Radstaak, G. 1998. Level 1 Fish and Fish Habitat Assessment of Dump, McLellan, Edlyn, Sato, Molly Walker, and Sawmill Creeks. Report Prepared for International Forest Products Ltd. Bio-Forest Consulting Services. ?pg.

I. List of Figures………………………………………………………………..….pg 5

II. List of Tables………………………………………………………………..….pg 6

III. List of Appendices………………………………………………………..……pg 7

IV. Acknowledgements………………………………………………………..…..pg 8

1. Introduction………………………………………………………………..……pg 9

2. Methods…………………………………………………………….…….….….pg 10

3. Creeks.

3.1. Dump……………………………………………………………….....pg 13

3.1.1. Description

3.1.2. Habitat Analysis

3.1.3. Population Analysis

3.1.4. Discussion and Restoration

3.2. McLellan…………………………………………………………..…..pg 19

3.2.1. Description

3.2.2. Habitat Analysis

3.2.3. Population Analysis

3.2.4. Discussion and Restoration

3.3. Edlyn………………………………………………………………..….pg 25

3.3.1. Description

3.3.2. Habitat Analysis

3.3.3. Population Analysis

3.3.4. Discussion and Restoration

3.4. Sato…………………………………………………………………..…pg 30

3.4.1. Description

3.4.2. Habitat Analysis

3.4.3. Population Analysis

3.4.4. Discussion and Restoration

3.5 Molly Walker……………………………………………………………pg 36

3.5.1. Description

3.5.2. Habitat Analysis

3.5.3. Population Analysis

3.5.4. Discussion and Restoration

3.6 Sawmill…………………………………………………………………..pg 42

3.6.1. Description

3.6.2. Habitat Analysis

3.6.3. Population Analysis

3.6.4. Discussion and Restoration

Figure 1. Map of Dump creek showing reach breaks and sample sites……………pg 13

Figure 2. Scatterplot of weight/length relationship in coho and cutthroat juveniles in Dump creek………………………………………………………………………… .pg 16

Figure 3. Map of McLellan creek showing reach breaks and sample sites………..pg 19

Figure 4. Scatterplot of weight/length relationship in coho and cutthroat juveniles in McLellan creek. pg 22

Figure 5. Map of Edlyn creek showing reach breaks and sample sites. pg 25

Figure 6. Scatterplot of weight/length relationship in coho and cutthroat juveniles in Edlyn creek…………………………………………………………………………..pg 28

Figure 7. Map of Sato creek showing reach breaks and sample sites…………..…pg 30

Figure 8. Scatterplot of weight/length relationship in coho and cutthroat juveniles in Sato creek………………………………………………………………………….…pg 33

Figure 9. Map of Molly Walker creek showing reach breaks and sample sites……pg 36

Figure 10. Scatterplot of weight/length relationship in coho and cutthroat juveniles in Molly Walker creek…………………………………………………………………pg 39

Figure 11. Map of Sawmill creek showing reach breaks and sample sites………pg 42

Figure 12. Scatterplot of weight/length relationship in coho and cutthroat juveniles in Sawmill creek……………………………………………………………………….pg 45

Table 1. Habitat parameters measures for each sample site………………………pg 11

Table 2. Habitat units and length measurements of Dump creek…………………pg 14

Table 3. Fish population density and measurements by species for representative type in Dump creek…………………………………………………………………………pg 15

Table 4. Habitat units and length measurements of McLellan creek……………..pg 20

Table 5. Fish population density and measurements by species for representative type in McLellan creek……………………………………………………………………..pg 22

Table 6. Habitat units and length measurements of Edlyn creek…………………pg 26

Table 7. Fish population density and measurements by species for representative type in Edlyn creek………………………………………………………………………….pg 27

Table 8. Habitat units and length measurements of Sato creek……………………pg 31

Table 9. Fish population density and measurements by species for representative type in Sato creek…………………………………………………………………………….pg 33

Table 10. Habitat units and length measurements of Molly Walker creek………..pg 37

Table 11. Fish population density and measurements by species for representative type in Molly Walker creek……………………………………………………………….pg 38

Table 12. Habitat units and length measurements of Sawmill creek………………pg 43

Table 13. Fish population density and measurements by species for representative type in Sawmill creek……………………………………………………………………pg 44

Appendix A: Fish habitat field survey forms

Appendix B: Sample site habitat descriptions and stream transect data forms

Appendix C: Channel characteristics of the 6 creeks averaged by habitat type and by reach

Appendix D: Substrate characteristics of the 6 creeks averaged by habitat type and by reach

Appendix E: Selected photos to illustrate habitat features

The completion of this report required the dedicated hard work of several people.

