EA1164

submitted to:

March 1998

EA1164

TABLE OF CONTENTS

TABLES

FIGURES

APPENDICES (under separate cover)

Table 1. Hydrologic regime of Big, Gaspard and Churn creeks.

An overview fish habitat assessment and watershed level riparian assessment were conducted for the Big, Gaspard and Churn creek watersheds. This consisted of: 1) a review of available literature and data, 2) a helicopter reconnaissance flight during which approximately 1:3,000 colour air photographs were acquired of visible streams, 3) a description of the current stream habitat and riparian conditions, and 4) recommendations regarding future assessment needs and opportunities for restoration. The study team consisted of two fisheries biologists and a vegetation ecologist with assistance from a stream hydrologist. In all cases, methods provided in publications by Forest Renewal B.C. on stream and riparian assessment procedures were followed (WRP 1996a,b).

Information on the distribution of fish species within the watershed is limited. Barriers to migration of anadromous salmon occur near the mouth of the Big and Gaspard creeks. Anadromous salmon do use the mainstem of the Churn Creek, but do not access West Churn Creek where some logging has occurred. A limited amount of sampling in streams and lakes in the study area indicate rainbow trout are common. Bull trout have been reported for Churn Creek only. Non-trout fish species have been reported in the lower mainstem of Churn Creek.

In general, stream habitats and riparian areas in the Big, Gaspard and Churn creeks have been impacted by logging and agriculture. Riparian buffers were not left in some areas. The primary impacts on streams in the study area include: 1) filling of pools and loss of surface flow as a result of bar growth and gravel deposition, 2) loss of large woody debris, 3) log jams, and 4) channel instability and widening where mature trees in the riparian area have been removed. The best opportunities for stream rehabilitation are in tributaries and the mainstem of central zones of Big and Gaspard creeks where logging impacts are evident, stream gradient is low, stream width and depth are favourable for instream work and fish occur in the vicinity.

The riparian zone has been impacted by logging and agricultural activity in the Big and Gaspard creeks. In locations where the riparian zone has been cut to the stream bank, the vegetation is in various stages of regeneration. In these areas, the riparian does not provide a source of large woody debris for streams. Site visits are recommended for sites where the riparian zone has been previously logged or roads occur within the riparian zone. The highest priority areas are in Big Creek, Twinflower Creek, Cooper Creek, Bambrick Creek, Gaspard Creek and Little Gaspard Creek.

We thank MOELP for providing digital TRIM and MOF for providing digital forest harvesting and road data in the Big, Gaspard and Churn creek watersheds. The following people contributed to the preparation of this report.

Bob Bocking, LGL Ltd. - Project Director

Ken Rood, Northwest Hydraulic Consultants - Hydraulic Assessment

Russ Frith, LGL Ltd. - Fish and Fish Habitat Assessment

Sergei Yazvenko, LGL Ltd. - Riparian Assessment

Robin Tamasi, LGL Ltd. - GIS Specialist

Lucia Ferreira, LGL Ltd. - GIS Technician

In 1997, LGL Limited was awarded a Forest Renewal B.C. (FRBC) contract (CCA-843) from the Ministry of Environment, Lands and Parks (MELP), Cariboo Region, to conduct an overview and Level 1 fish habitat assessment procedures (FHAP) and watershed and site level riparian assessment procedures (RAP) in the Big, Gaspard and Churn creek watersheds.

The goal of the overview fish habitat and watershed level riparian assessments reported here was to identify stream sections impacted by historic logging or with the potential to be impacted by historic logging activities. Stream sections were assigned a priority for further Level 1 fish habitat assessments, site level riparian assessments and possible rehabilitation. Natural and other industrial (e.g., agricultural) hazards to stream habitat were also noted.

In general, the following types of habitat degradation and rehabilitation opportunities were identified from 1:15000 and 1:3000 air photographs:

- bank erosion and sediment source areas;

- debris and sediment accumulation;

- lack of LWD;

- channel pattern effects;

- road crossings;

- torrented channels;

- potential ground water sources;

- opportunities for off-channel developments;

- potential barriers to fish migration;

- inadequate riparian conifer vegetation; and

- lack of riparian vegetation.

Based on the assessments conducted and Watershed Restoration Program Guidelines (WRP 1996a,b), a prioritized list and schedule of stream and riparian field assessments and rehabilitation opportunities was developed.

The Big, Gaspard and Churn creek watersheds are located approximately 60 km southwest of Williams Lake in the Williams Lake Forest District and within the Cariboo Forest Region (Figure 1). Big, Gaspard and Churn creeks drain the southwest portion of the Chilcotin Plateau and lie to the west of the Fraser River between Williams Lake and Lillooet. Gaspard Creek is wholly contained on the Chilcotin Plateau; however, upper Big Creek watershed extends into the east side of the Chilcotin Ranges of the Coast Mountains and part of Churn Creek lies in the Pavillion Ranges of the Cascade Mountains (Mathews 1986). The Big Creek mainstem flows for approximately 72 km in a northerly direction from its headwaters in the Eastern Chilcotin Ranges and then turns to the north-east for an additional 26 km before joining the Chilcotin River. Churn Creek flows for 40 km from its headwaters before turning to the east and flowing for an additional 30 km to join the Fraser River. Gaspard Creek flows for 24 km in a northerly direction and then turns sharply to the east to flow for 36 km before joining the Fraser River just north of Churn Creek.

The Chilcotin Plateau consists of gently rolling, undissected uplands. The study area drainage basins cover 4,237 km² with elevations generally between 1,300 and 1,600 m and rising to the west, reaching around 2,000 m near the Coast Mountains. Much of the plateau is underlain by volcanic flows, though the bedrock is mostly covered by glacial drift; drumlins and eskers provide local relief. The gently rolling surface has steep escarpments along creeks and major river valleys.

The Big, Gaspard and Churn creek watersheds are located mainly within the Interior Douglas-Fir and Sub Boreal Pine-Spruce biogeoclimatic zones at lower elevations and within the Engelmann Spruce-Subalpine Fir zones at higher elevations. Where the Gaspard and Churn creeks join the Fraser, the vegetation zone is grassland with common plant species of big sagebrush, rabbit-brush, bluebunch wheatgrass, needlegrasses, pasture sage and dropseed. Above the grass zone, Douglas-fir is the dominant tree with bluebunch wheatgrass understory, and lodgepole pine is dominant at higher elevations. Common understory species include common juniper, prickly rose, soopolallie, willows, kinnikinnick and aster. As elevation increases into the ESSF zone, Engelmann spruce and subalpine fir are dominant with understories of shrubs and grasses. Common plants include juniper, soopolallie, grouseberry, lupines, arnicas and lichens (Annas and Coupe 1979).

The Big Creek watershed was divided into seven sub-basins: Big, Tilton, Twinflower, Mons, Rushes Lake, Cooper, and Bambrick. The Big sub-basin was the residual area of the watershed remaining outside of the other tributary sub-basins. Only the portion of the Big watershed below Groundhog Creek was included in the study area as no logging has occurred above this point. Gaspard Creek watershed was divided into two sub-basins: Gaspard and Little Gaspard. Gaspard sub-basin is the residual area of the Gaspard watershed remaining outside of the Little Gaspard sub-basin, the main tributary to Gaspard Creek. The whole Gaspard watershed was included in the study area. For Churn Creek, the watershed area above West Churn Creek has not been historically logged and was not included in the study area. West Churn Creek is the largest tributary to Churn Creek and this sub-basin was the only sub-basin assessed for this study. Logging has occurred in other areas of the Churn watershed below West Churn Creek, but most logging occurred away from stream banks (Appendix A).

At Tatlayoko Lake, on the northwest side of the Plateau, almost 60% of the total annual precipitation falls between October and March, with the least falling in April and May. This is a result of Pacific storms in the fall and winter spilling over the Coast Mountains. In the eastern side of the Plateau, the trend differs. At Williams Lake and 100 Mile House most of the annual precipitation falls in June, July and August with a second minor peak in December and January. The Dog Creek AES station at the mouth of Gaspard and Churn creeks also shows two peaks in the annual precipitation pattern.

Snowfall is a major component of the total annual precipitation (30 to 50%) recorded at the Atmospheric Environment Services climate stations. Mean monthly temperatures stay below freezing from November through March throughout the region and, as a result, almost all the total precipitation falls as snow in these months. It is expected that a longer snowfall period occur at the higher elevations in the study area.

