Bonneville Power Administration Project # 1997-056-00

Figure 2. Haul Road Phase 3 pre-(left) and post-removal of the road surface, fill and recontouring of the floodplain (right).

 Phase 3

• Approximately 8,936 l.f. of embankment graded to enhance riverine and floodplain function:

• Fill removal (6 segments totaling 2,428 l.f.)
• Fill pullback (27 segments totaling 6,508 l.f.)
• Culvert Removal – cross drain (6)
• Culvert Removal – non-fish bearing tributaries (3)
• Culvert Removal – seasonal fish bearing tributary (1)
• Restore deformability of channel margins to permit lateral channel migration and serve as longterm LWD source
• Restore hillslope interaction

• Funded by: Salmon Recovery Funding Board Project #11-1428R and Bonneville Power Administration Project # 1997-056-00

Phase 4 (2013):

Goal:  Re-shape 1.57 miles of floodplain road embankment to restore connectivity of riverine, floodplain, and hillslope processes along 1.65 river miles of the Klickitat River.

 Project Scope:  The project area is located between river miles 22.20 to 23.85 of the Klickitat River.

Design:  The general approach is to remove or reduce the influence of the embankment on the active channel and valley bottom. Rip-rap will be removed from all segments to increase deformability and likelihood of the river reclaiming its historic channel migration zone. In general, where the embankment contacts the main or an active side channel, the toe of the embankment will not be modified beyond removal of rip-rap to simplify permitting and reduce project costs associated with de-watering.

 Deliverables:

Instream Habitat/Channel Reconfiguration and Connectivity:

• 9.78 acres of off-channel habitat reconnected
• 0.73 miles active streambank enhanced via embankment pullback
• 0.84 miles floodplain (not currently active streambank) enhancement via fill removal/pullback

Fish Passage Improvement:

• 1 road crossing removal to open .0.32 miles of tributary and side channel habitat

Riparian Habitat:

• 11.4 acres patrolled and controlled against non-native, invasive vegetation and planted with native species

 Treatments are generally described as follows:

• Fill removal (11 segments totaling 3,896 l.f.)
• Fill pullback (14 segments totaling 4,377 l.f.)
• Culvert Removal – cross drain (4)
• Culvert Removal – non-fish bearing tributaries (2)
• Culvert Removal – seasonal fish bearing tributary (1)
• Total treatment: 8,273 lineal feet of floodplain road

From FY07-09 Solicitation - Objective 4: Restore or enhance 70% (by length) in-channel and riparian habitat conditions to the greatest extent possible in priority geographic areas by 2015

Upper Klickitat River In-Channel and Floodplain Enhancement Project - Phase 2    (2008 - 2010)

Introduction:  The project addresses limiting features (channel confinement and habitat simplification) identified for this reach by the Klickitat Subbasin Plan and Klickitat Lead Entity Salmon Recovery Strategy (KLESRS).  The core Ecosystem, Diagnosis & Treatment (EDT) reach that encompasses the project sites ranks third overall in the Klickitat subbasin in restoration potential for combined performance of steelhead and spring Chinook (NPCC, 2004).  Project work addresses most of the top limiting factors identified for the reach between RM 70 and 74.5

Pre-project Problem: The primary problem is channel simplification.  The reach appears to have historically been a forced-pool and pool riffle morphology that had become plane-bed.  The channel had incised 1-2' and was largely armored with large cobble and small boulder material.  Pools had become infrrequent and where they did occur, residual pool depths were generally shallow (12-18").  The shift to a plane-bed is believed to have been triggered by realignment and filling of the channel and floodplain assoiciated with a construction of the 255 Road in the mid-1970's and subsequently magnified by flooding.  Prior to commencing project work there were six locations where the active channel contacted this arterial road and erodes the embankment.

In addition to the road’s influence on morphology and habitat, it seems likely that stream cleaning occurred at some point.  The Washington Department of Fisheries conducted a habitat survey between Castile Falls and McCormick Meadows in 1957 (LeMier, et al. 1957) and noted, “many log and debris jams caused by windfalls are present in the stream area covered ranging in size to 200 feet long, 50 feet wide, and 18 feet high.”  The report notes other conditions (depth and pool frequency) that were more favorable to salmonids than those observed pre-project.  In particular, the reach within which the Upper Klickitat Phase 2 project occurs contained, “The largest and most serious log jams.” The report went on to prescribe “…therefore, removal of these obstacles is mandatory if the [Castile] falls improvement work is undertaken.”  Stream cleaning was a common practice throughout the Pacific Northwest into the 1980s and the construction of the 255 Road would have made the reach much more accessible to the practice had it not occurred previously.  Given the absence of jams or older relics of jams on floodplain, it seems highly likely that stream cleaning occurred in the project reach.