Fisheries technicians, Andrew Andy and Casey Koop provided many long hours in the field collecting habitat and population data for the six creeks involved. They were also an integral part in the reporting phase of this project.

George Hall, Ed Theissen, Lloyd Brynildsen, Mryt Schendel, Bob Boucher, Bob Buchanan, Mrs. McInroy, Bob Lenci, and Sandi Maclaurin, local valley residents, allowed the project technicians access to the creeks on their property and provided valuable information to the historical and present status of the creeks involved.

Mike Ramsay, Ministry of Environment, Fisheries Specialist and Michael Parker, Ron Ptoelmy Senior River Specialist and Dave Burt, Fisheries Biologist, D. Burt and Associates provided technical input and guidance to the project.

Neil Oborne, Watershed Restoration Coordinator, International Forest Products Ltd., was the project administrator for this project and provided the aerial photos and color copying for the report.

And lastly, Annie Spec, from the B.C. Ministry of Forests and Mike Howard, Regional GIS Coordinator, for their help with the GIS TRIM maps.

In the fall of 1997, six creeks , Dump, McLellan, Edlyn, Sato, Molly Walker, and Sawmill Creek (herein referred to as "the six creeks") received a detailed Level 1 Fish and Fish Habitat assessment. The client, International Forest Products Ltd. in conjunction with the Watershed Restoration Program(WRP), Forest Renewal British Columbia (FRBC) and the Ministry of Environment Lands and Parks(MELP) administered the project.

Historically, these streams have been degraded through past logging and settlement practices. These streams are located in the populated lower end of the Bella Coola watershed. The goals of this project were to identify and record current habitat conditions, identify degraded sites, and provide a baseline of current fish population levels in anticipation of Level 2 assessment, and consequently stream rehabilitation, in order to restore stream function, hydrology and fish populations.

All six creeks are located within the lower Bella Coola watershed. Annual precipitation in the Bella Coola Valley averages 1300 mm(1996 data). Summers are temperate with temperatures averaging +14.1B C(May-Sept., 1996 data), winter are mild, averaging +2.5B C(Oct-April, 1996 data). The majority of the valley residents live in the lower 20 km of the valley, with the arable portions of the valley bottom used for agricultural and settlement purposes. The towns of Bella Coola and Hagensborg are located within this lower portion. Second or in some cases, third generation mixed forest consisting of alder, birch, cedar and spruce, forest the valley bottom.

Assessment of the habitat characteristics of the six creeks occurred in October and November. Creeks at that time were at flows similar to the low conditions found in the winter rearing months. Measurement of habitat characteristics followed methods as outlined in Johnston and Slaney (1996). Two field personal started at the mouths of the respective creeks and walked the total length of the fish assessable portions. Length of habitat units were delineated using a hip chain and recorded as pool, riffle, or run, and or other. Representative habitats were identified and the characteristics (Table 1) were measure and recorded on habitat data sheets(see Appendix A).

Table 1: Habitat parameters measured for each sample site.

A shore based setup was used in the electrofishing. An electrofisher(Smith-Root Type VIII A) was used to capture juvenile fish. Voltage and frequency varied, depending on water conductivity and temperature, between 450-850 V and 40-80 MHz, respectively. Areas of representative habitats were enclosed with stop nets (1 mm mesh diameter, 10 m in length). The first pass (C1) was done until it was felt that the area was covered extensively. Fish captured with a pole seine were then measure, weighed(using a Ohaus scale) and identified. The second pass (C2) captured any fish that escaped the first pass. Data was recorded on electrofishing data sheets(see appendix?).

The formula (Dee Leeuw, 1981) used to calculate the population of rearing species captured at electrofishing sites is as follows:

Where n =an estimate of the total population size at the site

C1 =number of individuals in first capture

C2 =number of individuals in second capture

Calculation of fish densities based on total area:

Where FPU_observed =observed fish per unit area

n =population estimate from equation 1

TA =total area of the electrofishing enclosure

Figure 1: Map of Dump creek showing reach breaks and sample sites.

