Page

Figure 1. Marjorie Creek basin hydrologic assessment................................................. back cover

Figure 2. Kleena Kleene temperature and precipitation (1942-68; 1980-95).................. 4

Figure 3. Lingfield Creek hydrograph, 1975-96............................................................ 6

Figure 4. Kleena Kleene annual precipitation fluctuation.............................................. 10

Figure 5. Lingfield Creek mean September unit discharge, 1975-96............................. 11

Figure 6. Beaver dams and ponds between Clearwater Lake and McClinchy Creek...... 21

Table 1. Beaver activity between Clearwater Lake and McClinchy Creek..................... 22

Appendix. Table 2. Descriptive statistics for Marjorie Creek sub-basin........................ 29

Residents of Kleena Kleene have observed a decrease in summer low flows in Marjorie Creek in recent years, jeopardizing water supply both for fish and irrigation use. During dry summer periods there has been ample flow in tributary channels within the cutblocks, but little or no flow in lower Marjorie Creek (Jansen, pers. comm.). The Kleena Kleene Resource Association (KKRA) initiated this study to evaluate the causes of this altered flow regime, with the effects of forest road development, logging, and beaver activity identified as particular concerns. Although the focus of this study is on Marjorie Creek basin, an additional concern is the impact of beaver activity on streamflow along the channel between Clearwater Lake and McClinchy Creek.

The objectives of this study were set out as follows:

• map the natural drainage network of Marjorie Creek basin (upstream of the Big Stick Lake F.S.R. bridge) and evaluate its alteration by logging roads and skid trails;
• evaluate the impact of logging and roads on snowmelt peak flows and summer low flows in Marjorie Creek;
• evaluate the impact of beaver activity on the streamflow regime of Marjorie Creek;
• identify the location of beaver dams that are significantly altering the streamflow regime between Clearwater Lake and McClinchy Creek;
• determine Marjorie Creek licensed irrigation water use and storage requirements;
• provide preliminary observations of channel conditions and requirements for channel restoration that may restore the former flow regime including identification of sites for wetland water control or drainage works to maintain low flows; and
• based on the available information, make recommendations relating to further study requirements and watershed restoration projects.

Prior to this study the KKRA had contracted three FRBC-funded assessments of the Clearwater Lake watershed - an Interior Watershed Assessment Procedure, a sediment source survey, and a fish habitat assessment (BioTerra Consulting; 1996a, 1996b, 1996c) - each at Level 1 as prescribed by Forest Practices Code guidebooks. These assessments were carried out to identify requirements for watershed restoration projects and to provide preliminary environmental information related to water supply concerns. The applicable results of these studies and information provided in discussions with the authors are incorporated in the present study.

Marjorie Creek (upstream of the Big Stick Lake F.S.R. bridge) drains a 45.8 km² area ranging in elevation from 1,000 m at the bridge to 1,900 m on the northern slopes of Mt. Nogwon. Slope gradients are generally less than 10 per cent, except for the uppermost, forested slopes and isolated, short slope segments elsewhere in the basin. The basin is drained by two major tributaries which flow from their headwaters easterly and southeasterly to converge in the large wetland complex upstream of the bridge (Tributaries D and G, Figure 1).

Surficial materials on the western, upper slopes of the watershed are composed mainly of gravelly, sandy loam glacial till. These materials are generally well drained, although springs, poorly-drained seepage areas, and small wetland fens are common, particularly along the lower reaches of the principal watercourses.

Roughly along Tributary G draining the eastern portion of the basin, there is a broad zone within which rapidly-drained, glacio-fluvial sand and gravel deposits frequently overlie the glacial till. These deposits occur in the forms of thick blankets, low ridges, and hillocks and have been excavated in some cases as gravel sources for roads.

The lower slopes of the study area are comprised of a benchland occupied in lower-lying portions by an extensive wetland complex. Surficial materials in this area of the basin are well-drained, glacio-fluvial sand and gravel on upland sites, overlain on lower-lying sites by organic materials of the wetland complex.