Project Goal:  Increase physical habitat complexity and reduce river-road interaction.  Enhance instream habitat and water quality to benefit mid-Columbia steelhead (ESA - Threatened) and spring Chinook (WDFW - Depressed) at three priority sites totaling 0.29 river miles (cumulative) along the Klickitat River between RM 70 and 74.5.  Roughly 3750 lineal feet of side channel will be reconnected. 

Design:  The general premise of the project was to convert the plane-bed morphology to forced-pool morphology.  There are currently a few isolated "islands" of recovering channel where large woody debris (LWD) recruited from bank mass-wasting has been deposited into jams and locally controls gradient and flow direction.  These areas tend to have fair to good pool formation immediatley upstream and downstream as well as accumulations of gravel.

The overall approach of the project is to mimic these areas and effectively fill the gaps in between them. YKFP staff developed the design in cooperation with Interfluve, Inc (Conley 2008).  We developed a 30% paper design based on collection of topographic data and a 1-dimensional hydraulic model.  Typical treatments were developed and continuous field supervision was provided to the construction contractor by YKFP and/or Interfluve staff.  Constructed jams were not installed at scour depth, but were built to accommodate scour and settling.  There were four main types of treatments:

•  Floodplain benches were constructed in Reaches 3 and 4 where the active channel contacts the road to provide a buffer between the toe of the road fill and active channel.  Excavation along the left (non-road) channel margin maintained channel capacity and provided a source for alluvial material to backfill the bench on the right-bank / road-side (Fig. 1).  A base layer of boulders and LWD was be placed to create the core of the new floodplain surface then backfilled with native cobbles and gravels using a dig-and-pitch approach (Fig 2).  This realigned the channel to be compatible with the bench treatment, yet maintain flow capacity.  The finished grade of the new floodplain was constructed to be inundated at approximately a 5-year recurrence (and greater) flood and provide a 10 to 25 horizontal foot buffer from road fill (Figures 3 and 4).  The new surface was planted with dormant hardwood cuttings.  Due to the greater hydraulic force in these areas, LWD was ballasted posts as well as with boulders using epoxy and cable.  In some cases pools and runs were excavated adjacent to the benches and LWD treatments. 

• LWD jams were constructed on the mainstem and side channel to encourage channel complexity and improve local hydraulic conditions to facilitate retention and sorting of sediments and pool formation/maintenance.  In particular, jams were constructed at sites 2, 3, and 4.  Jams consisted of 2-3 “key” pieces (>30” diameter) with additional members added as necessary.  Stability of the jams was provided by site selection, partial burial/keying, orientation and sizing of key pieces, as well as placement of additional members as ballast.  In some cases, cabling and ballasting with boulders, backfill, and/or posts was employed to increase stability.
• Channel reconnection occurred at site 4 where approximately 200’ of channel was constructed to reconnect a roughly 4000’ long side channel.  Excavated materials were used for backfill of LWD structures as well as graded into a nearby talus slope.
• Debris “barbs” with adjacent pools were constructed at Site 2 instead of a continuous floodplain bench to conserve materials and budget.

Construction:  Construction at sites 2, 3, and 4 occurred in the fall of 2009.  Construction at site 1 occured in the fall of 2010.

• Fall 2008 – logs of blowdown origin collected, transported and stockpiled at project sites
• Fall 2009 – logs of blowdown origin and boulders collected, transported and stockpiled at project sites; gravel fill produced; LWD jams completed; excavation of new channel and reconnection of historic side channel; temporary erosion control measures implemented
• Fall 2010 - LWD jams completed and side channel activated.

Construction was funded by YNFP sponsored grants from the Washington State Salmon Recovery Funding Board (SRFB) and the Pacific Coastal Salmon Recovery Fund (PCSRF).  KWEP provided funding for design, construction oversight and implementation (2010).  KWEP also funded non-LWD materials and supplies.  Significant components of the implementation include:

• 950 logs of blowdown origin and 690 boulders were collected and delivered to project sites
• Construction was completed at Sites 1, 2, 3, and 4B totaling approximately 1500' of bank.
• Production and delivery of filter rock for reaches 1, 2, 3, and 4
• Collection and delivery of boulder ballast for reaches 1, 2, 3, 4, and 6
• Construction of approximately 2900' of side channel at reaches 4A  & 4B, approximately 1300' of which is expected to be perennial
• Constructed 35 LWD jams
• Installed floodplain roughness at 5 locations
• Excavated 7 main-channel pools

Figure 3.  Upper Klickitat River Enhancement Phase 2 - site 4A pre- (top) and post-enhacement (bottom).