3.1. Dump Creek

Dump Creek(6 km east of Bella Coola) is located immediately east of Thorsen Creek. Dump creek is a tributary to Noohalk creek and joins immediately prior to where Noohalk creek enters Thorsen Creek. Source water for this system occurs from lateral dyke seepage from Thorsen Creek(Guy and Slaney, 1997). Dump Creek habitat is dominated by the presence of a large beaver impoundment (1802 m²). Adult salmonid access, in the past, has been restricted by gradient and beaver activity(Guy and Slaney, 1997, Maclaurin, pers comm.).

Table 2. Habitat units and length measurements of Dump Creek.

This reach is a single channel, 68 m long. Gradient for the first 40 m is low(<0.5%) but increases to levels in excess of 20% immediately prior to the culvert at Highway 20. Habitat in this reach consists of glide(60.3%), pool(5.9%) and riffle(33.8%). Stream cover is adequate with LWD(18%), boulder(<2%), cutbank(11.7%) and overstream vegetation(10%). Riparian vegetation is early seral stage consisting of alder and cottonwood(100%). Crown closure is moderate(51-70%). Photo 02-09 (Appendix E)is representative of this reach.

Reach 2 is entirely pond habitat. No habitat criteria were measured for this reach. Photo 01-18(Appendix E) is representative of this reach.

Reach 3 parallels Thorsen creek for 151 m. Gradient is low (<1%) with an average wetted width of 3.1 m. Habitat frequency consists of glide(58.8), pool(15.7%) and riffle(25.5%). Stream cover is moderate consisting of LWD(22.5%), cutbank (5%) and overstream vegetation(15%). There is no boulder cover within this reach. The riparian vegetation is comprised of mainly alder and cottonwood(70%) with some conifers(30%).

Table 3. Fish population density measurements by species for representative habitat types in Dump creek.

Figure 2. Scatterplot of weight/length relationship in coho and cutthroat juveniles in Dump creek.

At present, there appears to be a recent barrier to adult movement at the creek junction to Highway 20. Because of both gradient increases in reach 1 from past dyking activities of Thorsen Creek, the plugging of the culvert by beavers and the placement of wire screens(see photo 04-20 Appendix E), there appears that there was no coho adult or juvenile migration to reaches 2 and 3 in 1996. No fry(0+) were found above the culvert barrier. The placement of a beaver box this winter by DFO should alleviate beaver problems, although gradient restrictions should be assessed in reach 1. Reach 2, consisting of the beaver impoundment presently provides good rearing habitat for coho and Dolly Varden parr. It appears that coho juveniles do not utilize habitat in reach 3. This is probably a result of low utilization of existing good quality habitat (beaver pond) because of non-existent seeding, or immigration past the culvert barrier, as there are no barriers to fish movement from reach 2 to reach 3. Potential habitat for cutthroat and Dolly Varden juveniles exist in reach 3, but at present, densities are low. Spawning habitat within reach 3 is low caused by the high level of sedimentation(D₉₀=fines). Enhancement opportunity exists for extension of the length of Dump creek beside the Thorsen creek dyke, although water source (seepage from Thorsen creek) and quality (proximity to the Bella Coola landfill site(Summitt, 1996)) may be compromized.

Figure 3 Map of McLellan creek showing reach breaks and sample sites.

McLellan creek is located 8 km east of townsite Bella Coola. It is a tributary of Noohalk creek. Historically, McLellan creek was a side channel of the Bella Coola river following its length for 3.5 km. Currently, McLellan creek is separated from the Bella Coola river and its water source is from immediate drainage from the surrounding farmland and mixed-wood forests. Gradient throughout the system is less than 1% resulting in low water velocity.

Table 4 Habitat units and length measurements of McLellan creek.

*note Reach 4 contains 5.9% dry creek bed due to dewatering in the upper reach.

This is low gradient reach (<0.5%) with a single channel averaging 11.4 m in width. The entire length (1055 m) consist entirely of glide habitat(Table 4). Substrate is comprised almost entirely of fines and organics with a small amount of gravels (<2%) near the top end of this reach. Cover consists of LWD (10%), overhanging vegetation (15%), and cutbank (10%). Riparian coverage consisted of mature forests (>50 years) comprising of alder (70%) and conifers (30%). Canopy coverage ranged from low (20%, authors estimate) in the lower section of the reach to high (71-90%)in the upper section of the reach(Table??).