The study area falls in the rain shadow of the Coast Mountains and annual precipitation is consistently low. Normal annual precipitation over the Plateau varies from about 550 mm in the northeast to 400 mm in the southwest. Greater total precipitation is expected on the eastern slopes of the Coast Mountains, but no climate stations operate in this region and total precipitation is not known.

The greatest daily rainfalls (about 40 mm) are recorded in June, July or August in the eastern portion of the basin. Near the Coast Mountains, the greatest daily rainfalls are recorded in October and November, reaching a maximum of 70 mm at the Tatlayoko Lake climate station.

Big Creek above Groundhog Creek (08MB006) gauge was used to estimate the flow characteristics of the streams in the study area. Flow characteristics calculated for this gauge were transferred to the study streams, based on ratios of their watershed areas to that of Big Creek. As discussed earlier, the Big Creek watershed may not be very representative of the other study streams. We expect that this procedure will overestimate mean annual flow, mean annual runoff, summer seven-day low flows and peak flood discharges in the tributaries to Big Creek and in Gaspard and Little Gaspard Creeks.

Mean Annual Flow: The mean annual flow at the gauging station on Big Creek is 5.9 m³/s. When expressed as an equivalent depth over the watershed (1,304 km²) it translates to a mean annual runoff of 143 mm. Mean annual precipitation for the Chilcotin Plateau is 350 to 500 mm, suggesting that annual evapotranspiration and losses to ground water amount to 200 to 350 mm in the Big Creek Watershed. These losses are expected to be somewhat greater for the lower elevation tributaries. Table 1 summarizes the estimated mean annual flows for the study watersheds using an annual runoff of 143 mm.

Mean Annual and Larger Floods: Annual maximum discharges result mostly from snowmelt in May through July or, infrequently, from Pacific storms that spill over the Coast Mountains in October and November, prior to freeze up. Maximum discharges in the tributary streams were predicted from a ratio of drainage areas of the tributary and that at the Big Creek gauge, applied to the mean annual and twenty-year instantaneous discharge calculated for Big Creek above Groundhog Creek (08MB006) gauging station (Table 1).

There is considerable uncertainty in the estimates quoted on Table 1 because of differences in drainage area magnitude, relief and physiography of the catchment areas, and differences in annual precipitation across the region. We suspect that the flows quoted in Table 1 overestimate actual flood discharges, particularly in the smaller study streams.

Low Flows: Annual low flows occur under ice in January through March on the larger streams in the study area (Table 1). Seven-day low flows measured in Big Creek were transferred to the other study area basins based on the ratio of their drainage areas, then reduced for water extractions. Tributary basins with large lakes in them such as Rushes Lake and lower Mons Creeks may have higher summer seven-day low flows than other nearby streams. In the smaller streams, minimum annual flows may occur in the late summer during unusually dry years.

Gaspard Creek at Gang Ranch (08MD010) has a reported summer seven-day low flow of 0.16 m³/s for 1928 which is much less than would be expected, based on a ratio of drainage areas with the gauge on Big Creek. The much lower recorded value, which is listed on Table 1, results from a combination of lower flows from the drier Chilcotin Plateau, unusually dry summers in the 1920’s, and water manipulation in the basin. Churn Creek near Gang Ranch (08MD012) reports a mean summer seven-day low flow of 0.98 m³/s, which is consistent with that calculated by a ratio of drainage areas with the gauge on Big Creek. This occurs because both Churn and Big Creeks drain a portion of the Chilcotin Ranges.

Most of the study area lies on the Chilcotin Plateau, which is a flat to gently rolling, drift-covered surface at elevations of 1,200 m to 1,500 m. Thick accumulations of glacial drift overlie volcanic bedrock and the drift has been formed into drumlin-like shapes that provide much of the relief on the Plateau.

Deep valleys are found along the lower Reaches of Big, Gaspard and Churn creeks. These valleys occur where the main creeks are incised as they leave the plateau surface and drop steeply into the Chilcotin or Fraser Valleys. The results of this incision are steep-sided valley walls, often covered with unconsolidated glacial drift. Failures of the material on the valley walls occur as a result of river erosion along the toe of the wall and processes on the slopes. After failures occur, gullying, rilling and ravelling continue to contribute additional sediment to the river for an extended period of time. These unstable valley walls are an important source of coarse and fine sediment to the lower reaches of the three study streams.

On the Chilcotin Plateau, the main source of coarse and fine sediment is thought to be bank erosion (by meandering and by channel shifting) along the main rivers, and slope wash and other surface erosion processes.

The Big and Gaspard creeks are not known to support anadromous salmon. One falls in the lower Big Creek and a series of falls below the Little Gaspard confluence on the Gaspard Creek serve as barriers to anadromous salmon. Anadromous salmon have been reported in the mainstem of the Churn Creek, but only rainbow trout have been reported for West Churn Creek (Table 3). Rainbow trout are known to occur in the Big, Gaspard and Churn creeks. Electrofishing surveys conducted in 1996 reported frequent occurrence of rainbow trout in Big Creek (including many tributaries) and in Churn Creek (including West Churn Creek), but no other fish species (AIM 1996, Triton1997a,b,c). The FISS (1997) database reports the occurrence of rainbow trout in the Big and Gaspard creeks, and pink salmon, bull trout, coho salmon, rainbow trout, longnose sucker, chinook, longnose dace and sockeye salmon in Churn Creek. Bull trout are suspected to occur in Big Creek, but none have been reported.

Logging began in the Big, Gaspard and Churn watersheds prior to 1967 (Table 2). Harvest rates increased during the 1960's and 1970's and a steady rate of harvest has been maintained since the early 1980's. The percent area harvested within the three watersheds was greatest in Gaspard watershed at 19.70%, less in Big watershed at 8.78% and least in Churn watershed at 3.69%. However, due to the greater size of the Big watershed, the area harvested was greatest here at 208.4 km². Clearing from logging and farming occurred to the stream bank in some areas, but for the most part a riparian strip was maintained.

Logging history varies between tributaries (Table 2). In all sub-basins, the percent area harvested was less than 30%. In the Big watershed, the percent area harvested was greater than 20% in Twinflower and Mons sub-basins, between 10 and 20% in Tilton, Bambrick and Cooper sub-basins and less than 10% in Big and Rushes Lake sub-basins. In the Gaspard watershed, the area harvested was close to 20% for both the Gaspard and Little Gaspard sub-basins. In the Churn watershed, the area harvested was less than 10% in both the Churn and West Churn sub-basins.

Logging began in the Big Creek watershed before 1967, peaked in 1977 to 1986 and continues today. About 9% of the Big Creek watershed area has been cut. Equivalent clearcut areas are less than the total area cut because of hydrologic recovery, and are much lower than those required to alter the hydrologic regime.

Nearly all the harvesting in the Big Creek watershed occurs downstream of Groundhog Creek. As a result of this concentration of harvesting in the lower watershed, several tributaries to Big Creek have rates of forest harvesting which may affect their hydrologic regime. Tilton, Twinflower, Mons, and Cooper creeks all have rates of cut above 15% and may have increased peak flows from forest harvesting.

Tilton Creek has a total cut of 19.95%, with most harvesting between 1967 and 1986. The harvested areas have partially recovered and the equivalent clearcut area of the watershed is expected to be below the threshold where flood peaks are increased. Most of the cut blocks are distant from the stream and do not affect bank stability or LWD recruitment.

Twinflower Creek is the most heavily harvested tributary in the Big Creek watershed. Between 1967 and 1976 about 29% of the watershed was harvested. Hydrologic recovery since then has reduced equivalent clearcut areas, but for a number of years after 1976, following harvesting, we expect that flood peaks were increased.

Some channel instability may have resulted from the altered hydrologic regime in the late 1970's and early 1980's. Also, the middle reaches of Twinflower Creek were logged to their banks which will reduce bank stability and LWD recruitment in future decades.

Mons Creek is thought to have increased flood flows as a result of forest harvesting; 27.6% of the basin was logged between 1977 and 1992. Equivalent clearcut areas are similar to the percent harvested as little hydrologic recovery has occurred. Increased flood peaks may now be contributing to alteration of the channel in this tributary. Most of the harvested blocks are distant from the channel and do not affect bank stability or future LWD recruitment.

Cooper Creek, like Mons Creek, has recent harvesting, with about 15% of the basin cut since 1987. Little hydrologic recovery has occurred and equivalent clearcut areas are nearly equal to the percent cut. Harvesting levels are less than the threshold where changes to the hydrologic regime typically occur though minor changes in peaks flows may be observed, depending on the distribution of the cut blocks. There has been little streamside logging and bank stability or future LWD recruitment will not be affected.