From FY07-09 Solicitation - Objective 4: Restore or enhance 70% (by length) in-channel and riparian habitat conditions to the greatest extent possible in priority geographic areas by 2015

Upper Klickitat River In-Channel and Floodplain Enhancement Project - Phase 3  (2012-2013)

The project location is on a side channel of the Klickitat River in the vicinity of river mile 75.5 where a cross-valley alignment of the 255 Road interrupts floodplain connectivity.  The 255 Road is the major arterial road in the upper Klickitat watershed.  Currently, there is a single bridge crossing.  Several side channels exist up-valley from the road crossing.  The alignment of the larger of these channels (BFW ~20') is deflected where it contacts the fillslope from the 255 road where it then runs along the toe of the fill for approximately 300'.  A historic river alignment exists immediately down-valley of the road fill. 

The project reach is roughly 3300’ above sea level.  The contributing drainage area ranges from 70 mi² and is predominantly forested by Douglas fir, grand fir, ponderosa pine, and lodgepole pine.  Annual precipitation is approximately 65 inches and occurs primarily as snow.  Streamflows are primarily snowmelt driven, though the highest peak events on record (e.g. 1996) have been associated with large regional rain-on-snow events.  The study reach was identified as a priority by Yakama Nation Fisheries Program (YNFP) specialists based on observed interference of side-channel planform.

The project area is located upstream of Castile Falls where an improperly designed fish ladder precluded fish passage from roughly the mid-1960s until 2004.  The spring Chinook population upstream of the falls has been jump-started by out planting of surplus hatchery returns in some years with natural returns increasing each year since ladder modifications were completed in 2004.  Steelhead have been observed spawning upstream of Castile Falls and are currently recolonizing the area via straying.   Resident O. mykiss are observed in the area year-round.  Steelhead, spring Chinook, and resident rainbow trout will be the primary beneficiaries of this project, as it will improve spawning and rearing habitat.  Neither steelhead nor spring Chinook have been observed spawning within the proposed treated area.  O. mykiss have been observed rearing in the side channel 300’ downstream of the road crossing.  Spring Chinook have been observed spawning both upstream and downstream of the project reach.

Currently, the side channel upstream of the road is intermittent for approximately 1200 of its 2100’ length.  Debris accumulation and associated aggradations at the head of the side channel have been contributing to increasing flow duration over the past five years.  The average observed period for surface flow for the intermittent portion of the channel is March into June.  A headcut approximately 1100' (valley distance) upstream of the road crossing is resulting in increased channel capacity.  This is desirable from a habitat and stream function perspective as the current single-thread mainstem channel is incised and of marginal to low habitat value.  Current hydraulic conditions in the primary channel are unfavorable for habitat enhancement.  Once flows (particularly high flows) are more distributed as a result of the proposed project, restoration of conditions in the primary channel will be evaluated as a follow-up project.  The channel downstream of the road crossing has silted in for about 100-150’ immediately downstream.  The remaining 1300’ segment of downstream channel is perennial and gains flow from an unnamed tributary approximately 150’ downstream of the road.

The incision (and subsequent armoring) of the mainstem Klickitat River is believed to have been triggered by construction of the 255 Road (in the late 1970s).  The incision appears to have stabilized and mass wasting of banks is now prevalent through the reach.  Substrate is generally boulder to large cobble, except where LWD recruited from bank mass-wasting has been deposited into jams and locally controls gradient.  These isolated areas tend to be more of a forced-pool morphology with fair to good pool formation immediately upstream and downstream.  These areas also have accumulations of gravel indicating that the supply is present and there is potential for retention if the proper in-channel conditions exist.  The overall approach of the project is to mimic these areas and effectively fill the gaps in between them. 

The 255 Road is the primary route in the Upper Klickitat for tribal members accessing reservation lands for ceremonial, subsistence and economic purposes as well as for transporting forest products.  Relocation of the road would be a more desirable option and allow for evolution of a more stable planform and profile.  However, the size of the road and valley morphology make relocation cost prohibitive. 