Kleena Kleene has a cold, dry continental climate with an average annual temperature of 2.4 ^oC and average total precipitation of 345.6 mm measured at two Kleena Kleene weather stations (current elevation: 914 m) (Climate Data Services, no date). Figure 2 illustrates the variation of

temperature and precipitation (as rain and snow) through the year. Approximately 40 per cent of the annual precipitation is recorded as snow, mainly from November to February. This proportion would increase with increasing elevation in the study area. In Kleena Kleene the early spring months (March and April) are the driest of the year and the highest rainfall amounts occur in the summer (June, July, and August).

The streamflow regime of Marjorie Creek is dominated by snowmelt discharge that begins with low elevation melt runoff in late March and April and peaks in May with high elevation snowmelt. Streamflows then decline through the summer to minimum flows in July, August and September. A slight rise in fall flows is caused by reduced evapotranspiration by vegetation and evaporation from water bodies. Although the snowmelt period accounts for the maximum discharge volume, the annual peak streamflow may be produced by large rainstorms, occurring most often during summer and fall. The hydrograph for Lingfield Creek (draining the Potato Range) is shown in Figure 3 to illustrate the seasonal flow regime of creeks along these leeward slopes of the Coast Mountains (Water Survey of Canada, no date).

Logging began in upper Marjorie Creek basin with small areas clearcut in 1981 and 1982, however, most cutting was carried out from 1988 to 1992. In total 17.1 km² or 37 per cent of the 45.8 km² sub-basin area have been clearcut.

The road length within the basin totals 85.4 km, of which there are 9.4 km of active haul road (serving areas outside the basin), 1.4 km of road maintained for recreation and private land access, and 74.6 km of inactive or deactivated road used mainly for occasional recreational purposes. Bioterra Consulting (1996b) has described and mapped the condition of these roads in detail. In general, they are constructed on gently- to moderately-sloping, well-drained terrain with very little ditching or road cut and fill.

In the fall of 1994 the Ministry of Forests removed culverts and constructed fords and waterbars along many of the inactive logging roads in the basin. This road deactivation was carried out mainly to control surface runoff and erosion. The existing drainage structures allow 4x4 vehicle passage, but are sufficiently large and numerous to discourage many.

Hans Bettschen of Clearwater Lake Ranch holds two water licences on Marjorie Creek totaling 120 ac.-ft. (Conditional Licences 19417 and 29547). These licences authorize a 4-foot dam on Marjorie Creek (at a point about 300 m downstream of the bridge) and diversion of flow along a ditch to ranch hayfields during the period April 1 to September 30. The first licence also permits use of 1,000 g.p.d. year-round for domestic purposes. No water storage licences have been issued to support these irrigation and domestic uses.

In a letter to Hans Bettschen following his May 30, 1994 field inspection, Ken Kvist (Senior Allocation Technician, Water Management Branch) indicated that he found that beaver dams along the main channel upstream of the irrigation diversion structure and along the irrigation ditch itself could be restricting water supply for irrigation (letter of June 16, 1994). He also indicated that there is no requirement in the ranch’s water licences for maintenance of minimum flow in Marjorie Creek for fish. Regulations of the federal Fisheries Act would nevertheless supercede Provincial water licence allocations to control deleterious impacts on fish or fish habitat (Parker, pers. comm.).

Bettschen (pers. comm.) has pointed out that his principal concern is the very low summer flow in Marjorie Creek. He has found the flow along the main channel to be insufficient to meet even the minimum requirements for the downstream fishery and has not diverted irrigation flows for several years.

BioTerra Consulting’s (1996c: 2) Level 1 Fish Habitat Assessment determined that Marjorie Creek ‘has a high fisheries value within the watershed because it is the primary source of stream habitat available for spawning, rearing and recruitment’. In preliminary surveys, fish were recorded in Marjorie Creek only along reaches between Clearwater Lake and the road bridge. This length of channel was observed to have adequate rearing habitat but little spawning habitat, a factor suggested as a limitation to rainbow trout productivity in the system.

BioTerra identified low summer streamflows, fish passage obstruction and water impoundment by beaver dams, and limited spawning habitat as major limitations to the fishery. Stream sedimentation, apparently related to road construction, was noted along the reach downstream of site 1 (Figure 1), although no fish were found.

Nicklin (pers. comm.) has indicated that the entire lengths of the main tributary channels within the basin should be regarded as potential fish habitat, but that the eventual extent of fish use of the system cannot be known until beaver dam removal is carried out.