Forest harvesting began in the Gaspard Creek watershed before 1967 (Table 1). Hydrologic recovery of the areas harvested before 1967 have likely decreased the equivalent clearcut area from the total area cut reported in Table 1. Harvesting was concentrated in 1977 to 1992 and it appears that equivalent clearcut areas did not exceed typical thresholds for altering the hydrologic regime during this period. Minor increases in peak flows may have occurred which may, in turn, have altered channel morphology. Harvesting has been concentrated away from channels and bank stability and LWD recruitment have been minimally impacted.

Little Gaspard Creek has been intensively logged since 1977. The total percent cut is 20.59% with most of logging occurring between 1977 and 1992. Hydrologic recovery of the earlier blocks has likely reduced the equivalent clearcut area to less than the total area cut, though harvest levels are near that which may result in increased peak flows. As a result, changes to channel morphology may occur in some reaches of Little Gaspard Creek. The logging has been concentrated away from channels and bank stability and LWD recruitment have been minimally impacted.

Churn Creek and its tributary West Churn Creek have total harvested areas that are less than those expected to produce alterations to the hydrologic regime. The total areas harvested are 3.69% and 7.67% of the Churn and West Churn creek watersheds. The cut is not concentrated along the channels and does not affect bank stability or LWD recruitment.

Cattle ranching is the most common agricultural activity in the area. Lands have been cleared for grazing and farming of food crops for cattle, often adjacent to streams, and in some areas no riparian strip has been maintained. Trials that are probably use by cattle or farm vehicles cross through streams in some locations. Some bank erosion has resulted from these crossings and from clearing of vegetation to the stream bank.

Colour air photographs 1:15,000 (1991, 1996) and 1:3,000 (1997) were acquired for the mainstem and major tributaries in Big, Gaspard and Churn creek watersheds. The 1:3,000 colour 35 mm photographs were acquired during a reconnaissance helicopter overflight of the watershed in September 1997. Russ Frith (Fisheries Biologist) and Robert Bocking (Senior Fisheries Biologist) were present on the flight.

Air photograph analysis techniques were used to describe habitat characteristics of the stream channel. Methods of fish habitat assessment are taken from Watershed Restoration Technical Circular No. 8 (WRP 1996a). Any deviations from the methods outlined in this circular are described below.

Stream reaches were assigned based on consistent channel form, gradient and confinement viewed in air photos (Appendix A). Profiles derived from analyses of stream length and elevation change were evaluated and reach breaks adjusted to match significant changes in slope where appropriate. Stream length and elevation measurements were taken from 1:15,000 TRIM maps where stream lengths were measured using a map wheel.

Information on the distribution of fish species in the watershed was compiled from FISS and recent reports. The following individuals were contacted for further information on fish distributions:

Don Cadden, Regional Fisheries Biologist, Prince George

Michael Parker, WRP Coordinator, MELP, Williams Lake

DFO

The results of the fish distribution information are presented in Tables 3 and 4.

Habitat assessments were conducted according to WRP Technical Circular No. 8 (WRP 1996a) with no deviations in methodology. A preliminary habitat evaluation was conducted using qualitative measures of stream habitat and disturbance that were derived from features visible on 1:3,000 air photographs (see Table 5). Using the information gathered on fish use in the Big, Gaspard and Churn creeks and information acquired from air photographs on the distribution and the degree of disturbances for fish habitat, reaches were prioritized for Level 1 Field Assessment. Areas of disturbance that were potentially caused by logging activity (i.e., cutblocks or roads) were given higher priority than areas disturbed by other activities (i.e., agriculture). Areas with logging to the stream bank were clearly impacted directly by logging. The presence of disturbances like log jams and sediment deposits (unvegetated bars, sediment wedges, mid-channel bars) instream sections with abundant cutblocks adjacent to the stream (but not necessarily in the RMA) were assumed to indicate logging impacts.

The results of the habitat assessments are summarized in tabular form (see Tables 5 and 6).

Digital forest cover and road mapping was provided by Ministry of Forests, Williams Lake. Stream Reaches and riparian segments were marked on 1:20,000 TRIM cover maps and digitized.

Stream reaches for potentially impacted areas of the Big, Gaspard and Churn creek watersheds are shown on the map presented in Appendix A and on the air photo mosaics in Appendix B. A total of 129 reaches were assessed for physical habitat features and fish distributions. The total length of stream assessed was 196 km. Stream profiles for each reach are shown in Figures 2 to 11. Many smaller creeks, mostly unnamed, were not assessed due to poor visibility either from crown closure or small size. The lower portion of some of these creeks will be assessed during Level 1 Assessments where possible.

Anadromous salmon occur in the lower Churn Creek and possibly in the lower Big Creek. Access to the Big Creek is blocked to anadromous salmon by a falls at approximately 11 km upstream from the creek mouth (Reach 8; see Appendix A). Three barriers are identified in the FISS database for Gaspard Creek. Rapids are marked on TRIM for Gaspard Creek near the creek mouth and implies Gaspard is not accessible to anadromous salmon. The first barrier on Churn Creek occurs at approximately 21 km upstream from the creek mouth allowing access to anadromous salmon in the lower creek mainstem. Aquatic biophysical maps for the area indicate that sockeye, pink, coho and chinook salmon have been observed in the lower Churn Creek. Pink salmon are the only salmonid species observed with regularity, on odd years, in the Churn Creek (Table 3). The distribution of resident trout in the study area is known to a limited degree. Electrofishing sampling in 1996 and other studies throughout the Big, Gaspard and Churn creek watersheds indicate rainbow trout are widely distributed in these systems (Triton 1997a,b,c, AIM 1996, Table 4). Bull trout are known to occur in the Churn Creek mainstem and are suspected to occur in the Big Creek watershed (FISS). Sample sites for which rainbow trout were reported occur in Big Creek (Reach 1 and 26), Bambrick Creek (Reach 2, 5, 11, and 14), Twinflower (Reach 5 and 6), Rushes Lake (Reach 1, 2, and 3), Cooper (Reaches 3 to 6), West Churn Creek (Reach 1, 5, 11, and 13) and in most reaches of Gaspard Creek (Table 4).

Rainbow trout adults prefer riffles above pools for spawning and typically migrate into smaller tributaries from stream mainstems or lakes in spring to spawn (Scott and Crossman 1973). Juvenile rainbow trout prefer smaller streams with depths less than 80 cm and velocities ranging from 5 to 40 cm/s (Keeley and Slaney 1996). Cover and the lack of fines is also important to rainbow trout juveniles. Bull trout juveniles prefer shallow waters (30 - 40 cm) with cover and are often located close to wood or bottom substrates, primarily cobble and boulder (Baxter and McPhail 1996). Bull trout spawning habitat is very specific and once a site has been selected it is used from year to year with little variation. Bull trout spawning has been observed to occur in depths from 10 to 93 cm with pools, undercut banks and wood as the main cover. Low velocities are preferred ranging from 2.0 to 99.0 cm/s. Substrate preferences for spawning are gravels and cobbles (Baxter and McPhail 1996).

Logging occurred in the Big, Gaspard and Churn watersheds beginning in the early 1960's. The area logged since this time was greatest in Big (208.4 km²), less in Gaspard (178.5 km²) and least in Churn (35.3 km²) (Table 2, Appendix A). Logging in the Big and Churn watersheds has been restricted to the lower portions of these large watersheds. Due to their large areas, the percent of area logged in Big and Churn watersheds was less than 10% compared to close to 20% in Gaspard watershed (Table 2). Logging to the stream bank or within the RMA was most extensive in Big Creek, Twinflower Creek, Cooper Creek, Gaspard Creek and Little Gaspard Creek, whereas roads within the RMA were observed in most sub-basins. Erosion as a result of logging, grazing or other agricultural activities was commonly observed in air photographs from all sub-basins. Stream habitat in the watershed is generally lacking in LWD, pools and off-channel or side-channel habitats. Extensive riffle habitat and aggradation of boulder/cobble sediments were the most common impacts from logging observed throughout the watershed. The lack of LWD, pools, and off-channel and side-channel habitat are mainly limitations for rearing of rainbow trout, the target species for the region, although these habitats are also of importance to bull trout. Bull trout have not been captured throughout the study area, but they have been observed in Churn Creek and may occur in Big Creek and associated tributaries. Their potential occurrence throughout the Big and Churn watersheds should be kept in mind. The aggradation of sediments in the lower portions of tributaries may also impact the spawning habitats for rainbow and bull trout by limiting access to good spawning habitat or by degrading spawning habitat.

