Judy E. Hillaby

Western Forest Products Ltd.

Mainland/Islands Region,

#118 - 1334 Island Highway,

Campbell River, B.C.

V9W 8C9

Ministry of Environment, Lands and Parks,

680 Borland Street,

Williams Lake, B.C.

V2G 4T1

• From September 30 to October 8, and intermittently to October 24th, stream surveys were conducted throughout the lower sections of the Machmell River, in order to identify and describe potential sites that may be candidates for fish habitat restoration. Potential restoration sites were focused on the lower 20 km of mainstem, and its associated tributaries.
• A field crew of 17 individuals, primarily local residents familiar with the area and conditions, gathered information on stream habitat conditions. There were also 6 sites sampled for fish, mostly in tributaries isolated from the mainstem, both within and beyond the range of anadromous fish access.
• The general pattern affecting fish habitat is that of flood debris slowly shifting the lower Machmell towards the right bank, and causing flood protection problems for the camp infrastructure. On the left bank, side channel and tributaries are drying, causing a loss of fish habitat as these become elevated. The central delta area appears to be relatively unstable and comparatively barren, in ecological terms.
• Fish habitat has also been compromised by removal of the riparian zone in some areas, and by LWD movement and instability in others. Subsequent attention to riparian zone assessment is proposed.
• There were 10 sites proposed for further development to improve and create fish habitat, and others that could be developed with further consideration and site-specific information. For the 10 specific opportunities, costs and procedures are defined for possible implementation in 1998 and beyond, for at least $ 463,000, depending on the scale of hatchery and rearing/spawning channel opportunities.

The author wishes to recognize those individuals who conducted and assisted in the field work, during often adverse field and weather conditions. Mike Wigle, Mike Edmunds, Frank Hanuse Jr., and Patrick Johnson of the Gumbooters team; Lisa Forrington, Fred Smith, and John Johnson of the Happenings team; Ardice Neudorf, Lindsay Windsor, Ted Walkus and Rick Shaw of the Bravo team; and Wayne Jacob, Ivy Hanuse, Donovan LeBlonde and Clifford Smith of the Ol’ Chums team. Thanks are also due to Mark Rourke of Coast Forest Management, as well as Troy Larden, Mike Parker and Mike Howard of the Ministry of Environment, Lands and Parks in Williams Lake, for their contribution to organization, funding and quality control.

Extensive field support was required for this project, which was ably provided by Western Forest Products, through their workers at the Machmell and Sheemahant camps. These were headed by Dale Reinholz of G.B. Logging in Machmell camp and Nikki Skarvig of Western Forest Products in Ladysmith. Aircraft support was excellent, provided mostly by Forest Industries Flying Tankers, as well as by Vancouver Island Helicopters, Sharp Wings Ltd., Quatna Air, and Timberline Air. Considerable boat transportation was also provided by individual crew members, with the support of the Oweekeno Nation, which is gratefully acknowledged.

Considerable data organization was required to compile this report. Gordon Hanuse and Wayne Jacob, of the Oweekeno Nation, put in many long hours to this end and their efforts are greatly appreciated. Ardice Neudorf of Madrone Consultants very capably completed the hydrology overview and GIS integration, and contributed to the overview research and report as well. Carolyn Deering also provided great assistance at the word processing, photo archiving and final report assembly stage.

Special thanks also goes to Bob Brand and Bill Dwyer of Western Forest Products, as well as Percy Walkus and Alex Morris of the Oweekeno Nation, for ensuring that this project was properly initiated and brought to its conclusion.

At the request of Pacific Forest Products (now Western Forest Products), Forest Renewal B.C. has provided funding to undertake a Level I Fish Habitat Assessment on the Machmell River. In August and September of 1997, a Fish Habitat Overview study was conducted, which concluded that there were several species of fish that were potentially affected by fish habitat degradation, especially in the lower 20 km of the Machmell River where logging activity has been concentrated. With the cooperation of the Oweekeno Nation and with the technical and professional assistance of other consultants, field work was initiated in October 1997.

The overview assessment outlined approximately 40 km of stream length to be considered (considering both banks of the mainstem). Field work would be directed to approximately half of that which was identified as "high" or "medium" priority assessment. The initial restoration strategy outlined the potential for:

• Opportunities to develop side channel habitats that provide shelter from destructive flood events;
• Identification of stream side areas that have been damaged through destruction of the riparian zone, bank and gully wall destabilization, or by obstruction of the confluence area; and
• Identifying areas where eroding banks and upslope stability may affect mainstem and immediate peripheral fish habitat.

To provide directed field information that refined the opportunities identified in the overview assessment, a Level I Fish Habitat Assessment was conducted. Its general objectives were:

• to confirm or revise the identification of the nature, location, extent and severity of forest harvest impacts on fish habitat
• to provide sufficient information to identify and prioritize restoration options, and to identify initial project objectives and scope
• to identify the need for any level 2 assessments
• to prepare initial budgets and schedules for restoration projects.

In this case, the field studies focused on:

• walking creeks (one crew on each side, in the lower reaches of the mainstem) and up the immediate tributaries for short distances, to assess the quality of in-stream habitats, paying close attention to the condition of off-channel habitats, and identifying the water flow control points.
• Additional data collected included temperature, discharge, turbidity and bank material, specifically directed at off-channel habitats.
• Reach descriptions in the mainstem were on a subsampling basis, determined by a projected length of stream(s) to walk each day. Stream walking will record the habitat unit, chain distance, features, and habitat units. During the stream walk, a representative subsample length was chosen for more complete measurements, according to the criteria outlined in Technical Circular No. 8.

Consistent with the broader objectives of Forest Renewal B.C., it was also important to develop a closer working relationship between workers at the Oweekeno Nation and the forest licensees. Accordingly, this project provided the means to capture the knowledge and perspective of both local residents and forest industry workers.

The Machmell River is a glacially fed fourth order stream with a magnitude of 146. Peak flows are influenced by a high degree of seasonal streams (approximately 1,252 ephemeral streams were noted). Elevation ranges from just above sea level near the mouth of the river to 7600m at Mt. Buzza. The study area for this project extends to the Pashleth River; approximately 22 kilometers upstream. Upper reaches of the study area are characterized by sheer bedrock walls that form a canyon which runs for approximately three quarters of a kilometer. The morphology of the Machmell River in this area is dominated by cascades with a vegetated island at the confluence of the Pashleth River. Although the Pashleth joins the Machmell at the bottom of the canyon, most other tributaries generally drain from the hillsides onto a terraced platform and then plummet over the bedrock cliffs to the mainstem below. Downstream of the canyon, bedrock walls still dominate the landscape with a few areas of steep but sloping hills. The river’s width increases and riffle morphology dominates the channel with intermittent vegetated and unvegetated islands. From the mouth of the Machmell to approximately seven kilometers upstream, the area is characterized by a fluvial fan and flood plain. The lower reaches of this area also encompasses one of the major tributaries in the study area the Neechanz. Immediately to the northeast, the Genesee Creek/Walkus Lake watershed also shares the delta and the two systems have been periodically connected by a flood channel.