Gradient in this reach is similar to Reach 1 although the channel width (x=5.9m) is smaller. The length of this reach is 558 m. Habitat consists of glide (88.5%) and riffle(11.5%). Substrate is comprised of fines (13.7%), small gravels (33.3%), large gravel (50.0%) and cobble (3%). Cover consists of LWD(5.3%), overhanging vegetation(8.3%) and cutbank(3.3%). Riparian coverage consists of alder(43.3%) and conifer(56.7%). Canopy coverage was high (71-90%). Photo 02-11 (Appendix E) show general features of this reach.

This reach has been influenced by beaver (6 dams). The length of this reach is 1194 m Gradient is low (<1%), channel widths are variable(range=6.3-. 20 m(visual estimate))in this section due to the dams. Habitat consists of glide(64.8%) and pond (34.7%) with a small amount of riffle(0.5%) at the outlets of the beaver dams. Substrate is comprised of mainly of fines with a small amount of gravel in the riffle areas(visual observation). Cover consists LWD(2.5%), overhanging vegetation (5%) and cutbank (7.5%). Riparian coverage consist of alder(60%) and conifers (40%). Canopy coverage was variable with high (71-90%) closure in the glide portion between dams and low coverage around the dams. Photo 03-39 (Appendix E) is representative of this reach.

This reach splits into several smaller channels with low(<1%) gradient. At time of survey(late fall) these channel had become dewatered leaving low stagnant pools of water with very little or no flow. Due to probable low fisheries value no site were done in reach 4. Photo 03-17 (Appendix E) is representative of this reach.

Table 5. Fish population density measurements by species for representative habitat types in McLellan creek.

Figure 4. Scatterplot of weight/length relationship in coho and cutthroat juveniles in McLellan creek.

In reach 3, existing beaver dams had been breached for adult salmonid migration. At low water, pool habitat is limiting.

The breaching of dams for adult salmonid migration should be assessed for each specific situation. Breaching of dams in the fall may eliminated over-wintering pool habitat for existing juvenile salmonids. In this system, with existence of pools being one of the limiting factors, over-wintering habitat is critical. In McLellan creek, spawning gravels are non-existence in the upper reaches. Spawning success by adults in these reaches would by low.

The complexity and diversity of habitat in McLellan creek is low. Restoration activities should be considered to increase the frequency of habitat types. Glide habitat is the dominant habitat(94.6% by length). McLellan creek was once a side channel to the Bella Coola river at high water events. Channel characteristics of McLellan creek are a left over legacy of past high water events. In the past, water from the Bella Coola has flushed this system. But due to upstream dyking (Summit), lateral valley movement and/or lack of a high water event recently, water supply from the Bella Coola river to McLellan creek has ceased. Because of low velocity flow conditions(0.04 m/s), substrate is mainly fines(). Interstitial spaces of gravels have been filled in by the deposition of fines, limiting salmonid spawning areas. Currently, water source exists from local ground drainage. Present reduction of stream width through restoration activities would increase water flow and frequency of habitat types. There also exist the possibility of adding water flow from a back channel immediately east of the end of reach 4.

Opportunity exists to create access to wetland rearing areas located approximately 300m to the south of McLellan creek. At present, small diameter drainage pipes and culvert screens(see photos 03-03 and 03-06, Appendix E) connecting McLellan creek to these wetlands prevents access by both juvenile and adult salmonids.

Figure 5. Map of Edlyn creek showing reach breaks and sample sites.

3.3. Edlyn Creek

Edlyn creek is located 14 km east from townsite Bella Coola. This is a small creek that empties into a sidechannel of the Bella Coola river near the airport. Historically, this creek has been referred to locally as "George Hall creek".

Table 6. Habitat units and length measurements of Edlyn Creek.

Reach 1 is a single channel 540 m in length with low gradient(<1%). The average wetted width in this section is 2.5 m. The habitat within this reach consists of glide(79.4%), riffle(16.9%) and some pool(3.7%). Substrates were predominately fines(66.7%,) followed by gravel(20.0%), and cobbles(13.3%). Stream cover consists of a small amount of LWD(9.1%) and boulder(4%) with a higher amount of cutbank(11.3%) and overstream vegetation(7.6%). Riparian vegetation consist mainly of alder(62.3%) with some conifers(37.7%, Ss, Rc, Wh). Canopy closure is high(71-90%) as a result of the alder overstory. Photo 02-03(Appendix E) is representative of this reach.