The main harvesting-related impacts in the Big Creek basin occur in Tilton, Twin, Mons and Cooper Creeks. Farming and ranching also affect channel morphology and hydrology. Summer and winter low flows can be reduced by water extractions for irrigation, stream banks can be destabilized by cattle crossing streams and the elimination of the source of LWD for stream channels by clearing riparian zone of forest for agricultural use. Farming and ranching are mostly located in the middle reaches of Big Creek (Reaches 12 to 24).

The mainstem of Big Creek has been divided into 33 reaches from the mouth of the creek at the Chilcotin River to Groundhog Creek tributary. No historic or recent logging occurred upstream of this point and, therefore, no assessments were conducted upstream of Reach 33.

Most reaches (31 of 33) within Big Creek possess low slopes of less than 5% (exceptions are Reach 5 at 6.8% and Reach 8 at 8%) and 23 of 33 reaches have slopes less than 1% (Table 5 and Figure 2). Higher slopes occur in the first eight reaches where the creek is less sinuous and riffles are frequent. A chute occurs in Reach 8 which from visual observation appears to be a barrier to migratory salmon (Triton 1997a). Above this point, the stream meanders with various degrees of sinuosity, where exposed bars and off-channel habitat are common. Pools are expected to be abundant in this type of habitat given the size of the stream (10 - 15 m wide), the low gradient and sinuous pattern and the lack of riffles. Most reaches were assigned a 2 (26-50% pool area) or 3 (51-75% pool area) pool code. In general, woody debris and cover is lacking in this creek. The source of wood for streams has been reduced by logging and agricultural clearing to the stream bank. The main benefit of LWD in streams is for juvenile rearing habitat of rainbow trout and possibly bull trout.

Disturbances from logging and agriculture are common along the banks of Big Creek (Table 5). Land cleared by logging adjacent to the creek or impacting the stream banks was observed in 9 reaches (Reach 12, 17, 18, 19, 21, 22, 23, 24, and 28). Land cleared for agricultural use was seen in 6 reaches (Reach 14, 20, 23, 28, 29, and 30) although logging licences may have been issued for clearing the land initially (see Big Creek, Pages 1 - 61 of 61 in Appendix B). Not all cleared land adjacent to creeks observed in 1997 aerial photographs was recorded as a cutblock in MOF forest cover data. Land clearing to the stream bank increases the potential for erosion and the introduction of sediments to the stream. Logging and agricultural clearing probably increased sediment load to the Big Creek, but when this occurred and where sediments accumulated is unknown.

Tilton Creek was divided into two reaches (Table 5 and Figure 3). No logging or agricultural activity adjacent to the stream was apparent in photographs of the mainstem although close to 20% of this sub-basin has been harvested since 1967 (Table 2). Cover and LWD are limited and road access is poor (Table 5). The lack of LWD is probably natural given the lack of harvesting in the riparian zone (see Tilton Creek, Pages 1 - 4 of 4 in Appendix B). Rainbow trout are suspected to occupy Tilton Creek and the addition of cover could increase the carrying capacity of this creek.

Mons Creek was divided into two reaches beginning at the upper end of Mons Lake (Table 5 and Figure 5). Multiple channels were observed in both reaches with possible access problems to off-channel habitat (Table 5, see Mons Creek, Pages 1 - 2 of 2 in Appendix B). Pools are abundant and off-channel habitat is abundant. Few LWD occur which may limit the availability of cover. The lack of LWD and access problems to off-channel habitat are probably natural due to the lack of logging in the vicinity of Reaches 1 and 2. Road crossing are the main potential disturbance in this area. The quality of habitat for rainbow trout juveniles would be improved by the addition of cover and increased structural complexity in the creek.

Logging occurs adjacent to Mons Creek further upstream, but this area was not photographed due to difficulty in locating the stream from the air. Approximately 28% of Mons Creek sub-basin has been logged (Table 2), but most of the cut has occurred higher in the watershed (along the mainstem and on adjacent slopes of a major tributary to Mons Creek - an unnamed creek) and most harvest blocks leave a riparian strip. There is one small cutblock (1970 year of harvest) at the lower end of the unnamed tributary with cutting to the stream bank. The stream channel adjacent to cutblocks in the upper Mons mainstem appears to be disturbed by the accumulation of sediments. Logging in the watershed at and above this point may have increased the sediment load of Mons Creek and contributed to the observed sediment accumulation. Field assessment of this region is recommended to confirm the disturbance and assess the site for other logging related and other disturbances.

A total of six reaches were designated on Twinflower Creek (Table 5 and Figure 4). Agricultural land use adjacent to the stream occurs in Reach 2 and logging or fire has occurred to the stream bank in all other reaches (see Twinflower Creek, Pages 1 - 10 of 10 in Appendix B). Most logged areas within the RMA are in an early stage of regeneration (Table 7). Beaver dams and road crossings were disturbances observed in Reaches 4, 5 and 6 (Table 4). LWD is abundant in Reaches 3 and 4 and sparse elsewhere (Table 5). A moderate abundance of off-channel habitat occurs in Reaches 3 and 6 and is fair to poor in other Reaches. Cover and LWD are lacking in Reach 2. Instream cover and habitat complexity are important habitat for rainbow trout juveniles and adults.

Disturbances in Reach 2 are predominantly due to agricultural activities. The roads, buildings and tracks from harvesting equipment in fields adjacent to the stream suggest human activities inside and outside the RMA of Reach 2 are ongoing. The appearance of bare stream banks suggests erosion of sediments into the stream. Despite the intensity of logging (some within the RMA) upstream from Reach 2, LWD is common and abundant in some locations. Beaver activity may contribute LWD. Ponds, possibly formed by beaver dams, are abundant in Reaches 4 and 6 and some ponding occurs in Reach 3. However, ponding begins above old road crossings in Reaches 4 and 6. Poorly constructed roads may restrict stream flow impeding access to spawning areas by adult trout (rainbow and possibly bull trout) or restrict movement of rearing juveniles.

Rushes Lake Creek was divided into three reaches starting at the confluence with Big Creek (Table 5, Figure 6, see Rushes Lake Creek, Pages 1 - 4 of 4 in Appendix B). A road crossing occurs in Reach 1. Access to backwater habitat appears restricted in Reaches 1 and 2 (Table 5). Reach 3 is a lake. LWD and off-channel habitat is lacking instream reaches. Instream cover from LWD and to a less degree off-channel habitat are important habitat for rainbow trout juveniles.

Logging to the stream bank has resulted in a lack of vegetation source for LWD and the potential for bank erosion. The lack of LWD and riparian cover suggests holding and rearing habitat is poor for trout. Fish habitat in both reaches could potentially benefit from bank stabilization, planting of trees and other vegetation in the riparian zone and the placement of LWD and/or boulders in the stream to increase habitat complexity. A field visit is recommended to further assess the potential for habitat improvements.

Bambrick Creek was divided into 14 reaches from its confluence with Big Creek (Table 5 and Figure 7). No riffles were observed in 1:3,000 aerial photographs (35 mm) and stream habitat was dominated by pools and runs (Table 5). No barriers were observed except for one possible beaver dam in Reach 4.

The disturbances observed included eroding banks (Reaches 1, 9, 12, and 13), unvegetated bars (Reach 10) and multiple channels (Reach 7) (Table 5 and see Bambrick Creek, Pages 1 - 22 of 22 in Appendix B). Large woody debris was observed to be abundant in Reaches 2, 4, 6, 10 and 11. Off-channel habitat was commonly observed, but was only observed to be abundant in the multiple channel habitat of Reach 7. Agricultural land occurred adjacent to the creek in Reaches 1, 3, 5 and 9. Logging adjacent to the creek was observed in Reach 8 and may be present in Reaches 1 and 3 intermingled with agricultural land. Forest cover data indicates harvesting occurred in small patches and away from the stream banks in lower reaches (Reaches 1, 2 and 4) although not all areas that appeared to have been harvested in air photographs were present in the forest cover data. Larger cutblocks occurred higher in the watershed adjacent to Reaches 5, 8 and 11, but mostly away from the Bambrick mainstem. Historic logging may have contributed to disturbances in Reaches 10 and 12 although stream habitat in upper reaches was similar whether vegetation appeared harvested or not. Disturbances in Reaches 1 and 9 are related to agricultural activities. Impacts in Reaches 9 and 13 are probably due to natural causes. The more abundant logging at higher elevations may have contributed sediment to streams and increased flows during peak rainfall in some years. This logging activity may have contributed to the accumulation of sediments farther downstream. Areas with little or no stream bank vegetation or sediment bars and wedges are probably impaired for rearing and holding by rainbow trout.