The Machmell River is found in the Coastal Western Hemlock zone which is comprised of three biogeoclimatic subzones. The Central Moist Submaritime Coastal Western Hemlock Variant (CWHms2) subzone occurs in the lower elevations (below 700m) of the Machmell River. The climate in this area is characterized by moist, cool winters and cool but relatively dry summers. The vegetation in this area is dominated by mountain hemlock, Douglas-fir, western red cedar and amabilis fir. Common understorey species include Alaskan blueberry and a well developed moss layer. At elevation ranges from 700 to 1000m, the Montane Wet Submaritime Coastal Western Hemlock Variant (CWHws2) occurs. This subzone is characterized by moist, warm summers and cool to cold winters with relatively heavy amounts of wet snowfall. Large glaciers at the head of this drainage produce cold air that can influence the climate. The Winward Moist Maritime Mountain Hemlock Variant (MHmm1) is found at upper elevations. Climate at this subzone has long, wet cold winters and short cool moist summers. Total snowfall is usually high, which can result in substantial snowpacks that may persist well into July. Forests in this site are usually dominated by amabilis fir and mountain hemlock. Vegetation and stand characteristics are strongly influenced by local topography, which can in turn affect the timing and pattern of snowmelt.

From October 10^th to the 18^th, flow discharge and rainfall precipitation was measured and the results were recorded as indicated in Figure 1. Discharge was measured at the Machmell bridge (approximately 7km upstream) by calculating a cross section of the river and using a timed floating object over a predetermined length. These calculations were then multiplied by 0.75 to compensate for the increase in velocity found at the surface. There is a lapse in the data from October 15^th - 16^th due to the staff gauge being swept away during a peak flow period. Rainfall was collected from Clear Creek using an AES standard precipitation gauge. These data, although inconclusive, suggest that the Machmell River responds rapidly to a rainfall event.

There are no historical data for this area to demonstrate changes in peak flow as a result of clear-cutting, but studies of similar watersheds have shown this relationship to be true (Duncun, 1986, Garbrecht, 1991, Jones, 1994). Any change in the flow patterns are a concern to the fish population as it can lead to scouring of spawning habitat, an increase in sedimentation and degradation of rearing areas. In the upper Machmell there are few pools and off-channel habitat, although those that do exist are used by salmonids. Significant increases in flow can therefore severely limit fish populations by degrading the few areas that are available. Further studies should be initiated to determine to what degree the fish stocks of the Machmell River are being affected by hydrological impacts.

The mainstem Machmell River was not walked extensively, due to its large size and flood potential. Field studies were directed towards associated side channels, tributaries and back channels that may provide fish habitat, as well as to low-gradient tributaries. As defined in the overview assessment, surveys were primarily limited to the area just downstream of the Pashleth Creek confluence, to the top of Reach 5. During the field work, it was apparent that many areas of the mainstem would remain inaccessible and during high flood events, crews were directed into safer areas.

The Genesee Creek/Walkus Lake watershed shares a delta with the Machmell and is included in the study area. Reach 1 of Genesee Creek refers to a 1 km section that provides important sockeye spawning beds and grizzly bear refuge habitat. Its apparent dewatering is causing significant concern to DFO and to the Oweekeno Nation. Field surveys in this study were intended only to describe site-specific conditions to complement existing knowledge gained by other parties. Accordingly, a comprehensive, watershed-based approach was not conducted in this drainage.

The access approach was to walk up and assess the mainstem Machmell River, wherever possible, and to locate and assess the significant tributaries and side channels encountered. Both sides of the river were walked in Reaches 1 and 2, and spot-checked in Reach 3. Many off-channel habitats and tributaries were located in this manner. Three and four-member crews were dropped in by helicopter, and truck access was also used in the lower reaches. Where the access plan permitted, stream walking was normally conducted in an upstream manner. Some areas remained inaccessible, due to inaccessible canyons (Machmell Reaches 4 and 5; Pashleth Reaches 1, 2 and 3).

Crews gathered site-specific information on both tributary and mainstem habitats, focusing on off-channel and tributary habitats. Reach descriptions in the mainstem were on a subsampling basis, determined by a projected length of stream to walk each day. During the stream survey, subsampling by habitat type (1 in 3 riffles, 1 in 2 glides, all pools) was conducted, according to the criteria outlined in Technical Circular No. 8. Field data recorded was consistent with the FHAP Level I data form, but modified to include more detailed temperature and discharge measurements, as well as stream bank characteristics. Crew members also recorded a stream habitat log, correlated with hip chain measurements, and an accompanying diagram that explained more complex stream configurations and habitats. A photo log was also kept, and all photos were labeled and catalogued for archive purposes.

All crews had 1:20,000 Trim maps, compasses, and aerial photos, and were carefully oriented by truck or helicopter before beginning their surveys. Some crew members had GPS equipment, and these were used for positioning where appropriate. Crews walked from known positions on the map, charting their location by hip chain distances, assisted by laser rangefinder distancing where necessary and appropriate. The progress of each crew was charted on master maps after each day’s work.

Fish capture was conducted primarily in off-channel habitats. As many habitats as possible were sampled, with consideration for the cost-effectiveness of access into difficult locations.

Minnow traps were usually set for 24 hours, but occasionally for 5 hours if it was cost-effective to do so. Electrofishing was not conducted in this system, due mostly to weather constraints, but also as minnow trapping was more effective for the locations sampled (beaver ponds and back channels).

Species identification was frequently checked by the crew supervisor. Microhabitat locations and characteristics for reach trap site were recorded, species and life stage was determined for all fish captured, and fish length recorded to the nearest 5mm for all salmonids.

The first line of quality control was in coordinating field traverses and supervision of all crew members. There was a short training session prior to the start of the field program, when habitat recognition and measurement of stream characteristics was repeatedly practiced and data recorded. At the start of the field program, extra supervisory assistance was given by a roving biologist, who traveled at least one day with each team to answer questions and reinforce field measurements and strategies. Data sheets, field notes, equipment anomalies, etc. were checked daily by each team’s leader and by the senior biologist. Location and extent of field surveys were charted daily and field information generally reconciled with map information. Field surveys were structured so that crews did not overlap, and survey start and end points were flagged in the field.

In the field, every fifth day was a "data day" when crews reviewed their work, assembled data in different formats, completed photo records, checked equipment operation, sent film out for processing, etc. The field information was reviewed and discussed with the project biologist, and checked for anomalies and significant findings.

Since most of the field survey locations were accessed only by helicopter, it was not cost-effective to re-survey the same area by a different crew for quality control purposes. Instead, photo records were kept that were used to re-check interpretations of different habitat types and correlate general observations.

Quality control concern was appropriately directed towards flood events. The Machmell watershed is subject to extreme flood events which sometimes obscured stream habitat units. This system responds very quickly to rainfall events, such that a deluge in the morning can raise the river level as much as 2m by late afternoon (Figure 1). Accordingly, depth and surface velocity measurements were taken at the lower Machmell bridge and used to derive discharge, which is then used to qualify the data record. It is not appropriate to precisely correlate measured stream depth, etc. with flood records without more detailed analysis. However, the flood events provided valuable information on the fluvial processes affecting stream habitat, which is a subject in itself.