Project Goal and Objectives:

Goal: Enhance river and floodplain function and increase habitat quantity and quality for steelhead and spring Chinook along roughly 0.5 miles of the Klickitat River in the vicinity of river mile 75.5.

Objectives: 1) increase channel complexity, 2) reestablish floodplain connectivity, and 3) increase large woody debris levels.

Perforation of the embankment will distribute flow at channel-forming discharges.  This will facilitate: 1) development of the side channel and 2) reduce shear in the primary channel.  As the side channel continues to develop, rearing habitat will increase and spawning habitat (for O. mykiss) will develop.  LWD placed in the side channel as part of the project will increase the quality and hasten development of rearing habitat.

Construction:

• completed design
• completed excavation (~2000 cu-yd)
• completed hauling all rock materials (armor and top-course)
• transported all 3 bridge sections to the site
• pre-cast footings fabricated and delivered to site
• completed channel invert (~100 l.f.) through the bridge crossing
• completed 90% of slope armoring through bridge crossing
• roughed-in approximately 250 l.f. of new channel downstream of crossing
• completed the foundation (including subgrade compaction and footing placement) for the west abutment

From FY07-09 Solicitation - Objective 4: Restore or enhance 70% (by length) in-channel and riparian habitat conditions to the greatest extent possible in priority geographic areas by 2015

Objective 3: Restore hydraulic connectivity to the greatest extent possible in priority geographic areas by 2012.

Tepee Creek - IXL Meadows Restoration Project (2006 & 2007)

Background:  The project addresses limiting habitat features (bed degradation and pool structure) identified by the Subbasin Plan and KLESRS along 2000 feet of Tepee Creek.  Tepee Creek is a tributary to White Creek and provides important spawning and rearing habitat for ESA-listed Middle Columbia River steelhead and is a top geographic priority. The White Creek watershed as a whole is likely the most important spawning and rearing tributary watershed within the Klickitat subbasin. In recent years, the White Creek watershed has accounted for up to 40% of the observed steelhead spawning in the entire Klickitat subbasin.  Tepee Creek has accounted for up to 21% of the observed spawning in the Klickitat subbasin in recent years; however in most years it likely accounts for between 5 and 10%.  Extensive reaches of Tepee Creek have become incised and are now intermittent in many places that anecdotal information suggests were once perennial.

Problem:  In general, summer rearing habitat in the White Creek watershed is highly limited.  The project reach is incised, which has contributed to marginal spawning habitat and poor rearing conditions.  The project reach dried up in 4 out 5 years preceding project implementation.  Stranding and subsequent mortality of juvenile O. mykiss was a routine pre-project occurrence when the reach dried up (usually in early July).  Summer freshets have been observed to create continuous surface flow through the reach.  In one instance (August of 2004), flow reappeared temporarily (for 2-3 days) in the reach, prompting juvenile migration from upstream reaches and resulted in a second round of stranding/mortality for at least three age classes of O. mykiss within a single season. 

The primary mechanism for incision was initiated when an undersized bridge ½ mile downstream plugged causing Tepee Creek to capture the adjacent floodplain road.  Headward migration of the incision propagated upstream until it was arrested by the culvert inverts at the upstream end of the project reach.  Incision occurred largely within the planform of the historic channel, though there are a number of locations where meanders were cut off.  Hydraulic modeling of pre-project conditions indicated that a 10-year recurrence flood was required for overbank access.  Modeling correlated well with field indicators that suggested the channel had incised three to four vertical feet. 

A secondary mechanism for incision is related to increased peak flows.  The hydrologic effect of forest roads on peak discharges was modeled (using HEC-HMS) for a portion (~6.5 square-miles) of the contributing watershed in 2003.  The model showed a 7.3% increase in discharge for a 2.5-year storm event due to forest roads.  Unto itself, this increase creates additional energy resulting in additional erosive force on the stream bed and banks.  The effect is compounded by the limited generation of bedload-sized material by the watershed (discussed above) because the weathering and delivery rate of gravels isn’t keeping pace with the rate at which they are transported through and moved out of the system.  Road drainage and decommissioning treatments are being implemented in the watershed as part of a separate project.