Willan Creek, a tributary to Bambrick Creek, was not photographed at 1:3,000, but rainbow trout are known to occur in this creek (Triton 1997b) and logging activities occur in the vicinity. Extensive logging, some to the stream bank, occurred in 1985 on the north side of the lower portion of Willan Creek (below lake) and on tributaries at the top of the watershed. Logging impacts may have occurred in this area, but due to the low impacts observed in the Bambrick sub-basin in general and the importance of assessing other areas of the watershed, no field assessments were recommended for Willan Creek.

Cooper Creek was divided into seven reaches beginning at the confluence with Big Creek. Off-channel habitat and LWD was lacking in most reaches except for Reach 2 where a beaver dam was observed (Table 5 and see Cooper Creek, Pages 1 - 7 of 7 in Appendix B). Road crossings occur in Reaches 3, 5 and 6 with a potential problem culvert in Reach 6. Agricultural land adjacent to the creek is common and occurs in Reaches 1, 3 and 5. Logging adjacent to the creek was apparent in Reach 7.

Logging activity in this sub-basin reduced the source of wood adjacent to streams and contributed to the lack of LWD and riparian cover. Cover from LWD and riparian vegetation is particularly important habitat for juvenile trout (rainbow and possibly bull trout).

Gaspard Creek was divided into 28 reaches from below the confluence of Little Gaspard Creek up to Gaspard Lake. The high valley walls and unstable substrates in the lower Gaspard Creek (below Reach 1) made assessment and rehabilitation work difficult in this area. Overview assessments of fish habitat were not conducted in this portion of the creek.

Eroding banks were observed in Reaches 1, 11, 12, 15, 16, 19, 24 and 28 (Table 5 and see Gaspard Creek, Pages 1 - 51 of 51 in Appendix B). Bare soil and erosion are natural in Reach 1, 11, 12, 15, 16, 19 and 24, although paths or gravel roads, probably used by migrating cattle or ranch vehicles, occur close to the stream in Reach 11, 19 and 24 (see Pages 35 and 46 of 51 for Gaspard Creek in Appendix B) and cross the stream in some locations (Reach 11, Page 20 of 51 and Reach 23, Page 45 of 51 for Gaspard Creek in Appendix B). For reaches with eroding banks, logging is evident adjacent to Reaches 1, 11 and 19, but not within the RMA. A small cutblock can be seen adjacent to Reach 11 in the 1:3,000 photographs of Gaspard Creek that must have been harvested since 1993 (1993 1:15,000 air photographs and MOF fc1 data show no logging in this area - see Page 20 of 51 for Gaspard Creek in Appendix B). Gravel roads and cleared land contribute to bare soil and possible erosion in Reach 28 at the outflow to Gaspard Lake.

Access to backwater channels may be restricted in Reaches 6, 10 and 22 although the abundance of habitat that may be inaccessible to fish in these reaches is small compared to the abundance of off-channel habitat that exists in Gaspard Creek. Multiple channels were observed in Reaches 5, 7, 8, 20 and 21 and beaver dams in Reaches 13, 14, 16, 18, 19, 20, 22, 25 and 26 (Table 5). A 9 m falls occurs downstream of Reach 1, a 2 m falls occurs within Reach 2 and a 6 m falls occurs within Reach 3. According to FISS there is a cascade chute at the mouth of Gaspard Creek. Access to Reach 1 by anadromous salmon is clearly blocked by the 9 m falls if not at the mouth of the creek. A 2 m or greater falls is probably sufficiently high to block upstream passage of rainbow trout. Beaver dams may impede adult passage of rainbow trout whereas juveniles may be able to pass. Abundant LWD was observed only in Reaches 14 and 18 where beaver activity was evident in both reaches. A cutblock to the stream bank has been a source of abundant LWD in Reach 18. Off-channel habitat varied by reach and ranged from none to abundant. In Reaches 8, 9, 13 and 14 off-channel habitat was abundant. Low water may be a problem for rainbow trout during summer low flows, especially with the diversion of water from Gaspard Lake to the Little Gaspard Creek via Stobie Lake (Brian Chapman, MOELP, personal communication). Off-channel habitat and beaver dams may provide refuges during these periods. In some reaches, habitat complexity and cover are lacking which are particularly important for rearing by young rainbow trout.

Logging adjacent to the creek occurs in most reaches up to Reach 24 (see Map 1 in Appendix A). Harvesting to the stream bank or within the RMA occurs in Reaches 8, 9, 12, 13 and 20 and roads within the RMA are common. Paths or gravel roads used by cattle or vehicles are common within the RMA and in some locations cross through the creek (for example, see Reach 13, Page 24 of 51 for Gaspard Creek in Appendix B). Natural bare soil is also common. Logging, cattle ranching and natural sources are all potential contributors to sediment in the creek. Despite the multiple sources of sediments throughout Gaspard Creek, sediment bars are not common. Beaver dams are common throughout and do cause sediments to accumulate downstream of dams. For example, a large beaver dam in Reach 14 has created a sediment depositional zone downstream of the dam such that most of the stream has infilled and vegetation has established (see Page 26 of 51 for Gaspard Creek in Appendix B). A number of small rivulets maintain the stream flow downstream. This amount of sediment deposition does not appear unusual given the abundance of natural sources of sediments upstream.

The Little Gaspard Creek was divided into 20 reaches from its confluence with Gaspard Creek. Multiple channels were observed in Reach 2, eroding banks in Reaches 4, 5, 6, 7, 19 and 20, road crossings in Reaches 11 and 18 and a dam in Reach 20 (Table 5 and see Little Gaspard Creek, Pages 1 - 17 of 17 in Appendix B). Sand bars were evident in Reach 15 and sediment deposits at bends may occur in Reach 16. Road crossings and the dam at the mouth of Stobie Lake may be barriers to migration. LWD was observed to be abundant in Reaches 2, 5, 6 and 12 and off-channel habitat ranged from poor to good. In general, juvenile rearing habitat for rainbow trout appears to be of good quality, but could benefit from additional cover in some locations. In other areas, cover from LWD appears to be lacking and the addition of LWD would improve cover and increase habitat complexity.

The abundance of LWD and multiple channels observed in Reach 2 is not a system wide event, but a localised disturbance. Bare soil is observed adjacent to the mainstem and a side channel is present. No logging or roads occur in the area and the disturbances appear to be natural (see Page 2 of 17 for Little Gaspard Creek in Appendix B). Large areas of bare soil that appear to be natural occur adjacent to Reaches 5, 6, 7, 8 and 9. Most harvesting occurs away from the stream except for encroachment within the RMA in Reaches 8, 11, 12, 13 and 20 (see Riparian Table 7). The lack of sediment deposits in streams suggests that sediment input from adjacent poorly vegetated lands is not a major disturbance. Although bank stabilization may be beneficial in areas with bare stream banks, these areas are mostly in localized areas. Extensive stabilization of adjacent lands is not recommended. In general, Little Gaspard Creek lacks cover and habitat complexity, particularly important for rearing by rainbow trout. This impact would be off-set by the presence of stream-side boulders and cobbles. Field assessment is required to assess the quality of stream-side habitat not visible in photographs.

A water diversion channel splits off from the Little Gaspard Creek at the Reach 4 and 5 break. Flows in Reaches 1, 2, 3 and 4 are expected to be impacted by this diversion. Also, the Gang Ranch has a water license to supplement Little Gaspard Creek with water from Gaspard Lake via Stobie Lake at the upper end of the sub-basin. However, water from Little Gaspard Creek is diverted to Gang Ranch further downstream impacting the water available for fish habitat in this creek. Hydrological data indicates water flows in the Gaspard watershed have been significantly reduced by water use for agriculture. The amount of water during summer low flows may be the limiting factor for rainbow trout populations in the Little Gaspard Creek. The abundance of pool habitat and access to lakes will be important for survival during this period. The availability of good quality rearing habitat for juvenile rainbow trout in Little Gaspard Creek depends on the provision of adequate water flow, particularly during the summer low flow period.

The main impacts of logging on Little Gaspard Creek are contribution of sediments from clearcutting at higher elevations and bank erosion at road crossings. However, little evidence of sediment deposits in streams occurs. Cattle crossings, water diversion channels and natural erosion also contribute to degraded fish habitat. The provision of adequate water flow and additions of habitat cover from LWD are recommended for habitat improvements in this creek.