When high rain events caused the river to rise, field crews were directed towards tributary habitats that were more visible at this time. Mainstem surveys were not appropriate when the river was rising, primarily due to the physical difficulties of negotiating the banks of this system, but also due to the bear population which moved off the gravel bars and began moving through the roads and high-level riparian trails, as they became unable to fish.

The data has been assembled and calculated in two documents. The Volume I - Summary contains summarized data and proposed restoration opportunities. Volume II - Data Appendices contain reproductions of the field sheets, sketch maps of certain areas, rough field notes, as well as detailed fish capture results.

In this volume, Form 4, Level I Habitat Survey Data Form, has been summarized for discussion purposes. The field reference code corresponds to an information bundle that is detailed in Volume II and located on the enclosed project maps. The reader is directed to this source for site-specific details.

Summarized data has been assembled as follows:

Field surveys were directed first towards locating unmapped tributaries, sidechannels and other off-channel habitats. A secondary objective was to assess the condition of existing fish habitats in the mainstem where possible. Extensive traverses across the flood plain were avoided due to potential bear hazard.

The Machmell is a large river system and flooding in the mainstem is extreme (see Figure 2). Whether or not logging has affected this pattern must be addressed separately. It is clear that there are significant indicators of channel disturbance, including parallel debris jams, elevated mid-channel bars, eroding banks, etc. The bedrock constriction under the bridge at Reach 2-3 has a "nozzle" effect on stream flow and most of the channel disturbance is downstream of this point.

Generally, the mainstem channel is moving towards the right bank, to the extent that TRIM maps and 1991 airphotos are profoundly inaccurate. This movement is causing the left bank to dry out, as evidenced by encroaching vegetation and gravel compaction, shallow vegetated tributaries, and elevated sidechannel areas. The right bank is characterized by eroding, unstable banks, and new sidechannel formation.

Attempts to control flooding of the camp infrastructure have been made by construction of berms set back from the mainstem (right bank) and riprapping along the road. These have also contributed to channel formation patterns.

Besides the presence of some spawning area in the mainstem channel, fish habitat was limited to a few stable, off-channel areas:

• Intrusions from Oweekeno Lake, which appear to be old flood channels that have been cut off. These are not now stream habitats, but presumably were at one time.
• Stable sidechannel areas on the left bank (Reaches 1-1 to 1-4) that are in danger of drying out from channel movement away from this course;
• Tributaries at 5 km and 6 km on the right bank that have significant blockages at their confluence with the mainstem;
• Tributaries #1 and 2 on the left bank, flowing through and adjacent to an old cut block on the flood plain;
• Lower, confluence areas of steep slope tributaries, particularly Clear Creek on the left bank below the bridge crossing.

These two channels are to the north of the Machmell River mouth, adjacent to the Machmell camp. They appear to be flood channels that were cut off in the upstream end and are now fingers of lake intrusion. Field traverses found the first channel just northeast of the Machmell confluence (only 71m long). The second is farther along the lakeshore and extends 600m inland (see figure 5). Only the second channel has stream survey data.

• They are long slough habitats that are groundwater and lake-fed.
• The first channel extends 71m in from the lake. The second extends inland about 600m. Both provide good rearing habitat, but no spawning gravel is evident.
• LWD is concentrated at the mouths of the sloughs, and along the lake shore. There is virtually none upstream.

The mainstem Machmell River in these sections is confined between high bedrock walls, but has no fish obstructions until Reach 4. The shifting, shallow nature of the river bed also makes jet boat access impractical and unreliable (tested during the field survey). However, during low water conditions some of the mainstem bars were accessible by helicopter. Accordingly, limited surveys were carried out in the mainstem.

Field crews searched for and surveyed any available off-channel habitats that may be accessible to mainstem fish downstream of the Reach 4 obstructions. These areas would specifically include the mouths of the tributaries falling off the canyon walls. Any associated sidechannels or backchannels were to be surveyed and measurements taken that would help assess their condition. Measurements of the mainstem channel were also taken, where possible.

Due to the steep canyon walls, forest harvesting is nowhere near the mainstem channel. However, some roads and cut blocks have affected the banks and slopes of the steep tributaries and gullies on the left bank. While road work has stabilized most of these, the effect on fish habitats at the base of the cliff is important. A comparison of left bank (logged) and right bank (unlogged) shows that the left bank has had numerous slides, both new and old, that have inundated sidechannels and in-filled stable, complex stream habitats.

Nearly all of the tributary streams described on the TRIM maps do not provide fish habitat, in that they have in-stream gradients well over 20%. As expected, there is a very limited amount of fish habitat available:

• Mainstem Machmell River, Reaches 2-4 to 4-1. This area contains spawning salmon, probably some of the larger juveniles in selected areas, and would also briefly contain young of the year fry from mainstem spawners (in the spring and early summer).
• Off-channel habitats on the right bank (unlogged). These areas are formed by tributaries flowing off the steep slopes above, and serve as complex, clear, slow-water, nutrient capture zones that serves rearing juveniles well;
• Off-channel habitats on the left bank (logged). These areas differ from the right bank in that they have had some amount of debris torrenting over the years, which has altered the confluence habitat at the Machmell confluence.

Six different mainstem areas could be accessed by field crews, and five of these yielded some evidence of off-channel habitat. Reaches 3-4 to 5 were not accessible in the mainstem. The right bank has not been logged and these streams should be considered control systems, in their pristine condition (except for BR-01, Reach 2-4). Refer to Volume II, Appendices, for site sketches and field notes for further information on individual areas. Note that only the confluence area where off-channel habitat exists was surveyed: the upper portions of the streams are very steep and do not provide fish habitat beyond the canyon floor.

Note that the 1,089m of off-channel habitat walked and surveyed represents most of it on this section of the right bank. There are no low-gradient tributaries off this bank, and none that allow more than a few meters of fish access from the mainstem.

• Off-channel habitat just above the bridge in Reach 2-4 (BR-01) is mainstem-influenced. It provides poor substrate (sand) and is on the edge of a mainstem flood plain at the bottom of a seepage slope. An adjacent beaver dam functions as a nutrient sink but is probably not accessible to juvenile fish. There is limited juvenile rearing habitat and little or no spawning habitat.
• Off-channel habitat in Reach 2-5 (BR-02) provides complex and excellent fish habitat with clear water, clean gravel and good cover from overhanging vegetation. It has beaver dams and pools that are fed by upslope tributaries and drops gently to the mainstem to enter in a sheltered sidechannel area.
• Off-channel habitat in Reach 2-6 (BR-03) is a mainstem-influenced sidechannel with little cover, behind a cobble bar. Bedrock blocks and a scour log help form a good adult holding pool, and helped to form a more complex riffle, pool and glide sequence than is otherwise found in the mainstem.
• Off-channel habitat in Reach 2-7 (BR-07) is also stable sidechannel behind a vegetated island. It is primarily mainstem-influenced, with a slight contribution from one tributary. This sidechannel provides a sheltered spawning area and some low-gradient rearing area in the margins where the bedrock and LWD have allowed riffle-pool sequences to form. It has poor in-stream cover (no photos).
• Off-channel habitat in Reach 3-1 (BR-04) was surveyed for only a short distance but field notes and site diagrams indicate more extensive and complex fish habitats. This area consists of a slough about 70m downstream of a high waterfall off the bank. Both good spawning gravel and rearing habitats are present, and all are associated with beaver dams on at least 2 channels (no photos).