Reach 2 is 481 m in length comprising of two channels, one that goes sub-terrain. The gradient in this reach is also low(<1%) with an average wetted width of 2.8 m. Habitat is mostly glide(89.4%), with pool(3.3%) and riffle(7.3%) comprising of the remainder. Sedimentation is this reach is chronic with 80% of the habitat sites measured having D₉₀=fines. Stream cover consist mainly of over-hanging vegetation(24%) with some LWD(9.2%) and cutbank(11%). There is no boulder cover within this reach. Riparian vegetation is equally distributed between alder(54%) and conifers(46%, Rc, Ss,Wh). Photo 02-06 (Appendix E) is representative of this reach.

Table 7. Fish population density measurements by species for representative habitat types in Edlyn creek.

Figure 6. Scatterplot of weight/length relationship in coho, cutthroat and Dolly Varden juveniles in Edlyn creek.

Of all the 6 creek surveyed, the habitat of Edlyn creek was most complex (44 habitat units). Edlyn creek has a high level of fine deposition (D₉₀=fines at 80% of habitat sites) throughout both reaches. Spawning habitat for salmonids is limiting in the system due to interstitial filling of gravels by fines. Source of the fines appears to be localized as the headwaters of Edlyn are sub-surface. Temperature of the water remains above freezing keeping the creek ice free throughout the winter(Theisen, pers. comm.). New pool formation is limited by the lack of large wood debris in the system(93% of LWD is below 20 cm diameter, n=304). A fish barrier exists at a man-made dyke structure, 1022 m from the mouth, erected in the 1950’s(see photo 04-03 Appendix E). Spawning habitat above the barrier is low but some juvenile rearing habitat exists. Restoration opportunities exist in the placement of large woody debris and the strategic placement of diverters in order to flush fines at times of high water.

Figure 7 Map of Sato creek showing reach breaks and sample sites.

Sato creek is located approximately 19 km east from townsite Bella Coola. It is a small meandering low gradient creek that empties into a side channel of the Bella Coola river. There are two ponds within the Sato system, one natural and one man-made. Upper reaches contain high sediment loads. Historically, this creek had been fed from a channel to the south-east. Currently, water is drained from the mountain immediate to the south.

Table 8 Habitat units and length measurements of Sato Creek.

The gradient in this reach is low(<1%), with an average channel width of 2.2 m. The total length of reach 1 is 152 m. Sato Creek habitat consists of glide(63.2%), riffle(28.3%), and pool(8.5%). This reach has variety of substrate components (author estimate), consisting of gravel(40%), cobble(30%), and fines(20%). The over-stream cover is 15%, LWD is 10%, with no bolder cover. The stream canopy cover is mixed with conifers(60.7%, Ss, Wh, Rc, Df) and deciduous(40.3%, mainly alder). Photo 02-21 is representative of this reach (Appendix E).

The gradient in this reach is also low(<1%) and has an average wetted width of 2.8m. The total length of reach 2 is 302 m. Its habitat units consists of glide(72.8% ), pond(21.9%), and pool(5.3%). Substrate consists of mainly sand and small gravels(mean D₉₀ ranges from fines to 5 cm, n=4). The stream cover component is LWD(12%), and overhanging vegetation(13.4%) with no boulder cover. The crown closure 30% deciduous(alder) and 70% conifer(Rc, Wh, Ss).

The gradient(<1%) in reach 3 is low, with an average wetted width of 2.3 m. The total length of reach 3 is 245 m. The habitat is dominated by glide(97.6%), followed by riffle(2.4%). Sato creek has a high stream cover component of LWD(20%), and vegetation(30%), with no boulder cover. Crown closure is high(>90%), with deciduous(70%) being dominant, and conifer(30%) sub-dominant. Photo 01-01 is representative of this reach(Appendix E).

Gradient(1-3%) varies in the lower section of this reach. In the upper section the gradient increases(3-6%). The total length is 169 m with an average wetted width of 1.7 m. Habitat units are dominated by glide(81%), then riffle(19%). The stream cover components consist of LWD(10%), cutbank(10%), and vegetation(20%), with no boulder cover. Riparian vegetation was alder(60%) and conifers(Rc, Ss). The crown cover (author estimate) averages 50%.