Figure 4 illustrates the annual precipitation and a ‘three-year moving average’ of annual precipitation through the period of record at Kleena Kleene. (A ‘moving average’ smoothes the abrupt, year-to-year variations to depict major climatic trends more clearly.) Although there is a major gap in the record (from 1969-1979 inclusive), it is apparent that precipitation was generally below normal throughout the 1980’s and above normal from 1990 to 1995. The 1981-89 annual average precipitation was 313.2 mm and the 1990-95 average was 371.9 mm, respectively below and above the 1943-1995, 345.6 mm average.

Winter snow accumulations measured at the Tatlayoko Lake snow course (at 1,710 m on Potato Mtn.) demonstrate a similar pattern. The average water equivalent of the snowpack on April 1 was 213 mm during the 1980’s and 288 mm from 1990 to 1996 with the 44-year average being 252 mm (Department of Land and Water Management, 1980-1996).

Figure 5 shows the discharge per unit area for Lingfield Creek basin from 1975 to 1996 for September, the month of lowest flows in this system (Water Survey of Canada, no date). Although Lingfield Creek and Marjorie Creek basins are not directly comparable physiographically, the pattern of streamflow variation from year to year would be similar. The mean unit discharge for the 1980’s on Lingfield Creek was 0.124 m³/s/km² and for the 1990’s has 0.156 m³/s/km², again below and above the long-term average of 0.148 m³/s/km².

Both the precipitation and flow data suggest that summer low flows in the western Chilcotin Ranges of the Coast Mountains would be expected to be at or above the long-term average during the 1990’s. There does not appear to be a climatic explanation for any reduced summer low flow in Marjorie Creek during the 1990’s.

Research in a broad range of environments has shown that groundwater flow and baseflow (the low streamflow between runoff events) normally increase when forests are clearcut. This flow increase occurs because the forest canopy is no longer present to intercept rain and snowfall, a portion of which would evaporate, and because water loss to the atmosphere by transpiration of trees is not taking place. The popular belief that logging reduces low flow in streams may be related to the fact that the surface 30 cm or so of well-drained soils are often drier after forest removal, being more exposed to direct sunlight and wind.

Given the extensive clearcutting in the basin a small baseflow increase might occur along the tributary streams, although the degree of such change is generally found to be lower in well-drained terrain and drier climates as in the study area. In any case a baseflow increase would likely not be detectable downstream along the lower reaches of Marjorie Creek, particularly against the background of complicating factors such as flow impoundment in beaver ponds and year-to-year variation of streamflow. The key observation here is that, if anything, logging would cause a slight increase not a decrease in summer low flows.

In snow-dominated climates (as in the study area) the most pronounced hydrologic effect of clearcutting is the increased snowmelt runoff volumes that may occur. In forests some of the snow intercepted by the canopy is lost to sublimation and evaporation, whereas in clearcuts this interception loss is absent and more snow accumulates on the ground. Snow accumulations are generally 20 to 40 per cent greater in clearcut or burned openings than in adjacent forest.

Peak snowmelt does tend to occur 1 to 2 weeks earlier in clearcuts than in adjacent forest, varying mainly with slope aspect. However, the consequence of this change for streamflow is mitigated by the higher initial moisture content of cutblock soils and the greater overall snow accumulation. Streamflows from clearcut areas typically remain higher than from forested areas except, in some cases, late in the snowmelt period.

In most cases streamflow increases related to clearcutting are expressed as proportions of the total annual water yield from a basin, although most of the change is accounted for by snowmelt runoff. Research results are quite variable, however, increases are generally not significant until at least 20 per cent of a basin has been clearcut. At 37 per cent clearcut, the annual yield increase would typically be in the range 15 to 20 per cent.

The Interior Watershed Assessment Procedure (IWAP) (B.C. Ministry of Forests and B.C. Environment, 1995) includes calculation of a peak flow hazard index score to express the basin susceptibility to increased snowmelt peaks caused by clearcutting and roads. This rating is based on the proportion of the total basin that is in a condition equivalent to a clearcut (the ‘equivalent clearcut area’), the extent of clearcutting in the upper 60 per cent of the basin (above the ‘H₆₀ elevation’), and the density of roads above and below the H₆₀ elevation.