Seven separate survey areas yielded six different off-channel habitats, representing most of the off-channel habitat in this section. Each of these areas has some influence from the steep-slope tributaries flowing off the left bank. There is some mainstem fish access into these tributaries, but only within the canyon walls; further up, the stream gradient is too steep to support resident fish.

Some of the tributaries flow through old cut blocks (9.8 km and 14.5 km Creeks) and all tributaries are crossed by the mainline road on the slope above. Steep cliffs and slope stability issues have dictated road design and engineering throughout the study area and are not specifically addressed in this work.

Impacts on fish at the bottom of the cliffs was the primary focus. One limited survey was one on 14.5 km Creek to assess potential sources of bedload material and wood debris.

Note that off-channel habitats can be difficult to describe in linear terms. Please refer to the field notes and sketches in the Appendices (Volume II) for further information on specific sites.

• Off-channel habitat at Reach 2-6 (HP-01) consists of two creeks parallel to the mainstem that are influenced by both mainstem and upslope flows. ("249m" and "49m" creeks). Both streams contain attractive juvenile rearing habitat (many juvenile fish were observed) and are characterized by buffered flows, opportunities for nutrient capture as well as a good cover from overhanging vegetation and SWD. However, this area is downstream from a cut block and a creek crossing the road well upslope ("9.8 km" Creek) has had some bank erosion. In the fish habitat below, the substrate consists of cobble and gravel overlain with a layer of silt, indicating a loss of spawning habitat. There are two separate slides, each of which has connected directly with one of the two streams.
• Off-channel habitat in Reach 2-7 (HP-02) is an isolated backchannel area that is at the base of a slide and adjacent flooded area below a falls. The flooded area of 30m x 170m and up to 2m deep in standing timber at the base of the bedrock cliffs. There are two short seepage channels and one longer, isolated backchannel surveyed) for outflows. The backchannel is a low gradient mix of pool and glide habitat over fine sediment, with good in-stream cover. It offers some juvenile rearing habitat but no spawning and has good access to the mainstem, but is not connected to the flooded area.
• Off-channel habitat in Reach 3-2 (HP-04) is a 285m long sidechannel beside a vegetated cobble bar. It has limited quantities of spawning gravel (mostly cobble) and a gradient up to 5%. In-stream cover is from boulders and cutbanks, and from one large (56m) holding pool situated in the centre. It is probably most valuable as holding habitat for spawning adults.
• Some off-channel habitat in Reach 3-2 (BR-10) is at the base of the falls, below "14.5 km" Creek. This is a short section of stream habitat that is at least partly accessible to mainstem fish. From 0 - 30m there is good spawning habitat and fish can access to at least 50m, where gradients rise to 17%. Spawning sockeye were seen in a 1m pool in a sidechannel about 100m downstream of this tributary confluence. There has been some bedload movement over the falls, as evidenced by the LWD and sediment wedges. Some juvenile rearing habitat is also provided by the LWD complexing in the lower 30m (no photos).
• The upper part of "14.5 km" Creek (BR-11) indicates the source of much of the bedload material. It flows through a cut block that has a very narrow riparian zone and blowdown throughout the 400m surveyed section. Disturbance indicators are sediment wedges, eroding banks, braiding and dry channels. Large quantities of LWD are causing obstructions and changing channel morphology in a fairly high gradient zone (15 - 24%) (no photos).
• Off-channel habitat in Reach 3-3 (BR-05) consists of cascade-pool steps in two channels at the bottom of a large waterfall. Some spawning habitat is provided in one of the pools, but is above a 2m falls and is probably not accessible to fish. This habitat is steep (to 15%), bouldery, has very little in-stream cover, and provides little or no fish habitat. Slides, LWD jams and other disturbance indicators are not apparent.

There are several low-gradient tributaries to the Machmell and Pashleth River that flow through cut blocks above the canyons, where Dolly Varden are recorded. None of these tributaries are accessible to mainstem fish; resident fish populations are clearly isolated. Road access was available, and field crews surveyed sections of 3 creeks, named here as 20 km (Susan) Creek, 22 km Creek and Trib. #2 flowing into Pashleth Creek at the Reach 2/Reach 3 break. Fish capture efforts were directed towards establishing the distribution of fish in these tributaries, and in determining species mix.

All of these tributaries contained isolated populations of Dolly Varden, some living in very restricted habitats. Large portions of 20 km Creek, and 22 km Creek have been harvested to the stream edge with a consequent loss of riparian function. Slope stability has also compromised fish habitat in 20 km Creek. The most significant effect of forest harvesting is that there are significant in-stream barriers that are associated with LWD jams and debris piles that likely inhibit fish movement throughout the mainstem. Considering that these are relatively small creeks with small populations of fish that have very little ability to avoid unfavorable habitats, this effect is significant. Trib #2 flowing into the Pashleth system is an interesting control system, with mature coniferous forest, abundant LWD, alternating pool/riffle habitats, and relatively shallow water throughout its 1.1 km surveyed length.

Three streams were found that provide low-gradient fish habitat in the area of the Machmell - Pashleth confluence. Two streams flow through cut blocks, and all contain isolated populations of Dolly Varden. One stream (Pashleth Trib. #2) is in its pristine, unlogged condition. Another stream (22km Creek) has been logged to the stream edge and runs parallel to the road, but the stream channel remains stable, with good cover, and a rebuilding riparian zone. The third stream (20 km Creek) has significant stretches of eroding banks and other disturbance indicators and was surveyed in three separate locations.