The mainstem of the Churn possesses high valley walls with unstable lacustrine sediments. Logging occurs within the watershed, but mainly away from the Churn mainstem (Appendix A). Logging activity within the watershed may have contributed to sediment deposits in the lower mainstem where pink salmon spawn, but natural sources and mining activity in Fairless Creek are also potential contributors. The recent designation of the lower Churn Creek watershed as parkland will result in no additional logging in the area and will allow the impacted stream and adjacent lands to recover.

The lower 0.9 km of West Churn was not assessed due to high valley walls, entrenchment and lacustrine sediments. A 6 m falls is present at 0.9 km. The creek above the falls was divided into 14 reaches. Eroding banks were observed in Reaches 4, 9 and 11, sediment bars in Reaches 9 and 11 and a beaver dam may be a barrier to migration in Reach 11 (Table 5 and see West Churn Creek, Pages 1 - 13 of 13 in Appendix B). Logging occurs in the vicinity of West Churn Creek, but not within the riparian zone. The accumulation of sediments evident in some reaches may have been enhanced or caused by sediment inputs of logged areas in the watershed. Logging has occurred adjacent to reaches with sediment deposits. A recent cutblock is visible adjacent to Reach 9 in 1:3000 air photographs and cutblocks of the early 1990's occur adjacent to Reaches 4 and 11. In general, West Churn Creek is lacking in cover and would benefit from the addition of LWD. Cover and habitat complexity are important for adult holding and juvenile rearing of rainbow trout.

Disturbances to fish habitat on Fairless Creek have been reported (AIM Ecological Consultants Ltd. 1996). These impacts are probably due to mining activities and thus this creek is not eligible for further work. No logging has occurred in this sub-basin (Appendix A, Map 3).

Bull trout were reported from fish sampling in Dash Creek and the creek is apparently impacted by human activity. However, no logging activity has been reported for this sub-basin (Appendix A, Map 3) and, therefore, this creek is not eligible for further work.

The following stream impacts were identified for the Big, Gaspard and Churn creek watersheds:

Table 6 summarizes potential instream rehabilitation opportunities for the Big, Gaspard and Churn creek watersheds. All high priority sites and most, but not all, medium priority sites are recommended for Level 1 Field Assessments. Sites were selected based on their potential for improvement of juvenile rearing or adult spawning and holding habitat, and where impacts from logging were probable. Agricultural land use and logging often occurred together. Due to the difficulty in identifying the cause of a disturbance in areas with multiple sources, logging was assumed to be a potential cause of the disturbances when logging in the area had occurred.

Reaches were assigned no priority if no natural or anthropogenic disturbances were observed (Table 6). Reaches that were minimally disturbed or had poor access were given a low priority for assessment. Reaches that were significantly disturbed by logging or agricultural activities, had good access and warranted rehabilitation were designated as medium or high priority. The high priority was assigned to reaches with the greatest potential for improvement of fish habitat due to their proximity to other valuable habitat (off-channel habitat, tributaries, pools).

A total of 16 reaches were designated as no priority (Reaches 1 - 13, 15, 17, and 25), nine reaches were given a low priority for assessment (Reaches 14, 16, 20, 21, 22, 29, 30, 31, and 33), three reaches were designated as medium priority (Reaches 18, 19, and 28) and four as high priority (Reaches 23, 24, 26, and 32). Level 1 Habitat Assessments and fish sampling is recommended for all medium and high priority sites on Big Creek. All medium priority reaches had logging to the stream bank or were strongly impacted. Although some low priority sites had logging to the banks, impacts due to ongoing agricultural use reduced the potential for improvements.

A low priority was assigned to both reaches and no sampling was recommended (Table 6). Although cover was lacking and fish habitat could be improved with the addition of cover, no logging impacts were evident and access was poor.

A road crossing in Reach 2 requires checking. Level 1 Habitat Assessments are recommended in both Reaches (Table 6). Although no logging occurs in the vicinity of these two reaches, logging near and to the stream banks has occurred further upstream and may have impacted these two reaches. Accessible areas of Mons Creek in the vicinity of cutblocks further upstream and outside of the area photographed are recommended for field assessment.

High priority for rehabilitation was assigned to Reaches 2, 3 and 6 where logging occurs adjacent to the stream, habitat potential is high and cover is lacking. A medium priority was assigned to Reach 5 where a road crossing occurs and potential for LWD development exists. Although some logging has occurred adjacent to Reaches 1 and 4, no impacts were observed and a low or no priority was assigned. In general the greatest potential for habitat improvement is the addition of cover with LWD. Some bank stabilization would be beneficial in localized areas of Reaches 5 and 6, mainly road crossings. Agricultural land use occurs adjacent to Reach 2 and reduces the potential for improvements in this Reach.

Non-lake reaches (Reaches 1 and 2) were assigned a high priority due to their potential for habitat improvements and the presence of logging and agricultural activities adjacent to the creek (Table 6). Although Reaches 1 and 2 appear to be man-made, some rehabilitation opportunities may still exist.

Areas with logging or agricultural disturbances with potential for habitat improvements were assigned a high or medium priority. Reaches 5 to 13 were assigned high priorities for field assessment due to the presence of logging in the vicinity and the presence of disturbances including multiple channels, eroding banks, sediment bars, and lack of cover. Reaches 1 and 2 were also disturbed, but mainly due to agricultural activity and were assigned a medium priority. Habitat surveys and fish sampling are recommended for all medium and high priority reaches (250 m or greater sections). Areas with minimal or no disturbances were assigned a low (Reach 3) or no priority (Reaches 4 and 14).

Reaches 1, 3, 4 and 6 were given a high priority and recommended for Level 1 Habitat Assessments and fish sampling (Table 6). The remaining reaches were assigned a low priority. The high priority reaches showed the greatest potential for habitat improvements with the lack of cover from LWD being the major factor. Despite the prevalence of agricultural activity adjacent to the stream in most reaches, logging activity in the sub-basin may have reduced the availability of LWD to the stream. Addition of LWD would improve cover for juvenile rearing and habitat complexity for spawning and adult use.

Reaches 6, 9, 11, 12, 13, 19, 23 and 27 were assigned a high priority and Reaches 12, 14 and 20 were assigned a medium priority. All high priority reaches and Reach 12 were recommended for habitat assessment and fish sampling (Table 6). Reaches 14 and 20 were considered less of a priority due to a lower level of impact and close proximity to other assessment reaches. Small impacts were reported for Reaches 7, 8, 10, 15, 18, 24, 25 and 28 and these reaches were assigned a low priority. The remaining reaches were not impacted or experience natural, low level impacts and were assigned a no priority.

The main impacts on fish habitat quality, particularly for rainbow trout juveniles, are lack of LWD. Most high priority sites would benefit from the addition of LWD. There are some areas with localized erosion and road crossings that may benefit from bank stabilization. Access to off-channel habitat may have potential for improvement in some reaches.

Reaches 4, 5, 6, 7, 11, 18, 19 and 20 were assigned a high priority and are recommended for habitat assessment and fish sampling (Table 6). The remaining reaches were assigned a low or no priority and fish sampling is recommended in all reaches.

The diversion of water from Reaches 4 and 20 may impact fish habitat. The water level is a main determinant of habitat area and quality. Bank erosion problems require field assessment in Reaches 5, 6 and 7. Stabilization of banks with boulders and tree plantings may be required. Reaches 11, 17 and 19 require field assessment mainly because the stream was not visible in photographs. Road crossings in Reaches 11 and 18 require checking for erosion or channel alterations and the dam at the mouth of Stobie Lake in Reach 20 should be checked.

No further habitat assessments or fish sampling are recommended.

A high priority was assigned to Reaches 9 and 11 and Level 1 Habitat Assessment and fish sampling are recommended for these reaches only (Table 6). Reach 4 was assigned a medium priority and the remaining reaches were designated no priority. Logging is removed from the stream and although sediment deposits are present within the stream, these sediments appear to add habitat complexity and improve fish habitat. Cover is lacking in some reaches and fish habitat would be improved by the addition of LWD.

No habitat assessments or fish sampling are recommended.

No habitat assessments or fish sampling are recommended.

A total of 19.5 km of stream in 49 reaches is recommended for survey in the Big, Gaspard and Churn watersheds at the Level 1. There are also three Reaches in the Big Creek and 21 Reaches in Little Gaspard recommended for site assessments or fish sampling.