The equivalent clearcut area is derived from a relationship with tree height which expresses the extent to which the regenerating forest still functions, hydrologically, as a clearcut. The zone above the H₆₀ elevation is typically responsible for the peak snowmelt discharge from a watershed, thus the extent of modification of this zone by clearcutting is most important. Since roads can hasten surface runoff and reduce groundwater recharge, the road densities above and below the H₆₀ line are also used to rate the peak flow hazard.

Because most clearcutting in Marjorie Creek basin was during the past nine years the regenerating stands are generally under 3 m in height and the cutblocks thus remain in a condition equivalent to a recent clearcut (BioTerra, 1996a). (The small areas of older clearcutting which exceed 3 m were not considered significant at the basin scale.) BioTerra (1996a) calculated a peak flow hazard score of 0.69 for the entire Clearwater Lake basin upstream of the outlet channel’s confluence with McClinchy Creek - this indicates a moderate to high potential impact of logging on streamflow and channel stability. Recalculation of a peak flow index for the Marjorie Creek sub-basin (upstream of the Big Stick Lake F.S.R. bridge) yields a score of 0.8 (see Appendix) which suggests a high potential impact of logging activity. This high rating is produced by both the extensive clearcutting and the high density road network in the sub-basin. The principal hazard here is that streamflow peaks would be increased and channel stability reduced (see section 5.0). The rating does not relate to alteration of summer low flows.

One situation in which streamflow may be decreased by forestry activities is in the case of alteration of the natural drainage patterns by forest road construction. Forest roads, ditches, and culverts can divert surface runoff to areas that contribute less directly to streamflow (at the point of interest) or even across basin divides to other watercourses.

Forest roads can also serve to extend the drainage network and thereby expand the area that contributes storm runoff to streams. Such an increase in surface runoff could cause a corresponding depletion of the groundwater supply, the water source responsible for maintenance of baseflow in streams. Groundwater supply may also be reduced by its interception by road cuts and ditches and diversion to the drainage network by surface runoff.

Inspection of logging roads throughout the study area did not reveal any significant rerouting of natural drainage patterns, although there is evidence of considerable runoff along roads having occurred prior to the 1994 road deactivation work. While road surface runoff prior to 1994 may have increased stormflow volumes and peaks in the tributaries draining the clearcuts, no current examples were observed of flow diversion that would result in streamflow reduction at the basin scale.

During a helicopter overview flight prior to this study one KKRA member observed severe bank disturbance by machinery along one channel reach that had resulted in flow deflection from the channel to adjacent terrain (McCoy, pers. comm.). The writer also observed numerous cases of bank disturbance and flow re-routing due to rutting and deflection by roads, however, in these cases the general drainage pathways are maintained and basin-scale effects are negligible.

To reduce baseflow, road disruption of groundwater recharge and surface runoff patterns would have to be considerably more extensive than is the case in the study area. The location of roads on generally well-drained and moderately-sloping terrain with few ditches and no deep road cuts minimizes the risk of interception of groundwater flow. No sites were observed at which groundwater flow had been intercepted and deflected along roads.

For the most part the fords and waterbars that have been constructed have re-established the natural drainage pattern and considerably reduced surface runoff along roads. There are nevertheless sites where further deactivation work is required. BioTerra (1996b) documented numerous sites in the course of their sediment source survey. An additional example observed by the writer is located between sites 2 and 3 (Figure 1). Storm and snowmelt runoff flows a 1.3 km distance down the road between these points to discharge into an adjacent drainage line. Waterbars could easily control the runoff and the surface erosion that is resulting.

Soil compaction by harvesting and mechanical site preparation operations also have the potential to increase surface runoff at the expense of groundwater recharge by reducing rates of precipitation infiltration into the soil. In the cutblocks the imperfectly to poorly drained seepage sites, springs, and riparian zones have a very high susceptibility to compaction and, although only selected areas were inspected, soil compaction and rutting by skidder traffic appears to be commonplace. This disturbance is a significant site-scale problem, however, it would have to be considerably more extensive and severe to affect on watershed-scale streamflow.