• 20 km Creek has two forks. The left fork (HP-03) flows from the only non-glacial lake in the watershed ("One-eye Lake", inaccessible) over a high, bedrock falls and meanders downstream to join the right fork, just above the mainline road crossing, before plummeting over the cliff and into the mainstem Machmell in Reach 4-2. The right fork has less water (GB-1, 2) and meanders through mid-low gradient bench land before it joins the left fork. Both forks flow through previously-logged areas and both contain isolated populations of Dolly Varden where gradient permits.
• The left fork below the falls has many, significant channel disturbance indicators, including at least two zones of major eroding banks, at least two newly-formed log jams, sediment wedges, parallel debris, etc. High turbidity (.3m) was also noted in this section. Fish habitat seems limited firstly by upslope stability and secondly by future LWD supply.
• The right fork contains clear water and has some dewatered sections noted in the lower survey area (GB-02). Further upstream, disturbance indicators are lacking, but LWD is low and the canopy is mainly open.
• 22 km Creek (GB-03) also contains Dolly Varden and has been logged to the stream edge. It runs parallel to the mainline road before going over the cliff and into the mainstem Machmell in Reach 5. It provides good instream cover from LWD, no instream barriers, clear water and some canopy. Some eroding banks are noted and patches of spawning gravel mixed with fine substrates indicate that there may be some substrate degradation associated with this erosion.
• The "Pashleth tributaries" consist of three streams that join at the top of a high bedrock falls to flow into the mainstem Pashleth at the Reach P2/P3 break. One major tributary (Trib "0"), on the edge of a cut block, is too steep to provide fish habitat and was not surveyed. The second creek (Trib "1") was not logged and provides a short section of fish habitat, but is also much more ephemeral and contains Dolly Varden. The third stream (Trib "2) is a significant stream that flows through pristine, low-gradient forest land and contains Dolly Varden for at least 1.1 km of its length. It was briefly surveyed to provide some control information on the pre-logging condition for the Pashleth bench streams.
• Trib "2" is predominantly a sequence of riffle-pools, formed around LWD scour logs, although only 5% of the habitat was clearly identified as pools. The depth is shallow (to .65m) and goes to > 5m bankfull width in places. There is abundant LWD and a dense canopy of mature coniferous forest surrounding the stream. There are some associated wetlands and small tributaries that would provide a limited amount of off-channel habitat. Deep pools and similar overwintering habitats are lacking in the surveyed section. Fish were observed to 1.1 km upstream of the falls (end of the survey) and probably distribute further upstream from this point.

Previous information (FISS) indicated that cutthroat trout are found in Walkus Lake, so it was assumed that fish habitat exists in all downstream areas. With this in mind, there were three areas of interest to this study:

• Reach 1, where large numbers of spawning salmon are found. It is suspected that the channel has had a reduced flow in recent years and a level of field confirmation was required;
• Reach 3, where the stream passes through a cut block just above the falls (Reach 2). Field access is poor, as the roads are deactivated and no helicopter landing sites are in the immediate vicinity. However, landing upstream and a long hike downstream may yield a limited amount of survey information.
• Reach 5, at the Walkus Lake outlet on the southeast side. This section is too steep to provide fish habitat, but is of interest in terms of its connectivity with Walkus Lake and the Machmell River.

Reach 1 was walked in late September by a team of DFO and MOF workers. The object of the walk was to determine:

• if the Genesee was indeed drying out
• if sockeye spawning success was affected, and
• what could be done about it, if anything.

The outcome was that while some channel drying was evident, the effect on sockeye spawn was argumentative. Hydraulic sampling was proposed as a way of quantifying sockeye egg-to-fry survival rates in various stream areas. Assuming spawning habitat damage, there were three potential solutions proposed to increase flow into the channel (in order of "practical" priorities):

1. Consolidate the LWD and transient bedload at the base of the falls to direct the flow down the main channel;
1. Construct flow control works on the southeast end of Walkus Lake that would divert more water out the NW outlet;
1. Excavate an overflow channel that would restore an historic connection between the mainstem Machmell and the Genesee.

In this study, two field surveys were conducted:

1. At the base of the falls at the top of Reach 1. The purpose of this was to inspect the habitat condition at this point and compare it with similar locations in other steep slope tributaries that have been logged or unlogged;
1. In Reach 5, a traverse was made from Walkus Lake downstream to the Machmell River in order to inspect the lake outlet habitat, and view the condition of the stream itself.

Reach 3 was not field surveyed due to an inappropriate risk factor. Aerial surveys indicated no significant slope movement in the stream itself, and planted riparian zones and cut slopes seemed to be growing well. It is generally inaccessible for potential restoration purposes.

As noted above, surveyed areas were directed towards stream sections where critical information was needed. Fish habitat existing throughout Reaches 1 to 6; however, there is no fish habitat on the southeast lake side, as a stream connection is lacking. This is contrary to the TRIM map (1:20,000 scale) information.

This survey consisted of a few brief measurements and was cut short by bear activity. Photos and brief field notes indicate a disturbed area with piles of LWD forming a split channel that only partly re-consolidates back to the main channel. The age of the LWD seems to be approximately the age of the cut block above the falls. Some of the sidechannels contains good spawning gravel and sockeye spawners, and show signs of stabilizing (BR-12).

There is a dry stretch of 400m before a soggy area forms that eventually becomes a defined creek. The general stream character over the slope is that abundant and stable LWD is positioned in the stream, which actively retains bedload material and complexes stream habitats. It is much too steep to provide fish habitat and is important to fisheries resources in terms of slope stabilization and nutrient input (BR-06).

Most of the tributaries in this watershed are too steep to provide fish habitat. The lack of productive lakes (except for Walkus Lake) prohibits fish distribution in most high-gradient areas. Refer also to the accompanying maps that better illustrate fish distribution, watershed relief and contours.

Several significant obstructions occur in the Machmell canyon. The first of these is at the Reach 4, section 1/2 break that is a series of cascades thought to be impassable to all migrating fish. There are other cascades (Figure 23) just upstream at the Pashleth/Machmell confluence that would affect fish in both systems, and shortly (less than 1 km) upstream in both rivers. These are all at the bottom of a steep, narrow canyon and cannot be better assessed for fish passage. For our purposes, and in the absence of any evidence to the contrary, the cascades in Reach 4 are assumed to represent the limit of anadromous fish access. These obstructions are consistent with historic information provided by DFO and recorded in the FISS database.

Off-channel habitat to the mainstem was generally not sampled, as fish were frequently seen in these areas and fish access was assumed. In the lower Machmell, 5 km Creek on the right bank was minnow-trapped, which revealed that coho juveniles were able to penetrate this creek upstream of the beaver dam at the outlet (SS016). The tributary through the cut block on the left bank (entering at Reach 1-1, field reference OC-04), was not sampled and fish were seen in the upper reaches.

Despite the high bedrock cliffs at the stream mouths, FISS indicated that Dolly Varden were found in 20 km Creek. This was confirmed by minnow trapping (SS005, SS006) and local knowledge. Only one species was found, and the upper limit was not definitively found. New information can also be added to FISS, since Dolly Varden were also found in the other area tributaries (SS007 in 22 km Creek, SS001 in Pashleth Trib #1, and SS002 in Pashleth Trib #2). The upper limit to fish distribution was not identified and fish are suspected to inhabit the stream to the toe of the slope.

The water level in this system rose and fell significantly during the study, according to rainfall patterns (see Hydrology section). The effect of this during the study was primarily that mainstem habitats became inaccessible, for either survey or tributary access purposes. Due to the volatility of the Machmell watershed, fish habitat assessment was directed towards off-channel and tributary habitat, where fish and logging-related impacts were concentrated.

LWD data is subject to great interpretation by individual reviewers, especially so in rivers where large fluctuations in flow cause LWD to be high up on the bankfull area, or in large jams, so that its functionality and stability is questionable. To arrive at the habitat condition judgments in this case, photo records were used extensively.

The reader is reminded that judgments are in general categories, rather than precise calculations from individual data sets. Several forms of data were used to arrive at these judgments, including field data measurements, subjective field notes and sketches, as well as ground and aerial photos. Refer to the appendix document for further details on specific areas.