Possible logging impacts occur in all watersheds as well as impacts from cattle ranching and natural causes. The greatest impacts or factors contributing to low habitat quality were the lack of LWD and other sources of cover in creeks, lack of riparian cover and source for LWD and bank erosion. Impacts included logging on adjacent lands to the stream, logging and roads within the riparian zone and road crossings. The general lack of cover and stream-side structures in all three watersheds is particularly important for juvenile rainbow trout although LWD is important as cover for adults in deeper pools. The addition of structures like LWD would also increase habitat complexity leading to more pool habitat and velocity refuges, important to young and juvenile salmonids. Field assessments are necessary to assess the quality of habitat and degree of habitat complexity at greater detail.

Colour photographs at 1:15,000 (1991; 1996) and 1:3,000 (1997) scale were acquired for analysis. Air photograph analysis techniques were used to assess the riparian zone. Methods of riparian assessment were taken from WRP (1996b). Any deviations from the methods outlined in these reports are described below.

The stream banks within a reach were delineated into segments with similar riparian structure and the banks on either side of the stream were assessed separately. The 1:3,000 air photographs were mosaiced and viewed. Interpretation was supplemented by stereo viewing of 1:15,000 air photographs. Mosaics of the 1:3,000 air photographs were marked and riparian segment divisions transferred to TRIM and forest cover maps for digitizing. The riparian zone was defined by the area from the stream bank to a distance of 20 to 100 m from the stream, depending on the stream classification (WRP 1996b), and is referred to as the riparian management area (RMA).

Each riparian segment was assigned a code, a combination of the reach number and a letter code. For example, Riparian Segment 1A refers to the first riparian segment of Reach 1. The riparian segment on the lower bank of the reach starting at the lower reach break was assigned the letter A and each successive riparian segment on the lower bank was assigned a letter code progressing through the alphabet (A-Z). Upon reaching the upper reach break, segment letters were continued on the upper side of the reach starting again at the lower reach break and progressing up to the upper reach break (see Table 7 and Appendices A and B).

Riparian segments were assessed for structural stage, RMA class, reserve and management zone widths and assigned a visitation priority (Table 7). The structural stages used are based on those defined in the WRP (1996b) manual with some exceptions. Three new or modified structural stage classes were defined for (1) industrial areas (Ind), (2) shrub dominated communities (SHR) and (3) herb dominated communities(SHRh). Detailed definitions are provided at the bottom of Table 7. The RVC label was not included due to the limited value of this parameter in determining riparian function. The presence of conifers and maturity of trees are all that is necessary for determining the potential contribution of LWD from the riparian and the ability of the riparian community to filter sediments and stabilize banks. Species specific factors are rarely important at the overview level of assessment. Knowledge of understorey species is, however, important at the site level for the preparation of prescriptions.

The approach and decision criteria for prioritizing riparian segments for visitations and assessment of restoration concerns followed the Riparian Assessment and Prescription Procedures Field Guide (WRP 1996b). Riparian segments were assigned a high priority for site visitation if logging related disturbances were observed (i.e., harvesting within the RMA, roads within the RMA, bank erosion, sediment accumulation and lack of LWD) and the reach was known or suspected to be fish bearing.

Digital forest cover and road mapping was provided by the Ministry of Forest, Williams Lake. Stream Reaches and riparian segments were marked on TRIM maps and digitized.

Riparian segments and management areas (RMA) are illustrated on maps in Appendix A. A total of 517 riparian segments have been designated and defined based on vegetation, logging history and management interpretation and assigned a site visitation priority.

Riparian segment assessments and site visitation priorities are summarized in Table 7. There were 517 riparian segments defined. The dominant structural stage was initial stage (207) followed by mature forests (123), young forests (91), herb-shrub stage (75) and pole-sapling stage (21). Riparian segments were assigned structural stage, RMA class, reserve and management zone widths and a priority for visitation. Restoration opportunities were also noted.

Logging has occurred in Big, Gaspard and Churn creek watersheds beginning prior to 1967. Logging and road construction within the RMA occurs in most sub-basins. Bank stability, sediment input and lack of LWD are the main concerns. Cattle grazing and agricultural activities are also contributing factors to stream habitat degradation. The main riparian functions affected in these watersheds were stream shading, riparian cover, bank stability and source of LWD. Stands that were cut over 20 years ago are in various stages of regeneration, but represent a small portion of the total area cut.

Riparian forest vegetation is naturally sparse in this semi-arid region, so natural erosion is widespread. Agricultural and logging activities and roads occur within the riparian zone of Big Creek. Many riparian segments in Reaches 18-25 have been logged and are of high priority for the assessment of riparian vegetation (Table 7). Roads, grazing and farming impacts are of concern in Reaches 19-24, 28, and 29. Some priority sites are riparian segments where vegetation is lacking and banks are unstable; riparian segments 10A, 10B, 12A, 13B, 14A, 14C, 14E, 16B, 16D, 17B, 18C, 18F, 18H, 18K, 19A, 19D, 19F, 19I, 20B, 22B, 23C, 23H, 24E, 26B, 26D, 26H, 26J, 28F, 29C, 31B, 33A, 33D. In other riparian segments selected as priority sites, grasses occur, but no shrubs or trees: riparian segments 19G, 20A, 20E, 21A, 21C, 21E, 21F, 21H, 21J, 22E, 23A, 23G, 23K, 24A, 24C, 28C, 28H, 29D, 30A and 30C. The lack of riparian vegetation reduces bank stability, and impairs the LWD supply to the creek and shading function of trees. The remaining riparian segments with priorities for field assessments have impaired riparian functions due to roads or logging within the RMA. In these areas riparian function is impaired, but not entirely as some trees still stand. Partial reduction of bank stability, shading, LWD supply and sediment filtering occurs in these segments.

The vegetation in the riparian zone of Tilton Creek is mainly INIT, SHR and MFc. The vegetation appears to be natural and not the result of logging (see Tilton Creek, Pages 1 - 4 of 4 in Appendix B). Some natural erosion is evident and access is poor for visitation. No priorities were assigned for visitation. In riparian segments with INIT structural stage, riparian cover, source of LWD and shading are lacking. Segments with SHR structural stage are also lacking for a source of LWD. The lack of tress and shrubs in the riparian zone may have reduced bank stability and increased the potential for erosion in some locations (Riparian Segments 1A and 2D).

Grazing impacts are evident in most of the RMA of Reaches 2 and 3 (see Rushes Lake, Pages of 1 - 4 of 4 in Appendix B). Roads are common in the RMA of all Reaches. Initial successional stages are dominant with some shrubs and young trees. Roads within the RMA are the most common impact in riparian segments of Rushes Lake Creek causing loss of riparian structure and potential erosion. The loss of shrubs and trees due to roads results in lower stability of stream banks, loss of LWD supply, reduction of sediment filtering functions and less shading.

Logging and farming impacts are apparent in this creek. Farming occurs in Reaches 2 and 3 and grazing impacts are evident in Reach 2 (see Twinflower Creek, Pages 1 - 10 of 10 in Appendix B). Logging within the RMA was observed in Reaches 2, 3, 5, and 6. Roads occur within RMA of these reaches, as well as industrial activities and several stream crossings. Thus, riparian segments within Reaches 2, 5, and 6 received high priority for the assessment of riparian

vegetation, and most other segments were assigned medium priority (Table 7). The lack of stream-side vegetation from farming, grazing and logging has impaired the shading and cover function of the riparian zone and decreased the availability of trees for LWD. Roads within the RMA decrease the abundance of mature trees in the riparian and increase stability of the soils. Most of the impacted riparian segments had structural stages of INIT or PSc; thus, lacking mature trees.

Sections of Mons Creek upstream from Reaches 1 and 2 have been logged. Grazing and perhaps farming are present in the riparian zone of Reach 1, while a large fen, a valuable wildlife habitat, is located in Reach 2 (see Mons Creek, Pages 1 - 2 of 2 in Appendix B). Only one segment (1D) is of major concern due to intensive erosion (Table 7). The riparian zone farther upstream is in the vicinity of previous logging and the riparian vegetation function may have been impacted by logging. This area was not assessed at the watershed level due to the lack of 1:3000 photographs, but should be assessed in the field.