Table 9. Fish population density measurements by species for representative habitat types in Sato creek.

Figure 8. Scatterplot of weight/length relationship in coho, cutthroat, and Dolly Varden juveniles in Sato creek.

Sato creek has potential for increased spawning and rearing sites for salmonid species. Reach 1(see photo 02-21), like the majority of the other creeks, contains a high amount of glide(43%), although there is some pool(28%) and riffle(28%) within this section. Siltation also occurs within this reach but it is not as extensive as the other systems. D₉₀ ranges from fines to 13 cm. Reach 2 has pond habitat that is suitable for rearing, although the beaver dam may act as an obstacle to fish passage, especially at low water levels. Reach 3 contains a significant amount of sediments. LWD is limited both by size and abundance within this section. The channel complexity is simple. Diverters and addition of LWD should be added to increase the complexity of habitat and substrate within this section. In reach 4, there are several remnant wood bridges. These wood bridges need to be removed as they are creating obstacles to fish passage. Pool habitat within this reach is limited. Placement of LWD should be considered to created habitat complexity within this reach.

Figure 9 Map of Molly Walker creek showing reach breaks and sample sites.

Molly Walker creek is located approximately 40 km up valley from townsite Bella Coola. Molly Walker creek is of low gradient and is dominated by large beaver dams. Beaver activity along Molly Walker creek is current with new dams being established periodically. In the past Molly Walker Creek has been utilized to a large extent by spawning cutthroat trout(B. Buchanan, pers comm.)

Table 10 Habitat units and length measurements of Molly Walker Creek.

The gradient in reach 1 is low (<1%) with wetted width averaging 5.8 m. The length of this reach is 495 m consisting of glide(81.5%), riffle(14.6%) and pool(3.9%) habitat. Stream cover was moderate with LWD(10%), bolder(5%), cutbank(2%) and overstream vegetation(15%) comprising the makeup of the cover. Crown canopy cover was advanced(71%-90% canopy closure).

Reach 2 is characterized by a large beaver impoundment. The length of this reach is 103 m with low gradient(<.25%). Wetted width averages 1.4 m . The habitat is comprised pond(30%) and slough(70%). The riparian vegetation was mostly alder(70%) with some conifer(30%,Rc,Ss). Canopy closure throughout the entire reach length is low(<15%, estimate). Stream cover is was low with LWD(10%) providing some cover. Photo 05-06, Appendix E is representative of this reach.

Reach 3 also has a large amount of pond habitat(71.6%). A series of 4 consecutive dams within 64 m creates these pond conditions. The length of this reach is 218 m with a wetted width averaging 1.4 m. Habitat units consists of pond, and glide(23.9%) with no pool or riffle within this reach. Stream cover consists of a moderate amount of LWD(20%) with some overhanging vegetation(10%). Riparian vegetation in mainly alder(70%) with some conifers(30%, Rc, Ss, Wh).

Reach 4 is the longest reach(524 m) of Molly Walker creek. Gradient is low(<1%) with the wetted width averaging 2.7 m. Habitat consists of mainly glide(97.7%) with some pool(1.3%) and riffle(1.0%). Stream cover is abundant within this reach with LWD(33%), cutbank(8.3%) and vegetation(16.7%) providing the majority of the cover. Riparian vegetation is mainly deciduous(76.6%, alder, birch) with some conifers(23.4%, Ss, Rf). Photo 01-23, Appendix E is representative if this reach.

Table 11. Fish population density measurements by species for representative habitat types in Molly Walker creek.

Figure 10. Scatterplot of weight/length relationship in coho, cutthroat, and Dolly Varden juveniles in Molly Walker creek.

Molly Walker creek has potential to provide significant spawning and rearing habitat. Coho fry and parr were found extensively within the electrofishing sites(see Figure 10). All age classes of cutthroat trout were found within this system(see Figure 10). Presently, there is high siltation(D₉₀=fines) within the glide portions of Reach 1(ave-80%). Adult fish use of this creek is restricted by beaver dam activity. Dam management should be considered in the restoration phase of this creek. Since the field survey in the fall months, a new dam has been erected at the mouth(0+33m) of Molly Walker creek restricting fish access. Spawning of cutthroat trout in this system, this spring, will be limited to the immediate lower portion of Molly Walker creek with the continue existence of this dam. Recruitment of LWD to the system is adequate although size diameter classes are small (90% of stems range from 10-20 cm). Restoration activities should consider the placement of large diameter woody debris to facilitate pool-riffle complexes. Diverters could be used to flush this system at time of high water. A significant portion of reach 4 contains suitable spawning material although beaver dams within reach 3 block access to adult fish.