Tech. Circular #8 does not provide habitat evaluation standards for streams with 30 - 100m bankfull widths, as the Machmell is. For arriving at the specific ratings required for Form 6 (Habitat Diagnosis Summary Form), the small stream standards were used, with some modifications for pool habitats (eg. some glide habitats were considered rearing pools if this was supported by field notes and diagrams). Appendix E of Tech. Circular #8 was also used to provide guidance for the more general fish habitat evaluation. Fish habitat evaluation data is summarized in Form 6, located in the appendices.

This study assumed that salmonid production is limited by freshwater habitat conditions; in fact, this is not the case for sockeye salmon, at least. Directed commercial fisheries and changing ocean conditions can quickly increase or decrease the number of returning adults, so that fish habitat only appears limiting during years of high returns. While valley contours suggest obvious limits to the quantity of fish habitat, the value of that habitat may be underestimated for lack of population recruitment to colonize it.

The Overview Fish Habitat Assessment focused field studies on off-channel habitats, expecting that these were the most limited habitat types, rather than certain areas of the mainstem. However, fish move back and forth from the main channel as peripheral habitats become available (flow pattern, habitat alteration, etc.). Accordingly, some perspective on the value of the mainstem habitat is also warranted.

The distribution of adult salmon in the mainstem Machmell has not changed as a result of logging. Most of the tributaries have an obvious barrier not far from the mainstem and are unavailable to migrating fish. The long glide habitats and braided areas in the mainstem contain slow-water sections and adult holding pools are probably not limiting.

In the lower Machmell, adult migration into the low-gradient tributaries and some of the sidechannels is becoming restricted as a result of shifting of the mainstem channel. On the left bank, several peripheral areas are drying (oxbow and sidechannels; tributaries 1 and 2) and have either no or limited access to spawning fish. On the right bank, a combination of beaver dams and culvert crossings provide only partial fish access.

Extremely turbid, glacial conditions make fish observations in the mainstem difficult; however, all fish seen were located in the stable, marginal habitats on the edges of the mainstem. It is possible that if the channel instability in the mainstem is accelerating this would reduce the quantity of stable sockeye spawning habitat. Up to 80,000 sockeye are recorded spawning in the Machmell River (1982), dropping to 3,000 in 1996. This decline is thought to be related to ocean conditions, since all sockeye stocks in Rivers Inlet (and Smith Inlet, to the south) have experienced similar declines, in both logged and pristine watersheds. Accordingly, while spawning habitat does not appear limiting at this time, this can quickly change.

Dewatering in Genesee Creek is evident, although it was not fully examined in this survey. Since this system is an important sockeye spawning area it is likely that there has been some incubation loss as well as an increase in mortality from bear predation. Debris buildup at the base of the falls has probably contributed to this, which is consistent with observations elsewhere in the study area. Personal communication with other field staff indicates that there is no evidence of substrate degradation in the Genesee.

There is some evidence of spawning habitat degradation in some sections of 20 km and 22 km Creeks on the Pashleth bench. While it does not appear extensive, the fish populations in these creeks are confined to a very limited habitat and the relative impact of even a small gravel loss is significant.

The mainstem Machmell is a large river and characteristically does not contain much juvenile rearing habitat. Periodic flooding would wash out juveniles into refuge areas in off-channel habitats. LWD is not retained until it reaches the lower flood plain. There was no evidence of mainstem rearing of juvenile fish in this survey.

Off-channel habitat to the mainstem is limited in the Machmell system, due to the high canyon walls in the upper part of the study area, and the extreme flooding and channel instability in the lower. Juvenile coho were found in nearly all of the off-channel habitats that were surveyed.

Turbidity is high in the mainstem and while this limits fish observations, it also limits photosynthesis. Clear water, off-channel habitats are very limited and most contained juvenile coho (visually observed).

Comparison with the right bank indicates that there has been a loss of off-channel overwintering habitats on the left bank at the bottom of the canyon, due to upslope instability. Pockets of habitat have been lost at the bottom of 9.8 km Creek, 14 km Creek, and Clear Creek. This is likely an underestimate as other areas were not accessible to the survey crew.

There has been a loss of riparian habitat in some areas, which has an indirect effect on fish habitat, especially downstream of the lower bridge, where channel shifting has eroded the banks. This requires a larger-scale approach to channel stabilization, so that marginal habitats to the mainstem can revegetate enough to provide some rearing capacity.

Riparian zone stability is also a threat to valuable, isolated populations of Dolly Varden high up on the Pashleth bench. Tributaries to the Pashleth (20 km and 22 km Creeks, in particular), and others on the left bank of the Machmell canyon, have been logged to the stream edge. In 20 km Creek, sediment wedges, sloughing banks and numerous partial in-stream barriers probably limit fish survival. However, 22 km Creek has since stabilized, regenerated, and has abundant LWD and pool habitat and instream cover. It is possible that removal of the conifers in this case may have increased juvenile rearing potential.

The lower Machmell tributaries on the left bank have also lost some degree of their riparian function. They have been dewatered and require an improved riparian habitat to provide a long-term supply of LWD that would contribute to stream complexing.

Pool rearing areas has been reduced in some areas of 20 km Creek, also due to bank sloughing and upslope instability. The left bank tributaries through the cut blocks in the lower Machmell also have reduced pool areas; however, the cause of this is apparently due to the shifting main channel that has caused dewatering - without more flow, stream channels have no opportunity to form scour pools or stable, complex habitat.

These are not in order of priority. Since the Machmell River contains many species of fish, diverse habitat types and fragile populations, all the following are important.

It should be noted that the restoration opportunities described here are on the "short list". There are others that are not presented for lack of appropriate data support, locational problems, lack of techniques to apply, etc. It is important to recognize that some of these are well worthwhile and may become more fully developed as restoration proceeds, and workers are able to better define and elucidate restoration techniques for specific fish habitat scenarios. These proposed projects are a starting point for further work, and provide a profile of the kinds of activities that may be pursued in this watershed.

This study identified areas where eroding banks contributed to fish habitat degradation, either at the bank site, or further downstream. The Machmell River is a typical coastal watershed which is prone to upslope instability and forestry engineering must always take this into account. While slope stability is always a concern, the stability of stream banks, particularly through steep areas (gullies, rather than fish-bearing streams) is a different problem.

Re-stabilizing stream banks is a job for specialists. While costs for Level II assessments can be forecast, restoration costs cannot. A watershed-level Riparian Assessment is also underway which would help to focus these concerns more clearly.

It is recognized that upslope stability and stream bank stability are the responsibilities of MOF and MOELP, respectively, and to a certain extent the responsibility of the Licensee. In steep watersheds such as the Machmell, those responsibilities should be distinguished on a site-specific basis so that other funding decisions can be make. The following sections include provision for specialist expertise, which should allow restoration work to continue in a comprehensive manner.

Field Reference: M-01

Location: Reach 2, Clear Creek.

Impact: Trees have been removed to the edge of the stream, in some areas as a result of both harvesting and blowdown. Bank destabilization has occurred and both logs and bedload material have found their way over the falls and into fish habitat downstream. Rehabilitation of this habitat must begin with bank stabilization upstream.