Another possible explanation for reduced summer flows in Marjorie Creek is that water is lost within the wetland complex in the lower basin and in the numerous small beaver ponds upstream. This seems most probable, given the observed streamflow reduction between the clearcuts and the sub-basin outlet. Wetland terrain and ponds often help to maintain flows during low flow periods by slowly releasing water stored during higher flows, however, water supplied to the Marjorie Creek wetlands and ponds during the summer may be lost to groundwater, evaporation, and storage with limited surplus available for downstream supply.

As noted in section 2.1 the wetland complex is underlain by sand and gravel deposits that form a bench on the eastern margin of the basin lying 20-30 m above the adjacent McClinchy Creek valley bottom. Losses to groundwater may be taking place through these porous materials that by-pass the lower length of Marjorie Creek to Clearwater Lake.

Surface water inflow to the wetland from Marjorie Creek Tributary G is an obvious candidate for such groundwater loss. The results of land surveying carried out during the winter, indicate that this tributary can flow either east to a broad depression at site 4 or south to Pond A and the main channel of Marjorie Creek. Along this latter route there are two beaver dams which have held back small reservoirs (site 5), however they have both been opened, perhaps by humans to improve drainage of nearby wetland hayfields. Within the wetland upstream of these dams no defined channel was observed (during winter surveys) which would convey flow efficiently from Tributary G to Pond A. (A summer inspection would be required to confirm this observation.)

Along the main stem of Marjorie Creek (Tributary D) there are many beaver dams and ponds, the larger of which are shown in Figure 1. Beaver occupation of the area upstream of the road bridge reportedly did not take place until the late 1980’s (Dester; Jansen, pers. comm.). Prior to this, Pond A (McCormick Meadow) was a wetland fen with shrub and sedge vegetation on organic soils. At approximately 26 ha (64 ac.), this is the largest of the ponds in the sub-basin. Water levels are now up to 1.5-2 m deep in the central portion of the pond and dead pine on the periphery are standing in approximately 1 m of water. Ten to fifteen years ago, these pine would have been above the seasonal high water level and (although they were not dated) had apparently remained so for decaades.

There are insuficient data to calculate a ‘water balance’ for this wetland - that is, to apportion water inflow to storage, groundwater, evaporation, and outflow. However, first estimates of growing season evaporation from ponds and evapotranspiration from wetland vegetation indicate that these components represent less than half the total losses within the wetland. Water storage in beaver ponds in recent years would have increased evaporation and water storage quantities somewhat, however, groundwater losses are thought to remain the major mechanism of surface water withdrawal. Discounting the effects of recent beaver activity, summer outflow to lower Marjorie Creek would still be very low.

It should be noted that there is more than sufficient total discharge to Clearwater Lake from Marjorie Creek during spring to meet irrigation and fish habitat requirements during the summer low flow period. First estimates based on Lingfield Creek data indicate that there’s more than an order of magnitude more water discharged during April, May, and early June than would be required to meet these demands. The low flow concerns could thus be addressed by improved water storage and controlled water release. Preliminary inspection indicates adequate storage capacity could be developed in Pond A.

Two streamflow monitoring stations have been established downstream of the bridge on lower Marjorie Creek (Figure 1). These stations are to provide baseline streamflow data, both for flow prediction purposes and to assist with evaluation of the effects of land and water use on flows.

Station M1 is located at the Big Stick Lake F.S.R. bridge and consists of a staff gauge and a low timber sill for cross-section control. Station M2 is located at the upper end of the irrigation ditch approximately 400 m downstream of the bridge. This installation consists of a staff gauge installed in a uniform section of ditch. Measurement of flow diverted along the irrigation ditch permits calculation of the streamflow remaining along the lower length of Marjorie Creek.

Discharge measurements taken at Station M1 during this study are as follows:

- 29 October 1996 - 15:00 hr - 0.101 m³/s (0.0020 m³/s/km²)

- 28 January 1997 - 13:15 hr - 0.078 m³/s (0.0016 m³/s/km²)

- 28 March 1997 - 11:30 hr - 0.293 m³/s (0.0059 m³/s/km²)

- 26 April 1997 - 13:15 hr - 0.877 m³/s (0.019 m³/s/km²)

The values of unit discharge (in parentheses) are approximately equivalent to the monthly mean unit discharges for Lingfield Creek shown in Figure 3.