Project Goals: 

1)      Increase floodplain storage

2)      Reduce severity of active channel hydraulic conditions during high flows

3)      Enhance quantity and quality of steelhead spawning and rearing habitat

4)      Potentially restore base flows to this and downstream reaches

5)      Restore suitability of valley bottom for medicinal and traditional food plants

Design:  The conceptual intent was to raise the streambed elevation to regain floodplain storage and restore overbank flow frequency. The basic approach employed was to import gravels and reconstruct pool-riffle sequences through the project reach.  Large Woody Debris (LWD) was used to enhance pool quality and longevity as well as moderate bank erosion.  The majority of design work occurred in 2005 and early 2006.  The design team included the KWEP staff hydrologist and a geomorphologist and engineer from Interfluve, Inc.

Construction:  Construction was initiated in October 2006 and was implemented over two field seasons:

• Fall 2006 – gravel fill produced, hauled and stockpiled; riffles constructed to rough grade and dimension; coarsened grade-control riffle constructed; LWD debris collected, transported, and stockpiled; 95’ of new channel constructed; riparian vegetation salvage completed; roughly half of LWD jams completed; temporary erosion control measures implemented

•  Summer 2007 – riffles and pools graded to finished elevation and dimension, floodplain LWD treatments completed, LWD jams completed, re-vegetation completed, and livestock exclusion fence constructed

 Construction was funded by a YNFP-sponsored grant from the Washington State Salmon Recovery Funding Board (SRFB).  KWEP provided funding for design and construction oversight.  KWEP also funded non-LWD materials and supplies.   Significant components of the implementation include:

• Importation of gravel to raise bed elevation (~3’) and reconstruct pool/riffle sequences along 1740’
  
• Construction of 28 LWD jams constructed along channel margins to maintain pool depths, provide cover, and restrict bank erosion 
  
• Construction of 95’ of new stream channel to reconnect 135’ of historic channel and lengthen the overall reach to 1990’
  
• Removal of two culverts and related fill from an abandoned cross-valley road alignment to restore high-flow access into a historic channel alignment
• Construction of several dozen LWD placements constructed on floodplain
• Planted shrubs and trees on 2 acres (cumulative) of riparian area and floodplain
• Re-seeded approximately 4.5 acres of floodplain and riparian area
• Weed control on 7.8 acres
• Construction of 3250’ of livestock exclusion fence
• Construction of a 140’-long coarsened riffle at 3% grade to provide persistent grade control at the downstream end of the reach
  

Results:

• Flow Duration: 23 perennial pools maintained all 3 years since construction
• Groundwater: 2 - 4’ increase in summer water table
• High Flow Access: at bankfull or lower flows to four side channels totaling 835 l.f.
• Pools: increased from 15 to 23 (65%); greater depths & cover
• Wetlands: ~3100 ft² of emergent wetland created
• Riparian Vegetation: Rapid recovery, particularly of salvaged plant materials
• Spawning: five steelhead redds observed
• Rearing: 2x – 3x increase in juvenile O. mykiss abundance

Figure 4.  Tepee Creek IXL Meadows restoration  pre- (left) and post-treatment (right).

Figure 5.  Tepee IXL groundwater monitoring pre- and post-restoration groundwater elevations.

From 07-09 Solicitation -  Objective 1: Restore hydrologic function to the greatest extent possible in priority geographic areas by 2012.

Objective 2: Restore passage to meet WDFW criteria in priority geographic areas by 2008

Tepee and White Creek Fish Passage Restoration Projects  (2007 & 2008)

Introduction: The project addresses a limiting factor (passage) identified for this top geographic priority reach by the Subbasin Plan and KLESRS.  Tepee Creek, a tributary to White Creek in the Klickitat River subbasin, provides important spawning and rearing habitat for ESA-listed Middle Columbia River steelhead. The White Creek watershed as a whole is likely the most important spawning and rearing tributary watershed within the Klickitat subbasin. In recent years (2002-2007), the White Creek watershed on average accounts for 26% (0-52%) of the observed steelhead spawning in the entire Klickitat subbasin. Tepee Creek has accounted for up to 20% of the observed spawning in the Klickitat subbasin in recent years (2002-2007); however on average it accounts for 5%.  