Objectives: To keep banks from further sloughing into Clear Creek, and to allow rebuilding of the riparian zone where possible.

Priority: High

Work Plan Considerations: A site-level riparian assessment is required to address opportunities for bank stabilization. This is separate to silviculture in the block itself. Additional work is required from an appropriate specialist to develop a prescription for overall bank and slope stability.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Field Reference: BR-10, BR-11

Location: The area to be restored is throughout the upper reaches of the creek crossing the road at 14.5 km. This creek enters the Machmell River at Reach 3-2 on the left bank. The sensitive fish habitat is approximately 400m downstream of the bridge crossing, and consists of approximately 200m of side channel habitat.

Impact: 14.5 km Creek runs through a cut block up the slope, and has been logged to a thin riparian buffer zone. This stream in its upper reaches is characterized by sediment wedges associated with blowdown and eroding banks in some locations. Road construction near the lower reach (although well up on the bank above fish access) has a good bridge across and has not contributed to stream instability. Below the road the creek goes over a large fall and feeds a stable side channel to the Machmell River, which contains excellent spawning habitat. The lower portion (approx. 200m) is within anadromous fish access. While the fish habitat in the upper reach is too steep to provide fish habitat and is inaccessible, and may be in its pristine condition, there is potential for upslope debris torrenting to occur and inundate a valuable and limited, sheltered spawning area.

Objectives: To stabilize the upslope area using riparian reconstruction techniques and prevent potential debris torrenting.

Priority: Medium

Work Plan Considerations: Rehabilitate the riparian zone and examine the possibility of stabilizing sediment wedges and removing some log jams in the stream section flowing through the cut block. This is a preventative measure to protect the downstream fish habitat at the confluence with the Machmell River. Outright removal of the sediment wedges is not recommended due to the sensitivity of the downstream area to sediment input.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Field Reference: HP-03

Location: 20 km (Susan) Creek, Reach 4, flowing into the Machmell at Reach 4-2.

Impact: There is evidence of excessive bank erosion (especially at the 250m mark along the survey) and consequently, increased instability of this creek. This has caused a significant amount of channel braiding, as well as a reduction in pool abundance and rearing/holding habitat areas.

Objectives: To stabilize 2 x 25m sections of eroding banks and control the resulting woody debris and bedload movement into the creek.

Priority: High

Work Plan Considerations: This is the first phase of a more extensive rehabilitation of 20 km Creek. Given the height of the left bank erosion, specialist advice is required.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Field Reference: HP-01

Location: The area of bank instability is around the road crossing where it crosses the creek at 9.8 km. There is at least 300m of eroding bank to be stabilized.

Impact: At least two separate slide events have breached the sidechannels below, depositing debris (sediment and logs). It is apparent that logging has affected this as bucked logs were noted in the debris. At this time, SWD is inhibiting fish access to critical areas, and spawning habitat has been degraded. Flow has been diffused, creating isolated pools that trap fish which has made certain areas inaccessible (coho fry were observed in some areas).

Objectives: To stabilize and re-vegetate eroding banks, to control fine material entering the mainstem and to rebuild riparian habitats.

Priority: High

Work Plan Considerations: This is the first step to restoring the off-channel habitat at the bottom of the cliff. A second phase of this work will involve removing woody debris from the deposition zone.

Costing: Note that the costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Woody debris and bedload material that is transported downstream as well as channel alteration does particular damage to fish habitats when it affects the mainstem confluence area. In the Machmell River, these tributary mouths provide limited and valuable habitats that may be inundated by upslope material.

The required rehabilitation works for the confluence area is different for each creek, and depends on the precise configuration of the affected habitat. In the Machmell, field access is also important, as many areas are inaccessible to heavy equipment.

Field Reference: BR-13

Location: Top of Genesee Creek, Reach 1, from base of falls draining Walkus Lake towards Machmell logging camp.

Impact: Creek channel has abundant LWD and occasional jams that diffused the flow into multiple runoff channels. This valuable sockeye salmon spawning creek may have lost some spawning area.

Objectives: Re-open blocked channels and ensure adequate flow to current spawning channels. A secondary objective is to monitor potential debris movement from future forest harvesting upstream.

Priority: High

Work Plan Considerations: It is apparent that Genessee Creek has some degree of reduced flow, as several local residents and fisheries workers have noticed. A direct cause and effect is not apparent, but sockeye populations appear to be more susceptible to bear predation and possibly have reduced spawning area and lower incubation survival. One possible solution is to concentrate the flow at the base of the falls, where heavy equipment access is available. Concentrations of bears in the Genessee make stream surveys impractical, except in mid-winter. The high value spawning habitat downstream requires special consideration for silt control during construction.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Field Reference: M-02

Location: Machmell River, Reach 2-3 left bank, downstream of bridge crossing and adjacent to 8 km mark on Machmell Main road.

Impact: Road construction in Reach 1 of Clear Creek has consolidated tributary flows into one channel under the bridge crossing, just below the falls at the slope edge. Wetlands adjacent to the road no longer have fish access from the mainstem, and are no longer supplied with flow from Clear Creek itself. The resulting impact, primarily due to road positioning and road construction, is a loss of off-channel habitat at the confluence of Clear Creek and the Machmell River.

The lower portion of Clear Creek flows into a side channel of the Machmell River, which is somewhat sheltered from the mainstem instability by virtue of its position immediately downstream of the road crossing.

Objectives: To re-create about 200m² of stable, off-channel habitat that would provide winter flood refuge for rearing juvenile salmonids.

Priority: Medium

Work Plan Considerations: This project would create off-channel habitat by excavating the wetlands to a depth that would accommodate salmonids, and re-connect this habitat to the Machmell River in the lee of the bridge crossing.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Spawning habitat is being degraded by increasing channel instability. In particular, extreme flooding, channel shifting and eroding banks are reducing the amount of stable spawning area available to anadromous salmon and trout. Spawning channels can be designed by remove one or more of these limitations.

To restore the lower Machmell to its undisturbed condition would require greater overall channel stability. This is part of a larger-scale riparian rebuilding component yet to be established. Revegetation of unstable banks throughout the watershed will take many years. In the meantime, controlled-flow systems can be engineered and constructed to provide immediate, direct benefit to spawners. Three opportunities are outlined here.

The locations of the proposed spawning habitat improvements are not precise, due to the extensive and continuing process of mainstem channel shifting towards the right bank. Field positioning was done by helicopter orientation, followed by site sketches and map positioning as accurately as possible. In order to properly develop spawning channel opportunities, the location of the channels, flow control structures and existing flow channels all need to be mapped to reflect current conditions, and relative to one another. All the following projects should have their field positions re-confirmed, and specific routes for heavy equipment access outlined.

Field Reference: GB-05

Location: Behind the flood control berms off the right bank of Machmell River, reaches 1-2 and 1-3, adjacent to the camp.

Impact: The Machmell River is changing course towards the right bank resulting in installation of flood control measures. Large berms have been installed along this bank adjacent to old flood channels. Breaching the berms is not an option, as they are likely to be extended for flood control as needed. The channel outlet should be near the downstream end of the berms.