Individual flow measurements were also taken on 29 October at the numerous tributary streams upstream of the wetland and downstream of the logged areas. The total discharge from these upper basin channels was estimated to be less than the discharge at Station M1 on the same day. At that time no net loss of discharge in the wetland complex was apparent.

There has been no flow at Station M2 during the study period.

During this study brief observations were made of channel condition to provide indications of the extent of channel disturbance by logging activity and the necessity for further channel assessment and restoration works.

The Marjorie Creek tributaries within the logged areas maintain an irregular pattern within erodible materials of glacial till or alluvial deposits. The channels generally have a ‘cascade-pool’ morphology (B.C. Ministry of Forests and Ministry of Environment, 1996) with bed material of cobbles, gravel, sand, large woody debris, and finer organic matter. The extensive fallen woody debris, coarse materials, and streamside vegetation impart considerable stability to channel beds and banks.

BioTerra (1996a) measured logging to the streambank along 8 km of stream length (counting both banks) in the upper Marjorie Creek sub-basin. With about 40 km of stream length in this drainage area, this riparian buffer disturbance would be assigned a relatively high impact score of 0.7 under the Interior Watershed Assessment Procedure Level 1 analysis. Channel disturbance by machinery is common along these lengths of logged streambank. With only brief inspections, numerous sites were observed where channels had been filled with logging debris to facilitate machinery crossings and streambanks had been rutted.

Extensive road surface erosion prior to deactivation in 1994 and to some extent the deactivation works themselves have caused fine sediment to be deposited in pools along reaches downstream of many of the logging roads. In time, this sediment will be transported downstream to more permanent traps such as the beaver ponds within and upstream of the large wetland complex. This sediment movement will not have a significant impact on the lower reach of Marjorie Creek.

In spite of the observed channel disturbance and sedimentation the channels remain relatively stable, except where bank materials are mechanically disturbed. It is not expected that any progressive erosion or deposition will result from these impacts.

The length of channel between Clearwater Lake and McClinchy Creek was inspected during February 1997 to assess the degree to which beaver dams and ponds might be affecting streamflow regime. The locations of beaver dams and the largest ponds are shown in Figure 6 and selected site characteristics are summarized in Table 1.

In total 18 beaver dams were recorded along this length of channel with beavers present mainly along the upstream portion: from the lake to site 13 there are 13 actively maintained dams; and downstream to McClinchy Creek only minor beaver activity was observed and one dam is open. As noted in Table 1, machinery access to dams is easiest along the upper channel reach and relatively difficult downstream due to the extensive wet terrain and shrubland.

Many of the beaver dams impound small reservoirs, however, only the site 3 dam (at Hwy. 20) could be considered to be affecting seasonal streamflow regime. At its present height this dam appears to control the level of Clearwater Lake, flooding both the dam at site 2 and the site 1 lake outlet (Figure 6). This storage capacity would serve to prolong higher summer flows along the channel than would be the case if high snowmelt runoff flows were released directly to the outlet channel.

Fred Brink (pers. comm.), a long-time resident who ran a trap-line in the vicinity, reports that there was little beaver activity along this channel prior to the 1950’s. He recalls that natives operated a fish trap close to the bridge before beavers occupied the channel.

No conclusive evidence has been found to indicate that streamflow in lower Marjorie Creek has been lower in recent years than would naturally occur in this drainage system, however, summer flows have clearly been inadequate for irrigation and fisheries requirements. Although forestry activity has resulted in significant impacts on soils and stream channels, it is not thought to have reduced summer streamflow or snowmelt discharge volumes. If anything, these quantities would have been increased by clearcut logging. Climate and streamflow data also indicate that streamflow in the region would have been slightly above normal during the 1990’s.

The very low basin outflows in the summer are attributed to losses to groundwater, evaporation and storage in the ponds and wetland terrain in the lower portion of Marjorie Creek sub-basin. These losses would have been increased by beaver activity within the past 10-15 years, however summer flows are thought to have been very low prior to this period. Water storage in the wetland complex would prolong higher streamflows in spring and early summer, but as water in storage is depleted (particularly during dry years) the ponds and organic terrain would retain streamflow that would otherwise reach lower Marjorie Creek.