Pre-Project Problem:  Existing conditions in the watershed currently limit steelhead production in a variety of ways.  Extensive reaches of White Creek have become incised and have intermittent flow in many places that anecdotal evidence suggests were once perennial.  Fish passage barriers also exist at a number of locations in the watershed which limit post-emergent movement by steelhead fry and juveniles and limit their survival. Historically White Creek has been subjected to stream cleaning and riparian harvest that has resulted in low in-stream LWD abundance and subsequently decreased pool frequency and volume, this has contributed to down cutting of the channel bed.  The current tribal forest management plan requires adequate levels of in stream LWD.  Failures of old culverts and road prisms have resulting changes in channel morphology.  This process was exasperated by increased peak flows caused by runoff of concentrated flows from roads.  Channels have moved and incised.  Incision of the channel subsequently results in loss of floodplain connectivity and reduced recharge of groundwater.  In meadow areas, channel incisement and bed degradation has resulted in loss of connectivity to floodplain.  Consequently the effects of peak flows are more pronounced.

Tepee Creek/175 Road (2007)

Before:

• Two 7.8' pipe arches
• 1.6' outlet drop
• Debris plugging history at inlet

After:

• 40' span bridge
• Native bed material on natural (3%) slope
• Access restored to 2.5 miles of habitat

Figure 6.  Tepee Ck and 175 rd x-ing before (left) and after culvert replacement project (right).

E.F. Tepee Creek/175 Road (2008)

Before:

• Two 6' x 66' CMPs
• barrier: water depth & velocity
• restricted debris & bedload passage

After:

• 17' 1" x 5'-6" x 47.5' box culvert
• Roughened Channel with native bed material at 2.0% slope
• Access restored to 1.0+ miles of habitat

Figure 7.  East Fork Tepee Ck and 175 rd x-ing before (left) and after culvert replacement project (right).

White Creek/IXL Road (2008)

Before:

• Two 7.8' x 51' pipe arches
• barrier: water depth & velocity
• restricted debris & bedload passage

After:

• 40' x 16' bridge
• Native bed material on natural slope (2.2%
• Access restored to 2.77 miles of habitat

Figure 8.  White Ck and IXL rd x-ing before (left) and after culvert replacement project (right).

From FY07-09 Solicitation - Objective 4: Restore or enhance 70% (by length) in-channel and riparian habitat conditions to the greatest extent possible in priority geographic areas by 2015

Swale Creek River Mile 2 Enhancement (2008)

Background:  Swale Creek is a tributary of the Klickitat River.  In 1902 a railroad grade was constructed along Swale Creek Canyon.  In many areas the railroad prism reduced the width of the channel and floodplain. This created stream channel instability during normal high water and floods as the increased stream energy along the constricted floodplain and channel tried to restore the original dimensions characteristic of the watershed hydrology. Channel expansion occurred by eroding stream banks and the railroad prism.  Eroded sediments were transported downstream and deposited along reaches of lower energy. Continual flood control activities by the railroad left the channel in a severely degraded state.

Pre-project Problem:  Aerial photo interpretation (baseline photography from 1954) indicates that beginning with the 1969 photos, riparian vegetation appears to be transient.  Based on known events and discussion with landowners regarding occurrences between the 1954 and 1969 photos, the most probable disturbances responsible for the decrease in vegetation and pool structure are the 1964-flood and subsequent flood control actions taken by the railroad.  The photo sequence suggests that the stream has crossed a geomorphic threshold and is currently in a dis-climax cycle.  Even though post-1964 riparian vegetation conditions appear to be cyclical, pool structure did not recover to pre-1964 levels.   Field indicators, hydraulic modeling, and information contained in survey records also suggest this is likely the case. 

 Dis-climax patterns are well-known to occur in association with land-use disturbances.  Such cycles tend to be initiated by large runoff events enhanced by interaction with major land-use developments that confine, divert, or otherwise alter prior water and sediment transport relationships.  Dis-climax cycles will persist until either they are treated mechanically or flood(s) of sufficient magnitude returns a system to the stable side of the threshold.  The railroad ceased operation in the late 1980s, but it became evident following the 1996 floods that the dis-climax cycle persists.  Field observations of relatively stable downstream bank full dimensions and hydraulic analysis  indicate that bank full and lower discharge floods have completed as much “natural recovery” work as is possible. In the absence of mechanical intervention, further adjustments and lateral expansion of incised channel segments will happen during large floods which generally occur infrequently. 

Design and Construction:  YKFP staff developed the design in cooperation with Interfluve, Inc.  During the design phase a total of 32 representative cross sections were surveyed along the lower 12 miles of Swale Creek.  These representative cross sections were selected using both air photos and field based observations.  The surveyed cross sections and subsequent hydraulic analysis were utilized to develop a prioritization of project sites and 30% designs for three typical treatment types .  Swale River Mile 2 was determined to be a site with high enhancement potential.   The 30% design approach reduces cost and facilitates a “fit in the field” technique.  Continuous field supervision during construction was provided to the construction contractor by KWEP staff. 