Objectives: To recreate a stable, protected flood channel habitat accessible from the mainstem Machmell. To create a multi-species spawning (primarily) and rearing (secondarily) habitat, with complex in-stream habitat.

Priority: High

Work Plan Considerations: Heavy equipment access from the Machmell camp and a deposition plan for the excavated material (eg. increasing the length of flood protection berms ) makes this a workable, large-scale opportunity. Engineering surveys should establish the depth to groundwater, and also test groundwater quality and quantity. Once project scale and water quantity are established, biological design can contribute. This project should also include a bear migration corridor, as bears will be attracted to this area, and should not be encouraged to shortcut though the camp.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Site ID: MM16

Field Reference: OC-05, OC-06, OC-07

Location: Machmell River, left bank, reach 1-3 and 1-5. Channel on left side of islands immediately downstream and upstream of the oxbow.

Impact: Consistent with the gradual movement of the mainstem towards the opposite bank, this excellent spawning channel is in danger of becoming dry. There is abundant LWD and sediment buildup on the upstream end of this channel which is gradually rerouting the flow.

Objectives: To ensure adequate water flow through 300m of sockeye salmon spawning area.

Priority: High

Work Plan Considerations: This work requires an engineering inspection to best determine how flow could be maintained and banks stabilized through the channel. This is in an area of complex flow patterns and the existing spawning habitat must also be protected. Considerations such as the cost effectiveness of heavy equipment access through the flats, where to put the excavated material, how to keep the sediment load from impacting the existing spawning area, and the expected life of the project, all need to be addressed.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Site ID: MM10

Location: Machmell River, reaches 1-4, left bank.

Impact: Consistent with the gradual movement of the mainstem towards the opposite bank, the oxbow has been isolated from all but very high water events due to build up of sediments, beaver dams and LWD at the confluences with the mainstem Machmell River. Salmonid fry have been observed within the oxbow, but seem to be prevented from migrating out.

Objectives: To open the oxbow to the mainstem Machmell River and provide an additional 300m of spawning and rearing habitat.

Priority: High

Work Plan Considerations: This work requires an engineering inspection to best determine where and how flows could be restored into the oxbow. The cost effectiveness of heavy equipment access, costs for removal of excavated material, protection of surrounding spawning beds and the expected life of the project all need to be addressed.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

In-stream barriers play a role in restricting fish access within the stream channel itself, and are the primary focus of in-stream habitat restoration in the Machmell watershed.

In the case of 20 km Creek that is isolated from the mainstem and populated by resident Dolly Varden, most of these barriers are partial barriers, as fish are unable to move freely within different stream sections. It is expected that their mortality rate increases as they are not able to access suitable stream cover or food-producing areas, become more vulnerable to predation, or face increased competitive pressures as restricted habitats become more crowded. Some of these barriers have been documented, in terms of shifting channels and cut off peripheral areas, sediment wedges, etc. but it is beyond the scope of this study to examine more closely how in-stream survival may be affected.

In Reaches 1 and 2 of the Machmell River, in-stream habitat restoration is mostly focused on increasing fish access into and out of low-gradient tributaries, so that fish in the mainstem are able to access these habitats more freely. The primary species that would benefit from this is rearing coho salmon, which were observed in most of the off-channel habitats. These areas may also be used by other species (eg. chinook salmon, or trout fry) but they were not observed during the course of this study and may be seasonally present.

Removing barriers and increasing fish access, also provides opportunities to conduct some excavation, riparian zone rebuilding, and habitat complexing as well. In-stream habitats vary in the practical aspects of doing so, but the extent of the work and requirements for equipment access and excavation must be judged on a site-specific basis.

Site ID: MM7

Field Reference: GB-06

Location: Creek crossing Machmell mainline road at 5 km. Machmell River reach 2-2.

Impact: Road construction (blocked culverts) has created a stream-habitat area at the toe of a slope, which has been utilized by beaver to extend their habitat. Instability of the mainstem Machmell River, combined with beaver activity has inhibited fish access to this area. Very large coho have been found behind the beaver dam, and are suspected to be unable to migrate out.

Objectives: Improve fish access to this area, prevent road flooding (clean culverts). Extend fish habitat in the upper stream about 300m upstream, or to the practical limit of the excavation.

Priority: High

Work Plan Considerations: This is an opportunity to test the effectiveness of a "beaver box" design at the stream outlet, and / or at the culvert sites. Specific location in the upper stream should also be identified for excavation and stream complexing.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Location: Machmell River, Reach 1 -1 right bank, and Machmell River, Reaches 1 -1 to 1-3 left bank. Creek running through cut block at the confluence of the Neechanz and Machmell Rivers.

Impact: This creek has been logged to the stream edge which has not permitted accumulation of LWD cover in the logged reaches. Lower reaches appear to be very suitable fish rearing habitat, and rearing coho were seen in the cut block stretch.

Objectives: To provide fish refuge areas throughout the reaches within the cut block, and to improve habitat complexing to increase in-stream productivity.

Priority: Medium

Work Plan Considerations: Access to this site is very good. Roads through the cut block are still in a usable state. Addition of root wads and other LWD refuge areas within the creek reaches running through the cut block. Up to 700m of stream complexing (including some excavation) could provide greatly improved fish habitat. The practical extent of the excavation will set the limit on how much in-stream work should be done.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

Impact: Dolly Varden have been sighted in this creek which appears to have been logged close to the stream edge. There is evidence of excessive bank erosion (especially at the chain 250m mark along the survey path), and consequently increased instability of this creek leading to a reduction in pool abundance and rearing / holding habitat area.

Objectives: After stabilizing the eroding banks (see section 9.1.3) complex the in-stream habitat by removing some of the LWD to help create pools. Other measures to create off-channel habitats should be implemented.

Work Plan Considerations: While slope stability is generally not an issue in this location, bank stability and riparian zone destruction is. Accordingly, in-stream habitat complexing is a second phase for work in this high-priority stream. While removal of piles of LWD should be undertaken in 20 km Creek, it should also be more closely assessed, after bank stabilization, to outline a more specific plan for in-stream restoration. One important area to be restored is on the left fork, about 250m long, and consists of several recently-formed braided channels. Another area is in the right fork, where log jams and dewatered areas probably also inhibit fish movement.

Costing: Note that costs quoted are intended to indicate only the general scale of the work. Specific prescriptions will be more fully costed when Level II and on-site assessments are completed.

The Machmell River is a large stream (>20m channel width) that is characterized by extreme flood events and volatile mainstem channel morphology. Fish habitat, especially for rearing fish, is limited to a few pockets of off-channel habitat that have been protected from flooding by high canyon walls or road construction. Fish habitat damage is related to slope instability, bank and channel instability, drying sections near the mouth and streamside logging in certain tributaries.

Fish habitat rehabilitation opportunities are based on the creation or enhancement of off-channel habitats, as well as improving bank stability and sediment source control in isolated areas.

The total cost of these projects is at least $463,000. This should be phased over several years.