Grazing impacts are evident in Reaches 8-14. Disturbance of the riparian zone due to farming occurred in Reaches 3 - 5 and 9 (see Bambrick Creek, Pages 1 - 22 of 22 in Appendix B). Logging on a moderate scale occurred on adjacent slopes, but not within the RMA. Several riparian segments were assigned a high priority for visitation; Reaches 1, 3, 5, 8-10, 13, 14 (Table 7). The majority of riparian segments with impacts were assigned a dominant structural stage of INIT indicating that few mature trees were present in the riparian zone. Riparian functions that were impaired included shading, riparian cover, and a source for future LWD. No bank erosion was noted, but the lack of shrubs and trees impairs bank stability and increases the potential for erosion and the increased input of sediments into the stream.

Extensive farming and grazing to the stream bank occurs in Reaches 1, 3, and 5 (Table 7 and see Cooper Creek, Pages 1 - 7 of 7 in Appendix B). Logging occurred along Reaches 1 and 7 and roads in the RMA are common. A large fen in Reach 2 is an important wildlife habitat. Most impacted riparian segments were assigned a structural stage of INIT or SHR and thus were lacking mature conifers. As a result, the shading, riparian cover and source of LWD functions were impaired. In some segments mature forests were present, but were impacted by road crossings or roads within the RMA. Some bank erosion was evident.

Eroding banks, mainly naturally eroding, occur in almost all reaches due to sparse to almost absent forest cover. Recent logging is evident in some areas: Reaches 7-10, 13, 17, 18, and 20 (see Gaspard Creek, Pages 1 - 51 of 51 in Appendix B). Grazing is of significance in Reaches 6, 9, 10, 20-22, and 28. Road building has been extensive within the RMA in many segments (Table 7), which may cause excessive erosion and degradation of riparian vegetation and the stream. About one-third of all segments have been assigned medium or high priority for the assessment of riparian vegetation because of human impact on the vegetation. The dominant structural stage in impacted riparian zones was INIT although MF, YF and SHR were also common. Segments with INIT and SHR lacked mature trees and the riparian functions of shading, bank stability and source of LWD are impaired, as well as the lack of riparian cover in INIT segments.

Naturally eroding steep slopes exist in most reaches. This condition is aggravated by extensive road construction within the RMA and recent logging in Reaches 8, 9, 11, and 20 (see Little Gaspard Creek, Pages 1 - 17 of 17 in Appendix B). These reaches were assigned high priority for the assessment of riparian vegetation. Grazing is also evident in several reaches. Large fens and marshes in Reaches 7, 14, 15, and 16 are important wildlife habitats (Table 7). The structural stage of impacted sites were mainly INITn with some Mfc, Yfc, SHR, Psc and Ind. The lack of mature conifers along the banks of many impacted stream sections reduces shading, bank stability, stream cover and a source of LWD.

The riparian vegetation of this watershed is still in good condition despite grazing impacts evident in Reaches 1, 2, 3, 4, 7, 8, and 14 (see West Churn Creek, Pages 1 - 13 of 13 in Appendix B). Roads in the RMA are of concern in Reach 10 and logging probably occurred in Reaches 3 and 4. Eroding banks were observed in Reaches 4, 8, 9, 11, 12, and 13. Several segments in Reaches 3, 4, 10 and 11 have significant impacts in the RMA and received high priority for the assessment of riparian vegetation (Table 7). A large fen in Reach 14 is of high value as a wildlife habitat. Most impacted riparian segments have structural stages of INITn. The main functions impaired are loss of bank stability, lack of stream cover and shading, and a limited source of LWD.

Site visitation priorities, restoration opportunities and other comments are summarized in Table 7. Riparian segments most promising for fish habitat enhancement have received the highest site visitation priorities. Increasing the supply of LWD, riparian shading and cover and stabilizing banks at localised sites are the rehabilitation activities of primary importance throughout the fish-bearing portion of the watershed. Stabilization of adjacent slopes in order to prevent wasting events is not critical in the Big, Gaspard and Churn watersheds due to lower slopes. Most of the sites with high priorities for visitation have been cut or have roads within the RMA.

Enhanced growth of conifers is the best solution for increasing the long-term supply of quality LWD and for stabilizing banks. Depending on the results of site assessments, thinning of trees or other active measures may be suggested for some areas. However, riparian zone treatments commonly fail and a cautious approach is warranted.

The riparian segments highly recommended for site surveys include: R12-A; R13-A and B; R14-A, C, D and E; R15-A; R16-B and D; R17-B and F; R18-A, C, D, F, H, J and K; R19-A, B and D to J; R20-A, B, D and E; R21-A, C, E, F and H; 22-B, C and E; 23-A, C, E, G, H, J and K; 24-A, C, D and E; 25-B; 26-B; 28-C and H; 29-C; and 30-A. These segments were recommended as high priority for site visitation due to road crossings within the RMA, cutblocks to or close to the stream bank, road crossings and evidence of erosion. Segments given a medium for site visitation were: 10-A, B; 12-C; 14-F; 15-B; 16-F; 17-C, D, F; 18-I, L; 19-C; 21-D, J; 22-A; 23-B, F, I; 26-D, H, J; 28-D, F, G; 29-B, C; 30-C; 31-B, C, D; 33-A, C, D, and E. Medium designations were mainly due to exposed soil or erosion, old logging in RMA, NSR and roads in the RMA. Other causes included logging near the RMA, fields extending to the bank, a canal in the RMA and farming in the RMA.

No site visitations were recommended for Tilton Creek given that no forest harvesting, road routing or agricultural activity occurred within the RMA.

A high priority for site visitation was assigned to R1-A, B, C; 2-A, B; 3-A, B and a medium priority to R-3C. The impacts of concern within the RMA included logging, road routing, bridge crossings and bank erosion.

A high priority for site visitation was assigned to R1-D and a medium priority to R1-A, C, R2-C. Riparian impacts were due to erosion and roads within the RMA.

Riparian segments highly recommended for site visitation include R2-A, C, D, E; R5-B, F, G; and R6-C. A medium priority was assigned to R2-B; R3-A, C, D; R4-A, B; R5-A, D, E; R6-A, D, E. Riparian impacts are due to a road within the RMA, logging, road crossings, and some erosion.

High priorities for site visitation are recommended in R1-A, C; R3-A; R5-A; R8-A, C; R9-A, B; R13-A; R14-E. Medium priorities were assigned to R10-A, B; and R11-A, B. Most riparian impacts are due to erosion or lack of vegetation, agricultural lands cleared to the bank, and road crossings.

A high priority for site visitation was assigned to R1-A, B; R3-B, D, F; R5-A, B; R7-A, B and a medium priority to R1-D, E, F; R2-B; R3-E, H; R4-A, B; and R6-A, B. The causes for disturbance in the riparian zone included roads within the RMA, farming close to the banks, erosions and some logging.

Riparian segments given a high priority for site visitation were: R1-B; R2-A; R3-A; R4-A, B; R7-D; R8-D; R13-B, E, G; R17-E; R18-C; R20-A, B, C, and G; R21-F; R24-A; R25-A; R26-B; R27-A; R28-A; and a medium priority were given to: R5-B; R6-A, C; R7-B, E, F; R8-A, B, C; R9-A, B, C, D, E; R10-A, D; R11-A, B; R13-I; R15-D, F; R19-B, C, D; R20-E, F, H, I; R21-A, E, I; R22-B; R23-B, C, E; R25-B; and R26-A. Riparian impacts were frequently roads in the RMA, eroding slopes or banks, road crossing or logging impacts.

A medium priority for site visitation was assigned to R3-C; R4-A, B; R5-B, D; R6-A, C; R7-B, C; R8-B; R11-B; R14-A, C; R19-A and a high priority to R7-A; R8-A, D; R9-A, B, C; R10-A, C, D; R12-A, B; R13-C; R18-A, B; R20-A, D, and E. The main impacts on the riparian zone were bank erosion, roads within the RMA, recent logging and road crossings. Some bare banks and grazing impacts were noted in a few segments.

Riparian segments were assigned a medium priority for site visitation in R1-B; R8-B; R13-A, B; R14-A, and B; and a high priority in R3-B; R4-A; R10-A, B; R11-B; R12-A. Riparian impacts were mainly due to erosion or lack of vegetation, road crossings and some logging.

Annas, R.M. and R. Coupe. 1979. Biogeoclimatic zones and subzones of the Cariboo Forest Region. Ministry of Forests, Province of British Columbia. 103 pp.

Fish Information Summary System (FISS). 1997. Department of Fisheries and Oceans and B.C. Ministry of Environment, Lands and Parks.

Mathews, W.H. (compiler). 1986. Physiography of the Canadian Cordillera. Geological Survey of Canada Map 1701A (1:5,000,000).