Project Goal:  To reintroduce hydraulic and habitat complexity. The project used mechanical means to enhance in-channel conditions, re-establish a healthy riparian corridor, and improve habitat quality for fish and wildlife along roughly 600 feet of Swale Creek in the vicinity of river-mile 2.0.  The work performed will effectively speed channel recovery by simulating the effects of several large floods by reconstructing the floodplain and channel to emulate natural processes that occur over a long period of time. Pools adjacent to LWD jams were excavated to depths of 3 to 5 feet and banks were pulled-back to decrease hydraulic forces on the active channel and promote floodplain regeneration.    Floodplain surfaces generated by the project will provide growing environments that should increase survival of riparian vegetation during major flood events and assist in breaking the dis-climax cycle. 

Construction:   In August and September 2008, KWEP partnered with MCFEG to construct 5 LWD jams and create adjacent pools along 600’ of Swale Creek to reintroduce hydraulic and habitat complexity.   Valley-bottom railroad construction (1902) and 90 years of subsequent operation simplified channel conditions resulting in:  

• pool frequency of 7 pools per mile (4.2% of the habitat by channel length).
• an armored bed and has a simple, plane-bed  morphology
• dis-climax conditions for riparian vegetation

  The project occurred in the vicinity of river-mile 2.0 and involved:

• excavation of 5 pools
• construction of 5 LWD jams to promote pool persistence and enhance primary habitat for salmonids

Figure 9.  Swale Creek (RM2) post enhanement.

From FY07-09 Solicitation - Objective 4: Restore or enhance 70% (by length) in-channel and riparian habitat conditions to the greatest extent possible in priority geographic areas by 2015

 Lower Klickitat Riparian Re-vegeatation  (2008)

Background:  This project addresses limiting habitat features (poor riparian and floodplain vegetation) identified for this reach, a top geographic priority, as defined by the Subbasin Plan and Klickitat Lead Entity Salmon Recovery Strategy.  This reach is a migration and rearing corridor for nearly 100% of migratory fish in the Klickitat watershed and has accounted, on average, for 10% of observed basinwide steelhead spawning. The project area occurs within a reach identified by the Klickitat Technical Advisory Group (KTAG) as fourth out of 21 priority areas within the Klickitat Lead Entity's scope.  Riparian conditions in this reach are generally poor due to a combination of 1996 flood deposits and channel encroachment by highway and railroad fill. Many of the flood deposits are well above the 2-year flood surface and at a comparable elevation to surfaces that are well-vegetated and are generally stable.  Vegetation has been very slow in colonizing these coarse, well-drained substrates. Similar deposits from flooding in 1974 along Swale Creek (a Klickitat River tributary) are still bare.   A SRFB grant sponsored by the Mid-Columbia Regional Fisheries Enhancement Group (MCRFEG) funded the implementation of this project.  KWEP provided design, construction oversight, and monitoring support for the project.   KWEP and MCRFEG collaborated to revegetate 4 Lower Klickitat River sites with over 5,000 plantings in 2006 and 2008 (Conley and Lindley 2012).

Problem: The general problem is poor riparian and floodplain conditions within the project area.  Encroachment by road fills and residual flood deposits have reduced riparian cover and LWD recruitment potential. 

 Project Goal:  The goal of this project is to increase native riparian and floodplain vegetation, woody debris recruitment, and potential for trapping fine sediment between river miles 2.6 and 18.3 of the Klickitat River.  The first round of planting was completed in 2006 on five sites totaling approximately 6.6 acres.  Plantings consisted of willow, cottonwood, and dogwood livestakes.

2009 Activity:  Following planting in 2008, YN and MCFEG staff randomly selected and uniquely marked 741 of the plantings at 4 sites to monitor survival and growth (Conley and Lindley 2012).  KWEP staff performed the year-1 monitoring in May of 2009 and was able to relocate 707 plants uniquely marked plants.  Originally, a GPS point was collected for each plant with a Trimble GeoXT and external hurricane antenna.  Each plant was also originally marked with an aluminum tag inscribed with a unique ID number. 