Measures to increase summer flows and to restore watershed degradation related to logging are discussed below

Detailed prescriptions for treatment of road and landing sediment sources identified by BioTerra (1996b) and during the present project (e.g., sites 2-3) should be prepared and executed. This work will reduce storm runoff along roads, surface erosion, and stream sedimentation. Observations of road runoff processes during the snowmelt period would be valuable for planning deactivation measures, although work in the vicinity of streams should not be carried out until the late summer low flow period.

Brief channel inspection reveals considerable mechanical disturbance of banks, logging debris in channels, and stream sedimentation from logging road sources. These are significant impacts however they cannot be easily remedied by restoration work.

In some cases disturbed banks could be stabilized and minor flow diversions could be re-routed to their original drainage lines by careful machine work, hand work, and revegetation. Sites where such channel restoration would be feasible are infrequent however; in most cases, the channels should be allowed to re-stabilize without further interference.

The 4 km length of channel that has been logged to the banks should be walked to prescribe restoration requirements. Channel assessment should be limited to this purpose; since the channels are generally stable, a reach-by-reach classification of channel morphology and condition is not recommended. Channel restoration measures are expected to be minor and could be identified in the course of road deactivation planning.

Logging debris in channels is unsightly, however, in many cases it may be acting to trap sediment derived from roads and disturbed banks; its removal would hasten the downstream movement of this material. Stable bank materials might also be disturbed by such channel clearing operations. Unless sites are found that are causing channel instability or preventing fish passage, the debris in channels should be left in place.

Although sedimentation of channels downstream of logging roads is commonplace, this material is not causing significant channel instability. Channel condition will be gradually restored as these deposits move downstream to more permanent sediment sinks and road sediment sources are stabilized. No remedial measures are recommended.

Drainage ditches could be constructed through the wetland to increase summer flows downstream. A ditch from the outlet of Tributary G to the wetland to Pond A, for example, would reduce summer storage losses in the wetland and increase flows along lower Marjorie Creek. This approach, however, could result in considerable ecological change by reducing moisture availability for shrub and herbaceous growth and for maintenance of organic deposits. Winter observations revealed extensive use of the wetland by moose.

Numerous small beaver ponds could also be drawndown to increase downstream flows, however, this would still not produce sufficient water to meet downstream flow requirements. As well, one would be embarking on an indefinite beaver control programme with potentially undesirable ecological consequences.

Alteration of wetland drainage patterns and drawdown of small beaver ponds solely to increase summer streamflow is not recommended.

One clear solution to the problem of inadequate summer streamflow in lower Marjorie Creek is development of a larger storage reservoir at Pond A to supply water for irrigation and fisheries requirements. More than enough water is discharged in the snowmelt period for this purpose and there appears to be adequate storage capacity.

The dam would require a fishway and outlet control that screens fish passage and minimizes obstruction by beavers. The fishway would permit use of the pond for rearing and allow for fish dispersal upstream. The reservoir could also provide stable waterfowl habitat during the nesting period - an opportunity that may interest Ducks Unlimited (Arner, pers. comm.).

Given the presence of a beaver pond this reservoir could presumably be developed with relatively little alteration of wildlife habitat.

A study of project feasibility and costs should address: water supply; reservoir storage capacity and flood zone; engineering of the dam, control works, and fishway; reservoir water management; and fish and wildlife habitat impacts and benefits.

BioTerra (1996c) recommended removal of beaver dams between Pond A and Clearwater Lake to permit fish passage. There are dams with small impoundments in the few hundred metres downstream of the pond and along the reach immediately upstream of the lake. This length of channel is outside the study area and has not been inspected in detail, however, these small ponds would have little effect on seasonal flows and there is no hydrologic reason to retain them. Obviously all ponds should be drawndown gradually to avoid degradation of the downstream channel.

For the present it’s recommended that the Pond A dam remain in place. This will allow for planning of a storage reservoir or, failing that, for a system to maintain flows through the beaver dam itself. Ducks Unlimited in Ontario has installed pipes through beaver dams to allow for water release and still maintain a required pond level using an inverted outlet (Arner, pers. comm.). A valve could also be used to control water release. Such a system would retain the storage function of the beaver pond while providing for flow maintenance during the summer. Beaver dams cannot be licensed for irrigation storage, however, any improvement of summer water supply would benefit the licence-holder nonetheless. The disadvantages of this approach are that it would not allow for fish passage and it would not provide adequate water storage to supply downstream uses.