Figure 11. Map of Sawmill creek showing reach breaks and sample sites.

3.6. Saw Mill Creek

Sawmill Creek is located on the north side of the Bella Coola valley at the end of the Saloompt river road approximately 16 km east from townsite Bella Coola. Sawmill creek is a hydrological different stream than the previous five. Gradient throughout its entire anadromous length is generally steeper than the previous five creek. Substrate is also larger with boulder and cobbles dominating the stream bed.

Table 12. Habitat units and length measurements of Sawmill Creek.

Reach 1 is comprised of a single channel 113 m in length. The gradient is moderate ranging from 3-8%. The average wetted width is 6.9 m. The habitat in this reach is glide(53.9%), pool(20.5%), riffle(12.0%) and cascade(13.6%). Substrate in composed of 80% clay and 20% boulder and cobble(authors estimate)(see photo 01-08, Appendix E). Cover in this reach is low with LWD(<1%), cutbank(<2%) and overstream vegetation(4%) providing some cover. Riparian species are mainly alder(72%) with some conifer(28%, Rc, Ss, Wh, Rf), with low overall canopy closure(1-20%).

Reach 2 is comprised of a single channel 147 m in length. Gradient is variable ranging from 1-8%. It is also larger with a wetted width of 9.7 m. Habitat is mixed with glide(44.1%), cascade(29.4%) and riffle(26.6%). The majority of stream cover is boulder(40%, authors estimate) with vegetation(<1%) and cutbank(<2%) providing some cover. LWD(0%) is noticeably missing from reach 2. Riparian species are made up of alder(64%) and conifers (Ss, Rc, 36%), with canopy closure being moderate (41-70%).

The start of reach 3 breaks into 2 channels and the gradient in this section is variable(1-8%). The length of this reach is 145 m. The habitat components are glide(27.9%), pool(47.6%), riffle(2.5%) and cascade(22.0%). Stream cover is comprised of mainly boulder(70%, estimate) with LWD(<1%), cutbank(9%), vegetation(<1%) providing the remaining cover. The riparian vegetation is mainly coniferous(60%, Wh, Rc, Ss) with alder (40%) being the main deciduous species. Canopy closure is low(20-40%). A boulder/falls/cascade anadromous fish barrier exists at the end of this reach(see photo 01-05,Appendix E)

Table 13. Fish population density measurements by species for representative habitat types in Sawmill creek.

Figure 12. Scatterplot of weight/length relationship in coho and cutthroat juveniles in Sawmill creek.

A dominant prevalence of clay exists in the lower portions of this creek. The presence of clay substrate(see Photo 01-08 Appendix E) is natural and appears to be formed by high velocity runoff that has washed smaller diameter substrates from the system. Suitable spawning gravels are non-existent in the lower reaches. This stream, at present, appears to be more suitable for rearing of parr as the size classes of juveniles found were large(0 =93.9 mm, 13.3 g, CT; 61mm, 2.6 g CO). Large boulders(0 D₉₀=39.0 cm) exist throughout all three reaches providing suitable cover for rearing juveniles. High velocities in conjunction with the larger substrate(see Photo 01-06 Appendix E) tend to preclude most of the habitat for coho rearing. Cutthroat trout and Dolly Varden char and steelhead trout tend to utilize these types of habitats. Although no steelhead juveniles were found in our sampling, juveniles have been documented in past recent sampling of this system(Ptoelmy, unpubl. data, 1997). Some major erosion occurs within reach 1and 2 that has been caused by the destabilization of the bank from past cutting of riparian trees(see photo 01-09,01-12 Appendix E). An anadromous fish barrier exists at the end of reach three(see photo 01-05 Appendix E), resident trout may exist above this barrier but no sampling was done to verify this. Restoration opportunities, due to gradient and flow are limited, but bank erosion should be further assessed.

Appendix A Fish Habitat Form

Appendix C: Channel characteristics of the 6 creeks averaged by habitat type and by reach.

Appendix D: Substrate characteristics of the 6 creeks averaged by habitat type and by reach

Appendix E: Photos