Depth was defined as the vertical distance from the bottom of the installed cutting or rootmass to the ground surface.  The target depth for willows and cottonwoods was 3’, but this was frequently not met due to the rock content of the substrate.  Thus, plants were originally installed across a range of depths, but have been organized into bins of 1’ depth increments for presentation purposes (Table 3).  Species such as ponderosa pine and Oregon white oak are intolerant of burial of their root-crown and were planted at- or slightly below grade.  Pines were grown in 0.8 cu-ft containers that were 14” deep and filled with planting media to within 1” of the container rim.  Thus, upon planting, the deepest roots of the pines were 13-14” below ground surface.  Oaks were grown in containers that were about 6” tall and filled within 1” of the rim.  Roots of oaks would have been approximately 5-7” below grade upon installation.

Eight species were planted in 2008 (Tables 3 and 4), of which three involved both livestakes (a.k.a. dormant hardwood cuttings) and containerized stock.  Caliper size on all livestakes was between ¼” and 3/4”.  There were three pruning treatments applied to the cottonwoods and three willow species: cut below ground surface (cut-), cut 4 bud scales above ground (cut+), and uncut (uncut).  It was hypothesized that cutting hardwood stems below ground surface might aid survival in such highly exposed sites as ours as the method of planting (hydraulic stinger) leaves a conical depression.  This effectively means that more stem is exposed to produce vegetative shoots that could desiccate livestakes in particular and reduce establishment.

 Table 3. Plant survival (%) by depth class and pruning treatments for four non-crown sensitive species (sample size).

Table 4.  Survival for four plant species which lacked depth class as a variable.

Overall plant survival averaged 64% through the first year.  This was considered very good given the exposure of the sites to wind, flooding, and solar radiation combined with coarse, well-drained substrates that had largely precluded colonization by woody plants in the 12 years since the 1996 floods.  The 97.3% survival exhibited by ponderosa pine (Table 4) was highest rate for all species and is consistent with a census of planted pines in March 2009 that relocated 762 of 858 ponderosa pines planted at sites 17.24, 22.06, and 22.68 and found 96.7% survival.

Independent of depth, containerized stock had better survival than livestakes for coyote willow.  Overall, livestakes for the species toward the more hydric end of the continuum (coyote willow, Geyer’s willow, and black cottonwood) also tended to survive better proportional with the amount of pruning.  Conversely, survival of individuals of containerized origin for the same three species was inversely proportional to the amount of pruning received.  Survival of Scouler’s willow (the most drought tolerant of the riparian hardwood species planted) was basically the same between material types, both of which tended to survive better with no or less pruning. 

Independent of pruning treatments, survival averaged 21% greater for all species and material types planted deeper than 3’.  All types exhibited a ≥10% increase in survival with greater depth with the exception of containerized Scouler’s willow (+3.9%).  Containerized cottonwood (+43.7%) and coyote willow livestakes (+33.5%) had the most dramatic overall survival increases with depth.

For the site conditions and species in this study, it appeared to be important to install plant materials at least 3’ below ground.  However, professional judgment should be exercised when extrapolating or applying these results to other watersheds and depth thresholds can be expected to vary both regionally and locally.  Regional differences are will relate to major changes in geology and climate.  Local influences on subsurface hydraulic conditions will always require the greatest consideration, particularly with regard to: subsurface hydraulic control, seasonality and duration of alluvial aquifer stage, water-holding capacity of the substrate, and floodplain cross-sectional relief. 

Additionally, desiccation will not always be the limiting condition on plant establishment as it was on these sites.  For example, where plant establishment is limited by scour, high inundation frequency and/or duration, and/or sediment deposition the concept of a threshold depth for survival may be completely irrelevant.

Within depth classes, pruning treatment relationships generally mirrored overall relationships with less or no pruning being favorable to survival of individuals of containerized origin.  Pruning of containerized cottonwoods appeared to greatly diminish survival.  Scouler’s livestakes planted less than 3’ deep showed little response to pruning treatments, though individuals greater than 3’ deep showed increasing survival inverse to the amount of pruning.  Cottonwood livestake survival increased with greater pruning in both depth classes. 

The most interesting relationship  appears to be a depth-dependent reversal in the response of coyote willow livestakes to pruning.  Coyote willow livestakes planted <3’ deep exhibited a jump in survival when pruned below mean ground level.  Conversely, those planted ≥3’ deep showed improved survival inverse to the amount of pruning.  Coyote willow is the most hydric of the species planted and it appears that reducing the potential for initial vegetative production becomes important with installations at depths expected to have more marginal subsurface hydrology (i.e. greater depth to the water table).