Evaluation of the irrigation diversion system is also not within the study objectives, however, there are a few points that should be considered as improvements are made.

1) The irrigation diversion structure should be carefully screened to assure that fish cannot enter the irrigation ditch. The licence-holder indicates that the present intake is screened (Bettschen, pers. comm.).

2) Clearing of the irrigation ditch of beaver dams and other obstructions will minimize the quantity of water diverted from the stream.

3) An agreement or a requirement in the water licences should be established to provide for maintenance of a minimum flow in the creek. There is currently no such provision, although the present licence-holder is careful not to dewater the creek. Any improved diversion structure should also allow for a fixed minimum flow in the creek.

It’s recommended that periodic manual streamflow measurements be taken at stations M1 and M2 to determine snowmelt period water supply and summer streamflow requirements. At a minimum, streamflow should be meaured monthly from October to March and twice monthly from April to September. Allowance should be made for several additional measurements to be carried out as required.

There is no apparent hydrologic justification for removal of beaver dams between Clearwater Lake and McClinchy Creek, unless the higher lake levels are undesirable. It may be beneficial to remove dams and beavers to improve the fishery, however this would begin a long-term beaver control programme.

Arner, Brad. Pers. comm. Biologist, Ducks Unlimited, Prince George.

Bettschen, Hans. Pers. comm. Water licence holder, Clearwater Lake Ranch, Kleena Kleene.

BioTerra Consulting. 1996a. Clearwater Lake Interior Watershed Assessment. Final Summary of Level 1 Results. Prepared for Kleena Kleene Resource Association. Williams Lake, 16 p. + map.

BioTerra Consulting. 1996b. Clearwater Lake Watershed Sediment Source Survey. Prepared for the Kleena Kleene Resource Association. Williams Lake, 18 p. + appendices and maps.

BioTerra Consulting. 1996c. Overview and Level 1 Fish Habitat Assessment for the Clearwater Lake Watershed. Prepared for the Kleena Kleene Resource Association. Williams Lake, 25 p. + appendices and maps.

Brink, Fred. Pers. comm. Kleena Kleene Resource Association member.

British Columbia Ministry of Forests and Ministry of Environment. 1995. Interior Watershed Assessment Procedure Guidebook (IWAP). Level 1 Analysis. Forest Practices Code of British Columbia. 82 p.

British Columbia Ministry of Forests and Ministry of Environment. 1996. Channel Assessment Procedure Guidebook. Forest Practices Code of British Columbia. 37 p.

Climate Data Services. No date. Unpublished data. Environment Canada, Vancouver.

Department of Land and Water Management. 1980-1996. April 1 Snow Survey Bulletins. Water Management Branch, B.C. Environment, Victoria.

Dester, Mack. Pers. comm. Kleena Kleene resident.

Jansen, Ken. Pers. comm. Kleena Kleene Resource Association member.

McCoy, Carl. Pers. comm. Kleena Kleene Resource Association member.

Nicklin, Peter. Pers. comm. Resource Biologist, BioTerra Consulting, Williams Lake.

Parker, Mike. Pers. comm. Regional Fisheries Specialist, Fisheries Branch, B.C. Environment, Williams Lake.

Water Survey of Canada. No date. Unpublished data. Environment Canada, Vancouver.

Table 2. Descriptive statistics for Marjorie Creek sub-basin.

Basin area - 45.8 km²

Elevation range - 1,000 to 1,900 m

Stream channel length - 40 km

H₆₀ elevation - 1,180 m

Clearcut area - 17.1 km²

Logged length of channel bank (both sides) - 8 km

Road length - 85.4 km comprised of:

- 9.4 km active haul road

- 1.4 km of road maintained for recreation and private land access

- 74.6 km of inactive or deactivated road

Clearcut area above H₆₀ elevation - 8.8 km²

Clearcut area below H₆₀ elevation - 8.3 km²

Road length above H₆₀ elevation - 45.7 km

Road length below H₆₀ elevation - 39.7 km

IWAP peak flow hazard index score - 0.8 (high)

IWAP riparian buffer hazard index score - 0.7 (medium-high)